"Arts, Faculty of"@en . "Linguistics, Department of"@en . "DSpace"@en . "UBCV"@en . "Cook, Clare Elizabeth"@en . "2008-06-26T21:17:41Z"@en . "2008"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "This thesis proposes that there are two kinds of clauses: indexical clauses, which are evaluated with respect to the speech situation; and anaphoric clauses, which are evaluated with respect to a contextually-given (anaphoric) situation. Empirical motivation for this claim comes from the clause-typing system of Plains Cree, an Algonquian language spoken on the Canadian plains, which morpho-syntactically distinguishes between two types of clauses traditionally called INDEPENDENT and CONJUNCT orders. In the current analysis, the INDEPENDENT order instantiates indexical clauses, and the CONJUNCT order instantiates anaphoric clauses. \nAfter laying out the proposal (chapter 1) and establishing the morphosyntax of Plains Cree CPs (chapter 2), the remaining chapters discuss the proposal in detail.\nChapter 3 focusses on the syntax and semantics of indexical clauses (Plains Cree\u00E2\u0080\u0099s INDEPENDENT order). Syntactically, I show that there is an anti-c-command and an anti-precedence condition on indexical clauses. Semantically, I show that indexical clauses are always and only evaluated with respect to the speech situation, including the speech time (temporal anchoring), speech place (spatial anchoring), and speaker (referential anchoring).\nChapter 4 focusses on the syntax and semantics of anaphoric clauses (Plains Cree\u00E2\u0080\u0099s CONJUNCT order). Syntactically, I show that anaphoric clauses must always be either preceded or dominated by some other antecedent clause. Semantically, I show that the value of temporal/spatial/referential dependent elements within an anaphoric clause is determined by an antecedent.\nChapter 5 turns to the syntactic subclassification of Plains Cree\u00E2\u0080\u0099s CONJUNCT (i.e., anaphoric) clauses. I propose that there are three classes: chained clauses, adjunct clauses, and mediated argument clauses. I provide two kinds of diagnostics that distinguish these classes, and explore the consequences of this classification for argument clauses and complementation.\nFinally, Chapter 6 proposes a semantic subclassification of Plains Cree\u00E2\u0080\u0099s CONJUNCT (i.e., anaphoric) clauses. I propose that there is a direct mapping between the morphology and the semantics: one complementizer encodes presupposition of the proposition, the lack of a complementizer encodes a-veridicality of the proposition, and one complementizer is semantically unspecified (the elsewhere case). This means that Plains Cree\u00E2\u0080\u0099s clause-typing is fundamentally concerned with how the truth of the proposition is represented."@en . "https://circle.library.ubc.ca/rest/handle/2429/951?expand=metadata"@en . "5070199 bytes"@en . "application/pdf"@en . " THE SYNTAX AND SEMANTICS OF CLAUSE\u00E2\u0080\u0093TYPING IN PLAINS CREE by CLARE ELIZABETH COOK B.A., The University of Wisconsin. 2001 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Linguistics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) June 2008 \u00C2\u00A9 Clare Elizabeth Cook, 2008 ii Abstract This thesis proposes that there are two kinds of clauses: indexical clauses, which are evaluated with respect to the speech situation; and anaphoric clauses, which are evaluated with respect to a contextually-given (anaphoric) situation. Empirical motivation for this claim comes from the clause-typing system of Plains Cree, an Algonquian language spoken on the Canadian plains, which morpho-syntactically distinguishes between two types of clauses traditionally called INDEPENDENT and CONJUNCT orders. In the current analysis, the INDEPENDENT order instantiates indexical clauses, and the CONJUNCT order instantiates anaphoric clauses. After laying out the proposal (chapter 1) and establishing the morphosyntax of Plains Cree CPs (chapter 2), the remaining chapters discuss the proposal in detail. Chapter 3 focusses on the syntax and semantics of indexical clauses (Plains Cree\u00E2\u0080\u0099s INDEPENDENT order). Syntactically, I show that there is an anti-c-command and an anti- precedence condition on indexical clauses. Semantically, I show that indexical clauses are always and only evaluated with respect to the speech situation, including the speech time (temporal anchoring), speech place (spatial anchoring), and speaker (referential anchoring). Chapter 4 focusses on the syntax and semantics of anaphoric clauses (Plains Cree\u00E2\u0080\u0099s CONJUNCT order). Syntactically, I show that anaphoric clauses must always be either preceded or dominated by some other antecedent clause. Semantically, I show that the value of temporal/spatial/referential dependent elements within an anaphoric clause is determined by an antecedent. Chapter 5 turns to the syntactic sub-classification of Plains Cree\u00E2\u0080\u0099s CONJUNCT (i.e., anaphoric) clauses. I propose that there are three classes: chained clauses, adjunct clauses, and mediated argument clauses. I provide two kinds of diagnostics that distinguish these classes, and explore the consequences of this classification for argument clauses and complementation. Finally, Chapter 6 proposes a semantic sub-classification of Plains Cree\u00E2\u0080\u0099s CONJUNCT (i.e., anaphoric) clauses. I propose that there is a direct mapping between the morphology and the semantics: one complementizer encodes presupposition of the proposition, the lack of a complementizer encodes a-veridicality of the proposition, and one complementizer is semantically unspecified (the elsewhere case). This means that Plains Cree\u00E2\u0080\u0099s clause-typing is fundamentally concerned with how the truth of the proposition is represented. iii Table of contents Abstract...................................................................................................................................... ii Table of contents....................................................................................................................... iii List of tables ..............................................................................................................................xi List of symbols and abbreviations ............................................................................................ xii Acknowledgements ...................................................................................................................xv Citation ...................................................................................................................................xvii 1 Indexical versus anaphoric CPs ....................................................................................1 1.1 Proposal: Indexical vs. anaphoric CPs........................................................................1 1.2 Relation to previous work ..........................................................................................3 1.2.1 Connection to the matrix/embedded distinction ...........................................3 1.2.2 Connection to illocutionary force.................................................................3 1.2.3 Parallel between indexical CPs and indexical expressions............................4 1.2.4 Parallel between anaphoric CPs and anaphoric expressions .........................5 1.2.5 Connection to the INDEPENDENT/CONJUNCT contrast ....................................6 1.3 Methodology: Data collection and presentation..........................................................8 1.4 Plains Cree terminology...........................................................................................10 1.5 Layout of the thesis..................................................................................................12 2 Mapping indexical and anaphoric CPs onto Plains Cree\u00E2\u0080\u0099s morpho-syntax ..............15 2.1 Proposal: A one-to-one mapping in Plains Cree .......................................................15 2.2 Diagnostics for CPs in Plains Cree...........................................................................19 2.2.1 Ordering properties....................................................................................20 2.2.2 Peripheral agreement diagnoses CPs..........................................................23 2.2.3 Clause-typing diagnoses CPs .....................................................................31 2.2.4 Pronominal proclitics are complementary with clause-typing.....................34 2.2.5 Interim summary: Verbal complexes are CPs.............................................35 2.3 Diagnosing C vs. spec, CP .......................................................................................36 2.3.1 Selection of complement ...........................................................................37 iv 2.3.2 Substitution (does not) determine distribution............................................40 2.3.3 The significance of non-overtness..............................................................42 2.3.4 Interim summary .......................................................................................47 2.4 The indexical/anaphoric distinction \u00E2\u0089\u00A0 matrix/embedded distinction..........................48 2.4.1 Negation distinguishes matrix and embedded clauses ................................48 2.4.2 Interrogative c\u00C3\u00AE distinguishes matrix and embedded clauses.......................51 2.5 Summary .................................................................................................................52 3 Indexical clauses: Plains Cree\u00E2\u0080\u0099s INDEPENDENT order .................................................54 3.1 Proposal: The syntax and semantics of indexical clauses..........................................54 3.2 The structural context of indexical clauses ...............................................................56 3.2.1 Indexical clauses must be matrix clauses ...................................................57 3.2.1.1 Embedding predicates do not introduce indexical clauses ............58 3.2.1.2 Subordinating particles do not introduce indexical clauses ..........60 3.2.1.3 Embedded negation does not modify indexical clauses................62 3.2.1.4 Summary: indexical clauses cannot be subordinated....................63 3.2.2 Indexical clauses exclude cross-clausal dependencies ................................63 3.2.2.1 Relative roots..............................................................................63 3.2.2.1.1 Relative roots with predicate modifier antecedents........66 3.2.2.1.2 Relative roots with CP modifier antecedents.................69 3.2.2.1.3 Relative roots with cross-clausal antecedents ................70 3.2.2.2 Spatio-temporal variables must be bound in indexical clauses .....74 3.2.3 Pronominal proclitics are indexical............................................................76 3.2.3.1 Indexical proclitics cannot be bound ...........................................77 3.2.3.2 The absence of third-person proclitics .........................................79 3.2.3.3 Referents are deictically anchored in indexical clauses ................81 3.2.3.3.1 Distribution of overt nominals in a discourse ................82 3.2.3.3.2 Restrictions on switch reference in INDEPENDENT clauses...............................................87 3.3 The semantics of indexical clauses: Indexicality ......................................................90 3.3.1 Temporal deixis: Relating reference time to speech time ...........................90 v 3.3.1.1 Contrasting temporal interpretations of indexical and non-indexical clauses .......................................94 3.3.1.2 Indexical clauses present statives that hold at T0..........................98 3.3.1.3 Indexical clauses present activities that coincide with T0 ...........102 3.3.1.4 Indexical clauses present telic predicates whose result state coincides with T0.........................................105 3.3.1.5 Interim summary.......................................................................110 3.3.2 Referential deixis: The role of the speaker in indexical clauses ................110 3.3.2.1 Person-effects: Preference for indexical clauses when talking about self ...........................................................112 3.3.2.2 Indexical clauses are infelicitous in contexts of unconsciousness ................................................114 3.3.2.3 Events in indexical clauses must be directly perceived by the speaker .............................................117 3.3.2.4 \u00C3\u00AAsa has mirative force in indexical clauses ................................122 3.3.2.4.1 The interaction of clause-typing and \u00C3\u00AAsa .....................123 3.3.2.4.2 Mirativity as incongruent experience ..........................125 3.3.2.5 Speaker commitment to the proposition.....................................129 3.3.2.6 Subjective predicates convey speaker\u00E2\u0080\u0099s attitude.........................131 3.3.2.7 Interim summary.......................................................................132 3.4 Summary: Structural and semantic conditions on indexical clauses ........................133 4 Anaphoric clauses: Plains Cree\u00E2\u0080\u0099s CONJUNCT order ..................................................134 4.1 Proposal: Anaphoric clauses ..................................................................................134 4.2 From pronominal to clausal anaphora.....................................................................135 4.2.1 The forms of anaphoric elements.............................................................136 4.2.2 The relation between anaphor and antecedent ..........................................137 4.2.3 Conditions on antecedent-licensing: C-command and precedence............139 4.2.4 Summary: The properties of anaphoric clauses ........................................143 4.3 The distribution of anaphoric clauses .....................................................................144 4.3.1 Anaphoric clauses are subject to precedence and/or c-command..............144 vi 4.3.1.1 Anaphoric clauses that must be c-commanded are not subject to precedence ...................................................144 4.3.1.2 Anaphoric clauses with subordinating particles are not subject to precedence ..................................................148 4.3.1.3 Anaphoric clauses that are subject to precedence: Unembedded \u00C3\u00AA- clauses ...........................................................151 4.3.2 Long-distance precedence of antecedent for anaphoric clauses ................154 4.3.2.1 Out-of-the-blue contexts vs. established contexts ......................156 4.3.2.2 Contrastive focus ......................................................................158 4.3.2.3 The distribution of clause-typing in elicitation: A discourse effect ...................................................................160 4.3.3 Interim summary .....................................................................................164 4.4 Anaphoric dependencies in anaphoric clauses ........................................................165 4.4.1 k\u00C3\u00AE- is subject to c-command and precedence in CONJUNCT clauses............167 4.4.1.1 Precedence without c-command: k\u00C3\u00AE- is anaphoric ......................169 4.4.1.2 Precedence and c-command: k\u00C3\u00AE- is anaphoric.............................172 4.4.1.3 C-command without precedence: k\u00C3\u00AE- is anaphoric......................176 4.4.1.4 No precedence, no c-command: k\u00C3\u00AE- is not anaphoric..................178 4.4.2 -yi is subject to c-command and precedence in CONJUNCT clauses............181 4.4.2.1 C-command and precedence: -yi is licensed ..............................183 4.4.2.2 Precedence without c-command: -yi is licensed.........................184 4.4.2.3 C-command without precedence:-yi is licensed .........................186 4.4.2.4 No c-command, no precedence: -yi is not licensed.....................188 4.5 The cross-linguistic typology of anaphoric clauses.................................................191 4.5.1 Chained clauses are anaphoric clauses .....................................................192 4.5.1.1 The significance of asymmetric marking...................................193 4.5.1.2 Fixed relative order of the anaphoric clause and antecedent clause ..............................................................198 4.5.2 English modally subordinated clauses are anaphoric clauses....................200 4.6 Summary ...............................................................................................................205 vii 5 The syntax of anaphoric clauses................................................................................206 5.1 Proposal: Chained, adjoined, and mediated argument clauses.................................206 5.2 The diagnostics ......................................................................................................208 5.2.1 Exclusion tests.........................................................................................208 5.2.2 Island tests...............................................................................................208 5.3 Applying the exclusion tests...................................................................................209 5.3.1 Linear precedence....................................................................................209 5.3.1.1 Chained clauses must follow their antecedent............................210 5.3.1.2 Adjoined clauses can follow their antecedent ............................211 5.3.1.3 Mediated argument clauses usually (but not always) follow their antecedent ............................................................211 5.3.2 Ability to be a matrix clause based on morpho-syntax .............................213 5.3.2.1 Chained clauses are always capable of being matrix clauses......214 5.3.2.2 Adjoined clauses are not always capable of being matrix clauses ............................................................215 5.3.2.3 Mediated argument clauses are not always capable of being matrix clauses ...............................................215 5.3.3 Prosodification ........................................................................................215 5.3.3.1 Chained clauses require an intonational break ...........................216 5.3.3.2 Adjoined clauses do not require an intonational break ...............216 5.3.3.3 Mediated argument clauses do not require an intonational break ...............................................................217 5.3.4 Interim summary .....................................................................................217 5.4 Applying the island tests ........................................................................................218 5.4.1 Long distance wh-construal must be across mediated arguments .............219 5.4.1.1 Wh-words can be long distance with mediated argument clauses ......................................................221 5.4.1.2 Wh-words cannot be long distance with adjoined clauses ..........222 5.4.1.3 Wh-words cannot be long distance with chained clauses ...........223 5.4.2 Long distance quantifier fronting must be across mediated arguments .....224 viii 5.4.2.1 Mediated argument clauses permit long distance quantifier-fronting..............................................225 5.4.2.2 Adjoined clauses do not permit long distance quantifier fronting ..............................................226 5.4.2.3 Chained clauses do not permit long distance quantifier-fronting..............................................227 5.4.3 Long distance argument expression-fronting............................................227 5.4.3.1 Mediated argument clauses permit long distance argument-fronting ..............................................228 5.4.3.2 Adjoined clauses do not permit long distance argument-fronting ..............................................230 5.4.3.3 Chained clauses do not permit long distance argument fronting...............................................231 5.4.4 (Non-)obligatory switch reference picks out object-oriented clauses ........231 5.5 Consequences ........................................................................................................236 5.5.1 The non-existence of argument clauses....................................................237 5.5.1.1 \u00E2\u0080\u0098Subject\u00E2\u0080\u0099 clauses require overt nominal antecedent....................241 5.5.1.2 Clauses have different ordering properties than arguments ........244 5.5.1.3 No predicates subcategorize for clauses.....................................247 5.5.1.4 Predicates that subcategorize for nominals: AIt verbs................250 5.5.1.5 Incorporation is nominal ...........................................................252 5.5.1.6 Summary: Clauses do not sit in argument positions...................254 5.5.2 VP-complementation involves restructuring ............................................255 5.5.2.1 Restructuring involves a single set of agreement .......................256 5.5.2.2 Independent-order agreement is possible...................................257 5.5.2.3 Restructuring allows only one set of temporal marking .............258 5.5.2.4 Restructured clauses introduce a single set of arguments...........259 5.5.2.5 Restructured clauses only permit a single subject ......................261 5.5.2.6 Restructuring preverbs are category-insensitive.........................262 5.5.3 Copy-to-object constructions must be local agreement.............................263 ix 5.6 Summary ...............................................................................................................268 6 The semantics of anaphoric clauses ..........................................................................270 6.1 Proposal: Presuppositional, a-veridical, and unspecified clauses.............................270 6.2 \u00C3\u00AA- as the unspecified complementizer .....................................................................271 6.2.1 Distributional evidence for \u00C3\u00AA- being unspecified ......................................272 6.2.2 Interpretational evidence that \u00C3\u00AA- is unspecified.........................................274 6.3 k\u00C3\u00A2- as a presuppositional complementizer ..............................................................277 6.3.1 Relative clauses: k\u00C3\u00A2- and \u00C3\u00AA.......................................................................278 6.3.2 Wh-questions ..........................................................................................283 6.3.2.1 k\u00C3\u00A2- wh-questions as presuppositional ........................................285 6.3.3 Temporal modification ............................................................................290 6.3.4 Concessive clauses ..................................................................................292 6.3.5 Correlatives .............................................................................................295 6.4 \u00EF\u0081\u00B8 as a-veridical.....................................................................................................296 6.4.1 Mediated argument clauses split along a-veridicality ...............................299 6.4.1.1 Sensitivity to weak intensional predicates..................................299 6.4.1.2 Sensitivity to weak intensional meanings ..................................301 6.4.1.3 Sensitivity to negation...............................................................303 6.4.1.4 Insensitivity to factive-emotive predicates.................................304 6.4.1.5 Insensitivity to lack-of-speaker-knowledge ...............................305 6.4.2 Adverbial clauses split along a-veridicality..............................................307 6.4.2.1 Irrealis temporal modification ...................................................307 6.4.2.2 Unrealized alternatives..............................................................308 6.4.2.3 \u00E2\u0080\u0098before\u00E2\u0080\u0099 clauses ........................................................................309 6.4.2.4 Antecedents of conditionals ......................................................312 6.4.2.5 Purpose clauses .........................................................................313 6.5 Summary ...............................................................................................................314 7 Conclusions ................................................................................................................317 7.1 The syntax and semantics of clause-typing in Plains Cree ......................................317 x 7.2 The parallels between CPs and DPs .......................................................................319 7.3 (Im)possible analyses of k\u00C3\u00AE.....................................................................................319 7.3.1 k\u00C3\u00AE- marks disjunction and precedence.......................................................320 7.3.2 k\u00C3\u00AE- is not a deictic past tense ....................................................................322 7.3.3 k\u00C3\u00AE- is not a perfect ....................................................................................324 7.3.4 k\u00C3\u00AE- is not a perfective ...............................................................................328 7.4 Deconstructing modality: Clause-typing, irreality, and k\u00C3\u00AE .......................................330 7.4.1 The role of clause-typing: Circumstantial vs. deontic modality ................331 7.4.2 k\u00C3\u00AE- has existential modal force under negation..........................................334 7.4.3 Negation widens possible interpretations .................................................335 7.4.4 Embedding neutralizes modal distinctions ...............................................336 7.4.5 Summary.................................................................................................339 7.5 Variation in clause-typing across Algonquian ........................................................340 7.5.1 Variation of the pronominal proclitics: Plains Cree vs. Blackfoot ............340 7.5.2 Variation in initial change: Plains Cree vs. Blackfoot vs. Ojibwe.............343 References ..............................................................................................................................345 xi List of tables Table 1.1 Indexicality vs. embedding ..........................................................................................3 Table 1.2 Mapping the indexical/anaphoric contrast onto INDEPENDENT and CONJUNCT...............7 Table 1.3 The CONJUNCT modes ................................................................................................12 Table 2.1 Left-edge clause-typing proclitic determines distribution of clause ............................34 Table 2.2 Diagnostics for specifiers vs. heads............................................................................36 Table 2.3 Person marking in INDEPENDENT and possession paradigms.......................................37 Table 3.1 Summary of the INDEPENDENT order paradigm in Plains Cree ....................................54 Table 3.2 Distribution of subordinators by clause-type in Plains Cree........................................60 Table 3.3 Predicate modifier antecedents occur in both clause-types .........................................67 Table 3.4 CP-modifier antecedents are possible in both clause-types .........................................69 Table 3.5 No cross-clausal antecedents for indexical INDEPENDENT clauses ...............................71 Table 3.6 Distribution of spatio-temporal proform variables by clause-type...............................74 Table 3.7 Person marking in INDEPENDENT clauses vs. possessed nominals ...............................77 Table 3.8 (Un)attested co-occurrence of -yi in INDEPENDENT order (Wolfart 1973 .....................88 Table 3.9 Interpretation of INDEPENDENT clauses by aspectual class ..........................................94 Table 3.10 Interpretations of \u00C3\u00AAsa .............................................................................................123 Table 4.1 Structures which meet the c-command condition .....................................................139 Table 4.2 Structures which do not meet the c-command condition...........................................140 Table 4.3 Structures which meet the precedence condition ......................................................140 Table 4.4 Structures which do not meet the precedence condition............................................141 Table 4.5 Structures which are undefined for antecedent licensing of anaphoric elements........141 Table 4.6 Clause-typing in discourse-initial position ...............................................................161 Table 4.7 Sequencing of clause-types by discourse-type..........................................................161 Table 4.8 Deictic vs. anaphoric dependents .............................................................................167 Table 4.9 Distribution of -yi by anaphoric configuration..........................................................183 Table 4.10 Anaphoricity vs. embedding ..................................................................................191 Table 5.1 Morpho-syntactic classification of anaphoric CONJUNCT clauses ..............................206 Table 5.2 Diagnostic 1: Subjection to precedence....................................................................209 Table 5.3 Diagnostic 2: Matrix capability................................................................................213 Table 5.4 Only potential matrix clauses can be chained clauses ...............................................214 xii Table 5.5 Adjoined clauses need not have potential to be matrix clauses .................................215 Table 5.6 Mediated argument clauses need not have potential to be matrix clauses..................215 Table 5.7 Diagnostic 3: Intonational break ..............................................................................216 Table 5.8 Only mediated argument clauses allow long-distance wh-words ..............................220 Table 5.9 Only mediated argument clauses allow long-distance quantifier-fronting.................224 Table 5.10 Only mediated argument clauses allow long-distance argument expressions ..........227 Table 5.11 Restructuring preverbs in Plains Cree ....................................................................256 Table 6.1 \u00C3\u00AA-clauses are anaphoric in unembedded contexts......................................................273 Table 6.2 \u00C3\u00AA-clauses are non-presuppositional in relative-clause contexts..................................273 Table 6.3 \u00C3\u00AA-clauses are veridical in mediated argument-clause contexts...................................273 Table 6.4 Asymmetries between two wh-clause types .............................................................286 Table 6.5 Clause-typing and kiy\u00C3\u00A2m..........................................................................................293 Table 6.6 Comparison of Plains Cree simple CONJUNCT and Romanian subjunctive ................300 Table 6.7 Classification of predicates by clause-type of subordinate clause .............................300 Table 6.8 Unselective predicates .............................................................................................301 Table 6.9 Factive-emotive predicates in Plains Cree ................................................................305 Table 7.1 Patterning of k\u00C3\u00AE- relative to the English perfect.........................................................325 Table 7.2 Patterning of k\u00C3\u00AE- relative to perfective aspect............................................................328 Table 7.3 Modal interpretations...............................................................................................331 Table 7.4 Mapping of Plains Cree forms to modal interpretations............................................331 Table 7.5 Clause-typing affects modal base.............................................................................332 Table 7.6 k\u00C3\u00AE- codes existential force in modal contexts ............................................................333 Table 7.7 Modal base vs. quantification in Plains Cree ............................................................339 Table 7.8 Modal interpretations in Plains Cree ........................................................................340 Table 7.9 Diagnostics for determining the position of pronominal proclitics............................341 Table 7.10 Properties of initial change in Plains Cree, Blackfoot, and Ojibwe .........................343 xiii List of abbrevations and symbols Abbreviations 0 = inanimate (agreement) 1 = 1st person 1>2 = 1st person acts on 2nd person 1>3 = 1st person acts on 3rd person 2 = 2nd person 2>1 = 2nd person acts on 1st person 2>3 = 2nd person acts on 3rd person 3 = 3rd person 3>1 = 3rd person acts on 1st person 3>2 = 3rd person acts on 2nd person AN = animate APPL = applicative BEN = benefactive C1 = changed conjunct 1 C2 = changed conjunct 2 CONJ = conjunctive CONN = connective COME = directional towards origo DEIC = deictic DEM = demonstrative DIR = direct DJ = disjoint argument DS = different subject DUB = dubitative DUR = durative EMPH = emphatic EPEN = epenthetic EVID = evidential FUT = future GO = direction away from origo HAB = habitual HES = hesitation HORT = hortative IC = internal change IMP = imperfective INAN = inanimate INCEP = inceptive INV = inverse IRR = irrealis PERF = perfective PV = preverb marker INC = inceptive INDIC = indicative INT = intend IRR = irrealis LOC = locative MIDST = on-going action MOD = modifier NEG = negation NOM = nominalizer OBV = obviative ORIG = origin PL = plural Q = interrogative QUANT = quantifier PL = plural PREV = previous PV = preverb RAT = rationale RED = reduplication REFLX = reflexive REL = relative clause marker REMP = remote past RR = relative root SAP = speech act participant SG = singular SIM = simultaneous SUBJ = subjunctive SUBJ1 = subjunctive 1 SUBJ2 = subjunctive 2 TEMP = temporal TOP = topic marker USC = unspecified subject construction VAI = verb, animate intransitive VII = verb, inanimate intransitive VTA = verb, transitive animate VTI = verb, transitive inanimate WH = wh-word xiv Symbols * = string is ill-formed ! = string is well-formed, but cannot have the intended interpretation # = string is well-fomed, but infelicitous in the given context \u00E2\u0088\u0083 = existential quantifier \u00E2\u0088\u0080 = universal quantifier \u00C2\u00AC = negation < = temporally precedes [ ] = constituency brackets CP = complementizer phrase DP = determiner phrase F = function IP = inflectional phrase Op = operator p = proposition S = speaker s = situation variable T = time t = trace vbl = variable VP = verb phrase xv Acknowledgements Plains Cree is a beautiful language. My heartfelt thanks go first to the people that speak it who have been so patient and generous with me over the past five years as I tried both to learn it and to learn about it. I first heard Plains Cree when I took a conversational class at the University of Alberta; Dorothy Thunder always encouraged my efforts and made the class fun. Wally Awasis not only put up with my really long (and often bizarre!) constructed sentences, but told great jokes and sang beautiful songs. Josephine Small shared her time generously amid a schedule that would already be too much for most people. I appreciated Rita Daniels\u00E2\u0080\u0099 warm hospitality and openness to almost total strangers when we visited her (including the bannock lessons). She told lovely stories and made Plains Cree feel \u00E2\u0080\u0098real.\u00E2\u0080\u0099 The early morning conversations and work sessions with Joseph Deschamps in Alberta were the basis for many of the ideas in this thesis, and also made life seem better. Toni Cardinal always seemed to know what question I was trying to ask, and even better, how to answer it. Her clear thinking, patience, and steadfastness brought this work together. I hope some of what they have tried to teach me is captured in these pages. Amidst an amazingly busy schedule, my research supervisor, Rose-Marie D\u00C3\u00A9chaine, spent much time and energy answering my questions, explaining concepts (it always seemed so crystal clear when you said it!), playing devil\u00E2\u0080\u0099s advocate, and helping me work through the problems (obvious and not-so-obvious) of my analysis. She also taught me about a theoretical linguist\u00E2\u0080\u0099s life \u00E2\u0080\u0093 guiding me through everything from writing grants, abstracts and papers, to fieldwork, to giving talks, to surviving. Thank you. Thanks also to my other thesis committee members, Lisa Matthewson and Martina Wiltschko. Lisa carefully read every draft I gave her, no matter how sketchy, and gave detailed comments. I also particularly appreciated her enthusiasm for data and trying to understand how the language worked, even when we had no idea what was going on. Martina\u00E2\u0080\u0099s lectures on syntax first made me think I could understand it. Her appreciation of new ideas encouraged me to think about my data in fresh ways, and her insistence on clarity \u00E2\u0080\u0093 both in analysis and in writing \u00E2\u0080\u0093 definitely made this thesis more readable. My thesis benefited from being read by the three members of the external examination committee. Amy Dahlstrom, Nancy Hedberg, and Jessica de Villiers were generous enough to take the time to read the thesis and asked useful questions about all aspects of the thesis, from theoretical implementation to methodology to implications of the analysis for Plains Cree and other languages. Despite his more unofficial status, H.C. Wolfart played a key role in this thesis and in my development as a linguist. He has always been interested in both the big issues and the details (contentful and editorial), and he has been a sympathetic and encouraging audience even when he vehemently disagreed with the crazy things I was doing. He also set an example for how corpus linguistics can and should be done. I am also grateful to my undergraduate advisor Monica Macaulay. She first introduced me to Algonquian languages, taught my first class on fieldmethods, gave me suggestions about suitable graduate schools to consider and straight-forward practical advice about navigating it once I started, and oversaw my first conference presentation. I met and have benefited greatly from talking to a number of other Algonquian linguists. Thanks to Eleanor Blain, Phil Branigan, Carrie Dyck, Inge Genee, Jean Okim\u00C3\u00A2sis, Charlotte Reinholtz, Nicole Rosen, and Arok Wolvengrey. More generally, thanks to audiences at ACAL, xvi the Algonquian conference, ASA, CLA, NWLC, SSILA, WSCLA, and UBC for feedback on talks given in the development of this work. My early teachers at UW-Madison \u00E2\u0080\u0093 Murvet Enc, Marlys Macken, Jack McKeown, Victoria Pagan, Matt Pearson, Bozena Tieszen, and Andrew Wolpert \u00E2\u0080\u0093 gave me a love for language and linguistics. Without the UBC faculty \u00E2\u0080\u0093 Guy Carden, Henry Davis, Bryan Gick, Gunnar Hannsson, Doug Pulleyblank, Hotze Rullmann, Pat Shaw, Joe Stemberger, and Eric Vatikiotis-Bateson \u00E2\u0080\u0093 the linguistics department would not be nearly what it is. Edna Dharmaratne, the department secretary, helped make sure I got through the many paperwork hoops of a graduate degree, including getting registered and getting paid. I am very grateful for all the times she went the extra mile for me and others; I don\u00E2\u0080\u0099t know what I would have done without her. Fortunately, I didn\u00E2\u0080\u0099t have to go into financial debt to support going to school. This research was supported by a Theodore E. Arnold Fellowship, a University Graduate Fellowship, and a Josephine Berthier Fellowship. Thanks to the taxpayers of Canada. I was part of the largest incoming graduate class at UBC linguistics ever, and we were quite the group. Solveiga Armoskaite made lovely dinners and I enjoyed doing elicitation with her. Fiona Campbell was the first person in the department I met; she introduced me to Benny\u00E2\u0080\u0099s. Ramona MacDowell knew how to make decisions. I will not forget Jason Brown, Peter Jacobs, Jong-won Kim, Karsten Koch, Jeff Muehlbauer, Dominique Quis, and Christine Ravinski. I have no idea where you all will end up, but it will make for a great story. Many thanks to Rachel Wojdak and Sugunya (Add) Ruangjaroon for looking out for me as a new student, telling me what to expect in graduate school, and sharing many parts of their lives. Thanks to Florence Woo for afternoons of quilting and walking. Thanks to Ryan Waldie for knowing that when \u00E2\u0080\u009CX\u00E2\u0080\u009D is not explained, nothing is explained. Thanks to the Thursday night dinner & reading group. Thanks to students inside and out of UBC linguistics: Oladiipo Ajiboye, Mario Chavez-Peon, Chin Seok Koon, Yunhee Chung, Kerim Demirci, Atsushi Fujimori, Masaru Kiyota, Diana Gibraiel, Yoko Ikegami, Paola Quintanar, Kristin Speth, Tanya Slavin, Ian Wilson and Noriko Yamane-Tanaka. I didn\u00E2\u0080\u0099t ever expect to meet someone like Qin Shujun. Her search for what is right and willingness to try anything rekindles my idealism, and she makes everything more fun. Plus she introduced me to dim sum and heroically proofed my thesis (remaining typos testify only to how big the job was). Thanks to my parents, Dave and Kathy Cook. I am grateful to have a dad who taught me to read, think, and be interested in the world around me from an early age. The longer I live the more I realize the many ways my mom has set an example for me. Thanks to my brothers and sisters for interrupting me with mail, phonecalls, and visits: Sam, Denver, Martha, Aselefech, Ty, Laura, Marshall, Asrat, Tesfu, Nettie, Ellie, Rosie, Tommy, and Joe. Their schedules and lives helped keep mine in perspective. My extended family has also been extremely kind: Gladys Scherwitz, Judy Cook, Bill and Sue Scherwitz, Joe and Nella Cook, Caroline Cook, and Cy and Judy Shuster all provided much-needed encouragement (both personal and intellectual) for my endeavors. I would also like to remember my grandfathers, Philip Cook and William Scherwitz, who strongly believed in the value of an education, but passed away long before my schooling ended. Thanks to Gerald and Suzanne Muehlbauer, Ned and Debbie Wicker, and Kathy Ritenour for their support. Last but not least, I am grateful to Jeff. He more than anyone knows what the last six years have been; thank you for always asking what the truth of a flower is. xvii Nature gives most of her evidence in answer to the questions we ask her. Here, as in the courts, the character of the evidence depends on the shape of the examination, and a good cross-examiner can do wonders. He will not indeed elicit falsehoods from an honest witness. But, in relation to the total truth in the witness\u00E2\u0080\u0099s mind, the structure of the examination is like a stencil. It determines how much of that total truth will appear and what pattern it will suggest. C.S. Lewis The Discarded Image 1 CHAPTER 1 INDEXICAL VERSUS ANAPHORIC CPS 1.1 Proposal: Indexical vs. anaphoric CPs This thesis proposes that there is a fundamental division between sentence-types (CPs) which anchor to the speech act, and those which do not. I call the first type indexical CPs: these are sentences which are obligatorily interpreted deictically \u00E2\u0080\u0093 with respect to the speaker, the speech time, and the speech place (B\u00C3\u00BChler 1934, Bar-Hillel 1954, Fillmore 1975, 1982, Ehlich 1982, Kaplan 1989, Schlenker 2003). (1) CPindexical I call the second type of CP anaphoric. Anaphoric CPs are interpreted with respect to some other element; just as anaphoric expressions are interpreted relative to an antecedent (B\u00C3\u00BChler 1934, Ross 1967, Langacker 1969, Fillmore 1975, Reinhart 1976), so with anaphoric CPs. (2) XPantecedent \u00E2\u0080\u00A6 CPanaphoric Starting with the idea that a proposition cannot be evaluated until it is given a context (Austin 1950, Barwise 1981), I model the difference between indexical and anaphoric clauses within a situation semantics framework as a difference in the value of the situation. With indexical clauses the proposition is evaluated with respect to the speech situation (s0). With anaphoric clauses, the situation is not specified \u00E2\u0080\u0093 rather it is anaphorically given (s). a. INDEXICAL CP b. ANAPHORIC CP (3) CP 2 s0 2 C XP 5 \u00E2\u0080\u00A6 CP 2 s 2 C XP 5 2 The division between indexical and anaphoric CPs is motivated on the basis of the clause-typing system of Plains Cree, an Algonquian language spoken on the plains of western Canada and the United States. Plains Cree has an explicit clause-typing system whereby every clause is morpho-syntactically coded for its clause-typing status. For example, a clause can have two entirely different sets of inflectional morphology, depending on whether it is in the INDEPENDENT or CONJUNCT order. In (4a) there is a first-person morpheme ni- preceding the stem w\u00C3\u00A2pam \u00E2\u0080\u0098x sees animate\u00E2\u0080\u0099; a grammatical function coding morpheme -\u00C3\u00A2, and a third-person element -w. In (4b), the ni- has been replaced by the element \u00C3\u00AA-, and the two morphemes -\u00C3\u00A2 and - w have been replaced by a single morpheme coding a first-person subject and a third-person object. (4) Plains Cree (Algonquian) a. niw\u00C3\u00A2pam\u00C3\u00A2w atim INDEPENDENT ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w atim 1- see.VTA -DIR-3 dog \u00E2\u0080\u0098I see a dog.\u00E2\u0080\u0099 b. \u00C3\u00AA-w\u00C3\u00A2pamak atim CONJUNCT \u00C3\u00AA- w\u00C3\u00A2pam -ak atim C1-see.VTA -1>3 dog \u00E2\u0080\u0098\u00E2\u0080\u00A6 I see a dog.\u00E2\u0080\u0099 INDEPENDENT order clauses are restricted only to (a subset of) matrix clauses, and instantiate what I am calling indexical clauses. I will show that they are associated with a particular set of semantic properties deriving from their anchoring to the speech act: they are interpreted relative to the speech time, speech place, and the speaker. CONJUNCT clauses have a much wider distribution, which depends on a further subdivision determined by the affixes on the left and right edges of the clause. All CONJUNCT clauses can be embedded; those with the left-edge \u00C3\u00AA- morpheme may also occur in matrix clauses \u00E2\u0080\u0093 but, as I argue, without the deictic properties of INDEPENDENT clauses. CONJUNCT clauses, as a class, instantiate what I am calling anaphoric clauses: they are licensed either by a linguistic antecedent or by a shared context (cf. Fillmore 1975, Reinhart 2003, Kratzer 2007). Anaphoric CONJUNCT clauses differ in how this licensing is achieved; some specifically require subordination to an antecedent, while others do not. 3 1.2 Relation to previous work The clause-typing distinction between indexical and anaphoric CPs is connected to \u00E2\u0080\u0093 and therefore brings together \u00E2\u0080\u0093 several distinct fields that have had significant previous research, including clause-typing (\u00C2\u00A71.1.1, 1.1.2), indexicality (\u00C2\u00A71.1.3), anaphora (\u00C2\u00A71.1.4), and Algonquian linguistics (\u00C2\u00A71.1.5). 1.2.1 Connection to the matrix/embedded distinction The indexical versus anaphoric distinction cross-cuts the familiar division between matrix and embedded clauses (Hockett 1958) that has been discussed for many Indo-European languages. For Plains Cree, this means that the INDEPENDENT/CONJUNCT clause-typing system does not directly map onto the matrix/embedded distinction (cf. \u00C2\u00A72.3). Cross-linguistically, I expect that clauses which are morpho-syntactically or syntactically typed as matrix clauses will subclassify into indexical and anaphoric clauses, as in table 1.1. Matrix Embedded Indexical \u00E2\u009C\u0094 \u00E2\u009C\u0096 Anaphoric \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 1.1. Indexicality vs. embedding In addition, the properties of anaphoric clauses \u00E2\u0080\u0093 which can occur in both matrix and embedded contexts \u00E2\u0080\u0093 have properties in common with clauses that participate in clause-chaining (Longacre 1983, Finer 1985, Stirling 1993, among many others), but does not have to stipulate these chains as a special kind of clause (cf. Giv\u00C3\u00B3n 2001). Both of these issues are addressed in chapter 4. 1.2.2 Connection to illocutionary force The distinction between indexical and anaphoric clauses also cross-cuts illocutionary force (i.e., the distinction between declaratives, interrogatives, and/or imperatives; cf. Cheng 1991, 4 Chomsky 1995, Rizzi 1997, Portner 1999). This means that Plains Cree\u00E2\u0080\u0099s INDEPENDENT/CONJUNCT clause-typing system will not map directly onto illocutionary force; for example, both indexical and anaphoric clauses may in principle be (and in Plains Cree, are) declarative or interrogative. 1.2.3 Parallel between indexical CPs and indexical expressions Indexical expressions go back at least as far as B\u00C3\u00BChler (1934), and are defined for the purposes of this thesis as in (5), following B\u00C3\u00BChler (1934), Bar-Hillel (1954), Fillmore (1975), and Kaplan (1989), among many others. (5) Indexical expressiondef: a linguistic element whose interpretation requires identification of the speaker, speech time and/or speech location Typical English indexical expressions include pronominal I as well as spatial and temporal relation to the speech act (i.e., the here and now; cf. B\u00C3\u00BChler 1934, Bar-Hillel 1954). Since an indexical expression looks to the speech act for its interpretation, it is a particular kind of deictic (from Greek \u00CE\u00B4\u00CE\u00B5\u00CE\u00B9\u00CE\u00BA- \u00E2\u0080\u0098point out\u00E2\u0080\u0099) expression, as defined in (6) (cf. Fillmore 1975, Kaplan 1989, Green 1989, Nunberg 1993, Schlenker 2003). (6) Deictic expressiondef: a linguistic element whose interpretation requires pointing to some aspect of the context in which it is used Claiming that a clause is indexical therefore means that it has the same pointing function to the speech act that any other indexical expression has. In this thesis, I argue that Plains Cree\u00E2\u0080\u0099s INDEPENDENT order morpho-syntactically codes such a clause and does in fact have this pointing function. Claiming that clause-typing codes a relation to the speech act relates in significant ways to the debate about the syntactic representation of \u00E2\u0080\u0098speech-act\u00E2\u0080\u0099 elements. This is most clearly seen in the tense literature, where reference to the speech time goes back at least as far as Paul 1886, and has been used in literally countless syntactic and semantic analyses since (in chronological order: Reichenbach 1947, Klein 1994, Kamp 1981, Stowell 1982, Kamp & Rohrer 1983; Enc 1987, Stowell 1995, Abusch 1998, Demirdache & Uribe-Etxebarria 2000, among 5 many others). In the current work, temporal deixis on the speech time is taken to be part of the general deixis on the speech situation (cf. Kratzer 2007). Similarly, there is much current work on the representation of a speaker or a speech-act-phrase; this is apparent particularly in the evidential literature (cf. Cinque 1994, Rivero 1994, Rizzi 1997, Speas & Tenny 2003, 2005), but also in work on speaker-oriented truth (Lasersohn 2005, Stephenson 2007), and in the linguistic structure of discourse (Banfield 1982, Smith 2003). In Plains Cree, all of these phenomena are associated with a particular clause-type, and thus the work reported on here \u00E2\u0080\u0093 and the analysis pursued \u00E2\u0080\u0093 is an attempt to show how these concepts might be linked. In particular, I model these relations within a situation semantics framework, where every proposition must be evaluated with respect to a situation (Austin 1950, Barwise 1981, Barwise & Perry 1983, Barwise & Etchemendy 1987, Kratzer 2007a). I argue that in an indexical clause this situation as the speech situation1. Following Kratzer (1989, 2007a), I define a situation s as a partial world, where a partial world is a domain for truth evaluation that does not necessarily contain truth values for all possible propositions. (7) Situationdef: a partial world Within this framework, propositions are evaluated relative to situations. The speech situation s0 is simply a situation in which someone is speaking. When we think about what would be necessary for a situation to be called a speech situation, we would minimally need to include the individuals who are doing the speaking and hearing (i.e., the Speaker and Hearer); the time at which the speaking occurred (i.e., Speech Time), and the place at which the speaking occurred (i.e., Speech Location). The definition of a speech situation is given in (8). (8) Speech situationdef: a situation minimally involving (i) the Speaker/Hearer; (ii) the Speech Time; and (iii) the Speech Location A speech situation thus captures the relation between, for example, temporal effects and person effects. 1 See Kaplan 1989 for a treatment of indexicals in contexts, rather than situations. 6 1.2.4 Parallel between anaphoric CPs and anaphoric expressions Anaphoric expressions have also been an enormous research topic in linguistics. Working on pronominal forms in English, there are as many proposals about the relevant principles governing the licensing of anaphora as there are linguists, where licensing is defined as in (9). (9) Licensingdef: an element \u00CE\u00B1 is licensed iff there is some element \u00CE\u00B2 able to serve as its context of interpretation Fillmore (1975) provides discussion of how anaphora can be licensed both by an antecedent (antecedence licensing), and in the absence of a linguistic antecedent (shared context, termed symbolic licensing). There is general agreement that the latter case is not an entirely separate licensing mechanism from the first case (Heim & Kratzer 1998, Reinhart 2003); however, most of the work has tried to specify the conditions on the first mechanism, antecedence licensing. In particular, there have been disagreements about whether anaphora can be accounted for strictly by notions of hierarchy (e.g., the antecedent c-commands the anaphor; see Reinhart 1976, 1983, Kayne 1994), or whether both hierarchy and precedence (e.g., the antecedent precedes the anaphor) are relevant (Langacker 1967, Ross 1967, 1969, McCawley 1988, Carden 1986, Williams 1997). There is also disagreement as to whether anaphora have special properties (this seems to be the standard position), or whether they are an elsewhere case (as explicitly argued in McCawley 1970 and Williams 1997). This thesis contributes to the discussion in at least three ways. First, it extends the discussion of pronominal anaphora into the domain of clauses by claiming that there is a particular kind of clause that can host (both antecedence and symbolic) anaphoric relations. Second, it explicitly claims that anaphoric clauses are an elsewhere case: anaphoric clauses occur in contexts where an indexical clause fails to occur. Third, the data set considered here has the same licensing conditions as discussed by Carden (1986) and Williams (1997): i.e., c-command and precedence are split into separate conditions, with some anaphoric clauses needing only precedence, some needing only c-command, and others requiring both. 7 1.2.5 Connection to the INDEPENDENT/CONJUNCT contrast Within Algonquian linguistics, the distinction in clause-typing discussed in this thesis has been difficult to understand and analyze, despite a relatively long history of linguistic work on Plains Cree (Howse 1865, Lacombe 1874, Bloomfield 1928, Wolfart 1973, Dahlstrom 1991, Ogg 1991, Wolfart & Carroll 1996, Blain 1997, Long 1999, Hirose 1999) and related languages (in particular Reinholtz 1996, 1999, 2007 for Swampy Cree). It is hoped that the analysis developed here for Plains Cree\u00E2\u0080\u0099s INDEPENDENT/CONJUNCT distinction will serve as a basis for a more fine- grained description and analysis of the function of clause-typing across the Algonquian language family. In particular, while the INDEPENDENT/CONJUNCT contrast in Plains Cree maps in a one- to-one fashion to the indexical/anaphoric contrast (language 1 in table 1.2), it is possible that in other languages, the INDEPENDENT order extends across both clause-types (language 2), or alternatively, the CONJUNCT order extends across both (language 3). Clause type Language 1 (=Plains Cree) Language 2 (= ??) Language 3 (= ??) Indexical independent independent independent conjunct Anaphoric conjunct independent conjunct conjunct Table 1.2. Mapping the indexical/anaphoric contrast onto INDEPENDENT and CONJUNCT Within Plains Cree, Blain\u00E2\u0080\u0099s (1997) thesis on wh-questions proposes a structural analysis of two kinds of CONJUNCT clauses within a restricted set of contexts, but the extent to which the analysis can be generalized across the language has not been addressed. Although her analysis is only relevant for specific parts of this thesis, the findings here are consistent with her claims. Long\u00E2\u0080\u0099s (1999) thesis on complement clauses in Plains & Swampy Cree similarly proposes a structural analysis of one kind of CONJUNCT clause in \u00E2\u0080\u0098complement\u00E2\u0080\u0099 contexts, and provides a number of diagnostics to structurally distinguish complement clauses from adjunct clauses. The work in chapter 5 builds on this analysis, adding more clause-types, more diagnostics, and proposing a third syntactic clausal relation: chains. Other previous work on Plains Cree has primarily focussed on the (large amounts of) morpho-phonology (cf. Lacombe 1874, Bloomfield 1925, Wolfart 1973), its historical relation to 8 other Algonquian languages (Goddard 1967, 1974, Pentland 1979, 1999), and morpho-syntax (Dahlstrom 1991, D\u00C3\u00A9chaine 1999, 2001, 2002, 2003; Hirose 2000). This thesis has depended \u00E2\u0080\u0093 at times heavily \u00E2\u0080\u0093 on this previous work: without having it as a foundation, much of the current work could not have been done. At the same time, much of the data presented in this thesis \u00E2\u0080\u0093 looking at the co-occurrence restrictions between the verbal complex and particles, the structural relations between clauses, and the interpretation of utterances \u00E2\u0080\u0093 is novel, and many of the generalizations have not been previously discussed. Unless otherwise cited, all data comes from primary sources: either original elicitation fieldwork, or texts (i.e., transcribed recordings of narratives collected by others). This leads us to a discussion of methodology. 1.3 Methodology: Data collection and presentation This research is based on two methods of data collection. The first method is the elicitation of introspective speaker judgments. The data reported in this thesis was collected on the basis of work with six different fluent Plains Cree speakers from Alberta and Saskatchewan2. Elicitation sessions included a variety of tasks, including (i) translation tasks, where the speaker is either presented with an English sentence and asked to translate into Plains Cree or vice versa; (ii) judgment tasks, where the speaker is presented with a Plains Cree utterance and asked to judge its well-formedness and or felicity in a context; (iii) utterance-in-context tasks, where the speaker is asked to provide an appropriate utterance in a constructed context; and (iv) analytic tasks, where the speaker provides a reason for the (un)grammaticality or (in)felicity of the utterance (Cook & M\u00C3\u00BChlbauer 2006). Elicitation work is vital to this thesis in two ways: it provides linguists with information on what are impossible utterances, and with explicit information about the (im)possible meanings that an utterance may have. At the same time, it should be noted that \u00E2\u0080\u0093 as with all elicitation \u00E2\u0080\u0093 different tasks sometimes lead to quite different results. In particular, a Plains Cree utterance may be translated in a particular way into English (for example, past tense), but when tested in defined contexts, turn out not to have the same distribution or felicity 2 Relevant biographical information on consultants is as follows: S1 was male, mid-50s, from Thunderchild, SK; S2 was female, mid-40s, born Lac La Biche, AB, raised in Edmonton, AB; S3 was female, early 50s, from Ermineskin, AB; S4 was male, mid-50s, from Louis Bull, AB; S5 was female, early 40s, from Little Pine, SK; S6 was female, mid-60s, from Cold Lake, AB. 9 conditions as the English translation. Likewise, speakers may offer what turn out to be crucial insights in analytic tasks, but it is vitally important to test these insights via other tasks. The second method of data collection involves the use of corpus material. In particular, I draw from published narratives of four Plains Cree speakers that were recorded, then transcribed and (minimally) edited by Wolfart & Ahenakew3 (Whitecalf 1993, Minde 1997, K\u00C3\u00A2-Nipit\u00C3\u00AAht\u00C3\u00AAw 1998, Ahenakew 2000). These speakers are referred to in the thesis by the initials of their names: SW, EM, JKN, and AA, respectively. The narratives range in length from approximately one hour of speech (SW) to approximately two and a half hours of speech (each of EM, JKN, and AA). Generalizations from corpora are vital in that they show the possible utterances in an organically constructed context, and also indicate the robustness of a given phenomena. Following the premises of \u00E2\u0080\u0098cross-methodological validation\u00E2\u0080\u0099 (Carden & Dietrich 1982, Matthewson 2004), this thesis draws on both elicited and corpus data. Used together, these two methods of data collection can be used to cross-check the validity of the data set. For example, in elicitation contexts, speakers will sometimes rule utterances as impossible because the appropriate context has not been established; in such a case the corpora can (i) give evidence that the construction really does exist, and (ii) provide a ready-made context for that construction. Likewise, sometimes the construction being targeted will not appear in the corpora for accidental reasons (e.g., the appropriate context did not occur), but a fluent speaker can readily provide a judgment and context for the utterance when asked. In many cases, the two sources of data were combined, where pieces of the corpora were presented to the elicitation consultant and the consultant was asked about the meaning, or asked about possible permutations on the attested piece. This method was particularly useful for when judgments within a particular discourse context were needed, since the Plains Cree corpora provided a ready-made context without potential interference from a distinct framing language (in this case English). All Plains Cree data, whether from elicitation or textual sources, is presented in a (minimally) four-line format as follows (the lines enclosed in parentheses are given as relevant): 3 In essence, any transcription involves editorial decisions, even at the level of word breaks. Editorial work included marking of punctuation. False starts and hiccups were transcribed as they were heard. 10 (10) (context-of-utterance) Plains Cree data in standard Roman orthography Morphemic breakdown Morpheme-by-morpheme gloss \u00E2\u0080\u0098Free English translation.\u00E2\u0080\u0099 (comment(s) by consultant about utterance) In addition, text taken from textual sources is cited from the relevant text by speaker and paragraph number within the transcription4. Following the practices of Wolfart (e.g., 2000), textual data also includes the relevant contextual punctuation marking introduced by the editors (Ahenakew & Wolfart 1997, 1998; Wolfart & Ahenakew 1993, 2000) in the following way. Preceding the cited clause, ellipsis [...] indicates preceding linguistic material with no intervening period [.] or semicolon [;]. Any other intervening punctuation, including a comma [,], colons [:], and initial or ending quotations [\u00E2\u0080\u009C\u00E2\u0080\u009D] are marked. Following the cited clause, all punctuation is marked; if the punctuation is anything other than a period or semicolon, another ellipsis follows. Finally, it should be noted that because the translation is not always word-for-word, there are times where the punctuation of the English translation differs from the Plains Cree (for example, the Plains Cree clause may be in sentence initial position, but the English translation of that clause is in non-initial position). The internal morphological structure of stems is not usually given, since the relation between elements within the stem is different from the relation between elements external to the stem (Wolfart 1973). If necessary, stem-internal morpheme breaks are given within brackets [STEM-MORPHEMES]. A list of the abbreviations used in the morpheme-by-morpheme gloss is given in the front matter. As with any gloss, these are approximate and should not be taken as having any analytic or \u00E2\u0080\u0098real\u00E2\u0080\u0099 value. There are three symbols that may precede the Plains Cree line of an example: an asterisk [*], an exclamation point [!], or a pound sign [#]. The asterisk marks a string that was judged by one or more fluent speakers to be ill-formed \u00E2\u0080\u0093 i.e., an impossible utterance. The exclamation point marks an utterance that may be well-formed, but cannot have the relevant interpretation. Such utterances are often judged ungrammatical if presented in the context of the relevant 4 This annotation makes it easy to distinguish between data from textual sources and elicitation data. 11 structure (e.g., coordination), but judged grammatical in the context of some other construction (e.g., temporal modification). Finally, the pound sign marks a string that is grammatical, but infelicitous in a particular discourse context. 1.4 Plains Cree terminology Plains Cree, like other Algonquian languages, has three inflectional classes of words: verbs, which take one set of inflectional morphology, nouns, which take another set of inflectional morphology, and particles, which cannot be inflected (Wolfart 1973). Particles are a syntactically and semantically heterogeneous class which I will not deal with here (but see Ogg 1991 for discussion). Nominal stems may be inflected for possession and plurality (Lacombe 1874, Hockett 1966, Wolfart 1973, Dahlstrom 1991). Modifiers may attach to the left of the stem. A simplified template for nouns is given in (11). (11) Template for nominal stems [ POSS [ MOD [STEM] PL/OBV ] ] Depending on the context in which they occur, nouns fall into one of three referential categories: inanimate, animate, or obviative (a subclass of animate), but nouns are not inherently specified as to their category (see arguments in Wolfart 1973, M\u00C3\u00BChlbauer 2008). Verbal complexes consist of a stem, which almost always has internal structure (Wolfart 1973, Hirose 2000, D\u00C3\u00A9chaine 2003), including a root, a possible medial, and at least one final. (12) Template for verbal stems [STEM root \u00E2\u0080\u0093 (medial) \u00E2\u0080\u0093 final ] The finals are inflected for animacy and arguably code argument structure (i.e., the introduction of argument positions and the assigning of grammatical function) (cf. Hirose 1999, D\u00C3\u00A9chaine 2003). To the left of the stem is the pre-verb domain, which hosts, among other things modifiers and tense/aspect/modality markers (Edwards 1954, Wolfart 1973, Cook 2004). The left and right edges of the verbal complex external to the stem have person/number marking and, in the case of CONJUNCT clauses, a closed class of left-edge morphemes hosting an 12 ablaut process called initial change (IC, cf. Wolfart 1973); these latter will be of central concern to the thesis. As we saw at the beginning of the chapter, there are two agreement paradigms5. These are called orders: there are the INDEPENDENT and CONJUNCT orders, represented by the templates in (13) and (14) respectively. (The * in the template indicates the possibility of iteration; parenthesis indicates that the element is only sometimes present.) (13) template for INDEPENDENT order [VERBAL COMPLEX PERSON [ (PRE-VERB *) [STEM ] ] PERSON (NUMBER) ] (14) template for CONJUNCT order [VERBAL COMPLEX \u00C3\u00AA-/k\u00C3\u00A2-/IC [ (PRE-VERB *) [STEM ] ] PERSON (NUMBER) ] Orders may subclassify for modes; in Plains Cree, at least for the data set I have, the CONJUNCT order is the only one to have any modes (cf. Wolfart 1973, who documents three modes for the INDEPENDENT order in older forms of Plains Cree). These modes include (following Wolfart 1973) a primary division between simple CONJUNCT, and the changed CONJUNCT, and a further division depending on the suffixation of -i, yielding the subjunctive CONJUNCT, the and the iterative CONJUNCT. MODE FORM simple (ka-)nip\u00C3\u00A2t SIMPLE subjunctive nip\u00C3\u00A2ci \u00C3\u00AA-nip\u00C3\u00A2t k\u00C3\u00A2-nip\u00C3\u00A2t changed n\u00C3\u00AAp\u00C3\u00A2t CHANGED iterative n\u00C3\u00AAp\u00C3\u00A2ci Table 1.3. The CONJUNCT modes For the data set I am working with, the simple CONJUNCT is almost universally prefixed with the irrealis preverb ka-. Both the changed CONJUNCT formed by ablaut (n\u00C3\u00AAp\u00C3\u00A2t) and the iterative CONJUNCT (n\u00C3\u00AAp\u00C3\u00A2ci) are essentially absent from the data (i.e., no consultant recognized or produced the forms, and they were only attested a handful of times in the corpora). 5 There is a third paradigm \u00E2\u0080\u0093 the IMPERATIVE order. The imperative order cannot host most agreement, any of the elements on the far left edge, or most of the preverbs. I will not discuss it further in this thesis. 13 For a more in-depth discussion of Plains Cree\u00E2\u0080\u0099s grammar, see Wolfart (1973, 1996); Wolfart & Carroll (1981); and Dahlstrom (1991). For the purposes of this thesis, other Algonquian-specific terms will be introduced as necessary. 1.5 Layout of the thesis There are six chapters following the introduction. Chapter two presents a series of arguments that the sentence-level verbal complexes I am looking at in Plains Cree form a uniform syntactic class, which within the Principles & Parameters framework of Chomsky (1981) and Chomsky & Lasnik (1993) is called a CP. I further argue that in CONJUNCT clauses the left-edge elements head the CP (i.e., they are complementizers), while in INDEPENDENT clauses the left edge elements are in spec, CP. The next two chapters address the main proposal of the thesis, that the fundamental distinction in Plains Cree\u00E2\u0080\u0099s clause-typing system is between indexical and anaphoric clauses. Chapter three is concerned with structural and semantic contexts for indexical clauses, instantiated by Plains Cree INDEPENDENT order. Structurally, I show that indexical clauses (a) are subject to anti c-command (they cannot be c-commanded); and therefore (b) require all dependent elements to be resolved locally (within the clause). Semantically, I show that indexical clauses have indexical temporal and referential properties. Chapter four is concerned with what happens when the structural and semantic context is such that an indexical clause cannot occur. Here I claim that we get anaphoric clauses, instantiated by Plains Cree\u00E2\u0080\u0099s CONJUNCT order. Syntactically, I extend Williams\u00E2\u0080\u0099 (1997) analysis of anaphora, and show that anaphoric clauses are licensed by either a precedence condition or a c-command condition. Semantically, I show that the value of temporal/spatial/referential dependent elements within an anaphoric clause are determined by an antecedent. Finally, I use Fillmore\u00E2\u0080\u0099s (1975) contextual licensing principles of anaphora to derive the distribution of matrix anaphoric clauses. The next two chapters develop more specific syntactic and semantic analyses of anaphoric CONJUNCT clauses. 14 Chapter five argues that anaphoric clauses fall into three syntactic classes, defined by their relation to another clause: (i) chained clauses, which are governed solely by precedence and do not form a constituent with any other clause; (ii) adjunct clauses, which are governed by c-command and form a constituent of another clause; and (iii) mediated argument clauses, which are licensed by an argument-position (subject or object) and adjoined within the clause. Chapter six argues that anaphoric clauses also fall into three semantic classes, which cross-cut the syntactic classification of chapter 5, but map onto the form of the complementizer: (i) the complementizer k\u00C3\u00A2- introduces presupposed clauses, where the truth of the proposition being presented is assumed within the discourse; (ii) the \u00EF\u0081\u00B8 (null) complementizer introduces a-veridical clauses, where the truth of the proposition being presented is unevaluated within the discourse; and (iii) the complementizer \u00C3\u00AA- introduces an unspecified clause which does not carry any inherent semantic value. Chapter seven concludes by summarizing the main findings of the thesis, and pointing out possible directions for further research. 15 CHAPTER 2 MAPPING INDEXICAL AND ANAPHORIC CPS ONTO PLAINS CREE\u00E2\u0080\u0099S MORPHO-SYNTAX 2.1 Proposal: A one-to-one mapping in Plains Cree I claim that there are two clause-types in Plains Cree: indexical clauses and anaphoric clauses. Indexical clauses have an indexical speech situation (s0) in spec, CP, as in (1a), and anaphoric clauses have an anaphoric situation (s) in the same position (1b). a. INDEXICAL CP b. ANAPHORIC CP (1) CP 2 s0 2 C XP 5 CP 2 s 2 C XP 5 Syntactically and semantically, there is a one-to-one relation between the element in spec, CP (indexical vs. anaphoric), and the kind of dependencies which a CP may have. Indexical clauses code a proposition that is evaluated with respect to the speech situation (see chapter 3). Anaphoric clauses code a proposition that is evaluated to an anaphoric situation (see chapter 4). Morpho-syntactically, the distinction between indexical CPs and anaphoric CPs could logically have one of three patterns. The first possibility is that there is no morpho-syntactic distinction between different kinds of CPs; the distinction between them is contextually determined. The second possibility is that there is a morpho-syntactic differentiation between different kinds of CPs, but the distinction is cued to factors other than the indexical vs. anaphoric property. The third possibility is that there is a morpho-syntactic differentiation between different kinds of CPs which is specifically cued to the distinction between indexical and anaphoric 16 clauses. I claim that this is the pattern we see in Plains Cree (Algonquian): the two clause types are morpho-syntactically realized by two different clausal paradigms, called orders in the Algonquianist literature, and there is a direct mapping between the morpho-syntax, syntax, and semantics of indexical and anaphoric clauses. For Plains Cree, I show that the INDEPENDENT order instantiates indexical clauses. It is characterized by proclitics on the left edge of the clause (1st-person ni- in (2a); 2nd-person ki- in (2b)), and by a particular set of right edge agreement (e.g., the speech-act-participant suffix -n in (2a-b), and the third person suffix -w in (2c)). (2) Indexical clause = INDEPENDENT order a. nim\u00C3\u00AEcison ni-m\u00C3\u00AEciso -n 1-eat.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099ve eaten.\u00E2\u0080\u0099 b. kim\u00C3\u00AEcison ki-m\u00C3\u00AEciso -n 2-eat.VAI -SAP \u00E2\u0080\u0098You\u00E2\u0080\u0099ve eaten.\u00E2\u0080\u0099 c. m\u00C3\u00AEcisow m\u00C3\u00AEciso-w eat.VAI-3 \u00E2\u0080\u0098S/he\u00E2\u0080\u0099s eaten.\u00E2\u0080\u0099 For Plains Cree, I also show that the CONJUNCT order, of which there are several varieties, morpho-syntactically instantiates anaphoric clauses. What CONJUNCT clauses share with each other is a distinct set of right-edge morphology (for example, 3rd-person -t in (3a-c)). 17 (3) Anaphoric clauses = CONJUNCT order (citation forms) a. \u00C3\u00AA-w\u00C3\u00A2pam\u00C3\u00A2t \u00C3\u00AA- w\u00C3\u00A2pam -\u00C3\u00A2 -t C1-see.VTA-DIR-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6s/he sees him/her\u00E2\u0080\u0099 b. k\u00C3\u00A2-w\u00C3\u00A2pam\u00C3\u00A2t k\u00C3\u00A2- w\u00C3\u00A2pam -\u00C3\u00A2 -t C2-see.VTA -DIR-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6when s/he saw him/her\u00E2\u0080\u0099 c. wiy\u00C3\u00A2pam\u00C3\u00A2ci1 IC- w\u00C3\u00A2pam -\u00C3\u00A2 -t -i IC- see.VTA-DIR-3-PL \u00E2\u0080\u0098\u00E2\u0080\u00A6whenever s/he saw him/her\u00E2\u0080\u0099 On the left edge, the data in (3) illustrates a number of different clause-typing elements, all of which are associated with an ablaut process known in Algonquian linguistics as initial change (IC); these include the proclitics \u00C3\u00AA- and k\u00C3\u00A2-, (ablauted from i- and k\u00C3\u00AE-, respectively) and stem infixation (-iy-). In the absence of initial change, CONJUNCT clauses are termed \u00E2\u0080\u0098simple CONJUNCT\u00E2\u0080\u0099; they have an irrealis element (ka- or -i)2. These again share the right-edge agreement. (4) Anaphoric clauses = CONJUNCT order (citation forms) a. ka-w\u00C3\u00A2pam\u00C3\u00A2t ka- w\u00C3\u00A2pam -\u00C3\u00A2 -t IRR-see.VTA -DIR-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6him/her to see him/her\u00E2\u0080\u0099 b. w\u00C3\u00A2pam\u00C3\u00A2ci w\u00C3\u00A2pam -\u00C3\u00A2 -t -i see.VTA-DIR-3-SUBJ \u00E2\u0080\u0098\u00E2\u0080\u00A6if/when s/he see him/her\u00E2\u0080\u0099 1 Historically, ablaut of the stem could also take place without the subjunctive suffix \u00E2\u0080\u0093i (see, e.g., Wolfart 1973:46, who notes that it seemed to be disappearing in favor of the \u00C3\u00AA- proclitic). I have not found any examples of this kind in any of the corpora I have worked with, and none of the speakers I work with recognize or use these forms; even with the subjunctive marker, ablaut is now extremely restricted, and I have little to say about them. 2 Historically, a simple CONJUNCT clause like (4a) was reported to be possible without ka- (Wolfart 1973:46). However, I have not worked with a speaker who controls difference between a simple conjunct clause with vs. without the ka- proclitic. Writing 35 years ago, Wolfart (1973:45) comments that forms with ka- (or its alternant (ki)ta-) was by far more common than forms without it; since then, the completely bare verbal complex seems to have (all but?) disappeared. 18 The irrealis markers are different from the clause-typing elements in that they have a wider distribution: ka- can occur in both INDEPENDENT and CONJUNCT clauses, and -i can co-occur with the clause-typing elements \u00C3\u00AA-, k\u00C3\u00A2-, and internal change (IC). I will treat clauses like in (4) as having a null complementizer for reasons that will become clear later in the chapter. The mapping between the two clause types and the two orders in Plains Cree is thus represented as in (5).3 (5) TWO TYPES OF CLAUSES INDEXICAL CP (= INDEPENDENT) ANAPHORIC CP (= CONJUNCT) Structure: CP 2 s0 2 ni- C XP ki- 5 \u00EF\u0081\u00B8- CP 2 s 2 C XP k\u00C3\u00A2- 5 \u00C3\u00AA- IC \u00EF\u0081\u00B8- There are two parts to my claim which must be defended: (i) that what I\u00E2\u0080\u0099m talking about are CPs (rather than some smaller structure like an IP or a VP); and (ii) that there are two kinds of CPs (i.e., indexical and anaphoric). In the current chapter, I take up the claim that the two clause-types both have the properties of CPs. To the extent that this characterization is accurate, indexical and anaphoric CPs differ not in the amount of structure they have, but in the s vs. s0 contrast in spec, CP. This chapter lays the groundwork for the later chapters, where I address the second claim, analyzing the syntax and semantics of the two clause types in detail. 3 On this account, the ni-, ki-, and \u00EF\u0081\u00B8- proclitics are all in the same syntactic position as the s0 constant, in effect meaning there are three ways to spell out s0. If this is an accurate representation, we raise the interesting question of having many forms mapping to the same meaning. Since I take the the s0 constant to be a characteristic property of the INDEPENENT paradigm, rather than a property of a particular morpheme, the answer to the question is not crucial to the analysis. Alternatively, we could say that s0 is external to CP altogether (Lisa Matthewson, p.c.). Again, nothing hinges on this decision. 19 First, I lay out the evidence that both INDEPENDENT and CONJUNCT verbal complexes in Plains Cree are full CPs, and that the left-edge elements (pronominal proclitics and clause-typing proclitics introduced by initial change) in particular are hosted in CP. This is important because, in principle, both the pronominal proclitics and initial change could target any level of the clause, including VP, IP and CP, and this in fact seems to be a place where there is variation across the Algonquian family. For example, it has been argued that the pronominal proclitics in Blackfoot are hosted in the IP domain (Ritter & Wiltschko 2005, 2007), and similarly, the process of initial change in Ojibwa is associated with tense (James 1982, Blain 1999, Mathieu 2008). If the pronominal proclitics and clause-typing proclitics were in a lower position, we would not expect them to correlate with clause-typing distinctions. Second, I show that the pronominal proclitics ni- \u00E2\u0080\u00981st person\u00E2\u0080\u0099 and ki- \u00E2\u0080\u00982nd person\u00E2\u0080\u0099 differ from the clause-typing proclitics \u00C3\u00AA-, k\u00C3\u00A2- and IC in a number of respects. I model this difference as a difference in whether the element is introduced in spec, CP (i.e., for the pronominal proclitics) or head the CP-projection (i.e., the clause-typing elements). 2.2 Diagnostics for CPs in Plains Cree When we compare multiple elements in a class, we need a set of criteria which makes each element a member of that class. Here I am comparing members of the class of CP \u00E2\u0080\u0093 that is, things which are CPs (i.e., maximal clauses). Thus, in this section, I provide the set of criteria used in determining that Plains Cree verbal complexes are CPs. First, the ordering properties of the pronominal and clause-typing proclitics are consistent with having the highest position in the clause. Second, the distributional and interpretational properties of Plains Cree\u00E2\u0080\u0099s peripheral agreement are consistent with CPs, but not IPs or VPs (cf. D\u00C3\u00A9chaine 2001, 2002). Third, the sensitivity of verbal complexes to the matrix/embedded distinction is consistent with CPs, but not IPs or VPs. Finally, the complementary distribution of pronominal proclitics as opposed to clause-typing proclitics supports the claim that the pronominal proclitics are hosted by the same layer of the clause as the clause-typing proclitics. 20 Although no one of these criteria is conclusive evidence about the nature of Plains Cree verbal complexes, if they are taken together they present a coherent argument for the current analysis. 2.2.1 Ordering properties Aspect (temporal structure of the predicate), tense (temporal anchoring of the event), and modality ((ir)reality of the proposition) are taken to be in VP and IP domains of the clause (Pollock 1989, Hornstein 1990, Cinque 1999, Giv\u00C3\u00B3n 2001, Demirdache & Uribe-Etxebarria 2002, among many others). Cook (2003, 2004) has argued that the linear ordering of elements in Plains Cree\u00E2\u0080\u0099s preverb domain reflects hierarchical structure (cf. Baker 1986, Kayne 1994, Cinque 1999): if element \u00CE\u00B1 precedes element \u00CE\u00B2, then element \u00CE\u00B1 dominates element \u00CE\u00B2. This position leads us to expect that within Plains Cree\u00E2\u0080\u0099s verbal complexes, elements associated with the CP-layer of the clause must precede elements associated with the IP-layer of the clause. This is borne out in the data: the pronominal proclitics and clause-typing elements precede all tense, aspect, and modality preverbs. This is schematized in (6a) for INDEPENDENT clauses and (6b) for CONJUNCT clauses (cf. also Edwards 1954). (6) a. [ PERSON] [modality/temp.] [ aspect] [ STEM ] INDEPENDENT [ ni-, ki- ] [ka-, k\u00C3\u00AE-] [ati-, m\u00C3\u00AAkw\u00C3\u00A2-, w\u00C3\u00AE- ] [ \u00E2\u0080\u00A6 ] b. [ CLAUSE-TYPING ] [modality/temp.] [ aspect] [ STEM ] CONJUNCT [ \u00C3\u00AA-, k\u00C3\u00A2-, IC ] [ka-, k\u00C3\u00AE-] [ati-, m\u00C3\u00AAkw\u00C3\u00A2-, w\u00C3\u00AE- ] [ \u00E2\u0080\u00A6 ] A representative pair of data showing the position of the temporal shifting preverb k\u00C3\u00AE-4 relative to the pronominal proclitic ni- and clause-typing proclitic \u00C3\u00AA- is given in (7). 4 See chapters 3 and 7 for discussion of k\u00C3\u00AE- as a temporal shifting device. 21 (7) a. nik\u00C3\u00AE-w\u00C3\u00A2pam\u00C3\u00A2w ana n\u00C3\u00A2p\u00C3\u00AAw INDEPENDENT ni-k\u00C3\u00AE- w\u00C3\u00A2pam -\u00C3\u00A2 -w ana n\u00C3\u00A2p\u00C3\u00AAw 1-PREV-see.VTA-DIR-3 DEM.AN man \u00E2\u0080\u0098I saw that man.\u00E2\u0080\u0099 b. \u00C3\u00AAwakw \u00C3\u00A2nima p\u00C3\u00AAyak kis\u00C3\u00AAyiniw \u00C3\u00AA-k\u00C3\u00AE-nakiskawak,\u00E2\u0080\u00A6 CONJUNCT \u00C3\u00AAwakw anima p\u00C3\u00AAyak kis\u00C3\u00AAyiniw \u00C3\u00AA- k\u00C3\u00AE- nakiskaw -ak TOPIC DEM.INAN one old.man C1-PREV-meet.VTA -1>3 \u00E2\u0080\u0098I met a certain old man about that, ...\u00E2\u0080\u0099 (JKN 1.1) Likewise, (8) demonstrates that aspectual elements such as inceptive ati- always follow pronominal proclitics like ni- and clause-typing proclitics like \u00C3\u00AA-. (8) a. nitati-kinosin INDEPENDENT ni(t)- ati- kinosi -n 1- INCEP-tall.VAI-SAP \u00E2\u0080\u0098I am getting taller.\u00E2\u0080\u0099 b. ..., \u00C3\u00AAkos \u00C3\u00AAkwa, \u00C3\u00AA-ati-tipisk\u00C3\u00A2k \u00C3\u00AAkwa, ... CONJUNCT \u00C3\u00AAkosi \u00C3\u00AAkwa \u00C3\u00AA- ati- tipisk\u00C3\u00A2 -k \u00C3\u00AAkwa TOPIC and C1-INCEP-night.VII-0 and \u00E2\u0080\u0098And so, when it was getting to be night, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 2.2) The temporal anchoring preverb k\u00C3\u00AE- precedes the aspectual preverb ati-; this is consistent with k\u00C3\u00AE- occupying a higher position in the clause. (9) k\u00C3\u00AE- precedes ati- eight hours nik\u00C3\u00AE-ati-n\u00C3\u00B4ciht\u00C3\u00A2n\u00C3\u00A2n, ... eight hours ni- k\u00C3\u00AE- ati- n\u00C3\u00B4ciht\u00C3\u00A2 -n\u00C3\u00A2n eight hours 1- PREV-INCEP-pursue.VAI -1PL \u00E2\u0080\u0098It had taken us eight hours to go, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 3.4) The ordering of the preverbal elements with respect to the pronominal/clause-typing elements is consistent with the claim that the latter elements are in the highest position of all. Although the ordering does not tell us what that position is, the ordering is consistent with the claim I am making that the elements are in CP. 22 (10) CP 3IP [person] 3 VP [c-typing] k\u00C3\u00AE- 3 ati- 5 The irrealis preverb ka- interacts with the left edge in a more complex way: it is internal to the pronominal proclitics as in (11), and in complementary distribution with the CONJUNCT proclitics. (11) a. ..., \u00E2\u0080\u009C\u00C3\u00A2, \u00C3\u00AAkota nika-p\u00C3\u00B4sipayihon,\u00E2\u0080\u009D ... INDEPENDENT \u00C3\u00A2 \u00C3\u00AAkota ni- ka- p\u00C3\u00B4sipayiho -n INTERJ there 1- IRR-jump.VAI -SAP \u00E2\u0080\u0098..., \u00E2\u0080\u009CWell, I will jump on that,\u00E2\u0080\u009D \u00E2\u0080\u00A6 \u00E2\u0080\u0099 (AA 8.3) b. * \u00C3\u00AA-ka-p\u00C3\u00B4sipayihoy\u00C3\u00A2n CONJUNCT \u00C3\u00AA- ka- p\u00C3\u00B4sipayiho -y\u00C3\u00A2n C1-IRR- jump.VAI -1 --- (intended: \u00E2\u0080\u0098I will jump.\u00E2\u0080\u0099) However, unlike either the pronominal proclitics or the clause-typing proclitics, ka- can occur in both matrix INDEPENDENT and embedded CONJUNCT clauses. (12) Irrealis ka- across different clause-types a. ..., \u00E2\u0080\u009C\u00C3\u00A2, \u00C3\u00AAkota nika-p\u00C3\u00B4sipayihon,\u00E2\u0080\u009D ... INDEPENDENT \u00C3\u00A2 \u00C3\u00AAkota ni- ka- p\u00C3\u00B4sipayiho -n INTERJ there 1- IRR-jump.VAI -SAP \u00E2\u0080\u0098..., \u00E2\u0080\u009CWell, I will jump on that,\u00E2\u0080\u009D \u00E2\u0080\u00A6 \u00E2\u0080\u0099 (AA 8.3) b. nik\u00C3\u00AE-kw\u00C3\u00AAcim\u00C3\u00A2w Nettie ka-p\u00C3\u00AA-itoht\u00C3\u00AAt CONJUNCT ni- k\u00C3\u00AE- kw\u00C3\u00AAcim -\u00C3\u00A2 -w N ka- p\u00C3\u00AA- itoht\u00C3\u00AA -t 1- PREV-ask.VTA -DIR-3 N IRR-come-go.VAI-3 \u00E2\u0080\u0098I asked Nettie to come.\u00E2\u0080\u0099 The data in (12) shows that ka- crosscuts both the matrix/embedded distinction and Plains Cree\u00E2\u0080\u0099s INDEPENDENT/CONJUNCT distinction. In addition, ka- precedes k\u00C3\u00AE-, as in (13). (13) ..., nika-k\u00C3\u00AE-itw\u00C3\u00A2n \u00C3\u00AAwakw anima, ... ni- ka- k\u00C3\u00AE- itw\u00C3\u00A2 -n \u00C3\u00AAwakw anima 1- IRR-PREV-say.VAI-SAP TOPIC DEM.INAN \u00E2\u0080\u0098..., I can say that, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 2.1) 23 I take the distribution of ka- to be significant. In particular, even though ka- interacts with clause-typing elements, it is crucially not restricted to one of the clause-types (i.e., INDEPENDENT or CONJUNCT). Therefore, it must be sitting in a lower position than both the pronominal proclitics (with which it co-occurs) and the clause-typing proclitics (which are restricted to CONJUNCT order). Taking its ordering with respect to k\u00C3\u00AE-, its complementarity with the clause-typing elements \u00C3\u00AA- and ka-, and its irrealis meaning into account, I model ka- as a finiteness complementizer (Cfin; Rizzi 1997). This position is distinct from and lower than force complementizers (Cforce).5 (14) CPforce 3CPfin [person] 3 IP [c-typing] ka- 3 VP k\u00C3\u00AE- 3 ati- 5 The co-occurrence restrictions between ka- and the clause-typing elements can be seen as an instance of local head-to-head interaction (e.g., only one complementizer may be overt at a time). Summarizing, we see that the position of the pronominal proclitics and clause-typing elements is consistent with them being in the CP-layer of the clause. I now look at some evidence that they are in fact in this layer of the clause. 2.2.2 Peripheral agreement diagnoses CPs Algonquian languages are famous for the abundance of agreement they exhibit. Consider the INDEPENDENT clause in (15). From left to right, we see a pronominal proclitic ni-, the root w\u00C3\u00A2p, a valency marker -am that is codes the animacy of the internal argument, a valency marker -\u00C3\u00A2, a third person suffix -w, and a plural marker -ak. 5 I have not given evidence that ka- must be a finiteness complementizer, and in fact nothing in the following argumentation depends on it being such. 24 (15) niw\u00C3\u00A2pam\u00C3\u00A2wak INDEPENDENT ni- [w\u00C3\u00A2p- am] -\u00C3\u00A2 -w-ak 1- [light- eye.AN]see.VTA -DIR -3 -PL \u00E2\u0080\u0098I see them.\u00E2\u0080\u0099 The CONJUNCT clause in (16) has a left-edge clause-typing element \u00C3\u00AA-; it shows some of the same agreement on the right edge, including identical valency markers, but it has person (-t \u00E2\u0080\u00983rd') and number (-ik \u00E2\u0080\u0098pl\u00E2\u0080\u0099) agreement that is different from the INDEPENDENT clause. (16) \u00C3\u00AA-w\u00C3\u00A2pam\u00C3\u00A2cik CONJUNCT \u00C3\u00AA- [w\u00C3\u00A2p -am] -\u00C3\u00A2 -t -k C1- [light -eye.AN]see.VTA -DIR -3 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6I see them.\u00E2\u0080\u0099 The amount and kinds of agreement that these clauses exhibit provide evidence that these verbal complexes are structurally quite big. If a clause is composed of multiple domains (cf. Pollock 1989, Rizzi 1997, Cinque 1999, among others), then agreement may logically occur in any of these domains. For the purposes of this discussion, I will assume that a clause has at least the layers of VP (i.e., the predicate domain where theta-marking agreement occurs); IP (the inflectional domain where grammatical function agreement occurs); and CP (the clause-typing domain where discourse function agreement occurs). (17) CP \u00EF\u0083\u009F discourse function agreement 5 IP \u00EF\u0083\u009F grammatical function agreement 5 VP \u00EF\u0083\u009F theta-marking agreement 5 Agreement that remains constant across distinct clause-typing environments (e.g., matrix vs. embedded clauses or declarative vs. interrogative clauses) is not a good candidate for CP- agreement, and I will not discuss it here (see D\u00C3\u00A9chaine 2001, 2002 for discussion). However, for several reasons, the peripheral person and number agreement which alternates in Plains Cree\u00E2\u0080\u0099s INDEPENDENT and CONJUNCT order provide evidence that the verbal complexes under analysis are full CPs, rather than a reduced constituent (e.g., an IP or a VP): 25 (i) the agreement does not map directly onto either theta-roles or grammatical-function; (ii) the agreement does not interact with finiteness; and (iii) the agreement correlates with clausal embeddedness. First, person and number agreement are insensitive to grammatical function (i.e., subject vs. object vs. indirect object). In (18), the third person -w suffix of the INDEPENDENT order can act as a subject (18a), an object (18b), or an indirect object (18c). (18) a. m\u00C3\u00A2tow m\u00C3\u00A2to -w cry.VAI-3 \u00E2\u0080\u0098s/he is crying/cries\u00E2\u0080\u0099 b. ninakiskaw\u00C3\u00A2w ni- nakiskaw -\u00C3\u00A2 -w 1- meet.VTA -DIR-3 \u00E2\u0080\u0098I met him/her.\u00E2\u0080\u0099 c. niwihtamaw\u00C3\u00A2w ni- wihtam -aw -\u00C3\u00A2 -w 1- tell.VTA-BEN-DIR-3 \u00E2\u0080\u0098I tell this to him/her.\u00E2\u0080\u0099 Similarly, in (19) the third person -t suffix of the CONJUNCT order can mark subjects (19a), objects (19b), or indirect objects (19c). (19) a. \u00C3\u00AA-m\u00C3\u00A2tot \u00C3\u00AA- m\u00C3\u00A2to -t C1-cry.VAI-3 \u00E2\u0080\u0098s/he is crying\u00E2\u0080\u0099 b. \u00C3\u00AA-m\u00C3\u00A2miton\u00C3\u00AAyihtamikot \u00C3\u00AA- m\u00C3\u00A2miton\u00C3\u00AAyihtam -iko -t C1-trouble.VTA -INV -3 \u00E2\u0080\u0098it (inan.) troubles him/her\u00E2\u0080\u0099 c. \u00C3\u00AA-it\u00C3\u00B4t\u00C3\u00A2kot \u00C3\u00AA-it\u00C3\u00B4t\u00C3\u00AA-aw-iko-t C1-do.VTA-BEN-INV-3 \u00E2\u0080\u0098it did this to him/her\u00E2\u0080\u0099 26 The plural markers which occur on the far right edge of the verbal complex may also be associated with either the subject or object position6. For example, -ak in the INDEPENDENT order may be associated with a subject (20a) or an object (20b). (20) a. w\u00C3\u00A2pam\u00C3\u00AAwak SUBJECT w\u00C3\u00A2pam -\u00C3\u00AA -w -ak see.VTA-DIR-3 -PL = \u00E2\u0080\u0098They see him/her.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098S/he sees them.\u00E2\u0080\u0099 b. niw\u00C3\u00A2pam\u00C3\u00A2wak OBJECT ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w -ak 1- see.VTA-DIR-3 -PL \u00E2\u0089\u00A0 \u00E2\u0080\u0098We see him/her.\u00E2\u0080\u0099 = \u00E2\u0080\u0098I see them.\u00E2\u0080\u0099 The -k plural suffix of the CONJUNCT order may also be associated with a subject (21a) or an object (21b). (21) a. \u00C3\u00AA-w\u00C3\u00A2pam\u00C3\u00A2cik SUBJECT \u00C3\u00AA- w\u00C3\u00A2pam -\u00C3\u00A2 -t -k C1-see.VTA-DIR-3 -PL = \u00E2\u0080\u0098 \u00E2\u0080\u00A6 they see him/her.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098\u00E2\u0080\u00A6s/he sees them.\u00E2\u0080\u0099 b. \u00C3\u00AA-w\u00C3\u00A2pamakik OBJECT \u00C3\u00AA- w\u00C3\u00A2pam -ak -k C1-see.VTA-1>3-PL \u00E2\u0089\u00A0 \u00E2\u0080\u0098\u00E2\u0080\u00A6we see him/her.\u00E2\u0080\u0099 = \u00E2\u0080\u0098\u00E2\u0080\u00A6I see them.\u00E2\u0080\u0099 Finally, the INDEPENDENT order pronominal proclitics ni- \u00E2\u0080\u00981st\u00E2\u0080\u0099 and ki- \u00E2\u0080\u00982nd\u00E2\u0080\u0099 may be associated with either a subject or an object argument. In (22), ni- and ki- are associated with the subject of a classically unergative predicate p\u00C3\u00A2hpi- \u00E2\u0080\u0098laugh.VAI\u00E2\u0080\u0099. 6 The determination of which argument the plural is associated with is complex and depends on multiple factors. In general, they are associated with an object if a speech act participant is the subject, and associated with subject elsewhere (including cases where 3PL acts on 3PL). The point here is that it does not mark only one type of argument. 27 (22) a. nip\u00C3\u00A2hpin ni- p\u00C3\u00A2hpi -n 1- laugh.VAI-SAP \u00E2\u0080\u0098I laugh.\u00E2\u0080\u0099 b. kip\u00C3\u00A2hpin7 ki- p\u00C3\u00A2hpi -n 2- laugh.VAI-SAP \u00E2\u0080\u0098You laugh.\u00E2\u0080\u0099 In (23), ni- and ki- are associated with the subject of a classically unaccusative predicate n\u00C3\u00AAstosi- \u00E2\u0080\u0098tired.VAI\u00E2\u0080\u0099. (23) a. nin\u00C3\u00AAstosin ni- n\u00C3\u00AAstosi -n 1- tired.VAI-SAP \u00E2\u0080\u0098I am tired.\u00E2\u0080\u0099 b. kin\u00C3\u00AAstosin ki- n\u00C3\u00AAstosi -n 2- tired.VAI-SAP \u00E2\u0080\u0098You are tired.\u00E2\u0080\u0099 In (24), ni- and ki- are associated with the subject of a transitive verb miskam- \u00E2\u0080\u0098find.VTA\u00E2\u0080\u0099; notice that the theme-sign immediately following the stem is the direct form -\u00C3\u00A2. (24) a. nimiskam\u00C3\u00A2w ni- miskam -\u00C3\u00A2 -w 1- find.VTA-DIR-3 \u00E2\u0080\u0098I found him/her.\u00E2\u0080\u0099 b. kimiskam\u00C3\u00A2w ki- miskam -\u00C3\u00A2 -w 2- find.VTA -DIR-3 \u00E2\u0080\u0098You found him/her.\u00E2\u0080\u0099 In (25), however, ni- and ki- are associated with the object of the transitive verb; here the subject/object association has been reversed by the use of the inverse marker (Dahlstrom 1991, D\u00C3\u00A9chaine and Reinholtz 1997, 2008). 7 Note that the second-person forms are pragmatically very odd, since the contexts where a statement about the addressee can be made felicitously are extremely restricted (cf. Ross 1970; Rutherford 1970 for discussion of this problem in English). These forms are given to demonstrate their formal grammaticality. 28 (25) a. nimiskamik ni- miskam -ik -w 1- find.VTA-INV-3 \u00E2\u0080\u0098S/he found me.\u00E2\u0080\u0099 b. kimiskamik ki- miskam -ik -w 2- find.VTA-INV-3 \u00E2\u0080\u0098S/he found you.\u00E2\u0080\u0099 The above data shows that the peripheral elements in Plains Cree verbal complexes is independent of the subject or object position. Plains Cree does have elements which are associated with subjects (e.g., the different subject marker -yi (M\u00C3\u00BChlbauer 2007) and the inverse marker -ik (D\u00C3\u00A9chaine & Reinholtz 1997, 2008)), and elements associated with objects (direct theme signs, D\u00C3\u00A9chaine & Reinholtz 2008). The insensitivity of peripheral person marking to these positions must therefore mean that it is not in the IP domain (i.e., subject domain) or the VP-domain (i.e., object domain). Combining this evidence with the ordering facts discussed above, I conclude that these peripheral elements must be external to IP. Another fact about the peripheral person marking in Plains Cree which suggests it is external to the IP-domain is that there is no correlation between agreement and finiteness of clauses. The irrealis clauses which are used in dependent clauses for commands, wishes, etc. (i.e., the contexts where English has non-finite clauses) show exactly the same kind of agreement as other kinds of embedded clauses. The (non-)contrast is shown in (26), where (26a) shows a realis clause, introduced by a factive predicate and (26b) shows an irrealis clause (with the irrealis ka-) introduced by a predicate of desire (ninitaw\u00C3\u00AAyim\u00C3\u00A2w). The right-edge agreement (3rd-person -t) stays constant. (26) a. \u00C3\u00AA-wankisiy\u00C3\u00A2n Clare \u00C3\u00AA-nip\u00C3\u00A2t FINITE \u00C3\u00AA- wankisi -y\u00C3\u00A2n C \u00C3\u00AA- nip\u00C3\u00A2 -t C1-forget.VAI-1 C C1-sleep.VAI-3 \u00E2\u0080\u0098I forgot Clare was sleeping.\u00E2\u0080\u0099 b. ninitaw\u00C3\u00AAyim\u00C3\u00A2w Jeff ka-nip\u00C3\u00A2t NON-FINITE ni- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -w J ka- nip\u00C3\u00A2 -t 1- want.VTA -DIR-3 J IRR-sleep.VAI-3 \u00E2\u0080\u0098I want Jeff to sleep.\u00E2\u0080\u0099 This is different from some other Algonquian languages, such as Blackfoot, where person 29 marking changes in some irrealis conditions (Frantz 1991, Ritter & Wiltschko 2007). For example, in (27a) the 2nd-person pronominal proclitic ki- occurs on the left edge of the embedded finite clause; in (27b), there is no pronominal proclitic, and the agreement is on the right edge of the embedded non-finite clause. This is one factor that has led Ritter & Wiltschko to posit that person marking is hosted in the IP domain. (27) Blackfoot person marking changes relative to finiteness. a. nits\u00C3\u00ADkohtaahs\u00C3\u00AD\u00E2\u0080\u0099taki kik\u00C3\u00A1\u00C3\u00B3\u00E2\u0080\u0099toohsi FINITE nit- ik- oht- yaahs -i\u00E2\u0080\u0099taki k- ik\u00C3\u00A1\u00C3\u00A1- o\u00E2\u0080\u0099too -hs -yi 1- very-source-good -feel.VAI 2- PERF- arrive.VAI-conj-conj \u00E2\u0080\u0098I\u00E2\u0080\u0099m glad that you have arrived.\u00E2\u0080\u0099 (Frantz 1991:112) b. ikkam\u00C3\u00A1yo\u00E2\u0080\u0099kainoainiki, nit\u00C3\u00A1akahkayi NON-FINITE ikkam -\u00C3\u00A1 -yo\u00E2\u0080\u0099kaa -inoainiki nit- y\u00C3\u00A1ak- wa:hkayi if -dur-sleep.VAI -2PL.SUBJ 1- FUT- go^home \u00E2\u0080\u0098If you (pl) are sleeping, I\u00E2\u0080\u0099ll go home.\u00E2\u0080\u0099 (Frantz 1991:113) The final distinguishing characteristic of Plains Cree\u00E2\u0080\u0099s peripheral agreement which is consistent with placing it in the CP layer is its sensitivity to the matrix/embedded distinction. A verbal complex with the INDEPENDENT agreement -w in (28a) cannot occur in an embedded clause (notice here that the INDEPENDENT clause has no pronominal proclitic, but still is restricted to matrix clauses). (28) Only CONJUNCT agreement allowed in embedded clauses a. * ninitaw\u00C3\u00AAyim\u00C3\u00A2w n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan (ka)-m\u00C3\u00AEcisow ni- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -w ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan m\u00C3\u00AEciso-w 1- want.vta -DIR-3 1- friend eat.VAI-3 --- (intended: \u00E2\u0080\u0098I want my friend to eat.\u00E2\u0080\u0099) b. ninitaw\u00C3\u00AAyim\u00C3\u00A2w n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan ka-m\u00C3\u00AEcisot ni- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -w ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan ka- m\u00C3\u00AEciso -t 1- want.VTA -DIR-3 1- friend IRR-eat.VAI-3 \u00E2\u0080\u0098I want my friend to eat.\u00E2\u0080\u0099 It is also impossible for the pronominal proclitics (ni- \u00E2\u0080\u00981st-person\u00E2\u0080\u0099 and ki- \u00E2\u0080\u00982nd-person\u00E2\u0080\u0099) to occur in an embedded clause, regardless of the right-edge agreement. In (29), the first-person ni- cannot occur in an embedded clause with right-edge INDEPENDENT agreement (29a), nor can it 30 occur in an embedded clause with right-edge CONJUNCT agreement (29b). The embedded clause must have CONJUUNCT agreement without the pronominal proclitic (29c). (29) Person proclitics are impossible in embedded clauses a. * Sam nitaw\u00C3\u00AAyihtam ni-(ka)-nikamon INDEPENDENT S nitaw\u00C3\u00AAyihtam -w ni-ka- nikamo -n S want.VTI -3 1-IRR-sing.VAI -SAP --- b. * Sam nitaw\u00C3\u00AAyihtam ni-(ka)-nikamoy\u00C3\u00A2n CONJUNCT S nitaw\u00C3\u00AAyihtam -w ni-ka- nikamo -y\u00C3\u00A2n S want.VTI -3 1-IRR-sing.VAI -1 --- This again differs from Blackfoot, where the pronominal proclitics occur in both matrix (30a) and embedded (30b) clauses. (30) a. kit\u00C3\u00A1akahkayi MATRIX kit- \u00C3\u00A1ak- ahkayi 2- FUT- go^home \u00E2\u0080\u0098You\u00E2\u0080\u0099re going home.\u00E2\u0080\u0099 (Frantz 1991:15) b. nits\u00C3\u00ADkohtaahs\u00C3\u00AD\u00E2\u0080\u0099taki kik\u00C3\u00A1\u00C3\u00B3\u00E2\u0080\u0099toohsi EMBEDDED nit- ik- oht- yaahs -i\u00E2\u0080\u0099taki k- ik\u00C3\u00A1\u00C3\u00A1- o\u00E2\u0080\u0099too -hs -yi 1- very-source-good -feel.VAI 2- PERF- arrive.VAI-conj-conj \u00E2\u0080\u0098I\u00E2\u0080\u0099m glad that you have arrived.\u00E2\u0080\u0099 (Frantz 1991:112) In summary, then, the peripheral agreement in Plains Cree is associated with the CP domain: it is sensitive to the matrix/embedded distinction, a characteristic property of CP-level elements; and it is insensitive characteristic properties of IP-level elements such as the subject/object distinction and finiteness. In both of these respects, there is variation across the Algonquian language family, meaning that it is important to establish the position of agreement for each individual language. 31 2.2.3 Clause-typing diagnoses CPs If the CP-layer associates the proposition to a larger structure (Rizzi 1997), this implies that elements whose presence conditions the distribution and interpretation of a clause relative to its larger context must be in the CP-layer of the clause. If the element can only occur in matrix clauses, or only in embedded clauses, that element invokes a CP structure. For example, the complementizer that in English is taken to be in C: it specifies the clause as a complement clause and can only be found in embedded structures (Rosenbaum 1967, Emonds 1976). (31) a. * that I\u00E2\u0080\u0099m tired. [CP [C that [IP I\u00E2\u0080\u0099m tired ] ] ] b. I told my brother that I\u00E2\u0080\u0099m tired. I told my brother [CP [C that [IP I\u00E2\u0080\u0099m tired ] ] ] In Plains Cree, the form of the verbal complex is dependent on the matrix/embedded distinction. As we have already seen, INDEPENDENT clauses cannot be embedded. (32) INDEPENDENT clauses are sensitive to the matrix/embedded distinction a. nin\u00C3\u00AAstosin MATRIX ni-n\u00C3\u00AAstosi-n 1-tired.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m tired.\u00E2\u0080\u0099 b. * nik\u00C3\u00AE-w\u00C3\u00AEhtamaw\u00C3\u00A2w nis\u00C3\u00AEmis nin\u00C3\u00AAstosin EMBEDDED ni- k\u00C3\u00AE- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w ni- s\u00C3\u00AEmis ni-n\u00C3\u00AAstosi -n 1- PREV-tell.VTA -DIR-3 1- sibling 1- tired.VAI-SAP --- (intended: \u00E2\u0080\u0098I told my younger brother that I\u00E2\u0080\u0099m tired.\u00E2\u0080\u0099) One could argue that some overt complementizer or subordinator is needed to subordinate INDEPENDENT clauses (e.g., if one tried to posit matrix clauses as IPs). However, adding an overt subordinating particle like os\u00C3\u00A2m \u00E2\u0080\u0098because\u00E2\u0080\u0099 does not help, as (33) shows. 32 (33) os\u00C3\u00A2m \u00E2\u0080\u0098because\u00E2\u0080\u0099 can only introduce INDEPENDENT clauses a. * nawac \u00C3\u00AA-k\u00C3\u00AE-cihk\u00C3\u00AAyihtahkik, os\u00C3\u00A2m k\u00C3\u00AE-s\u00C3\u00A2kihitowak INDEPENDENT nawac \u00C3\u00AA- k\u00C3\u00AE- miyaw\u00C3\u00A2tam -k -k, os\u00C3\u00A2m k\u00C3\u00AE- s\u00C3\u00A2kih -ito -w -ak more C1-PREV-happy.VTI -0 -PL because PREV-love.VTA-REFL-3 -PL --- b. ..., nawac \u00C3\u00AA-k\u00C3\u00AE-miyaw\u00C3\u00A2tahkik, os\u00C3\u00A2m \u00C3\u00AA-k\u00C3\u00AE-s\u00C3\u00A2kihitocik, \u00E2\u0080\u00A6 CONJUNCT nawac \u00C3\u00AA- k\u00C3\u00AE- miyaw\u00C3\u00A2tah -k -k os\u00C3\u00A2m \u00C3\u00AA- k\u00C3\u00AE- s\u00C3\u00A2kih -ito -t -k more C1-PREV-happy.VTI -0 -PL because C1-PREV-love.VTA-REFL-3-PL \u00E2\u0080\u0098..., they had been happier even when they were poor, because they used to love one another, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 6) In CONJUNCT clauses, the presence and particular form of clause-typing element determines the distribution of the verbal complex as a whole with respect to matrix and embedded environments. The element k\u00C3\u00A2- and internal change both restrict the verbal complex to embedded clauses. (34) k\u00C3\u00A2-clauses must be embedded a. * atim k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-nip\u00C3\u00A2t MATRIX atim k\u00C3\u00A2- m\u00C3\u00AAkw\u00C3\u00A2- nip\u00C3\u00A2 -t dog C2- MIDST- sleep.VAI-3 --- b. nik\u00C3\u00AE-atosk\u00C3\u00A2n atim k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-nip\u00C3\u00A2t EMBEDDED ni-k\u00C3\u00AE- atosk\u00C3\u00A2 -n atim k\u00C3\u00A2- m\u00C3\u00AAkw\u00C3\u00A2- nip\u00C3\u00A2 -t 1-PREV-work.VAI-SAP dog C2- MIDST- sleep.VAI-3 \u00E2\u0080\u0098I worked while the dog slept.\u00E2\u0080\u0099 (35) IC-clauses must be embedded a. * kiy\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2ci p\u00C3\u00AAyak w\u00C3\u00A2skahikan MATRIX iy-k\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2 -t -i p\u00C3\u00AAyak w\u00C3\u00A2skahikan IC-finish.VAI-3-SUBJ one house --- b. ..., kiy\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2ci p\u00C3\u00AAyak w\u00C3\u00A2skahikan, kotakihk \u00C3\u00AA-itoht\u00C3\u00AAt; EMBEDDED iy-k\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2 -t -i p\u00C3\u00AAyak w\u00C3\u00A2skahikan kotak -ihk \u00C3\u00AA- itoht\u00C3\u00AA -t IC-finish.VAI-3-SUBJ one house other -LOC C1-go.VAI -3 \u00E2\u0080\u0098..., and when he had finished one house, he went to the next;\u00E2\u0080\u0099 (AA 1.9) 33 Substituting the clause-typing proclitic \u00C3\u00AA- for k\u00C3\u00A2- or internal change correlates with a change in the distribution of the CONJUNCT verbal complex; with \u00C3\u00AA-, the verbal complex may occur in both matrix and embedded environments. (36) \u00C3\u00AA-clauses allow both matrix and embedded clauses a. atim \u00C3\u00AA-nip\u00C3\u00A2t MATRIX atim \u00C3\u00AA- nip\u00C3\u00A2 -t dog C1- sleep.VAI -3 \u00E2\u0080\u0098...the dog is sleeping.\u00E2\u0080\u0099 b. nik\u00C3\u00AE-w\u00C3\u00A2paht\u00C3\u00AAn atim \u00C3\u00AA-nip\u00C3\u00A2t EMBEDDED ni- w\u00C3\u00A2paht\u00C3\u00AA-n atim \u00C3\u00AA- nip\u00C3\u00A2 -t 1- see.VTI -SAP dog C1- sleep.VAI -3 \u00E2\u0080\u0098I saw that the dog was sleeping.\u00E2\u0080\u0099 Finally, if the left-edge process of initial change is absent (so-called simple CONJUNCT), the verbal complex is also restricted to embedded clauses. A simple CONJUNCT clause can only be introduced by higher predicates, as in (37). (37) Simple CONJUNCT clauses must be embedded a. * (ka-)m\u00C3\u00AEcisot MATRIX ka- m\u00C3\u00AEciso -t IRR-eat.VAI-3 --- b. ninitaw\u00C3\u00AAyim\u00C3\u00A2w n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan ka-m\u00C3\u00AEcsot EMBEDDED ni- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -w ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan ka- m\u00C3\u00AEciso -t 1- want.VTA -DIR-3 1- friend IRR-eat.VAI -3 \u00E2\u0080\u0098I want my friend to eat.\u00E2\u0080\u0099 (38) Subjunctive CONJUNCT clauses must be embedded a. * w\u00C3\u00A2pamaki Jeff MATRIX w\u00C3\u00A2pam -ak -i J see.VTA-1>3-SUBJ J --- b. w\u00C3\u00A2pamaki Jeff, nika-w\u00C3\u00AEhtamaw\u00C3\u00A2w k\u00C3\u00A2-itw\u00C3\u00AAyan EMBEDDED w\u00C3\u00A2pam -ak -i J ni- ka- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w k\u00C3\u00A2- itw\u00C3\u00AA -yan see.VTA-1>3-SUBJ J 1- IRR-tell.VTA -DIR-3 C2- thus.say.VAI-2 \u00E2\u0080\u0098Should I see Jeff, I\u00E2\u0080\u0099ll tell him what you said.\u00E2\u0080\u0099 34 In CONJUNCT clauses, then, it is the left-edge element that determines the distribution of the verbal complex. The table summarizing the distribution is given below. Matrix Embedded k\u00C3\u00A2- \u00E2\u009C\u0096 \u00E2\u009C\u0094 IC- \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00C3\u00AA- \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00EF\u0081\u00B8 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 2.1. Left-edge clause-typing proclitic determines distribution of clause Since it is the choice of clause-typing element that correlates with distribution in matrix vs. embedded environments, these are the elements that look much like that in English \u00E2\u0080\u0093 their distributional effect is consistent with putting them in C. (39) [CP [C \u00C3\u00AA- / k\u00C3\u00A2- / IC [IP \u00E2\u0080\u00A6 ] ] ] 2.2.4 Pronominal proclitics are complementary with clause-typing In the last section we saw that the left-edge proclitics in the CONJUNCT order have a clause-typing function. If we compare these proclitics to the pronominal proclitics, we see additional evidence that pronominal proclitics are in CP: they are in complementary distribution with the clause- typing elements (including \u00C3\u00AA-, k\u00C3\u00A2-, and internal change) (cf. Wolfart 1973, Blain 1997). This is illustrated in (16-17), with the first-person marker ni- and the clause-typing element \u00C3\u00AA-. We have already seen that ni- and \u00C3\u00AA- can both occur in matrix clauses, However, as (40) shows, ni- and \u00C3\u00AA- cannot co-occur with right-edge INDEPENDENT order agreement -n. 35 (40) a. * ni(t)-\u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2ton nit- \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00A2to -n 1- C1-PREV-cry.VAI-SAP --- (intended: \u00E2\u0080\u0098I cried.\u00E2\u0080\u0099) b. * \u00C3\u00AA-ni-k\u00C3\u00AE-m\u00C3\u00A2ton \u00C3\u00AA- ni- k\u00C3\u00AE- m\u00C3\u00A2to -n C1-1- PREV-cry.VAI-SAP --- (intended: \u00E2\u0080\u0098I cried.\u00E2\u0080\u0099) And as (41) shows, ni- and \u00C3\u00AA- cannot co-occur with right-edge CONJUNCT order agreement -y\u00C3\u00A2n. (41) a. * ni(t)-\u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2toy\u00C3\u00A2n nit-\u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2toy\u00C3\u00A2n nit-\u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2to-y\u00C3\u00A2n --- (intended: \u00E2\u0080\u0098\u00E2\u0080\u00A6I was crying\u00E2\u0080\u0099) b. * \u00C3\u00AA-ni-k\u00C3\u00AE-m\u00C3\u00A2toy\u00C3\u00A2n \u00C3\u00AA-ni-k\u00C3\u00AE-m\u00C3\u00A2to-y\u00C3\u00A2n C1-1-PREV-cry.VAI-1 --- (intended: \u00E2\u0080\u0098\u00E2\u0080\u00A6I was crying\u00E2\u0080\u0099) There is no semantic reason why a first-person marker should be incompatible with a complementizer, since many languages permit this (e.g., English: John told me that I was going to win.). Thus, we can conclude that the incompatibility of the person prefixes and complementizers is a syntactic problem: their complementarity arises from the fact that they are both within the same layer of the clause (i.e., the CP). 2.2.5 Interim summary: Verbal complexes are CPs I have presented a four-part argument that verbal complexes in Plains Cree are CPs, with the pronominal proclitics and clause-typing proclitics specifically hosted in the CP-layer of the clause: (i) pronominal proclitics and clause-typing proclitics precede all tense/aspect and modality preverbs; (ii) peripheral agreement has does not have any properties associated with IP (e.g., subjecthood, sensitivity to finiteness); 36 (iii) the distribution of the clause in matrix vs. embedded contexts is determined by the presence of pronominal proclitics and/or choice of clause-typing proclitics; (iv) pronominal proclitics and clause-typing proclitics are in complementary distribution. In the next section I distinguish between the different verbal complexes, showing that INDEPENDENT clauses have an overtly filled specifier position (spec, CP), that CHANGED CONJUNCT clauses have an overt complementizer (i.e., C), and that SIMPLE CONJUNCT clauses have a covert complementizer (C \u00EF\u0081\u00B8 ). 2.3 Diagnosing C vs. spec, CP In this section, I give evidence to support the claim that the pronominal forms (ni- \u00E2\u0080\u00981st\u00E2\u0080\u0099 and ki- \u00E2\u0080\u00982nd\u00E2\u0080\u0099) are in spec, CP while the clause-typing proclitics are in C. The diagnostics I use to determine whether pronominal proclitics are heads or not are: (a) whether they select for a complement; (b) whether substitution of elements within the same class changes the distribution of the clause; and (c) the (non-)significance of a covert element. In each case, the pronominal proclitics diverge from the clause-typing proclitics; the former are consistent with specifiers, the latter with heads. DIAGNOSTIC PRONOMINAL PROCLITICS (=SPEC) CLAUSE-TYPING PROCLITICS (=HEAD) select for complement? \u00E2\u009C\u0096 \u00E2\u009C\u0094 substitution determines distribution? \u00E2\u009C\u0096 \u00E2\u009C\u0094 covertness significant? \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 2.2. Diagnostics for specifiers vs. heads I discuss each of these diagnostics in turn in the following subsections. 37 2.3.1 Selection of complement Turning our attention first to the pronominal proclitics, we see that they are not specific to clauses. Rather, they are analogous to the paradigm for possessors in nominals (Wolfart 1973, Ahenakew 1987, Dahlstrom 1991), and in fact their use in the verbal is argued to be a historical extension from the nominal domain (Goddard 1967). In (42a), we see the pronominal prefix attaching to a verbal stem and associated with one of the arguments of the predicate; in (42b), this same prefix attaches to a nominal stem and marks the possessor: (42) a. nicihk\u00C3\u00AAyiht\u00C3\u00AAn ni- cihk\u00C3\u00AAyiht\u00C3\u00AA -n 1- happy.VTI -SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m happy.\u00E2\u0080\u0099 b. nimaskisin ni- maskisin 1- shoe \u00E2\u0080\u0098my shoe\u00E2\u0080\u0099 The full paradigms are given in Table 2.3. For the first and second person, both nominals and clauses have a left edge ni- or ki-; and share the same right-edge plural marking, including 1pl. excl. -n\u00C3\u00A2n; 1-2.pl -naw; and 2.pl -w\u00C3\u00A2w (note that the verbal version of the latter has an extended form \u00E2\u0080\u0093n\u00C3\u00A2w\u00C3\u00A2w). In the third person, the verbal complex lacks the prefix o-, (a fact I will return to in \u00C2\u00A73.2.3.2) but again the verbal and nominal paradigms have the same right-edge marking in the 3.obv form (-yiwa). PERSON CATEGORY INDEPENDENT ORDER NOMINAL POSSESSION 1.sg. ninip\u00C3\u00A2n nimis 2.sg. kinip\u00C3\u00A2n kimis 1.pl. excl. ninip\u00C3\u00A2n\u00C3\u00A2n nimisin\u00C3\u00A2n 1.pl. incl. kinip\u00C3\u00A2naw kimisnaw 2.pl kinip\u00C3\u00A2n\u00C3\u00A2w\u00C3\u00A2w kimisiw\u00C3\u00A2w 3.sg. nip\u00C3\u00A2w omisa 3.pl nipawak omisiw\u00C3\u00A2wa 3.obv nip\u00C3\u00A2yiwa omisiyiwa Table 2.3. Person marking in INDEPENDENT and possession paradigms 38 Pronominal proclitics do not select for a particular kind of XP: they are neutral with respect to the distinction between DPs and CPs. (43) a. CP b. DP 2 2 ni- 2 ni- 2 ki- C 4 ki- D 4 Syntactically, this is a classic difference between heads and specifiers: heads must select for a complement, but a specifier does not. In the case of complementizer C heads, the head selects for a clausal constituent. However ni- and ki- are not selecting for a clausal constituent \u00E2\u0080\u0093 the presence of ni- or ki- does not identify the constituent it attaches to as a clause. This is evidence that they are hosted in spec, CP, rather than C. While pronominal proclitics are found in both verbal and nominal contexts, the clause- typing proclitics occur only in verbal contexts. For example, in (44), the proclitic \u00C3\u00AA- may not attach to a nominal stem, but it may attach to a verbal stem (identifiable by the -t agreement). (44) a. * \u00C3\u00AA-min\u00C3\u00B4s \u00C3\u00AA- min\u00C3\u00B4s C1-cat -- b. \u00C3\u00AA-min\u00C3\u00B4sit \u00C3\u00AA- min\u00C3\u00B4s \u00E2\u0080\u0093i -t C1-cat -EPEN-3 \u00E2\u0080\u0098S/he is a cat.\u00E2\u0080\u0099 Thus, the clause-typing proclitics do identify a clause. Notice that categorial properties of the verbal stem are identifiable by the right-edge agreement, and recall that we saw reason to suppose that this agreement is very high in the clause, above the IP-layer of the clause. Following D\u00C3\u00A9chaine (2001), I therefore take this agreement to occupy the phrase which is selected by the complementizer, as in (45): the whole verbal predicate m\u00C3\u00A2to- \u00E2\u0080\u0098cry.VAI\u00E2\u0080\u0099 rises from its lower position to sit in the spec of the Agr Phrase, giving rise to the discontinuity of the two elements. 39 (45) [CP [C \u00C3\u00AA- [XPverb m\u00C3\u00A2toy\u00C3\u00A2n ] ] ] CP 3 3 AgrP C 3 \u00C3\u00AA- XPi 3 IP 4 Agr 5 m\u00C3\u00A2to -y\u00C3\u00A2n \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 Since nominal stems lack this verbal agreement, the clause-typing proclitic cannot select for an appropriate complement, yielding ungrammaticality, as with min\u00C3\u00B4s \u00E2\u0080\u0098cat\u00E2\u0080\u0099. (46) [CP [C \u00C3\u00AA- [XPnoun min\u00C3\u00B4s ] ] ] Adding the appropriate selectional material (i.e., clausal agreement) satisfies selection. Notice that when the nominal stem is framed by the clause agreement, we get a predicate reading: \u00E2\u0080\u0098he is a shoe\u00E2\u0080\u0099; this is a common strategy in Plains Cree. (47) \u00C3\u00AA-min\u00C3\u00B4sit \u00C3\u00AA- min\u00C3\u00B4s -i -t C1-cat -EPEN-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6s/he is a cat.\u00E2\u0080\u0099 (48) [CP [C \u00C3\u00AA- [AgrP min\u00C3\u00B4sit ] ] ] CP 3 3 AgrP C 3 \u00C3\u00AA- XPi 3 4 Agr IP min\u00C3\u00B4s -t 5 \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 With respect to selection, the clause-typing proclitics exhibit behaviour that is quite distinct from the behaviour of the pronominal proclitics: the former behave like heads, the latter do not. 40 2.3.2 Substitution (does not) determine distribution Specifier positions also differ from heads in that interchanging the form of the former should not necessarily change the clause\u00E2\u0080\u0099s distribution relative to external linguistic structure, while interchanging the form of the latter (i.e., the complementizer) should. For example, there are several wh-words that may move to spec, CP in English (Ross 1967, Huang 1982, Richards 1997). (49) a. What did you do yesterday? [CP whati [C did [IP you do ti yesterday ] ] ] b. Why did you do that? [CP whyi [C did [IP you do it ti] ] ] c. Where did you find it? [CP wherei [C did [IP you find it ti] ] ] Substituting what for who will change the question that is being asked, but it does not change the fact that the clause is a wh-clause with wh-syntax. However, the element in C is invariant across this movement: it is always an auxiliary moved from I. We cannot substitute a different complementizer, such as that, or while. (50) a. * What that you did yesterday? [CP whati [C that [IP you did ti yesterday] ] ] b. * What while you did yesterday? [CP whati [C while [IP you did ti yesterday] ] ] If we apply this logic to Plains Cree clauses, we get a split between pronominal proclitics and clause-typing proclitics. Clause-typing proclitics look like English complementizers: substitution changes the distribution of the clause. For example, the \u00C3\u00AA- proclitic obligatorily occurs in clauses associated with an object position. Absence of the \u00C3\u00AA- proclitic (51b) or replacement with a different proclitic (51c) result in a clause that cannot be interpreted as an object clause. (51) a. Jeff \u00C3\u00AA-wanikiskisit \u00C3\u00AA-m\u00C3\u00AEcisot [CP [C \u00C3\u00AA- [XP m\u00C3\u00AEcisot ] ] ] J \u00C3\u00AA- wanikiskisi -t \u00C3\u00AA- m\u00C3\u00AEciso -t J C1- forget.VAI -3 C1-eat.VAI-3 \u00E2\u0080\u0098Jeff forgot that he had eaten.\u00E2\u0080\u0099 41 b. ! Jeff \u00C3\u00AA-wanikiskisit m\u00C3\u00AEcisoci [CP [C \u00EF\u0081\u00B8 [XP m\u00C3\u00AEcisoci ] ] ] J \u00C3\u00AA- wanikiskisi -t m\u00C3\u00AEciso -t -i J C1-forget.VAI -3 eat.VAI-3 -SUBJ --- (intended: \u00E2\u0080\u0098Jeff forgot that he was eating.\u00E2\u0080\u0099) c. ! Jeff \u00C3\u00AA-wanikiskisit k\u00C3\u00A2-m\u00C3\u00AEcisot [CP [C k\u00C3\u00A2- [XP m\u00C3\u00AEcisot ] ] ] J \u00C3\u00AA- wanikiskisi -t k\u00C3\u00A2-m\u00C3\u00AEciso -t J C1-forget.VAI -3 C2-eat.VAI -3 --- (intended: \u00E2\u0080\u0098Jeff forgot that he was eating.\u00E2\u0080\u0099) Pronominal proclitics, on the other hand, do not distinguish between clause-types. For example, substituting the pronominal form does not change the CP\u00E2\u0080\u0099s inability to be embedded. In (52), the clause hosting the proclitic ni- cannot be embedded (52b). (52) a. niw\u00C3\u00A2pamik MATRIX ni- w\u00C3\u00A2pam -ik -w 1- see.VTA-INV-3 \u00E2\u0080\u0098He saw me.\u00E2\u0080\u0099 b. * Jeff niw\u00C3\u00AEhtam\u00C3\u00A2k niw\u00C3\u00A2pamik EMBEDDED J ni- w\u00C3\u00AEhtam -aw -ik ni- w\u00C3\u00A2pam -ik J 1- tell.VTA -BEN-INV 1- see.VTA-INV --- In (53), ki- has been substituted for ni-, and the clause is still unable to be embedded (53b). (53) a. kiw\u00C3\u00A2pamik MATRIX ki- w\u00C3\u00A2pam -ik 2- see.VTA-INV \u00E2\u0080\u0098He saw you.\u00E2\u0080\u0099 b. * Jeff niw\u00C3\u00AEhtam\u00C3\u00A2k kiw\u00C3\u00A2pamik EMBEDDED J ni- w\u00C3\u00AEhtam -aw -ik ki- w\u00C3\u00A2pam -ik J 1- tell.VTA-BEN-INV 2- see.VTA -INV --- 42 Finally, in (54) there is no proclitic at all, and the clause is again unable to be embedded (54b). (54) a. nim\u00C3\u00A2ma w\u00C3\u00A2pamik MATRIX ni- m\u00C3\u00A2ma w\u00C3\u00A2pam -ik 1- mom see.VTA-INV \u00E2\u0080\u0098My mother saw him/her.\u00E2\u0080\u0099 b. * Jeff niw\u00C3\u00AEhtam\u00C3\u00A2k nimama w\u00C3\u00A2pamik EMBEDDED J ni- w\u00C3\u00AEhtam -aw -ik ni- m\u00C3\u00A2ma w\u00C3\u00A2pam -ik J 1- tell.VTA-BEN-INV 1- mom see.VTA-INV --- I therefore posit a structure in which the pronominal proclitics are in spec, CP, rather than C: substitution of the pronominal proclitic (i.e., ni- vs. ki- vs. \u00EF\u0081\u00B8-) does not change the distribution of the clause. (55) CP 2 ni- 2 ki- C 5 \u00EF\u0081\u00B8- The behaviour of the clauses with a \u00EF\u0081\u00B8- proclitic leads us to the next point: the significance of non-overtness. 2.3.3 The significance of non-overtness Just as substitution of different forms has different consequences for specs vs. heads, so the absence of a form has consequences. For heads, the absence of a form means either that the head is gone (resulting in less structure), or that there is a null head (which should affect the form and function of the phrase it projects to (56)). (56) a. XP b. CP 5 2 2 \u00EF\u0081\u00B8 XP 5 43 The structure in (56b) more accurately captures what we see with the clause-typing proclitics. In (57a) we see a clause w\u00C3\u00A2pahtam Jeff \u00E2\u0080\u0098Jeff saw it\u00E2\u0080\u0099 introducing a dependent proposition with the clause-typing proclitic \u00C3\u00AA-. If this clause-typing proclitic is absent, the utterance becomes ungrammatical. (57) Clauses associated with object position require \u00C3\u00AA- proclitic a. w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-k\u00C3\u00AE-mispohk w\u00C3\u00A2pahtam -w J \u00C3\u00AA- k\u00C3\u00AE- mispon -k see.VTI -3 J C1-PREV-snow.VII-0 \u00E2\u0080\u0098Jeff saw it had snowed.\u00E2\u0080\u0099 b. * w\u00C3\u00A2pahtam Jeff k\u00C3\u00AE-mispohk w\u00C3\u00A2pahtam -w J k\u00C3\u00AE- mispon -k see.VTI -3 J PREV-snow.VII-0 --- Likewise, clauses without an overt clause-typing proclitic (used, for example, in some conditionals) become ungrammatical if an overt clause-typing proclitic is added. This is demonstrated in (58). (58) Antecedents of conditionals require \u00C3\u00AA- proclitic a. ksp\u00C3\u00AE n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan sipw\u00C3\u00AAht\u00C3\u00AAci w\u00C3\u00A2pahki, nika-kask\u00C3\u00AAyiht\u00C3\u00AAn8 ksp\u00C3\u00AE ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan sipw\u00C3\u00AAht\u00C3\u00AA -t -i w\u00C3\u00A2pah -k -i ni-ka- kask\u00C3\u00AAyiht\u00C3\u00AA -n if/when 1- friend leave.VAI-3-SUBJ dawn.VII-0 -SUBJ 1- IRR-lonely.VTI -SAP \u00E2\u0080\u0098If my friend leaves tomorrow, I will be lonely.\u00E2\u0080\u0099 b. * ksp\u00C3\u00AE n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA-sipw\u00C3\u00AAht\u00C3\u00AAci w\u00C3\u00A2pahki, nika-kask\u00C3\u00AAyiht\u00C3\u00AAn ksp\u00C3\u00AE ni-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA- sipw\u00C3\u00AAht\u00C3\u00AA -t -i w\u00C3\u00A2pah -k -i ni- ka- kask\u00C3\u00AAyiht\u00C3\u00AA -n if/when 1- friend C1-leave.VAI-3-SUBJ dawn.VII-0-SUBJ 1- IRR-lonely.VTI -SAP --- This bi-directional implication between the presence of a phonologically overt clause-typing proclitic versus the functional and distributional properties of the clause is behaviour that is accounted for by representing them as C heads. 8 The ksp\u00C3\u00AE element in these examples is regularly used in these construction by one of the consultants I worked with. It is not clear to me if this is a morpho-phonological permutation of k\u00C3\u00AEspin \u00E2\u0080\u0098if\u00E2\u0080\u0099, a morpho-phonological permutation of \u00C3\u00AAkosp\u00C3\u00AE \u00E2\u0080\u0098then\u00E2\u0080\u0099, or an entirely different particle. I have thus left this particle in its surface form. 44 There is an additional complication with simple CONJUNCT clauses in that the preverb ka- and suffix -i have a different distribution. For example, the -i clause cannot be substituted for the ka- clause in (59). (59) a. nikw\u00C3\u00AAcim\u00C3\u00A2w Jeff ka-nikamot ni-kw\u00C3\u00AAcim -\u00C3\u00A2 -w J ka- nikamo -t 1-ask.VTA -DIR-3 J IRR-sing.VAI-3 \u00E2\u0080\u0098I asked Jeff to sing.\u00E2\u0080\u0099 b. * nikw\u00C3\u00AAcim\u00C3\u00A2w Jeff nikamoci ni-kw\u00C3\u00AAcim -\u00C3\u00A2 -w J nikamo -t -i 1-ask.VTA -DIR-3 J sing.VAI-3-SUBJ \u00E2\u0080\u0098I asked Jeff to sing.\u00E2\u0080\u0099 However, I do not treat them as complementizers on par with the clause-typing proclitics because they do not have the distributional restrictions that the true clause-typing proclitics have. As we saw earlier in the chapter, the irrealis preverb ka- can occur in matrix and embedded clauses, and in both INDEPENDENT and CONJUNCT clauses. The relevant data is repeated in (60): in (60a), ka- is occurring in an INDEPENDENT clause to indicate a future event, and in (60b) it is in a CONJUNCT clause to indicate a clausal relation akin to the Indo-European subjunctive (see chapter 6 for details). (60) Interpretations of irrealis ka- across Plains Cree\u00E2\u0080\u0099s orders a. ..., \u00E2\u0080\u009C\u00C3\u00A2, \u00C3\u00AAkota nika-p\u00C3\u00B4sipayihon,\u00E2\u0080\u009D ... INDEPENDENT \u00C3\u00A2 \u00C3\u00AAkota ni- ka- p\u00C3\u00B4sipayiho -n future INTERJ there 1- IRR-jump.VAI -SAP \u00E2\u0080\u0098..., \u00E2\u0080\u009CWell, I will jump on that,\u00E2\u0080\u009D \u00E2\u0080\u00A6 \u00E2\u0080\u0099 (AA 8.3) b. nik\u00C3\u00AE-kw\u00C3\u00AAcim\u00C3\u00A2w Nettie ka-p\u00C3\u00AA-itoht\u00C3\u00AAt CONJUNCT ni- k\u00C3\u00AE- kw\u00C3\u00AAcim -\u00C3\u00A2 -w N ka- p\u00C3\u00AA- itoht\u00C3\u00AA -t irrealis 1- PREV-ask.VTA -DIR-3 N IRR-come-go.VAI-3 \u00E2\u0080\u0098I asked Nettie to come.\u00E2\u0080\u0099 Likewise, the suffix -i, which is glossed as a subjunctive marker in simple CONJUNCT clauses (61a), also appears in CONJUNCT clauses that have clause-typing proclitics. In the latter cases it indicates plurality of inanimate referents, as in (61b), or plurality of realis events (61c) (see M\u00C3\u00BChlbauer 2008 for discussion). 45 (61) Interpretations of \u00E2\u0080\u0093i across Plains Cree\u00E2\u0080\u0099s CONJUNCT modes a. miyw\u00C3\u00A2siki \u00E2\u0080\u00A6 SIMPLE SUBJUNCTIVE miyw\u00C3\u00A2si -k \u00E2\u0080\u0093i non-singular unrealized event nice.VII -0 -SUBJ \u00E2\u0080\u0098if it/they are nice\u00E2\u0080\u0099 b. \u00C3\u00AA-miyw\u00C3\u00A2siki maskisina \u00C3\u008A-CONJUNCT \u00C3\u00AA- miyw\u00C3\u00A2si -k -i maskisin -a non-singular inan. referent C1-nice.VII -0-PL shoe -PL \u00E2\u0080\u0098...the shoes are nice.\u00E2\u0080\u0099 c. \u00C3\u00AAkwa piy\u00C3\u00AA-takoht\u00C3\u00AAtw\u00C3\u00A2wi IC-CONJUNCT \u00C3\u00AAkwa iy- p\u00C3\u00AA- takoht\u00C3\u00AA -tw\u00C3\u00A2w -i non-singular realized event and IC-COME-arrive.VAI-3PL -PL \u00E2\u0080\u0098And when the men would come home m\u00C3\u00A2na n\u00C3\u00A2p\u00C3\u00AAwak k\u00C3\u00AE-k\u00C3\u00AEsowihkasowak, \u00E2\u0080\u00A6 m\u00C3\u00A2na n\u00C3\u00A2p\u00C3\u00AAw -ak k\u00C3\u00AE- k\u00C3\u00AEsowihkaso -w -ak usually man -PL PREV-warm.self.VAI-3 -PL they used to warm themselves \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 50) This means that, for ka- and -i, the choice of clause-type affects the interpretation of the proclitic, rather than the proclitic affecting the choice of clause-type. I take the data above to signify that ka- or the -i are in a position immediately below the clause-typing domain; concretely, I call this Cfiniteness, following Rizzi (1997), and call the clause- typing domain where the pronominal proclitics and clause-typing proclitics sit Cforce. (62) a. CPforce 2 2CPfiniteness C 2 \u00EF\u0081\u00B8 2 ka- 5 -i Let us now turn to the specifier position. If a specifier position is phonologically null, the projection as a whole does not change, and we expect that the function and distribution of the constituent also will not change. 46 (63) CP 2 \u00EF\u0081\u00B8 2 C For the pronominal forms, we observe this latter pattern. While the pronominal proclitics are obligatory if a speech act participant (1st or 2nd person) is one of the participants in the event, there is no pronominal proclitic at all if no speech act participant is an event participant. Thus (64) shows two examples that differ only in the presence/absence of a phonologically overt pronominal proclitic: if there is an overt pronominal proclitic, it gets interpreted (in this example) as a subject (64a); if there is none, then the subject is obligatorily unspecified (thus denoted by the passive translation to English in 64b) (cf. D\u00C3\u00A9chaine & Reinholtz 1999 on unspecified subject constructions). (64) a. niw\u00C3\u00A2pam\u00C3\u00A2w ni-w\u00C3\u00A2pam -\u00C3\u00A2 -w 1- see.VTA-DIR-3 \u00E2\u0080\u0098I see him/her b. w\u00C3\u00A2pam\u00C3\u00A2w w\u00C3\u00A2pam -\u00C3\u00A2 -w see.VTA-DIR-3 \u00E2\u0080\u0098s/he was seen.\u00E2\u0080\u0099 Other morpho-syntactic configurations of the INDEPENDENT order, which involve only third persons, obligatorily lack an overt pronominal proclitic, as in (65). (65) a. w\u00C3\u00A2pam\u00C3\u00AAw w\u00C3\u00A2pam -\u00C3\u00AA -w see.VTA-DIR-3 \u00E2\u0080\u0098s/he saw him/her.\u00E2\u0080\u0099 b. * niw\u00C3\u00A2pam\u00C3\u00AAw ni- w\u00C3\u00A2pam -\u00C3\u00AA -w 1- see.VTA-DIR-3 --- A phonologically null \u00E2\u0080\u0093 or absent \u00E2\u0080\u0093 pronominal proclitic does not change the external syntax of the verbal complex. The form w\u00C3\u00A2pam\u00C3\u00AAw \u00E2\u0080\u0098s/he sees him/her\u00E2\u0080\u0099, which lacks a pronominal proclitic, 47 cannot suddenly appear in an embedded clause. In this sense, the pronominal proclitics behave like elements in a specifier, rather than a head, position. 2.3.4 Interim Summary The preceding pages have shown several ways in which the pronominal proclitics in the INDEPENDENT order and the clause-typing proclitics in the CONJUNCT order differ; I have argued that these differences correspond to the split between heads and specifiers. This means that the mapping between Plains Cree\u00E2\u0080\u0099s morpho-syntax and the distinction between indexical versus anaphoric clauses is quite transparent. Given that the INDEPENDENT order corresponds to indexical clauses (the topic of chapter 3), indexical clauses in Plains Cree may host an element in spec, CP. (66) INDEXICAL CLAUSE \u00EF\u0083\u00A0 CP \u00EF\u0083\u00A0 2 s0 2 C XP 5 Plains Cree INDEPENDENT CP 2 ni- 2 ki- C XP \u00EF\u0081\u00B8- 5 w\u00C3\u00A2pam\u00C3\u00A2w And given that the CONJUNCT order corresponds to anaphoric clauses (the topic of chapter 4), anaphoric clauses in Plains Cree may host a complementizer in C that corresponds with the anaphoric situation. 48 (67) ANAPHORIC CLAUSE \u00EF\u0083\u00A0 CP \u00EF\u0083\u00A0 2 s 2 C XP 5 Plains Cree CONJUNCT CP 2 s 2 C XP \u00C3\u00AA- 5 k\u00C3\u00A2- w\u00C3\u00A2pam\u00C3\u00A2t IC- \u00EF\u0081\u00B8- 2.4 The indexical/anaphoric distinction \u00E2\u0089\u00A0 matrix/embedded distinction The final point I want to make in this chapter is that, although Plains Cree\u00E2\u0080\u0099s clause-typing split interacts with the matrix/embedded distinction in many ways, it does not pick out the matrix/embedded distinction. In this section, I briefly look at two elements in Plains Cree that do distinguish matrix vs. embedded contexts: negation, and the interrogative marker c\u00C3\u00AE. The relevant point is that the distribution of both of these elements is not determined by the morpho-syntactic INDEPENDENT/CONJUNCT distinction in Plains Cree, but rather by the syntactic matrix/embedded distinction. 2.4.1 Negation distinguishes matrix and embedded clauses Plains Cree has two negative elements: \u00C3\u00AAk\u00C3\u00A2 and nam\u00C3\u00B4ya9 (Lacombe 1874, Wolfart 1973, Dahlstrom 1991, D\u00C3\u00A9chaine & Wolfart 1998, 2000). Unlike English negation, the form of negation in Plains Cree is sensitive to the distinction between matrix and embedded clauses (Wolfart 1973, Reinholtz & Wolfart 1996, Reinholtz 1999, D\u00C3\u00A9chaine & Wolfart 1998, 2000; D\u00C3\u00A9chaine & Wiltschko 2002, 2006). 9 This negator actually has multiple morpho-phonological forms, including nama, ma, m\u00C3\u00B4ya, and m\u00C3\u00B4y. In general, the m\u00C3\u00B4y(a) forms are most commonly found with clausal negation (as opposed to constituent negation), but more work is needed to understand the interaction of form with function and distribution. 49 The nam\u00C3\u00B4ya form occurs in matrix environments. For example, the verbal complexes under negation in (68a-b) have the same form (CONJUNCT), but differ as to whether they are embedded. The m\u00C3\u00B4y form of negation cannot be used in embedded clauses. (68) a. m\u00C3\u00B4y \u00C3\u00AA-kisk\u00C3\u00AAyimak MATRIX m\u00C3\u00B4y \u00C3\u00AA- kisk\u00C3\u00AAyim -ak NEG C1-know.VTA-1>3 \u00E2\u0080\u0098I didn\u00E2\u0080\u0099t know him.\u00E2\u0080\u0099 b. * nit\u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AAn m\u00C3\u00B4y \u00C3\u00AA-kisk\u00C3\u00AAyimak EMBEDDED ni(t)- \u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AA -n m\u00C3\u00B4y \u00C3\u00AA- kisk\u00C3\u00AAyim -ak 1- consider.difficult.VTI-SAP NEG C1-know.VTA-1>3 --- (intended: \u00E2\u0080\u0098It was hard that I didn\u00E2\u0080\u0099t know him.\u00E2\u0080\u0099) comment: in this sentence, \u00C3\u00AAk\u00C3\u00A2y feels better In (69), the matrix negator m\u00C3\u00B4ya occurs with both INDEPENDENT clauses (69a) and CONJUNCT (69b). This means that the form of negation does not map onto a particular morpho-syntactic form in Plains Cree. (69) a. m\u00C3\u00B4y nin\u00C3\u00AAstosin INDEPENDENT m\u00C3\u00B4y ni- n\u00C3\u00AAstosi -n NEG 1- tired.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m not tired.\u00E2\u0080\u0099 b. \u00E2\u0080\u00A6, nam\u00C3\u00B4y \u00C3\u00AA-m\u00C3\u00B4hcw\u00C3\u00AAyimakik, \u00E2\u0080\u00A6 CONJUNCT nam\u00C3\u00B4ya \u00C3\u00AA- m\u00C3\u00B4hcw\u00C3\u00AAyim -ak -k NEG C1-consider.VTA-1>3-PL \u00E2\u0080\u0098\u00E2\u0080\u00A6, I do not consider them stupid, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (JKN 1.3) The \u00C3\u00AAk\u00C3\u00A2ya form of negation occurs in embedded clauses or in clauses that have the irrealis marker ka-.10 Without ka-, the negator \u00C3\u00AAk\u00C3\u00A2ya is prohibited from matrix clauses, whether they be INDEPENDENT (70a) or CONJUNCT (70b). 10 This is a long-standing puzzle in Plains Cree syntax and semantics: why do these environments pattern together? A third environment where \u00C3\u00AAk\u00C3\u00A2ya negation is used is in imperatives, which, like clauses with ka-, have an irrealis flavor; this suggests that, whatever the answer, the puzzle is not specific to the morpheme ka-. Thus, on the one hand, \u00C3\u00AAk\u00C3\u00A2ya\u00E2\u0080\u0099s distribution is syntactically conditioned (by the matrix/embedded split), and on the other hand it is semantically conditioned (by the realis/irrealis split) (D\u00C3\u00A9chaine & Wolfart 1998, 2000). 50 (70) \u00C3\u00AAk\u00C3\u00A2 negation cannot occur in matrix clauses a. * \u00C3\u00AAk\u00C3\u00A2 nikisk\u00C3\u00AAyim\u00C3\u00A2w INDEPENDENT \u00C3\u00AAk\u00C3\u00A2 ni-kisk\u00C3\u00AAyim-\u00C3\u00A2-w NEG 1-know.VTA-DIR-3 --- (intended: \u00E2\u0080\u0098I don\u00E2\u0080\u0099t know him/her.\u00E2\u0080\u0099) b. * \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA-kisk\u00C3\u00AAyimak CONJUNCT \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- kisk\u00C3\u00AAyim -ak NEG C1-know.VTA-1>3 --- (intended: \u00E2\u0080\u0098I don\u00E2\u0080\u0099t know him/her.\u00E2\u0080\u0099) Since INDEPENDENT clauses are never allowed in embedded contexts, they are unsurprisingly bad here too (71a); in a CONJUNCT embedded clause, \u00C3\u00AAk\u00C3\u00A2 negation is fine (71b). (71) \u00C3\u00AAk\u00C3\u00A2 negation occurs in embedded clauses a. * nit\u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AAk\u00C3\u00A2 nikisk\u00C3\u00AAyim\u00C3\u00A2w INDEPENDENT ni(t)- \u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AAk\u00C3\u00A2 ni- kisk\u00C3\u00AAyim -\u00C3\u00A2 -w 1- consider.difficult.VTI-SAP NEG C1- know.VTA -DIR-3 \u00E2\u0080\u0098It was hard because I didn\u00E2\u0080\u0099t know him.\u00E2\u0080\u0099 b. nit\u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AAk\u00C3\u00A2 e-kisk\u00C3\u00AAyimak CONJUNCT ni(t)- \u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- kisk\u00C3\u00AAyim -ak 1- consider.difficult.VTI-SAP NEG C1-know.VTA-1>3 \u00E2\u0080\u0098It was hard because I didn\u00E2\u0080\u0099t know him.\u00E2\u0080\u0099 (AA 2.1, presented in elicitation) Proof that \u00C3\u00AAk\u00C3\u00A2 negation is not selecting for CONJUNCT clauses can be found when we look at clauses with the modal ka- (cf. Lacombe 1874, D\u00C3\u00A9chaine & Wolfart 1998). In (72a-b), we see examples of \u00C3\u00AAk\u00C3\u00A2 co-occurring with an INDEPENDENT clause hosting ka-. (72) INDEPENDENT + ka- + \u00C3\u00AAk\u00C3\u00A2 negation a. \u00C3\u00AAk\u00C3\u00A2 ka-kimiwan \u00C3\u00AAk\u00C3\u00A2 ka- kimiwan NEG IRR-rain.VII \u00E2\u0080\u0098It better not rain!!\u00E2\u0080\u0099 b. \u00C3\u00AAk\u00C3\u00A2 nika-m\u00C3\u00AEcison \u00C3\u00AAk\u00C3\u00A2 ni- ka- m\u00C3\u00AEciso -n NEG 1- IRR-eat.VAI-SAP \u00E2\u0080\u0098I won\u00E2\u0080\u0099t eat (right now).\u00E2\u0080\u0099 51 Summing up, the distribution of negation in Plains Cree is sensitive to (although not entirely determined by) the matrix/embedded distinction. Relevant to the current discussion is the fact that the distribution of negation is not sensitive to the distinction between Plains Cree\u00E2\u0080\u0099s INDEPENDENT and CONJUNCT orders. 2.4.2 Interrogative marking distinguishes matrix and embedded clauses A second element that is sensitive to the matrix/embedded distinction in Plains Cree is the interrogative marker c\u00C3\u00AE, which can only occur in matrix clauses. This is not surprising, given that interrogative force is a kind of illocutionary force (Searle 1965, Austin 1950), which in turn is thought to be a function of the CP-domain (Cheng 1991, Chomsky 1995, Portner 1999). Again, the point I want to make here is that c\u00C3\u00AE picks out matrix clauses, which are a heterogeneous class in terms of Plains Cree\u00E2\u0080\u0099s morpho-syntax. Both INDEPENDENT and CONJUNCT verbal complexes host Plains Cree c\u00C3\u00AE, as shown in (73): in (73a), c\u00C3\u00AE follows an INDEPENDENT verbal complex, and in (73b), it follows an \u00C3\u00AA-CONJUNCT verbal complex. (73) a. kim\u00C3\u00AEcison c\u00C3\u00AE INDEPENDENT ki-m\u00C3\u00AEciso-n c\u00C3\u00AE 2-eat.VTA-SAP Q \u00E2\u0080\u0098Have you eaten?\u00E2\u0080\u0099 b. \u00C3\u00AA-n\u00C3\u00AAstosiyan c\u00C3\u00AE CONJUNCT \u00C3\u00AA-n\u00C3\u00AAstosi-yan c\u00C3\u00AE C1-tired.VAI-2 Q \u00E2\u0080\u0098\u00E2\u0080\u00A6are you tired?\u00E2\u0080\u0099 However, c\u00C3\u00AE may not be embedded under a higher predicate. Thus, in (74a), c\u00C3\u00AE is in second position and has scope over the clause it follows \u00E2\u0080\u0093 the matrix clause kiw\u00C3\u00A2pam\u00C3\u00A2w \u00E2\u0080\u0098you saw her\u00E2\u0080\u0099. In (74b), which was an attempt to form an embedded interrogative, c\u00C3\u00AE is ungrammatical. 52 (74) a. kiw\u00C3\u00A2pam\u00C3\u00A2w c\u00C3\u00AE Rose-Marie \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00A2hpit MATRIX ki- w\u00C3\u00A2pam -\u00C3\u00A2 -w c\u00C3\u00AE RM \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00A2hpi -t 2- see.VTA-DIR-3 Q RM C1-PREV-laugh.VAI-3 \u00E2\u0080\u0098Did you see that Rose-Marie laughed?\u00E2\u0080\u0099 b. * Rose-Marie m\u00C3\u00B4y niw\u00C3\u00A2pam\u00C3\u00A2w [\u00C3\u00AA-sipw\u00C3\u00AAht\u00C3\u00AAt c\u00C3\u00AE] EMBEDDDED RM m\u00C3\u00B4ya ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w \u00C3\u00AA- sipw\u00C3\u00AAht\u00C3\u00AA -t c\u00C3\u00AE RM NEG 1- see.VTA -DIR-3 C1-leave.VAI-3 Q --- (intended: \u00E2\u0080\u0098I didn\u00E2\u0080\u0099t see if/whether Rose-Marie left.\u00E2\u0080\u0099) Similarly, in (75) we observe that k\u00C3\u00AEspin \u00E2\u0080\u0098if\u00E2\u0080\u0099 is used to introduce an indirect yes/no question (75a), and that it is ungrammatical to replace k\u00C3\u00AEspin \u00E2\u0080\u0098if\u00E2\u0080\u0099 with c\u00C3\u00AE \u00E2\u0080\u0098Q\u00E2\u0080\u0099 (75b). (75) a. nikw\u00C3\u00AAcim\u00C3\u00A2w Rose-Marie k\u00C3\u00AEspin \u00C3\u00AA-w\u00C3\u00AE-itoht\u00C3\u00AAt ni- kw\u00C3\u00AAcim -\u00C3\u00A2 -w RM k\u00C3\u00AEspin \u00C3\u00AA- w\u00C3\u00AE- itoht\u00C3\u00AA -t 1- ask.VTA-DIR-3 RM if C1-INT-go.VAI-3 \u00E2\u0080\u0098I asked Rose-Marie if/whether she was coming.\u00E2\u0080\u0099 b. * nikw\u00C3\u00AAcim\u00C3\u00A2w Rose-Marie c\u00C3\u00AE \u00C3\u00AA-w\u00C3\u00AE-itoht\u00C3\u00AAt ni- kw\u00C3\u00AAcim -\u00C3\u00A2 -w RM c\u00C3\u00AE \u00C3\u00AA- w\u00C3\u00AE- itoht\u00C3\u00AA -t 1- ask.VTA -DIR-3 RM Q C1-INT-go.VAI-3 --- In summary, then, c\u00C3\u00AE picks out matrix clauses, but it does not pick out INDEPENDENT clauses. Together, negation and the interrogative c\u00C3\u00AE provide evidence that the morpho-syntactic division between INDEPENDENT and CONJUNCT in Plains Cree does not correspond to the matrix/embedded distinction. 2.5 Summary This chapter has been concerned with how the indexical vs. anaphoric division in clauses maps onto Plains Cree\u00E2\u0080\u0099s morpho-syntax. I first argued that the left-edge pronominal proclitics and clause-typing proclitics are hosted in the CP-domain in Plains Cree, a place where Plains Cree differs from at least some other Algonquian languages, and that these were thus candidates for cuing the indexical/anaphoric division, which I claim is hosted in spec, CP. 53 a. INDEXICAL CP b. ANAPHORIC CP (76) CP 2 s0 2 C XP 5 CP 2 s 2 C XP 5 I then argued that the pronominal proclitics in the INDEPENDENT order are in spec, CP, while the clause-typing proclitics in the CONJUNCT order are complementizers. a. INDEXICAL CP b. ANAPHORIC CP (77) CP 2 ni- 2 ki- C XP \u00EF\u0081\u00B8- 5 CP 2 s 2 \u00C3\u00AA- XP k\u00C3\u00A2- 5 IC- \u00EF\u0081\u00B8- This results in a one-to-one mapping between the indexical vs. anaphoric clauses on the one hand, and Plains Cree\u00E2\u0080\u0099s clause-typing morpho-syntax on the other. I now turn to the external syntax and the semantics of each of these clauses. 54 CHAPTER 3 INDEXICAL CLAUSES: PLAINS CREE\u00E2\u0080\u0099S INDEPENDENT ORDER 3.1 Proposal: The syntax and semantics of indexical clauses In chapter 2, we looked at the internal structure of indexical clauses, and I argued that they have an indexical speech situation variable in spec, CP. (1) Internal structure of an indexical clause CP 2 s0 2 ni- C XP ki- 5 w\u00C3\u00A2pam\u00C3\u00A2w In Plains Cree, an indexical clause is instantiated by the INDEPENDENT order. In the summary given in table 3.1, we see that INDEPENDENT clauses are characterized by left-edge 1st and 2nd person marking, and by a unique set of right-edge person marking. PERSON CATEOGRY INDEPENDENT ORDER 1.sg. ninip\u00C3\u00A2n 2.sg. kinip\u00C3\u00A2n 1.pl. excl. ninip\u00C3\u00A2n\u00C3\u00A2n 1.pl. incl. kinip\u00C3\u00A2naw 2.pl kinip\u00C3\u00A2n\u00C3\u00A2w\u00C3\u00A2w 3.sg. nip\u00C3\u00A2w 3.pl nipawak 3.obv nip\u00C3\u00A2yiwa Table 3.1. Summary of the INDEPENDENT order paradigm in Plains Cree 55 In this chapter I turn to indexical clauses\u00E2\u0080\u0099 external properties; i.e., how an indexical clause relates to clause-external linguistic material. I make claims about the reflexes of indexicality in these clauses\u00E2\u0080\u0099 structural, semantic, and discourse properties. Structurally, I take there to be two ways in which clauses can be related: the hierarchical notion of c-command as defined in (2), and the linear notion of precedence as defined in (3). (2) C-commanddef: A constituent \u00CE\u00B1 c-commands \u00CE\u00B2 iff \u00CE\u00B2 is dominated by the lowest node of a major category that dominates \u00CE\u00B1. (3) Precedencedef: A constituent \u00CE\u00B1 precedes \u00CE\u00B2 iff constituent \u00CE\u00B1 is linearly ordered before \u00CE\u00B2 within a given domain. Given these conditions1, I claim that indexical clauses are subject to anti-c-command and anti- precedence. Thus they are prohibited in configurations like (4) where the indexical clause is dominated by another clause. (4) * CP 5 CPIND 5 They are also prohibited in configurations like (5), where the indexical clause is non-initial within the domain (indicated by the square brackets). (5) * [ CP ... CPIND ... ] In \u00C2\u00A73.2, I show that these conditions on indexical clauses derive the distribution of Plains Cree\u00E2\u0080\u0099s INDEPENDENT order clauses. The anti-c-command condition derives the fact that Plains Cree\u00E2\u0080\u0099s INDEPENDENT clauses are always matrix clauses. The anti-precedence condition derives the fact that variables introduced in Plains Cree\u00E2\u0080\u0099s INDEPENDENT order clauses must have a clause-internal antecedent \u00E2\u0080\u0093 i.e., even a non-c-commanding antecedent is ruled out. In other words, indexical clauses \u00E2\u0080\u009Cmust be free.\u00E2\u0080\u009D 1 There is much disagreement in the literature about whether c-command and precedence are two separate conditions (Ross 1967b, Carden 1986, Williams 1997) or whether one can be derived from the other (cf. Reinhart 1983, perhaps laid out most explicitly in Kayne 1994). As we will see in this chapter and chapter 4, Plains Cree exhibits patterns that are best captured by positing both conditions. 56 Semantically, indexical clauses have a privileged relation to the speech act (cf. Banfield 1982): they are indexed to it (cf. B\u00C3\u00BChler 1934, Bar-Hillel 1954, Kaplan 1989 on indexical expressions). Indexicality is a subset of deixis that picks out the speaker, the speech time and/or the speech location.2 We therefore expect that an indexical clause will have the following particular deictic properties: (i) referentially, they are anchored to the speaker; (ii) temporally, they are anchored to the speech time; and (iii) spatially, they anchored to the speech place. In \u00C2\u00A73.3, I show that these properties account for the restricted interpretation of Plains Cree\u00E2\u0080\u0099s INDEPENDENT order clauses: they are evaluated relative to speech time, relative to the speaker, and relative to the speech location. Modelling this within a situation semantics framework, where every proposition must be evaluated with respect to a situation (Austin 1950, Barwise 1981, Barwise & Perry 1983, Kratzer 1989, 2007), I argue that in an indexical clause this situation is the speech situation. As discussed in chapter 1, a situation s is a partial world; the speech situation s0 is simply a situation in which someone is speaking. The speech situation minimally must include the individual who is doing the speaking (i.e., the speaker I); and the temporal/spatial location of the speaking (i.e., speech time now and speech place here). Therefore, if the truth of a proposition expressed by an indexical clause is evaluated relative to the speech situation, this logically entails that the clause be evaluated relative to both the individual (speaker) and temporal/spatial location to be coded. In \u00C2\u00A73.3 I look at the referential, temporal, and spatial anchoring properties of indexical clauses to empirically motivate the semantic claims about them. 3.2 The structural context of indexical clauses In this section I discuss the structural contexts of indexical clauses, focussing on the implications of the claim that indexical clauses are subject to anti-c-command and anti-precedence. We expect 2 The confluence of speaker, speech time, and speech place is called the origo in some treatments (e.g., B\u00C3\u00BChler 1934; Garrett 2001). 57 that the exponent of indexical clauses in Plains Cree, INDEPENDENT clauses, will be excluded from all embedded contexts (\u00C2\u00A73.2.1). However, being a matrix clause is not enough to ensure an indexical clause. The implication goes only one way: indexical clauses must be matrix clauses, but there can be matrix clauses which are not indexical. In order for a matrix clause to be an indexical clause it must also satisfy anti-precedence: it cannot be preceded by another clause within its domain. This means, for example, that an indexical clause cannot enter into cross-clausal dependencies. Thus, in a language that morpho-syntactically marks indexical clauses (such as Plains Cree\u00E2\u0080\u0099s INDEPENDENT order), we expect that all and only indexical matrix clauses will lack cross-clausal dependencies (\u00C2\u00A73.2.2). 3.2.1 Indexical clauses must be matrix clauses In this section I show that indexical INDEPENDENT clauses cannot be introduced by higher predicates or by subordinating particles. Further, elements which can be independently argued to be restricted to embedded clauses will be ungrammatical with INDEPENDENT clauses; for example, I show that the embedded negator \u00C3\u00AAk\u00C3\u00A2ya cannot be used with an INDEPENDENT clause. English clauses are unspecified with respect to the indexical/non-indexical distinction; there is no morpho-syntactic marking to distinguish them. The form of an English matrix clause can be morpho-syntactically identical to its embedded counterpart. (6) a. I\u00E2\u0080\u0099m happy. b. I told her I\u00E2\u0080\u0099m happy. In Plains Cree, however, these two contexts are morpho-syntactically distinguished: an indexical INDEPENDENT clause can occur in a matrix context (7a), but is replaced by a non-indexical CONJUNCT clause in the corresponding embedded context (7b). 58 (7) a. nicihk\u00C3\u00AAyiht\u00C3\u00AAn INDEPENDENT ni- cihk\u00C3\u00AAyiht\u00C3\u00AA -n 1- happy.VTI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m happy.\u00E2\u0080\u0099 b. niw\u00C3\u00AEhtamaw\u00C3\u00A2w \u00C3\u00AA-cihk\u00C3\u00AAyihtam\u00C3\u00A2n CONJUNCT ni- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w \u00C3\u00AA- cihk\u00C3\u00AAyihtam -\u00C3\u00A2n 1- tell.VTA -DIR-3 C1-happy.VTI -1 \u00E2\u0080\u0098I told him/her I\u00E2\u0080\u0099m happy.\u00E2\u0080\u0099 3.2.1.1 Embedding predicates do not introduce indexical clauses Many verbs in Plains Cree may introduce an embedded clause, but indexical clauses (Plains Cree\u00E2\u0080\u0099s INDEPENDENT order) are impossible in an embedded position; another clause type (Plains Cree\u00E2\u0080\u0099s CONJUNCT order) must be used. (8) CP1 6 matrix CPi 5 \u00E2\u009C\u0094 CONJUNCT \u00E2\u009C\u0096 INDEPENDENT The examples below illustrate. In (9) the embedded clause is a simple CONJUNCT clause (9a); an indexical INDEPENDENT clause is ungrammatical (9b). (9) a. ninitaw\u00C3\u00AAyim\u00C3\u00A2w n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan ka-m\u00C3\u00AEcisot CONJUNCT ni- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -w ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan ka- m\u00C3\u00AEciso -t 1- want.VTA -DIR-3 1- friend IRR-eat.VAI-3 \u00E2\u0080\u0098I want my friend to eat. b. * ninitaw\u00C3\u00AAyim\u00C3\u00A2w n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan m\u00C3\u00AEcisow INDEPENDENT ni- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -w ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan m\u00C3\u00AEciso-w 1- want.VTA -DIR-3 1- friend eat.VAI-3 --- (intended: \u00E2\u0080\u0098I want my friend to eat.\u00E2\u0080\u0099) Likewise, in (10) we observe an embedded \u00C3\u00AA-conjunct clause (10a); again the indexical INDEPENDENT counterpart is ungrammatical. 59 (10) a. nik\u00C3\u00AE-w\u00C3\u00A2paht\u00C3\u00AAn \u00C3\u00AA-kimiwahk CONJUNCT ni- k\u00C3\u00AE- w\u00C3\u00A2paht\u00C3\u00AA -n \u00C3\u00AA- kimiwan -k 1- PREV-see.VTI -SAP C1-rain.VII -0 \u00E2\u0080\u0098I saw that it was raining.\u00E2\u0080\u0099 b. * nik\u00C3\u00AE-w\u00C3\u00A2paht\u00C3\u00AAn kimiwan INDEPENDENT ni-k\u00C3\u00AE- w\u00C3\u00A2paht\u00C3\u00AA -n kimiwan 1-PREV-see.VTI -SAP rain.VII --- (intended: \u00E2\u0080\u0098I saw that it was raining.\u00E2\u0080\u0099) Similarly, predicative particles (e.g., piko \u00E2\u0080\u0098be.necessary.that\u00E2\u0080\u0099) cannot introduce an INDEPENDENT clause (cf. Wolfart 1973, Ahenakew 1987). Rather, they always introduce a CONJUNCT clause. This is illustrated in (11), where both simple CONJUNCT clauses (prefixed with the irrealis marker ka-) and changed CONJUNCT clauses (prefixed with the complementizer \u00C3\u00AA-) are grammatical (11a- a\u00E2\u0080\u0099), but INDEPENDENT clauses are not (11b). (11) a. piko ka-w\u00C3\u00A2pamak ana n\u00C3\u00A2p\u00C3\u00AAw SIMPLE CONJ piko ka- w\u00C3\u00A2pam -ak ana n\u00C3\u00A2p\u00C3\u00AAw be.necessary IRR-see.VTA-1>3 DEM.AN man \u00E2\u0080\u0098I have to see that man.\u00E2\u0080\u0099 a\u00E2\u0080\u0099. piko \u00C3\u00AA-w\u00C3\u00A2pamak ana n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u008A- CONJUNCT piko \u00C3\u00AA- w\u00C3\u00A2pam -ak ana n\u00C3\u00A2p\u00C3\u00AAw be.necessary C1-see.VTA -1>3 DEM.AN man \u00E2\u0080\u0098I have to see that man.\u00E2\u0080\u0099 b. * piko niw\u00C3\u00A2pam\u00C3\u00A2w ana n\u00C3\u00A2p\u00C3\u00AAw INDEPENDENT piko ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w ana n\u00C3\u00A2p\u00C3\u00AAw be.necessary 1- see.VTA-DIR-3 DEM.AN man -- In summary, Plains Cree INDEPENDENT clauses cannot be embedded. This is a way in which Plains Cree\u00E2\u0080\u0099s INDEPENDENT clauses are more restricted than clauses that appear in matrix contexts in English: the latter can occur in embedded contexts without any change in the morpho-syntax, whereas the indexical INDEPENDENT clauses cannot. 60 3.2.1.2 Subordinating particles do not introduce indexical c\lauses In addition to embedded clauses, there are a number of subordinators which introduce different kinds of adjoined dependent clauses. These subordinators are uninflected particles which sit external to and precede the verbal complex; they act as restrictors on the complementizer of the clause they introduce, specifying the type of embedded clause. Syntactically, I posit that these particles are complementizers. Since indexical clauses are by hypothesis subject to anti-c-command, we expect that Plains Cree\u00E2\u0080\u0099s INDEPENDENT clauses should never occur with these subordinators. This expectation is fulfilled by the data as exemplified in the following table. The distribution of each subordinator according to clause-type is given for four different speakers (AA, EM, JK, SW).3 None of the subordinators introduce an indexical INDEPENDENT clause for any of the speakers, while all of them may introduce an anaphoric CONJUNCT clause (the numbers give the number of attested examples for each speaker). indexical: INDEPENDENT anaphoric: CONJUNCT Subordinator JK SW AA EM JK SW AA EM os\u00C3\u00A2m \u00E2\u0080\u0098reason\u00E2\u0080\u0099 -- \u00E2\u009C\u0096 -- \u00E2\u009C\u0096 -- \u00E2\u009C\u0094 (5) -- \u00E2\u009C\u0094 (24) iyikohk \u00E2\u0080\u0098as far as\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (12) \u00E2\u009C\u0094 (4) \u00E2\u009C\u0094 (47) \u00E2\u009C\u0094 (46) kiy\u00C3\u00A2m \u00E2\u0080\u0098although\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (4) \u00E2\u009C\u0094 (6) \u00E2\u009C\u0094 (5) \u00E2\u009C\u0094 (5) p\u00C3\u00A2mway\u00C3\u00AAs \u00E2\u0080\u0098before\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (2) \u00E2\u009C\u0094 (2) \u00E2\u009C\u0094 (3) \u00E2\u009C\u0094 (10) mayaw \u00E2\u0080\u0098as soon as\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (7) \u00E2\u009C\u0094 (3) \u00E2\u009C\u0094 (4) \u00E2\u009C\u0094 (2) Table 3.2. Distribution of subordinators by clause-type in Plains Cree The distributional facts of INDEPENDENT clauses in the context of subordinators are quite striking: they simply do not occur. By contrast, every subordinator introduces some form of an anaphoric CONJUNCT clause4. This data was confirmed in elicitation sessions, where speakers rejected utterances where an indexical INDEPENDENT clause had been substituted for the anaphoric CONJUNCT clause: minimal pairs are given in (12-13). For example, concessive clauses are introduced by kiy\u00C3\u00A2m \u00E2\u0080\u0098although\u00E2\u0080\u0099, and always appear in the k\u00C3\u00A2-conjunct (12a), usually followed by the particle \u00C3\u00A2ta \u00E2\u0080\u0098even\u00E2\u0080\u0099. An indexical INDEPENDENT clause is ungrammatical (12b). 3 There are a limited number of subordinators that occur with any regularity. The table is intended to be an exhaustive list of those which occur across multiple speakers. 4 See chapter 6 for details. 61 (12) kiy\u00C3\u00A2m \u00E2\u0080\u0098although\u00E2\u0080\u0099 only in CONJUNCT a. kiy\u00C3\u00A2m \u00C3\u00A2ta k\u00C3\u00A2-pipok, \u00C3\u00A2hci piko m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-y\u00C3\u00AEkinik\u00C3\u00AAt nik\u00C3\u00A2w\u00C3\u00AEn\u00C3\u00A2n. CONJUNCT kiy\u00C3\u00A2m \u00C3\u00A2ta k\u00C3\u00A2-pipon -k although even C2-winter.VII-0 \u00C3\u00A2hci piko m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- y\u00C3\u00AEkinik\u00C3\u00AA -t ni-k\u00C3\u00A2w\u00C3\u00AE -n\u00C3\u00A2n still usually C1-PREV-milk.VAI-3 1- mother-1.PL \u00E2\u0080\u0098Even during the winter our mother would still milk the cows.\u00E2\u0080\u0099 (EM 17) b. * kiy\u00C3\u00A2m \u00C3\u00A2ta pipon, \u00C3\u00A2hci piko m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-y\u00C3\u00AEkinik\u00C3\u00AAt nik\u00C3\u00A2w\u00C3\u00AEn\u00C3\u00A2n. INDEPENDENT kiy\u00C3\u00A2m \u00C3\u00A2ta pipon \u00C3\u00A2hci piko m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- y\u00C3\u00AEkinik\u00C3\u00AA -t ni-k\u00C3\u00A2w\u00C3\u00AE -n\u00C3\u00A2n although even winter.VII still usually C1-PREV-milk.VAI-3 1- mother-1.PL --- Degree clauses are introduced by the element iyikohk \u00E2\u0080\u0098so\u00E2\u0080\u0099. As a degree marker,5 iyikohk \u00E2\u0080\u0098so\u00E2\u0080\u0099 always introduces an anaphoric CONJUNCT clause (13a); the corresponding indexical INDEPENDENT clause is ungrammatical (13b). (13) iyikohk \u00E2\u0080\u0098so\u00E2\u0080\u0099 only in CONJUNCT a. \u00E2\u0080\u009C\u00E2\u0080\u00A6 \u00C3\u00AAkotowahk m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEciy\u00C3\u00A2n, iyikohk \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAy\u00C3\u00A2n,\u00E2\u0080\u009D \u00E2\u0080\u00A6 CONJUNCT \u00C3\u00AAkotowahk m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00AEci -y\u00C3\u00A2n iyikohk \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -y\u00C3\u00A2n that.kind usually C1-PREV-eat.VTI-1 DEG C1-hungry.VAI-1 \u00E2\u0080\u0098 \u00E2\u0080\u009C\u00E2\u0080\u00A6I was so hungry that I would eat that kind,\u00E2\u0080\u009D \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 71) (alt. trans. \u00E2\u0080\u009CI used to eat the kind because I was so hungry.\u00E2\u0080\u009D) b. * \u00C3\u00AAkotowahk m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEciy\u00C3\u00A2n, iyikohk nik\u00C3\u00AE-noht\u00C3\u00AAhkat\u00C3\u00A2n6 INDEPENDENT \u00C3\u00AAkotowahk m\u00C3\u00A2na \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEci-y\u00C3\u00A2n iyikohk ni-k\u00C3\u00AE-n\u00C3\u00B4ht\u00C3\u00AAhkat-\u00C3\u00A2-n that.kind usually C1-PREV-eat.VTI-1 DEG 1-PREV-hungry-VAI-SAP --- (intended: \u00E2\u0080\u0098I was so hungry that I would eat that kind.\u00E2\u0080\u0099) 5 Note that, like many particles, iyikohk occurs in a number of varied contexts, with a number of interpretations. While some of these contexts do allow INDEPENDENT clauses, these contexts do not have the dependence of the degree clauses given above. See chapter 6 for further discussion. 6 In this example, I have presented the INDEPENDENT order clause with the temporal sequencer k\u00C3\u00AE-, since k\u00C3\u00AE- is necessary to get a time disjoint from utterance time (cf. \u00C2\u00A77.1), which is what we have in the preceding \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEciy\u00C3\u00A2n clause. The INDEPENDENT is also bad if the k\u00C3\u00AE- is omitted: (i) * \u00E2\u0080\u00A6\u00C3\u00AAkotowahk m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEciy\u00C3\u00A2n, iyikohk ninoht\u00C3\u00AAhkat\u00C3\u00A2n \u00C3\u00AAkotowahk m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00AEci -y\u00C3\u00A2n iyikohk ni- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00A2 -n that.kind usually C1-PREV-eat.VTI-1 DEG 1- hungry.VAI-SAP --- (intended: \u00E2\u0080\u0098I was so hungry that I would eat that kind.\u00E2\u0080\u0099) 62 3.2.1.3 Embedded negation does not modify indexical clauses The last embedded context presented here is one specific to Cree \u00E2\u0080\u0093 the interaction of clause-type with negation. As we saw in chapter 2, Plains Cree has two forms of negation: nam\u00C3\u00B4ya and \u00C3\u00AAkaya. These two forms are sensitive to the matrix/embedded distinction. nam\u00C3\u00B4ya occurs in unembedded contexts. The \u00C3\u00AAk\u00C3\u00A2ya form of negation occurs only in embedded clauses (cf. D\u00C3\u00A9chaine & Wiltschko 1998, 2006)7 ; the relevant contrast is shown in (14). (14) a. nit\u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AAk\u00C3\u00A2 e-kisk\u00C3\u00AAyimak ni(t)- \u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- kisk\u00C3\u00AAyim -ak 1- consider.difficult.VTI-SAP NEG C1-know.VTA-1>3 \u00E2\u0080\u0098It was hard because I didn\u00E2\u0080\u0099t know him.\u00E2\u0080\u0099 (AA 2.1, presented in elicitation) b. * nit\u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AAn m\u00C3\u00B4y \u00C3\u00AA-kisk\u00C3\u00AAyimak ni(t)- \u00C3\u00A2yim\u00C3\u00AAyiht\u00C3\u00AA -n m\u00C3\u00B4y \u00C3\u00AA- kisk\u00C3\u00AAyim -ak 1- consider.difficult.VTI-SAP NEG C1-know.VTA-1>3 --- (intended: \u00E2\u0080\u0098It was hard because I didn\u00E2\u0080\u0099t know him.\u00E2\u0080\u0099) Thus we expect that it will not be possible to negate an indexical INDEPENDENT clauses with \u00C3\u00AAk\u00C3\u00A2. This is correct, as shown in (15): replacing m\u00C3\u00B4y negation with \u00C3\u00AAk\u00C3\u00A2 negation yields ungrammaticality. (15) a. m\u00C3\u00B4y ninoht\u00C3\u00AAhkat\u00C3\u00A2n m\u00C3\u00B4y ni- noht\u00C3\u00AAhkat\u00C3\u00A2 -n NEG 1- hungry.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m not hungry.\u00E2\u0080\u0099 b. * \u00C3\u00AAk\u00C3\u00A2 ninoht\u00C3\u00AAhkat\u00C3\u00A2n \u00C3\u00AAk\u00C3\u00A2 ni- noht\u00C3\u00AAhkat\u00C3\u00A2-n NEG 1- hungry.VAI-SAP -- Thus, negation provides further evidence for the indexical analysis: we see that the only type of negation available for INDEPENDENT clauses is the negation that is restricted to unembedded environments. 7 Unless the irrealis ka- preverb is present; see chapter 2. 63 3.2.1.4 Summary: Indexical clauses cannot be subordinated We have seen three independent pieces of evidence that indexical INDEPENDENT clauses are subject to anti-c-command: they cannot be introduced by a higher predicate, they cannot be introduced by a subordinating particle, and they cannot be negated by the \u00C3\u00AAk\u00C3\u00A2y\u00C3\u00A2 negator. In the next section I turn to the anti-precedence condition on indexical clauses and show how this condition accounts for the lack of cross-clausal dependencies in indexical INDEPENDENT clauses. 3.2.2 Indexical clauses exclude cross-clausal dependencies The purpose of this section is to show that cross-clausal dependencies, such as the binding of a variable by a clause-external variable, are excluded from indexical clauses. First, I examine a class of variables known in the Algonquianist literature as relative roots (\u00C2\u00A73.2.2.1), and show that the antecedence relation is affected by clause-type, an observation which, regardless of whether the current analysis is correct or not, offers an important insight into the grammar of Algonquian languages. Second, I examine temporal and locative proforms and show that unless the proform has a morphologically marked deictic component, they are ungrammatical in indexical INDEPENDENT clauses (\u00C2\u00A73.2.2.2). Third, I discuss how reference to argument expressions is restricted in indexical INDEPENDENT clauses (\u00C2\u00A73.2.2.3) 3.2.2.1 Relative roots Relative roots are a class of proforms (locative, manner, temporal, etc.) found across all Algonquian languages (Bloomfield 1962; Wolfart 1973; Valentine 2001; Rhodes 1976). They are termed roots because they may be found in the root position of a stem (even though they may also be found in places where they are not in a root position). They are relative because they do not have an independent interpretation, but rather are interpreted relative to the an antecedent which is obligatory for the utterance in which they occur to be well-formed (Bloomfield 1962, Wolfart 1973). More generally, they are variables that are quite unspecified as to their features: 64 their specific function is determined in part by its position in the clause (there are at least three possible positions), and in part by nature of its antecedent. For each of the two relative root variables that I look at here, there are at least three kinds of antecedents. Although they have been widely discussed in the literature (cf. Bloomfield 1928, 1946, 1962; Wolfart 1973, Rhodes 1976, 2003, Pentland 1979, Dahlstrom 1991, Bruening 2001, among others), the principles that determine (im)possible antecedence relations remain very poorly understood. In Plains Cree, the inventory of these antecedent-dependent elements is a closed class and includes: it/isi \u00E2\u0080\u0098thus\u00E2\u0080\u00998; oht/ohci \u00E2\u0080\u0098originating.from\u00E2\u0080\u0099 isko- \u00E2\u0080\u0098to.such.an.extent\u00E2\u0080\u0099, and tahto- \u00E2\u0080\u0098so.many\u00E2\u0080\u0099 (Wolfart 1973:66). In this thesis I have chosen to look at the two relative roots which are found across all of the speakers I have worked with: the relative root of manner: it/is(i) \u00E2\u0080\u0098thus\u00E2\u0080\u0099 and the relative root of origin: oht/ohc(i) \u00E2\u0080\u0098originating from\u00E2\u0080\u0099. Both relative roots may be found in a variety of positions within the clause. In the following examples, I have bolded the relative root and underlined its antecedent (the element without which the utterance would be ungrammatical). First, they may occur in a root position: in (16a), isi- \u00E2\u0080\u0098thus\u00E2\u0080\u0099 is the root and Jane is the antecedent; in (16b), ohc- \u00E2\u0080\u0098origin\u00E2\u0080\u0099 is in the root position and Calgary is the antecedent. (16) ROOT POSITION a. Jane isiy\u00C3\u00AEhk\u00C3\u00A2sow J isiy\u00C3\u00AEhk\u00C3\u00A2so -w J THUS.be.called.VAI-3 \u00E2\u0080\u0098Her name is Jane.\u00E2\u0080\u0099 b. Calgary nitohc\u00C3\u00AEn C nit- ohc\u00C3\u00AE -n C 1- ORIG.VAI-SAP \u00E2\u0080\u0098I am from Calgary.\u00E2\u0080\u0099 8 These glosses are meant only to give a rough idea of their meaning; as will become clear, their semantics are underspecified. The addition of -i causes a palatalization of both relative roots: it \u00EF\u0083\u00A0 isi; and oht \u00EF\u0083\u00A0 ohci (cf. Piggott 1971, Wolfart 1973). Due to morpho-phonological processes which lead to the deletion of \u00E2\u0080\u0093i (for example, vowel hiatus), many times the surface form will be palatalized but not have -i. Finally, ohc(i) alternates with \u00C3\u00B4h- based on factors that are as yet undescribed in the literature. The alternations do not seem to have any direct correlation to the syntactic and semantic generalizations presented here (although I have never seen the \u00C3\u00B4h- form with an \u00C3\u00AA-CONJUNCT verbal complex), so I will not be concerned further about which form shows up. 65 Second, relative root variables may occur in a preverbal position, as in (17)9. (17) PREVERBAL POSITION a. m\u00C3\u00A2ka kahkiyaw p\u00C3\u00A2h-p\u00C3\u00AEtos kitis-\u00C3\u00A2y\u00C3\u00A2n\u00C3\u00A2naw. m\u00C3\u00A2ka kahkiyaw p\u00C3\u00A2h- p\u00C3\u00AEtos kit- is- \u00C3\u00A2y\u00C3\u00A2 -n\u00C3\u00A2naw but all RED- different 2- THUS-be.VAI-2.PL \u00E2\u0080\u0098but we are all different.\u00E2\u0080\u0099 (EM 19) b. mistahi m\u00C3\u00A2n \u00C3\u00A2ya, t\u00C3\u00B4ht\u00C3\u00B4s\u00C3\u00A2poy nik\u00C3\u00AE-ohci-pim\u00C3\u00A2cihikon\u00C3\u00A2n \u00C3\u00AAkwa aya, ... mistahi m\u00C3\u00A2na aya t\u00C3\u00B4ht\u00C3\u00B4s\u00C3\u00A2poy ni- k\u00C3\u00AE- ohci- pim\u00C3\u00A2cih -iko -n\u00C3\u00A2n \u00C3\u00AAkwa aya a.lot usually CONN milk 1- PREV-ORIG-sustain.VTA-INV-1PL and CONN \u00E2\u0080\u0098She used to have lots of milk on which to sustain us, ...\u00E2\u0080\u0099 (EM 16) Finally, a relative root variable can be an adposition. With verbs of motion, isi indicates motion towards goal (e.g., towards waskahikanihk \u00E2\u0080\u0098the house.LOC\u00E2\u0080\u0099 in 18a) and ohci indicates motion from the origin (e.g., away from waskahikanihk \u00E2\u0080\u0098the house.LOC\u00E2\u0080\u0099 in 18b) (cf. Edwards 1954). (18) ADPOSITION \u00E2\u0080\u0093 VERB OF MOTION a. nipimoht\u00C3\u00A2n w\u00C3\u00A2skahikanihk isi ni- pimoht\u00C3\u00A2 -n w\u00C3\u00A2skahikan -ihk isi 1- walk.VAI -SAP house -LOC THUS \u00E2\u0080\u0098I\u00E2\u0080\u0099m walking towards the house.\u00E2\u0080\u0099 b. nipimoht\u00C3\u00A2n w\u00C3\u00A2skahikanihk ohci ni- pimoht\u00C3\u00A2 -n w\u00C3\u00A2skahikan -ihk ohci 1- walk.VAI-SAP house -LOC ORIG \u00E2\u0080\u0098I\u00E2\u0080\u0099m walking from the house.\u00E2\u0080\u0099 If the verb is not a verb of motion, the adpositional relative root indicates manner for isi, as in (19a), and instrumental for ohci, as in (19b). 9 I take the clausal material occurring external to the verbal complex to be part of the CP constituting the verbal complex; e.g., in (17) I take the adverbial p\u00C3\u00A2h-p\u00C3\u00AEtos \u00E2\u0080\u0098different\u00E2\u0080\u0099 to be a modifier of the verb ay\u00C3\u00A2- \u00E2\u0080\u0098be\u00E2\u0080\u0099. I do not know of any good analysis of the mechanisms driving some clausal elements to be external to the verbal complex, and others to be internal to it (although see Dahlstrom 1995, M\u00C3\u00BChlbauer 2003, and D\u00C3\u00A9chaine 2007 for a more detailed description of the issue). On a very broad view, the issue seems to be one of non-concatenative morpho-syntax (cf. non-concatenative morpho-phonology in Semitic; Arad 2000). 66 (19) ADPOSITION \u00E2\u0080\u0093 OTHER VERBS a. ..., \u00C3\u00A2ta \u00C3\u00AA-k\u00C3\u00AE-kisk\u00C3\u00AAyihtahkik \u00C3\u00A2h-\u00C3\u00A2y\u00C3\u00AEtaw isi maskihkiy, ... \u00C3\u00A2ta \u00C3\u00AA- k\u00C3\u00AE- kisk\u00C3\u00AAyihtam -k -k \u00C3\u00A2h- \u00C3\u00A2y\u00C3\u00AEtaw isi maskihkiy even C1-PREV-know.VTI -0 -PL RED-side THUS medicine \u00E2\u0080\u0098..., although they used to know both sides of medicine, ...\u00E2\u0080\u0099 (AA 10.1) (Lit: \u00E2\u0080\u0098... they knew both sides of medicine that way.\u00E2\u0080\u0099) b. m\u00C3\u00B4hkom\u00C3\u00A2n ohci \u00C3\u00AA-w\u00C3\u00AE-mansam\u00C3\u00A2n m\u00C3\u00B4hkom\u00C3\u00A2n ohci \u00C3\u00AA- w\u00C3\u00AE- mansam -\u00C3\u00A2n knife ORIG C1-INT-cut.VTI -1 \u00E2\u0080\u0098I am going to cut it with a knife.\u00E2\u0080\u0099 Notice that in all these examples the underlined antecedent precedes the bolded relative root it binds: for example the antecedent m\u00C3\u00B4hkom\u00C3\u00A2n \u00E2\u0080\u0098knife\u00E2\u0080\u0099 must precede the adposition ohci \u00E2\u0080\u0098with\u00E2\u0080\u0099. This is a context where there is an fixed ordering between two elements in Plains Cree10 (Wolfart 1973, see also Rhodes 2003 for Ojibwa). As we will see, however, this pattern is part of a more general principle about the relation that must hold between a dependent element and its antecedent (cf. chapter 4). Now that we have seen the different positions where a relative root position may be introduced, I turn to the different possible antecedents. I focus on relative roots in the preverbal position because it is this position that (i) shows the most variation in possible antecedents, but (ii) has antecedents that are both clause-internal and clause-external, allowing us to test the claim about indexical INDEPENDENT clauses. 3.2.2.1.1 Relative roots with predicate modifier antecedents One type of antecedent that preverbal relative roots may be anaphoric on is a predicate modifier (i.e., an adverbial or oblique argument). Syntactically, predicate modifiers are usually assumed to be introduced quite low in the clause, either in the vP, or the functional domain (AspP or TP). 10 There are some speakers for whom the ordering some examples is not fixed. In particular, when the relative root is stem-internal, it does not require that the antecedent precede the stem, as in (i) volunteered by a consultant. (i) nitisiy\u00C3\u00AEhk\u00C3\u00A2son Clare ni(t)- isiy\u00C3\u00AEhk\u00C3\u00A2so -n C 1- THUS.be.called.VAI-SAP C \u00E2\u0080\u0098My name is Clare.\u00E2\u0080\u0099 I take this to be a separate grammar, where the stem is now opaque \u00E2\u0080\u0093 it has \u00E2\u0080\u0098word-level\u00E2\u0080\u0099 properties in the sense of DiSciullo & Williams (1987), and thus the relative root is not available for syntactic operations. See also Hirose (2000) for discussion of variation with respect to the syntactic visibility of stems in Plains Cree. 67 Plains Cree\u00E2\u0080\u0099s predicate modifiers are consistent with this claim: they are linearly internal to elements in information-structure positions (i.e., topic/focus), negation, and quantifiers (cf. Dahlstrom 1995, M\u00C3\u00BChlbauer 2003). For concreteness, I place them as modifiers to the vP. In (20), I give a proposed structure, where the relative root variable (RR.vbl) is associated with a predicate modifier (indicated by the coindexation). (20) CP 2 NEG TP / AspP SUBORD. 2 ASP/QUANT vP 2 vP PRED.MODi 5 RR.vbli- Since predicate modifiers are within the same CP as the relative root variable they are associated with, we expect them to be possible antecedents for both INDEPENDENT and CONJUNCT clauses. This is accurate for both relative roots, as shown in table 3.3. indexical INDEPENDENT anaphoric CONJUNCT Antecedent JK SW AA EM JK SW AA EM isi- \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Predicate modifier ohci- \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 3.3 Predicate modifier antecedents occur in both clause-types The preverbal isi- can have lexical manner adverbs as antecedents, as in (21a-b). Both indexical INDEPENDENT and anaphoric CONJUNCT clauses allow manner adverbs such as kwayask \u00E2\u0080\u0098proper\u00E2\u0080\u0099 to be antecedents to isi-. (21) a. ..., pik \u00C3\u00B4ma ka-m\u00C3\u00A2maw\u00C3\u00B4hkam\u00C3\u00A2toyahk, CONJUNCT piko \u00C3\u00B4ma ka-m\u00C3\u00A2maw\u00C3\u00B4hkama-ito-yahk necessary DEM.INAN IRR-work.VTA -RECIP-21pl kwayask ka-kakw\u00C3\u00AA-isi-pim\u00C3\u00A2tisiyahk, ... kwayask ka-kakw\u00C3\u00AA-isi-pim\u00C3\u00A2tisi-yahk proper IRR-try-THUS-live.VAI -21 \u00E2\u0080\u0098..., we must work together to try to lead a good life, ...\u00E2\u0080\u0099 (EM 37) 68 b. ..., matw\u00C3\u00A2n c\u00C3\u00AE kwayask nika-k\u00C3\u00AE-isi-t\u00C3\u00A2hk\u00C3\u00B4t\u00C3\u00AAn \u00E2\u0080\u00A6 INDEPENDENT matw\u00C3\u00A2n c\u00C3\u00AE kwayask ni-ka- k\u00C3\u00AE- isi- t\u00C3\u00A2hk\u00C3\u00B4t\u00C3\u00AA -n EVID Q proper 1- IRR-PREV-THUS-discuss.VTI-SAP \u00E2\u0080\u0098..., I wonder if I will be able to discuss it with proper faithfulness, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (JKN 6.1) The other relative root, ohci, introduces instrumental adjuncts11. In (22a), the topic element \u00C3\u00AAwako \u00E2\u0080\u0098that\u00E2\u0080\u0099 is the antecedent for ohci and represents the means of washing the floor. In (22b), the deictic element \u00C3\u00AAkoni is the antecedent for ohci and indicates the means of blessing the addressee. The former is an anaphoric CONJUNCT clause; the latter an indexical INDEPENDENT clause. (22) a. ..., pihko \u00C3\u00AA-sisw\u00C3\u00AAw\u00C3\u00AApinahkik \u00C3\u00AAkwa CONJUNCT pihko \u00C3\u00AA- sisw\u00C3\u00AAw\u00C3\u00AApinam -k -k \u00C3\u00AAkwa ash C1-sprinkle.VTI -0 -PL and \u00E2\u0080\u0098Some I even saw sprinkle ashes about and \u00C3\u00AAwako \u00C3\u00AA-ohci-w\u00C3\u00A2piskahahkik aya, \u00E2\u0080\u00A6 \u00C3\u00AAwako \u00C3\u00AA- ohci- w\u00C3\u00A2piskaham -k -k aya TOPIC C1-ORIG-wash.VTI -0 -PL CONN use that to wash the floor-boards \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 82) (lit: \u00E2\u0080\u0098...and wash the floor-boards with that ...\u00E2\u0080\u0099) b. \u00E2\u0080\u009Ch\u00C3\u00A2w, \u00C3\u00AAkoni \u00C3\u00B4hi, k-\u00C3\u00B4h-saw\u00C3\u00AAyimitin n\u00C3\u00AEst \u00C3\u00B4ma, \u00E2\u0080\u00A6 INDEPENDENT h\u00C3\u00A2w \u00C3\u00AAkoni \u00C3\u00B4hi ki- oh- saw\u00C3\u00AAyim -iti -n n\u00C3\u00AEsta \u00C3\u00B4ma indeed DEIC.TOPIC DEM.INAN 2- ORIG-bless.VTA-1>2-SAP 1.EMPH DEM.INAN \u00E2\u0080\u0098 \u00E2\u0080\u009CIndeed, with these I myself will bless you, \u00E2\u0080\u00A6 \u00E2\u0080\u0099 (JKN 7.2) 11 The preverbal ohci- can also introduce directional adjuncts, just like the adpositional ohci. Directional adjuncts also being predicate modifiers, they can occur with either indexical (INDEPENDENT) or anaphoric (CONJUNCT) clauses, as shown in (i). Notice that with the CONJUNCT example, there is a demonstrative intervening between the locative element \u00C3\u00B4t\u00C3\u00AA \u00E2\u0080\u0098there\u00E2\u0080\u0099 and the verbal complex; this is indicative of a cleft construction (cf. Blain 1997) and significantly, is absent in the INDEPENDENT example. (i) a. ..., \u00C3\u00B4t\u00C3\u00AA k-\u00C3\u00B4h-os\u00C3\u00A2pamikow\u00C3\u00A2w. \u00C3\u00B4t\u00C3\u00AA ki- oh- os\u00C3\u00A2pam -iko -w\u00C3\u00A2w there 2- ORIG-watch.jealously.VTA-INV -2PL ... , that they are [jealously] watching you from over there, ... (JKN 3.17) b. \u00C3\u00B4t\u00C3\u00AA ana \u00C3\u00AA-ohci-kit\u00C3\u00A2pamiht, ... \u00C3\u00B4t\u00C3\u00AA ana \u00C3\u00AA- ohci- kit\u00C3\u00A2pam -ih -t there DEM.AN C1-ORIG-watch.over.VTA-USC-3 he is watched over from there, ... (JKN 4.9) 69 3.2.2.1.2 Relative roots with CP-modifier antecedents CP-modifiers can also act as antecedents to a preverbal relative root variable, including the deictic topic marker \u00C3\u00AAkosi and negation (both nam\u00C3\u00B4ya and \u00C3\u00AAk\u00C3\u00A2ya). The topic marker is part of information structure, and on independent grounds, negation is a CP-modifier (D\u00C3\u00A9chaine & Wiltschko 2002). (23) CP 3 NEGi TP / AspP TOPICi 2 ASP vP 2 vP PRED.MOD 5 vbli- Since negation and topic markers are CP-modifiers, again we expect that they are possible antecedents in both INDEPENDENT and CONJUNCT clauses. This is borne out in all four corpora, as shown in table 3.4. indexical INDEPENDENT anaphoric CONJUNCT Antecedent JK SW AA EM JK SW AA EM isi- (oblique) \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Predicate modifier ohci- (oblique) \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 isi- (topic) \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 CP- modifier ohci- (negation) \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 3.4. CP-modifier antecedents are possible in both clause-types The topic-marker \u00C3\u00AAkosi \u00E2\u0080\u0098that way\u00E2\u0080\u0099 is an antecedent for isi (24); consistent with being a CP- modifier, it occurs in a clause-initial position (Baker 1985, Cinque 1999; Cook et. al 2003).12 12 \u00C3\u00AAkosi ends in the segmental sequence si-; following Wolfart (1973), I analyze this element as having a bi-partite structure \u00C3\u00AAkw \u00E2\u0080\u0098deixis\u00E2\u0080\u0099 and isi \u00E2\u0080\u0098thus\u00E2\u0080\u0099. 70 (24) a. \u00C3\u00AAwakw \u00C3\u00A2nim [CP \u00C3\u00AAkos \u00C3\u00AA-k\u00C3\u00AE-isi-pim\u00C3\u00A2cihocik kay\u00C3\u00A2s ayisiyiniwak ]. CONJUNCT \u00C3\u00AAwakw anima \u00C3\u00AAkosi \u00C3\u00AA- k\u00C3\u00AE- isi- pim\u00C3\u00A2ciho -t -k kay\u00C3\u00A2s ayisiyiniw -ak topic DEM.INAN DEIC.so C1-PREV-THUS-live.VAI -3 -PL long.ago person -PL \u00E2\u0080\u0098That is how the people made a living long ago.\u00E2\u0080\u0099 (EM 12) b. \u00E2\u0080\u00A6, [CP \u00C3\u00AAkosi m\u00C3\u00AEna m\u00C3\u00A2na nik\u00C3\u00AE-isi-m\u00C3\u00A2miton\u00C3\u00AAyiht\u00C3\u00AAn ]. INDEPENDENT \u00C3\u00AAkosi m\u00C3\u00AEna m\u00C3\u00A2na ni- k\u00C3\u00AE- isi- m\u00C3\u00A2miton\u00C3\u00AAyiht\u00C3\u00AA -n DEIC.so also usually 1- PREV-THUS-think.about.VTI-SAP \u00E2\u0080\u0098\u00E2\u0080\u00A6, and that is how I used to think about it.\u00E2\u0080\u0099 (EM 29) Negation is an antecedent for ohci in both anaphoric CONJUNCT and indexical INDEPENDENT clauses13; here it is suppletive with the temporal preverb k\u00C3\u00AE-14 and has a past orientation. In (25a), \u00C3\u00AAk\u00C3\u00A2 \u00E2\u0080\u0098NEG\u00E2\u0080\u0099 antecedes ohci- in a CONJUNCT clause; in (25b) m\u00C3\u00B4y negation antecedes \u00C3\u00B4h- in an INDEPENDENT clause. (25) a. \u00E2\u0080\u00A6, \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA-ohci-nisitaw\u00C3\u00AAyimakik ayisiyiniwak. CONJUNCT \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA -ohci- nisitaw\u00C3\u00AAyim -ak -k ayisiyiniw -ak NEG C1-ORIG- know.VTA -1>3-PL person -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6, because I did not know people.\u00E2\u0080\u0099 (EM 8) b. m\u00C3\u00A2ka, niya wiya m\u00C3\u00B4y n\u00C3\u00B4h-pakw\u00C3\u00A2t\u00C3\u00AAn anima ... INDEPENDENT m\u00C3\u00A2ka niya wiya m\u00C3\u00B4ya ni-\u00C3\u00B4h- pakw\u00C3\u00A2t\u00C3\u00AA -n anima but 1 EMPH NEG 1-ORIG-hate.VTI-SAP DEM.INAN \u00E2\u0080\u0098But I did not mind [it]...\u00E2\u0080\u0099 (EM 8) 3.2.2.1.3 Relative roots with cross-clausal antecedents Like other kinds of variables, relative roots can have an antecedent in another clause, creating a cross-clausal dependency15. These include clause-external wh-words (Blain 1997; Cook 2003, 13 In this discussion, I claim that NEG is an antecdent to ohci- in the sense that (i) the presence of NEG is sufficient to license ohci-; and (ii) the presence of NEG yields a particular interpretation of ohci-. An alternative hypothesis is that ohci- under negation is a negative polarity item. I think that the NPI analysis is not mutually exclusive to the discussion here \u00E2\u0080\u0093 e.g., as an NPI, ohci- would be licensed by negation, which is consistent with the discussion here. The main point I am trying to make here is that ohci- must always be licensed by something. There is still the question of how ohci gets the interpretation it does \u00E2\u0080\u0093 i.e., how is the \u00E2\u0080\u0098past\u00E2\u0080\u0099 interpretation related to the directional and instrumental readings. For this question, I think a consideration of the very abstract ORIGIN concept would be useful, although I do not have time and space to consider it here. Finally, if ohci- is analyzed as an NPI, it would have to be a strong NPI in that negation is the only context that triggers it (e.g., ohci- cannot be triggered in interrogatives, relative clauses, etc., at least in Plains Cree). Since the NEG\u00E2\u0080\u00A6ohci pattern is robust across the Algonquian family, it might be worthwhile to look at NPI contexts across different languages to see if there is variation on this account. Thanks to A. Dahlstrom (p.c.) for bringing this to my attention, and to L. Matthewson (p.c.) for discussion. 14 See \u00C2\u00A77.4 for a discussion on the role of negation and modality in the interpretation of k\u00C3\u00AE-. 15 These cross-clausal dependencies lead Bruening (2001) to posit a relative root phrase. 71 2004) as in (26a), clause-external non-wh antecedents (Wolfart 1973) as in (26b), and discourse antecedents (Bloomfield 1928, 1946) as in (26c). (26) a. XP 5 WHI CP 5 vbli b. \u00E2\u0080\u00A6XPi \u00E2\u0080\u00A6 CP 5 vbli- c. CPi 5 CP 5 vbli Now, if indexical clauses have an anti-c-command condition, there will be no higher clause to host the antecedent. We therefore expect that cross-clausal antecedents will only be possible with anaphoric CONJUNCT clauses; indexical INDEPENDENT clauses should be ungrammatical. This is correct, as summarized in table 3.5. INDEPENDENT CONJUNCT Antecedent JK SW AA EM JK SW AA EM isi- (oblique) \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Predicate modifier ohci- (oblique) \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 isi- (deictic) \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 CP-modifier ohci- (negation) \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 isi- (manner) \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Cross-clausal wh- ohci- (reason) \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 isi- (manner) \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Cross-clausal non wh- ohci- (reason) \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Superordinate clause isi- \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 3.5. No cross-clausal antecedents for indexical INDEPENDENT clauses Let us look at each case in turn. Starting first with the wh-antecedents, we observe that the relative root isi- may be bound by the manner wh-word t\u00C3\u00A2nisi \u00E2\u0080\u0098how\u00E2\u0080\u0099 (27a). Similarly, the relative root ohci- may be bound by the reason wh-word t\u00C3\u00A2n\u00C3\u00AAhk\u00C3\u00AE \u00E2\u0080\u0098why\u00E2\u0080\u0099 in (27b). 72 (27) a. [CP [pred t\u00C3\u00A2nisii ]j [DP proi [CP Opi ... isii ... ] tj ] t\u00C3\u00A2nisi \u00C3\u00AA-isi-pimipayik, ... t\u00C3\u00A2nisi \u00C3\u00AA-isi-pimipayi-k Q.manner C1-THUS-go.VII-0 \u00E2\u0080\u0098how it happened\u00E2\u0080\u0099 (SW16) b. [CP [pred t\u00C3\u00A2n\u00C3\u00AAhk\u00C3\u00AEi ]j [DP proi [CP Opi \u00E2\u0080\u00A6 \u00C3\u00B4hi \u00E2\u0080\u00A6 ] tj ] \u00E2\u0088\u0092 t\u00C3\u00A2n\u00C3\u00AAhk \u00C3\u00A2wa, k-\u00C3\u00B4h-ihtakot \u00C3\u00B4ta? t\u00C3\u00A2n\u00C3\u00AAhk\u00C3\u00AE awa k\u00C3\u00A2- \u00C3\u00B4h- ihtako -t \u00C3\u00B4ta Q.RAT DEM.AN C2-ORIG-exist.VAI-3 here \u00E2\u0080\u0098why does this one here exist?\u00E2\u0080\u0099 (JK3:13) Relative root variables may also be bound by non-wh antecedents. In (28) we have a bi- clausal structure: the \u00C3\u00AAkos \u00C3\u00A2nima sequence is a kind of nominal predication structure (D\u00C3\u00A9chaine 1997, Blain 1999) roughly equivalent to \u00E2\u0080\u0098the way is this\u00E2\u0080\u0099. The anaphoric CONJUNCT clause modifies the subject anima \u00E2\u0080\u0098this\u00E2\u0080\u0099. Here the relative root variable isi- in the embedded modifying clause has the deictic manner element \u00C3\u00AAkosi \u00E2\u0080\u0098this way\u00E2\u0080\u0099 in the higher clause as its antecedent. (28) [CP [pred \u00C3\u00AAkosii ]j [DP animai [CP Opi \u00E2\u0080\u00A6 isii \u00E2\u0080\u00A6 ] tj ] ..., \u00C3\u00AAkos \u00C3\u00A2nim \u00C3\u00AA-isi-kit\u00C3\u00A2pam\u00C3\u00A2t; (EM38) \u00C3\u00AAkosi anima \u00C3\u00AA- isi- kit\u00C3\u00A2pam -\u00C3\u00A2 -t TOP DEM.INAN C1-THUS-look.VTA-DIR-3 ..., that is the way she looks upon them; Similarly, the relative root variable ohci- can have a cross-clausal antecedent like the \u00C3\u00AAwak\u00C3\u00B4hci (from \u00C3\u00AAwakw \u00E2\u0080\u0098that\u00E2\u0080\u0099 + ohci \u00E2\u0080\u0098originate\u00E2\u0080\u0099) in (29b), which is again arguably acting as the subject of a higher nominal predication structure (D\u00C3\u00A9chaine 1997, Blain 1999). Such antecedents are fine with an anaphoric CONJUNCT clause, but not with indexical INDEPENDENT clauses. 73 (29) [CP [pred \u00C3\u00AAwak\u00C3\u00B4hci ]j [DP proi [CP Opi \u00E2\u0080\u00A6 ohi \u00E2\u0080\u00A6 ] tj ] a. \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AA-os\u00C3\u00AAhk\u00C3\u00AAmit, \u00C3\u00AAwak\u00C3\u00B4hci k\u00C3\u00B4h-\u00C3\u00A2tosk\u00C3\u00AAt Jeff CONJUNCT \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AA-os\u00C3\u00AAhk\u00C3\u00AAmi -t \u00C3\u00AAwakw ohci k\u00C3\u00A2- \u00C3\u00B4h- \u00C3\u00A2tosk\u00C3\u00AA -t J C1-WANT-have.car.VAI-3 TOPIC ORIG C2-ORIG-work.VAI-3 J \u00E2\u0080\u0098He wants a car, that is why Jeff is working.\u00E2\u0080\u0099 b. * \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AA-os\u00C3\u00AAhk\u00C3\u00AAmit, \u00C3\u00AAwak\u00C3\u00B4hci \u00C3\u00B4hc-atosk\u00C3\u00AAw Jeff INDEPENDENT \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AA-os\u00C3\u00AAhk\u00C3\u00AAmi -t \u00C3\u00AAwakw ohci \u00C3\u00B4hc- \u00C3\u00A2tosk\u00C3\u00AA -w J C1-WANT-have.car.VAI-3 TOPIC ORIG ORIG-work.VAI-3 J --- (intended: \u00E2\u0080\u0098He wants a car, that\u00E2\u0080\u0099s why Jeff is working.\u00E2\u0080\u0099) The final kind of cross-clausal antecedent is specific to the relative root variable isi-. This antecedent is not a word- or phrase-level constituent, but rather the a preceding (set of) clause(s) (cf. Bloomfield 1928). For example, in (30), the narrator is describing of the things they had to do, and she then says that, through those actions they were able to avoid starvation. Thus, all of the things described in the initial clauses serve as an antecedent to the manner variable isi- in the purpose clause. (30) piko mitoni --~ t\u00C3\u00A2pitawi pikw \u00C3\u00AA-~ \u00C3\u00AA-w\u00C3\u00AE-kakw\u00C3\u00AA-t\u00C3\u00B4tam\u00C3\u00A2hk k\u00C3\u00AEkway, piko mitoni t\u00C3\u00A2pitawi piko \u00C3\u00AA- w\u00C3\u00AE- kakw\u00C3\u00AA-t\u00C3\u00B4tam -\u00C3\u00A2n -k k\u00C3\u00AEkway QUANT much truly QUANT C1-INT-TRY- do.VTI -1 -PL thing \u00E2\u0080\u0098we very much had to try and do things at all time k-\u00C3\u00AAsi-pihkohtam\u00C3\u00A2sohk ka-m\u00C3\u00AEcihk. ka- isi- pihkohtam\u00C3\u00A2so -hk ka- m\u00C3\u00AEci -hk IRR-THUS-manage.VAI -IMP IRR-eat.VTI-IMP in order to manage to have something to eat.\u00E2\u0080\u0099 (AA 9.1) Once again, since the antecedent is external to the clause, the behaviour of isi- exemplified in (30) is unattested with indexical INDEPENDENT clauses. To sum up, relative roots show that indexical INDEPENDENT clauses exclude cross-clausal anaphoric relations that are possible in anaphoric CONJUNCT clauses. This is important because it is consistent with our expectation that dependency relations must be resolved locally (i.e., clause- internally) with indexical clauses. 74 3.2.2.2 Spatio-temporal variables must be bound in indexical clause Plains Cree has dedicated spatial and temporal proform variables, including ita \u00E2\u0080\u0098where\u00E2\u0080\u0099; it\u00C3\u00AA \u00E2\u0080\u0098where\u00E2\u0080\u0099 and isp\u00C3\u00AE \u00E2\u0080\u0098when\u00E2\u0080\u0099; these occur on the far left edge of the clause, and must be bound by an antecedent. The anti-c-command and anti-precedence conditions predict that they will be excluded from indexical clauses. This is correct: as shown in table 3.6, they are unattested. indexical INDEPENDENT anaphoric CONJUNCT Variable JK SW AA EM JK SW AA EM it\u00C3\u00AA \u00E2\u0080\u0098where\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (5) \u00E2\u009C\u0094 (3) \u00E2\u009C\u0094 (7) \u00E2\u009C\u0094 (8) ita \u00E2\u0080\u0098where\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (46) \u00E2\u009C\u0094 (8) \u00E2\u009C\u0094 (9) \u00E2\u009C\u0094 (27) isp\u00C3\u00AE \u00E2\u0080\u0098when\u00E2\u0080\u0099 \u00E2\u009C\u0096 -- \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (4) -- \u00E2\u009C\u0094 (7) \u00E2\u009C\u0094 (9) Table 3.6. Distribution of spatio-temporal proform variables by clause-type Elicitation data confirms that locative it\u00C3\u00AA and ita, and temporal isp\u00C3\u00AE are incompatible with indexical INDEPENDENT clauses. (31) locative proform variables are bad in INDEPENDENT a. * it\u00C3\u00AA itoht\u00C3\u00AAwak INDEPENDENT it\u00C3\u00AA itoht\u00C3\u00AA -w -ak LOC go.VAI-3 -PL --- (intended: \u00E2\u0080\u0098They are going there/somewhere.\u00E2\u0080\u0099) b. * ita itoht\u00C3\u00AAwak INDEPENDENT it\u00C3\u00AA itoht\u00C3\u00AA -w -ak LOC go.VAI-3 -PL --- (intended: \u00E2\u0080\u0098They are going somewhere.\u00E2\u0080\u0099) (32) temporal proform variable is bad in INDEPENDENT a. isp\u00C3\u00AE k\u00C3\u00A2-p\u00C3\u00AEhtikw\u00C3\u00AAt, \u00E2\u0080\u00A6 CONJUNCT isp\u00C3\u00AE k\u00C3\u00A2- p\u00C3\u00AEhtikw\u00C3\u00AA -t TEMP C2-go.in.VAI -3 \u00E2\u0080\u0098when she went in, \u00E2\u0080\u00A6\u00E2\u0080\u0099 b. * isp\u00C3\u00AE p\u00C3\u00AEhtikw\u00C3\u00AAw INDEPENDENT isp\u00C3\u00AE p\u00C3\u00AEhtikw\u00C3\u00AA -w TEMP go.in.VAI -3 --- (intended: \u00E2\u0080\u0098When/then she went in.\u00E2\u0080\u0099) 75 However, if these variables are bound by the deictic element \u00C3\u00AAkw- (Wolfart 1973), they become perfectly acceptable, as demonstrated by the locative proforms with \u00C3\u00AAkw- in (33) and the minimal pair of temporal proforms in (34). (33) Locative proforms bound by \u00C3\u00AAkw- are good in INDEPENDENT a. itoht\u00C3\u00AAwak \u00C3\u00AAkot\u00C3\u00AA itoht\u00C3\u00AA -w -ak \u00C3\u00AAkot\u00C3\u00AA go.VAI-3 -PL there \u00E2\u0080\u0098They went over there.\u00E2\u0080\u0099 b. itoht\u00C3\u00AAwak \u00C3\u00AAkota itoht\u00C3\u00AA -w -ak \u00C3\u00AAkot\u00C3\u00AA go.VAI-3 -PL there \u00E2\u0080\u0098They went there.\u00E2\u0080\u0099 (34) Temporal proforms require \u00C3\u00AAkw- in INDEPENDENT a. * isp\u00C3\u00AE kimiwan \u00C3\u00AAkot\u00C3\u00AA k\u00C3\u00A2-itoht\u00C3\u00AAy\u00C3\u00A2hk isp\u00C3\u00AE kimiwan \u00C3\u00AAkot\u00C3\u00AA k\u00C3\u00A2- itoht\u00C3\u00AA -y\u00C3\u00A2n -k time rain.VII there C2- go.VAI -1 -PL --- b. \u00C3\u00AAkosp\u00C3\u00AE kimiwan \u00C3\u00AAkot\u00C3\u00AA k\u00C3\u00A2-itoht\u00C3\u00AAy\u00C3\u00A2hk \u00C3\u00AAkosp\u00C3\u00AE kimiwan \u00C3\u00AAkot\u00C3\u00AA k\u00C3\u00A2- itoht\u00C3\u00AA -y\u00C3\u00A2n -k then rain.VII there C2-go.VAI -1 -PL \u00E2\u0080\u0098At that time it was raining, when we went there.\u00E2\u0080\u0099 Syntactically, \u00C3\u00AAkw- acts as an antecedent to the variable, precluding the necessity of a cross- clausal antecedent. Semantically, recall from chapter 1 that deictic expressions point to their referent (Fillmore 1975, Kaplan 1989, a.o.); thus the presence of \u00C3\u00AAkw- fixes the reference of the spatio/temporal proform in a given context. In fact, \u00C3\u00AAkw- is a general-purpose deictic antecedent. For example, we saw earlier that the deictic topic marker \u00C3\u00AAkosi \u00E2\u0080\u0098that way\u00E2\u0080\u0099 was one of the possible antecedents for the isi variable in indexical clauses. The relevant example is repeated in (35). 76 (35) ..., \u00C3\u00AAkos \u00C3\u00A2nim \u00C3\u00AA-isi-kit\u00C3\u00A2pam\u00C3\u00A2t; \u00C3\u00AAkosi anima \u00C3\u00AA- isi- kit\u00C3\u00A2pam -\u00C3\u00A2 -t TOP DEM.INAN C1-THUS-look.VTA-DIR-3 \u00E2\u0080\u0098..., that is the way she looks upon them;\u00E2\u0080\u0099 (EM 38) It is also used for deictic referents, in combination with the referential ani (cf. Reinholtz & Wolfart 2001). For example, in (36) \u00C3\u00AAkoni is associated with n\u00C3\u00A2p\u00C3\u00AAwa \u00E2\u0080\u0098man\u00E2\u0080\u0099, and is used with either a demonstrative gesture (deixis), or to refer back to man previously talked about (anaphora). (36) \u00C3\u00AA-w\u00C3\u00A2pam\u00C3\u00A2t iskw\u00C3\u00AAw \u00C3\u00AAkoni n\u00C3\u00A2p\u00C3\u00AAwa \u00C3\u00AA- w\u00C3\u00A2pam -\u00C3\u00A2 -t iskw\u00C3\u00AAw \u00C3\u00AAkoni n\u00C3\u00A2p\u00C3\u00AAw -a C1-see.VTA-DIR-3 woman that.one man -OBV \u00E2\u0080\u0098The woman saw that man.\u00E2\u0080\u0099 The morphological bi-partite structure of \u00C3\u00AAkw- affixed elements (cf. Wolfart 1973:38) has syntactic and semantic significance as well: the initial \u00C3\u00AAkw- morpheme acts as an antecedent to a variable introduced by the second unit of the demonstrative. (37) 3 DEMP CP 3 5 DEM XP niw\u00C3\u00A2pam\u00C3\u00A2w \u00C3\u00AAkw- 2 pro 2 LOC ita/it\u00C3\u00AA TEMP isp\u00C3\u00AE ARG ani MANNER isi 3.2.3 Pronominal proclitics are indexical I have so far shown that indexical clauses have a particular set of structural properties: (i) they cannot be embedded; and (ii) dependent elements must have their dependency resolved clause- internally or be deictic. I have claimed that these properties reflect the syntax of indexicality, here implemented as anti-c-command and anti-precedence conditions. A third outcome of the 77 indexical analysis is that indexical clauses should not have anaphoric pronominal forms. In particular, the pronominal proclitics in Plains Cree INDEPENDENT clauses should have a deictic/indexical dependency, rather than an anaphoric one. Turning to the INDEPENDENT mode paradigm in table 3.7, this includes 1st-person ni- and 2nd-person ki- . In Plains Cree, unlike many other Algonquian languages, it does not include 3rd-person o- PERSON CATEOGRY INDEPENDENT MODE NOMINAL POSSESSION 1.sg. ninip\u00C3\u00A2n nimis 2.sg. kinip\u00C3\u00A2n kimis 1.pl. excl. ninip\u00C3\u00A2n\u00C3\u00A2n nimisin\u00C3\u00A2n 1.pl. incl. kinip\u00C3\u00A2naw kimisnaw 2.pl kinip\u00C3\u00A2n\u00C3\u00A2w\u00C3\u00A2w kimisiw\u00C3\u00A2w 3.sg. nip\u00C3\u00A2w omisa 3.pl nipawak omisiw\u00C3\u00A2wa 3.obv nip\u00C3\u00A2yiwa omisiyiwa Table 3.7. Person-marking in INDEPENDENT clauses vs. possessed nominals If these pronominal proclitics are indexical, they should have a more restricted behaviour than general pronominals: (a) indexical pronominals cannot be bound16, and (b) indexical pronominals cannot lack referential features. 3.2.3.1 Indexical proclitics cannot be bound 1st- and 2nd- person pronominal forms are anaphoric variables; rather, they are a sub case of deixis: they directly point to the speech act participants (speaker and hearer). Thus, in possession, nimin\u00C3\u00B4sim \u00E2\u0080\u0098my cat\u00E2\u0080\u0099 is the cat of the speaker (38a); in the clause nin\u00C3\u00AAstosin \u00E2\u0080\u0098I am tired\u00E2\u0080\u0099 it is the speaker who is tired (38b). 16 As we will see, English forms like I and you can be bound in some contexts. Although they are often considered prototypical indexical forms, I would argue following Heim (1991) that the binding facts mean I and you cannot be dedicated indexicals; i.e., they only have an indexical function in some contexts. 78 (38) a. nimin\u00C3\u00B4sim ni- min\u00C3\u00B4s -im 1- cat -DISJ \u00E2\u0080\u0098my cat\u00E2\u0080\u0099 b. nin\u00C3\u00AAstosin ni-n\u00C3\u00AAstosi -n 1- tired.VAI-SAP \u00E2\u0080\u0098I am tired.\u00E2\u0080\u0099 In Plains Cree, the pronominal forms are dedicated indexicals, where pronominal forms in English are not, as can be seen by looking at the contexts in which first- and second person may function as variables (cf. Heim 1991, Partee 1989, Kratzer 1998, Rullmann 2003, 2004, among others). For example, for at least some speakers of English, the second \u00E2\u0080\u0098I\u00E2\u0080\u0099 in (39) is a bound- variable; in ellipsis contexts it can be bound by the higher subject (e.g., John). (39) Only I got a question I understood; John didn\u00E2\u0080\u0099t. = (i) John didn\u00E2\u0080\u0099t get a question John understood. Bound-variable = (ii) John didn\u00E2\u0080\u0099t get a question I understood. Indexical (adapted from Rullmann 2004) In Plains Cree however, the bound variable reading of \u00E2\u0080\u0098I\u00E2\u0080\u0099 and \u00E2\u0080\u0098you\u00E2\u0080\u0099 must be represented by anaphoric CONJUNCT agreement. Ellipsis is done as in (40), with the contrastive conjunction m\u00C3\u00A2ka \u00E2\u0080\u0098but\u00E2\u0080\u0099, matrix negation m\u00C3\u00B4ya, and the subject Jeff17: (40) niya niw\u00C3\u00A2pam\u00C3\u00A2w wacask, m\u00C3\u00A2ka m\u00C3\u00B4ya Jeff. niya ni-w\u00C3\u00A2pam-\u00C3\u00A2-w wacask m\u00C3\u00A2ka m\u00C3\u00B4ya J 1.EMPH 1see.VTA-DIR-3 muskrat but NEG J \u00E2\u0080\u0098I saw a muskrat but Jeff didn\u00E2\u0080\u0099t (see a muskrat).\u00E2\u0080\u0099 In order to get a bound-variable reading of 1st-person, the CONJUNCT form in (41a) is used; when this utterance was presented to the consultant, the bound-variable reading was volunteered, and the consultant strongly dispreferred the non-bound-variable reading. When the INDEPENDENT form is substituted, the utterance is ruled ungrammatical \u00E2\u0080\u0093 in other words, contexts that allow a bound variable reading prohibit indexical clauses. 17 Ellipsis structures have not, to my knowledge, been discussed in the Plains Cree literature previously. 79 (41) a. niya niw\u00C3\u00A2pam\u00C3\u00A2w atim k\u00C3\u00A2-nitonak, m\u00C3\u00A2ka m\u00C3\u00B4ya Jeff. CONJUNCT niya ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w atim k\u00C3\u00A2- niton -ak m\u00C3\u00A2ka m\u00C3\u00B4ya J 1.EMPH 1- see.VTA -DIR-3 dog C2- look.for.VTA-1>3 but NEG J = \u00E2\u0080\u0098I saw the dog I was looking for, but Jeff didn\u00E2\u0080\u0099t see the dog he was looking for.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 ? \u00E2\u0080\u0098I saw the dog I was looking for, but Jeff didn\u00E2\u0080\u0099t see the dog I was looking for.\u00E2\u0080\u0099 b. * niya niw\u00C3\u00A2pam\u00C3\u00A2w atim niniton\u00C3\u00A2w, m\u00C3\u00A2ka m\u00C3\u00B4ya Jeff. INDEPENDENT niya ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w atim ni- niton -\u00C3\u00A2 -w m\u00C3\u00A2ka m\u00C3\u00B4ya J 1.EMPH 1- see.VTA -DIR-3 dog 1- look.for.VTA-DIR-3 but NEG J --- Likewise, the counterpart examples discussed by Lakoff (1968) are obligatorily translated into the CONJUNCT order. For example, in (42), the dreamer is dreaming that s/he is someone else, and as that other person, kisses the dreamer. In such a context, the English I in \u00E2\u0080\u0098I kissed myself\u00E2\u0080\u0099 is not indexical. The indexical clause in Plains Cree is ruled ungrammatical. (42) context: Speaker is describing a dream in which s/he was someone else (Bridget Bardot), and as that other person, kisses the dreamer a. CONJUNCT \u00C3\u00AA-pakwatam\u00C3\u00A2n aw\u00C3\u00A2s-tipskaw Bridget Bardot \u00C3\u00AAsa niya \u00C3\u00AAkwa \u00C3\u00AA-oc\u00C3\u00AAmisoy\u00C3\u00A2n \u00C3\u00AA- pakw\u00C3\u00A2tam -\u00C3\u00A2n awas-tipisk\u00C3\u00A2w BB \u00C3\u00AAsa niya \u00C3\u00AAkwa \u00C3\u00AA- oc\u00C3\u00AAmiso -y\u00C3\u00A2n C1-dream.VTI -1 last-night BB EVID 1.PRO and C1-kiss.VAI -1 \u00E2\u0080\u0098I dreamt I was Bridget Bardot, and I kissed myself.\u00E2\u0080\u0099 b. INDEPENDENT ! \u00C3\u00AA-pakw\u00C3\u00A2tam\u00C3\u00A2n awas-tipskaw Bridget Bardot \u00C3\u00AAsa niya ekwa nitoc\u00C3\u00AAmison \u00C3\u00AA- pakw\u00C3\u00A2tam -\u00C3\u00A2n awas-tipisk\u00C3\u00A2w BB \u00C3\u00AAsa niya \u00C3\u00AAkwa \u00C3\u00AA- oc\u00C3\u00AAmiso -y\u00C3\u00A2n C1-dream.VTI -1 last-night BB EVID 1.PRO and C1-kiss.VAI -1 --- comment: \u00E2\u0080\u0098I\u00E2\u0080\u0099m not familiar with this [nitoc\u00C3\u00AAmison] form.\u00E2\u0080\u0099 Thus, the pronominal proclitics in Plains Cree have a fixed reference: they do not introduce a bound-variable dependency. 3.2.3.2 The absence of third-person proclitics Unlike 1st-and 2nd-person pronominals, which are deictic on the the speech act, third-person pronominals such as her/him/she/he/it must be assigned reference by an antecedent or by some 80 corresponding deixis (i.e., pointing) (cf. Postal 1969, Ritter 1995, Heim & Kratzer 1998, D\u00C3\u00A9chaine & Wiltschko 2002). They are also only accidentally part of the speech situation \u00E2\u0080\u0093 there is nothing in the features of a third person that necessarily links them to a speech act. In possession, the dependence of third-person pronominals can be seen by the infelicity of uttering omin\u00C3\u00B4sima \u00E2\u0080\u0098his/her cat\u00E2\u0080\u0099 without specifying the antecedent of o- (e.g., Fred in (43)). (43) #(Fred) w\u00C3\u00A2pam\u00C3\u00AAw omin\u00C3\u00B4sima w\u00C3\u00A2pam-\u00C3\u00AA-w F o- min\u00C3\u00B4s -im -a see.VTA-DIR-3 F 3-cat -DISJ-OBV \u00E2\u0080\u0098Fred saw his cat\u00E2\u0080\u0099 However, notice that the o- is necessary to show the relation between Fred and min\u00C3\u00B4s (44). The prefix o- thus codes reference, but does not identify the referent. (44) * Fred w\u00C3\u00A2pam\u00C3\u00AAw min\u00C3\u00B4sima F w\u00C3\u00A2pam -\u00C3\u00AA -w min\u00C3\u00B4s -im -a F see.VTA-DIR-3 cat -DISJ-OBV --- This means that the 3rd-person prefix o- should not be present in an indexical clause, where it would be located in spec, CP. This is correct. (45) a. Fred n\u00C3\u00AAstosiw F n\u00C3\u00AAstosi -w F tired.VAI-3 \u00E2\u0080\u0098Fred is tired.\u00E2\u0080\u0099 b. * Fred on\u00C3\u00AAstosiw F o- n\u00C3\u00AAstosi -w F 3- tired.VAI-3 --- Notice that this argument does not say that there are no third-person arguments in INDEPENDENT clauses (notice that 45a is completely grammatical), or even that there is no agreement for third person arguments (45a has the third-person subject agreement suffix -w). Rather, the point is 81 that there is no third-person pronominal form sitting in spec, CP. Since this is the position occupied by the indexical speech situation, this is the position that is of relevance here18. (46) CP 2 ni- 2 ki- 5 * o- This raises the question of how third persons in general (i.e., apart from the pronominal proclitic) behave in indexical clauses, which I turn to next. 3.2.3.3 Referents are deictically anchored in indexical clauses We have so far considered the properties of first- and second-person referents, and I have shown that they are always indexical in indexical clauses. I have not so far considered the properties of 3rd-person referents. In particular, while 1st and 2nd person referents, who I take to be speech act participants (cf. Fillmore 1975, Benviniste 1950) and thus licensed by the s0 speech situation, 3rd person referents have no indexical properties since they are not speech act participants. The current analysis therefore predicts a different structure for third-person referents in indexical vs. non-indexical clauses, with accompanying distributional and interpretation differences. There are two difficulties with testing this prediction: first, there is in general a lack of criteria that could be used to distinguish different kinds of referents; second, reference-tracking in Plains Cree specifically is not at all well-understood. What I have to say here will be tentative in nature; this is a huge topic for further research. Minimally we see that the forms used to refer to third-person referents differ between indexical INDEPENDENT clauses and anaphoric CONJUNCT ones: in the former we have the suffix - w, and the latter we have -t (3rd.animate) and -k (3.inanimate). Of course until we have an idea 18 This predicts that Algonquian languages which do have the third-person prefix (e.g., Ojibwa, Blackfoot) would either lack the indexical/anaphoric distinction described here for Plains Cree, or that the third-person prefix would on independent grounds be in some other position (e.g., in IP). 82 about the semantics of these agreement markers, this does not provide evidence of what the difference is: but at least these facts are consistent with my claim. Let us suppose that referents can be defined over some unit of discourse (i.e., they are the topic of that unit) (cf. Longacre 1979, Fox 1987a, b, Smith 2003 for the correspondence between reference-tracking and topicality in English and other languages; see also the discussion of reference tracking in centering theory: Brennan, Friedman & Pollard 1987, Grosz, Joshi & Weinstein 1995, etc.). We could then say that INDEPENDENT clauses only pick out referents which are topics; if the referent they refer to is not a topic, an overt nominal will be necessary. There is some preliminary evidence that this is on the right track, although much more work would need to be done to work out this analysis in detail. I discuss three cases from the corpora that I have found: (i) cases where the indexical INDEPENDENT clause uses an overt nominal even when the nominal was also present in the previous CONJUNCT clause; (ii) cases where an indexical INDEPENDENT clause requires re-introduction of a referent (via an overt nominal) in a subsequent clause; and (iii) cases where an indexical INDEPENDENT clause has no overt nominals, but refers to the main characters of a story. I also consider the properties of the disjoint-subject marker -yi, and show that it has only occurs in two restricted environments in indexical INDEPENDENT clauses, both of which are consistent with the analysis of INDEPENDENT clauses. While the data presented here is not conclusive, and by no means a full account of overt nominals, it may serve as a step towards understanding the structure of discourse in Plains Cree. 3.2.3.3.1 Distribution of overt nominals in a discourse The first piece of evidence suggesting that indexical INDEPENDENT clauses do not contain anaphoric reference to referents comes from data like (47). Here we find a CONJUNCT order clause accompanied by the overt nominal awa n\u00C3\u00A2p\u00C3\u00AAsis \u00E2\u0080\u0098little boy\u00E2\u0080\u0099; this is followed by an indexical INDEPENDENT clause, which is also accompanied by the same overt noun phrase (awa n\u00C3\u00A2p\u00C3\u00AAsis \u00E2\u0080\u0098the little boy\u00E2\u0080\u0099). 83 (47) overt nominal + CONJUNCT \u00E2\u0080\u00A6, awa n\u00C3\u00A2p\u00C3\u00AAsis \u00C3\u00AAkwa awa \u00C3\u00AA-niht\u00C3\u00A2wikit, awa n\u00C3\u00A2p\u00C3\u00AAsis \u00C3\u00AAkwa awa \u00C3\u00AA- niht\u00C3\u00A2wiki -t DEM.AN boy and DEM.AN C1-born.VAI -3 \u00E2\u0080\u0098\u00E2\u0080\u00A6, and when the little boy was born, overt nominal + INDEPENDENT nipamih\u00C3\u00A2w m\u00C3\u00A2n \u00C3\u00A2wa n\u00C3\u00A2p\u00C3\u00AAsis \u00C3\u00AA-~ \u00C3\u00AA-~, \u00E2\u0080\u00A6 ni- pamih -\u00C3\u00A2 -w m\u00C3\u00A2na awa n\u00C3\u00A2p\u00C3\u00AAsis 1- care.VTA-DIR-3 usually DEM.AN boy I would look after him too.\u00E2\u0080\u0099 (AA 5.5) Based on observations made about English and theories of anaphora, we would expect the main clause to be fine without this referent (cf. Ross 1967, Langacker 1969, Reinhart 1976, 1983, and the accompanying English translation). Based on the analysis of indexical clauses, however, we expect this kind of \u00E2\u0080\u0098repetition\u00E2\u0080\u0099, since the reference to a third-person should be defined within the clause (i.e., by the overt nominal). Another suggestive piece of evidence has to do with what happens when an overt nominal is introduced by an indexical INDEPENDENT clause. Here the following INDEPENDENT clauses can also refer to awa n\u00C3\u00A2p\u00C3\u00AAsis \u00E2\u0080\u0098the little boy\u00E2\u0080\u0099, but only as long as each indexical INDEPENDENT clause successively refers to him. Thus, in (48), all three indexical INDEPENDENT clauses refer to him. (48) ..., nipamih\u00C3\u00A2w m\u00C3\u00A2n \u00C3\u00A2wa n\u00C3\u00A2p\u00C3\u00AAsis \u00C3\u00AA-~ \u00C3\u00AA-~, k\u00C3\u00A2-sipw\u00C3\u00AAht\u00C3\u00AAcik m\u00C3\u00A2n \u00C3\u00B4ki, ni- pamih -\u00C3\u00A2 -w m\u00C3\u00A2na awa n\u00C3\u00A2p\u00C3\u00AAsis k\u00C3\u00A2-sipw\u00C3\u00AAht\u00C3\u00AA -t -k m\u00C3\u00A2na \u00C3\u00B4ki 1- care.VTA-DIR-3 usually DEM.AN boy C2-leave.VAI -3-PL usually DEM.AN \u00E2\u0080\u0098..., I would look after him too, when they [his parents] went out, niya m\u00C3\u00A2na nikanaw\u00C3\u00AAyim\u00C3\u00A2w \u00C3\u00AAkwa \u00C3\u00AA-pamihak. niya m\u00C3\u00A2na ni- kanaw\u00C3\u00AAyim -\u00C3\u00A2 -w \u00C3\u00AAkwa \u00C3\u00AA- pamih -ak 1.EMPH usually 1- care.VTA -DIR-3 and C1-care.VTA-1>3 I kept him and looked after him. \u00C3\u00AAkosi piyis aci-misikicisiw, ... \u00C3\u00AAkosi piyis aci- misikicisi -w TOPIC finally INCEP-big.VAI -3 So at last he was getting quite big, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 5.5-6) 84 However, when the speaker then switches to clauses referring only to herself, and then mentions the boy again in the INDEPENDENT, an overt nominal is again used. The following example picks up immediately after the last example left off. The INDEPENDENT clauses in the first line (bolded) refer only to the speaker; the INDEPENDENT clause at-\u00C3\u00B4hpikiw \u00E2\u0080\u0098he was growing up\u00E2\u0080\u0099 in the second line (bolded) refers to the little boy again, and has the overt nominal phrase n\u00C3\u00A2p\u00C3\u00AAsis awa \u00E2\u0080\u0098the little boy\u00E2\u0080\u0099 (bolded and underlined). (49) \u00E2\u0080\u00A6, \u00C3\u00AAkwa \u00C3\u00AAkos \u00C3\u00AAkwa nik\u00C3\u00AE-p\u00C3\u00B4natosk\u00C3\u00A2n \u00C3\u00AAkwa n\u00C3\u00AEt\u00C3\u00AA k\u00C3\u00A2wi nik\u00C3\u00AE-isi-k\u00C3\u00AEw\u00C3\u00A2n. \u00C3\u00AAkwa \u00C3\u00AAkosi \u00C3\u00AAkwa ni-k\u00C3\u00AE- p\u00C3\u00B4natosk\u00C3\u00AA -n \u00C3\u00AAkwa n\u00C3\u00AEt\u00C3\u00AA k\u00C3\u00A2wi ni- k\u00C3\u00AE- isi- k\u00C3\u00AEw\u00C3\u00AA -n and TOPIC and 1- PREV-stop.work.VAI-SAP and there again 1- PREV-thus-gohome.VAI-SAP \u00E2\u0080\u0098\u00E2\u0080\u00A6, and then I quit my job and went back home over there. k\u00C3\u00AEtahtaw\u00C3\u00AA k\u00C3\u00A2-p\u00C3\u00AEhtam\u00C3\u00A2n aya (-- at-~ at-\u00C3\u00B4hpikiw awa n\u00C3\u00A2p\u00C3\u00AAs-~, n\u00C3\u00A2p\u00C3\u00AAsis awa, ... k\u00C3\u00AAtahtaw\u00C3\u00AA k\u00C3\u00A2- p\u00C3\u00AAhtam -\u00C3\u00A2n aya ati- \u00C3\u00B4hpiki -w awa n\u00C3\u00A2p\u00C3\u00AAsis awa suddenly C2- hear.VTI-1 CONN INCEP-grow.VAI-3 DEM.AN boy DEM.AN Later I heard (the little boy was growing up, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 5.6) Here, it appears that once reference has been shifted away from the boy, it cannot be picked up again with an indexical INDEPENDENT order clause; an overt nominal phrase is used to re- establish the referent. The same pattern happens again in (50) (taken from later in the same corpus). Here an indexical INDEPENDENT clause is accompanied by an overt nominal phrase an \u00C3\u00AEskw\u00C3\u00AAw \u00E2\u0080\u0098that woman\u00E2\u0080\u0099, and the next indexical INDEPEPENDENT clause also refers to her. In the third and fourth lines, we have indexical clauses which refer only to the speaker19, and when the next indexical clause refers again to the woman, the overt nominal phrase is re-used. (50) \u00E2\u0080\u00A6, \u00C3\u00AAkwa nik\u00C3\u00AE-~ k\u00C3\u00AE-\u00C3\u00A2cim\u00C3\u00A2w an \u00C3\u00AEskw\u00C3\u00AAw an \u00C3\u00A2yi, \u00C3\u00AA-asiwasot \u00C3\u00B4t\u00C3\u00AA Battleford, \u00C3\u00AAkwa k\u00C3\u00AE- \u00C3\u00A2cim -\u00C3\u00A2 -w ana iskw\u00C3\u00AAw ana ayi \u00C3\u00AA- asiwaso -t \u00C3\u00B4t\u00C3\u00AA B and PREV-tell.VTA-DIR-3 DEM.AN woman DEM.AN CONN C1-be.locked.up.VAI-3 there B \u00E2\u0080\u009C\u00E2\u0080\u00A6, and then that woman was said to be locked up over there at Battleford, \u00C3\u00AAtikw\u00C3\u00AA k\u00C3\u00AE-~ k\u00C3\u00AE-k\u00C3\u00AEskw\u00C3\u00AAy\u00C3\u00AAyihtam, \u00C3\u00AAtikw \u00C3\u00A2nim \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00A2skisw\u00C3\u00A2t on\u00C3\u00A2p\u00C3\u00AAma. \u00C3\u00AAtikw\u00C3\u00AA k\u00C3\u00AE- k\u00C3\u00AEskw\u00C3\u00AAy\u00C3\u00AAyihtam -w \u00C3\u00AAtikwe anima \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00A2skisw -\u00C3\u00A2 -t o- n\u00C3\u00A2p\u00C3\u00AAm -a DUBIT PREV-crazy.VTI -3 DUBIT DEM.INAN C1-PREV-kill.VTA -DIR-3 3-husband-OBV she must have gone mad, I suppose, upon killing her husband. 19 I do not count the clauses in the direct quote, since those are part of a separate discourse (cf. Banfield 1982). 85 \u00C3\u00AA-w\u00C3\u00AE-~ \u00C3\u00AA-~ nik\u00C3\u00AE-itoht\u00C3\u00A2n \u00C3\u00AA-nitawi-nitonawak, ni-k\u00C3\u00AE- itoht\u00C3\u00A2 -n \u00C3\u00AA- nitawi- nitonaw -ak 1-PREV-go.VAI-SAP C1-go- look.for.VTA-1>3 I went there to go and look for her, \u00C2\u00ABmw\u00C3\u00A2c, m\u00C3\u00B4y \u00C3\u00B4ta ay\u00C3\u00A2w, m\u00C3\u00B4y \u00C3\u00B4hci-pim\u00C3\u00A2tisiw,\u00C2\u00BB nik\u00C3\u00AE-itikawin, namw\u00C3\u00A2c m\u00C3\u00B4ya \u00C3\u00B4ta ay\u00C3\u00A2 -w m\u00C3\u00B4y ohci- pim\u00C3\u00A2tisi -w ni-k\u00C3\u00AE- it -ikawi -n NEG NEG here be.VAI-3 NEG PREV-live.VAI -3 1- PREV-say.VTA-USC -SAP and I was told \u00E2\u0080\u009CNo, she is not here, she has died,\u00E2\u0080\u009D \u00C3\u00AAkosi m\u00C3\u00B4y n\u00C3\u00B4h-w\u00C3\u00A2pam\u00C3\u00A2w an \u00C3\u00AEskw\u00C3\u00AAw. \u00C3\u00AAkosi m\u00C3\u00B4y n- \u00C3\u00B4h- w\u00C3\u00A2pam -\u00C3\u00A2 -w ana iskw\u00C3\u00AAw topic NEG 1-PREV-see.VTA-DIR-3 DEM.AN woman and so I never did get to see that woman.\u00E2\u0080\u009D (AA 5.7) The final set of data I talk about has to do with topics of the story. M\u00C3\u00BChlbauer (2007) argues that when we look at the distribution of nominals in a Plains Cree discourse, we find that there are two different kinds of nominals, introduced at two different stages of the discourse. In the initial stage of a discourse, speakers introduce a set of referents, and identify the relation of each referent to the speaker, via a kin-term, some intermediate individual, or by shared space/time. In the second stage of a discourse, the speaker talks about the events surrounding these referents (cf. also Janzen 2004 for similar patterns in American Sign Language). Referents that are introduced in the intial stage I will refer to as topic referents. For example, in the following piece of narrative, taken from an earlier point of the same narrative as the past two pieces of data) the narrator (Alice Ahenakew) starts with an intransitive verb in the INDEPENDENT order, then introduces her relation to the two \u00E2\u0080\u0098main characters\u00E2\u0080\u0099 m\u00C3\u00B4niy\u00C3\u00A2wak \u00E2\u0080\u0098white people\u00E2\u0080\u0099 via the transitive verb \u00C3\u00AA-k\u00C3\u00AE-atoskawakik \u00E2\u0080\u0098I worked for them.\u00E2\u0080\u0099 (51) \u00C3\u00AAkwa \u00C3\u00B4t\u00C3\u00AA m\u00C3\u00AEna m\u00C3\u00A2na nik\u00C3\u00AE-atosk\u00C3\u00A2n \u00C3\u00B4t\u00C3\u00AA isi, \u00C3\u00AAkwa \u00C3\u00B4t\u00C3\u00AA m\u00C3\u00AEna m\u00C3\u00A2na ni- k\u00C3\u00AE- atosk\u00C3\u00AA -n \u00C3\u00B4t\u00C3\u00AA isi and here also usually 1- PREV-work.VAI-SAP here DIR \u00E2\u0080\u0098And then I also used to work over in this direction, m\u00C3\u00B4niy\u00C3\u00A2wak \u00C3\u00AA-k\u00C3\u00AE-atoskawakik. m\u00C3\u00B4niy\u00C3\u00A2w -ak \u00C3\u00AA- k\u00C3\u00AE- atosk-aw -ak -ik white -PL C1-PREV-work-BEN-1>3-PL I used to work for White people.\u00E2\u0080\u0099 (AA 5.5) These two main characters are then identified independently of any events in the story, as given in (52). 86 (52) p\u00C3\u00AAyak ana, aya, Irish an[a] \u00C3\u00AEskw\u00C3\u00AAw, \u00C3\u00AAkwa ana n\u00C3\u00A2p\u00C3\u00AAw German, ... p\u00C3\u00AAyak ana aya Irish ana iskw\u00C3\u00AAw \u00C3\u00AAkwa ana n\u00C3\u00A2p\u00C3\u00AAw G one DEM.AN CONN Irish DEM.AN woman and DEM.AN man G \u00E2\u0080\u0098The one, the woman was Irish, and the man was German,\u00E2\u0080\u0099 (AA 5.5) And then the narrator proceeds to tell the story of working for these two people, and how, eventually, some time after she stopped working for them, the woman ends up killing her husband. Several other referents come into this story at different points: e.g., the couple\u00E2\u0080\u0099s children (a boy and two girls), several other nouns are also used; however the story centers around these two individuals. These I call the topic referents of the story. In fact, other than the instance we saw above with the awa n\u00C3\u00A2p\u00C3\u00AAsis \u00E2\u0080\u0098little boy\u00E2\u0080\u0099, all other instances of INDEPENDENT clauses in this narrative that lack overt nominals for all their arguments refer to one of these two referents. There are three such cases, distributed throughout the story. The first one refers to the woman and has no nominal phrases outside the verbal marking (53). (There is no overt nominal in the previous clause either.) (53) INDEPENDENT refers to text-level woman \u00E2\u0080\u00A6, nitaw-\u00C3\u00B4p\u00C3\u00AAp\u00C3\u00AEmiw \u00C3\u00AAkwa ... nitaw- op\u00C3\u00AAp\u00C3\u00AEmi -w \u00C3\u00AAkwa go- have.baby.VAI-3 and \u00E2\u0080\u00A6, she went to have her baby, \u00E2\u0080\u00A6 (AA 5.5) The second case is a transitive verb with the man as the subject and the woman as the object. Here there is a possessive form w\u00C3\u00AEwa \u00E2\u0080\u0098his wife\u00E2\u0080\u0099 referring to the woman as the man\u00E2\u0080\u0099s wife. While the demonstrative ana \u00E2\u0080\u0098that.AN\u00E2\u0080\u0099 refers to the man, the demonstrative without an accompanying noun does not have enough information to identify the referent. The man has not been brought up since line 2. (54) INDEPENDENT refers to text-level man ..., \u00C3\u00AAkot\u00C3\u00AA \u00C3\u00AAkwa itohtah\u00C3\u00AAw ana w\u00C3\u00AEwa; \u00C3\u00AAkot\u00C3\u00AA \u00C3\u00AAkwa itohtah -\u00C3\u00AA -w ana w- \u00C3\u00AEw -a there and take.VTA-DIR-3 DEM.AN 3- wife-OBV \u00E2\u0080\u0098\u00E2\u0080\u00A6, then that man took his wife there;\u00E2\u0080\u0099 (AA 5.5) Finally, the last line of the story contains just a bare INDEPENDENT clause. The verb is transitive, and is the punch line of the story: the woman introduced at the beginning of the story has killed 87 the man introduced at the beginning of the story (notice, for example, the emphatic flavour of the English with the affirmative did. (55) INDEPENDENT refers to both text-level individuals \u00E2\u0080\u00A6, k\u00C3\u00AE-nipah\u00C3\u00AAw. k\u00C3\u00AE- nipah -\u00C3\u00AA -w PREV-kill.VTA-DIR-3 \u00E2\u0080\u00A6, and she did kill him.\u00E2\u0080\u0099 (AA 5.6) To sum up then \u00E2\u0080\u0093 INDEPENDENT clauses referring to third persons seem to have a specific set of properties with them. First, they will repeat overt nominals even when the relevant referent was introduced in the previous anaphoric clause. Second, a referent introduced in an indexical clause is reintroduced after an intervening indexical clause. Third, other indexical INDEPENDENT clauses lacking overt nominals correspond to referents that are the main \u00E2\u0080\u0098topics\u00E2\u0080\u0099 of the story. These observations provide an opening into much further research. 3.2.3.3.2 Restrictions on switch-reference in INDEPENDENT clauses Plains Cree has a switch-reference marker -yi, which marks a subject disjoint from an argument of some other predicate (different subject (DS); Dahlstrom 1991, M\u00C3\u00BChlbauer 2007, in prep). Because -yi requires that arguments across two predicates be evaluated, the current analysis predicts that -yi in indexical INDEPENDENT clauses will be either ill-formed or have a deictic function. The latter prediction is borne out and actually predicts a pattern has not generally been noticed in the language. Thus, teaching grammars such as Ahenakew (1987) and Hunter, Karpinski, & Mulder (2001) show, as part of the standard paradigm, independent forms with the -yi suffix (usually termed obviative agreement), but it is not coincidental that all of their examples use a CONJUNCT form of the verb. In fact, a look at Wolfart (1973) shows that, in running speech, -yi is only attested in some of the expected forms in the INDEPENDENT order, (Wolfart 1973:41; Ahenakew 1987 also gives text counts which show that -yi is very restricted in the indexical INDEPENDENT clauses). Table 3.8 replicates Wolfart\u00E2\u0080\u0099s findings. 88 Syntactic context Expected form Attested in INDEPENDENT order? Intransitive DS STEM-yiwa \u00E2\u009C\u0094 DS > 1 ni-STEM-ikoyiwa \u00E2\u009C\u0096 DS > 2 ki-STEM-ikoyiwa \u00E2\u009C\u0096 Transitive DS > 3OBV STEM-\u00C3\u00AAyiwa \u00E2\u009C\u0094 Table 3.8. (Un)attested co-occurrence of -yi in INDEPENDENT order (Wolfart 1973) Even when we limit ourselves to the forms which do occur in running speech, which I will discuss below, it is difficult to reproduce these forms in elicitation contexts for reasons that will become clear below. Consider the following pair, both of which were presented in elicitation. With the anaphoric CONJUNCT form, the consultant found the sentence infelicitous without context, but when asked how it would be interpreted, (e.g., if accidentally overheard) could translate it (56a). By contrast, consultants do not even recognize the form in (56b) - it is uninterpretable. (56) Context: presentation of different subject marking in CONJUNCT and INDEPENDENT forms a. # n\u00C3\u00A2p\u00C3\u00AAwa \u00C3\u00AA-nikamoyit CONJUNCT n\u00C3\u00A2p\u00C3\u00AAw -a \u00C3\u00AA- nikamo -yi -t man -OBV C1-sing.VAI-DS-3 \u00E2\u0080\u0098Someone\u00E2\u0080\u0099s guy is singing.\u00E2\u0080\u0099 comment: who are you talking about? b. ! anihi n\u00C3\u00A2p\u00C3\u00AAwa nikamoyiwa INDEPENDENT anihi n\u00C3\u00A2p\u00C3\u00AAw -a nikamo -yi -w -a dem man -OBV sing.vai-DS -3 -OBV -- comment: I\u00E2\u0080\u0099ve never heard that before. The inability of consultants to recognize the latter form in elicitation contexts highlights the difference between indexical clauses, and anaphoric clauses, which even when unembedded, can be interpreted with respect to some previous antecedent. When we look at running speech, we do see INDEPENDENT clauses with the -yi suffix on them \u00E2\u0080\u0093 but only in two specific contexts (M\u00C3\u00BChlbauer 2007, in prep). The first is when the subject of the verb is possessed, as in (57). 89 (57) Clare om\u00C3\u00A2m\u00C3\u00A2wa ka-nikamoyiwa INDEPENDENT C o-m\u00C3\u00A2m\u00C3\u00A2 -a ka- nikamo -yi -w -a C 3-mother-obv IRR-sing.VAI-DS -3 -OBV \u00E2\u0080\u0098Clare\u00E2\u0080\u0099s mother will sing.\u00E2\u0080\u0099 In this case, there is internal structure of the DP subject Clare om\u00C3\u00A2m\u00C3\u00A2wa \u00E2\u0080\u0098Clare\u00E2\u0080\u0099s mother\u00E2\u0080\u0099; in particular, there are two referents: Clare and om\u00C3\u00A2m\u00C3\u00A2wa \u00E2\u0080\u0098her mother\u00E2\u0080\u0099. The different subject marks disjunction between the subject of the verb om\u00C3\u00A2m\u00C3\u00A2wa \u00E2\u0080\u0098her mother\u00E2\u0080\u0099 and the subject of the possession construction Clare as represented in (58); crucially, there is no cross-clausal antecedence relation between the two subjects. (58) -yi: (Subjom\u00C3\u00A2m\u00C3\u00A2wa \u00E2\u0089\u00A0 SubjClare ) XP 3 DPi IP 2 2 DP2 DP1 2 5 5 -yi XP Clare om\u00C3\u00A2m\u00C3\u00A2wa This context accounts for almost all -yi marked INDEPENDENT clauses, and it should be underlined that this is the only context I know of where a consultant has accepted a -yi marked INDEPENDENT clause in an elicitation context. The second context where -yi can occur is when it marks what again look like text-level referents. For example, in (59), the speaker is telling a joke about a dead prairie-chicken found by a woman going to church. The prairie chicken is introduced as an obviative referent relative to the woman, and is the referent on which the joke hangs. At the point of the story where this clause is uttered, there are only two possible referents in the discourse: n\u00C3\u00B4cikw\u00C3\u00AAsiw \u00E2\u0080\u0098old woman\u00E2\u0080\u0099 and pih\u00C3\u00AAwa \u00E2\u0080\u0098(obviative) prairie chicken\u00E2\u0080\u0099. Further, previous to this clause, there have been no switching of subjects between clauses. In this situation, then, we have an utterance like (59).20 20 When these kinds of examples are presented to speakers in elicitation contexts, they are accepted as fine, but when the speakers are asked to reproduce them, the context seems to evaporate and the different subject marking disappears. 90 (59) context: middle of story about woman and prairie chicken ..., m\u00C3\u00A2k \u00C3\u00AAtikw\u00C3\u00AA miy\u00C3\u00A2kosiyiwa, ... m\u00C3\u00A2ka \u00C3\u00AAtikw\u00C3\u00AA miy\u00C3\u00A2kosi -yi -w -a but EVID stink.VAI -DS -3 -OBV \u00E2\u0080\u0098..., but it [the prairie chicken] must have been smelling already, ...\u00E2\u0080\u0099 (AA7.1) In this case we can say that the -yi is not anaphorically dependent on the previous clause, but rather it is deictically pointing to the (only) obviative referent in the story. Thus the subordinate subject marker -yi, has deictic behaviour in INDEPENDENT clauses (cf. its anaphoric behaviour in CONJUNCT clauses, discussed in chapter 4). 3.3 The semantics of indexical clauses: Indexicality We now turn from the structural conditions on indexical INDEPENDENT clauses to the implications of the clause being anchored to the speech act (via the situation variable in spec, CP); in particular focussing on the temporal implications (i.e., that indexical INDEPENDENT clauses are evaluated with respect to speech time) and referential implications (i.e., that indexical INDEPENDENT clauses are evaluated with respect to the speaker). First, I show that indexical clauses have a privileged temporal relationship to the speech time (\u00C2\u00A73.3.1). Second, I show that indexical clauses have a privileged referential relationship to the speaker (\u00C2\u00A73.3.2). 3.3.1 Temporal deixis: Relating reference time to speech time In this section, I consider the temporal properties of indexical clauses. Within a Reichenbachian framework (cf. Paul 1886, Reichenbach 1947, Hinrichs 1986, Enc 1987, Hornstein 1990, Kamp & Reyle 1993, Klein 1994, Kratzer 1998, Demirdache & Uribe-Etxebarria 2002, among others), there are traditionally three times: the speech time (aka utterance time), the reference time (aka topic time), and the event time (aka situation time). These are defined in (60) and related to the corresponding linguistic structures. 91 a. Speech Time (T0) the time of speaking speech act b. Reference Time (Tref) the time the sentence makes a claim about propositional structure (60) c. Event time (Tsit) the time of the event or situation predicate structure From these times, a large number of relations can be made to model different tense/aspectual distinctions (see, for example Klein 1994). The relation between these times can be sequencing in nature, i.e., x precedes y, x follows y. It may also be a relation of inclusion: x includes y or conversely, y includes x21. Following recent work in the tense/aspectual literature, I model the relations between these times as a [\u00C2\u00B1 coincidence] relation, where [- coincidence] captures the sequencing relation, and [+ coincidence] captures the inclusion relation (cf. Demirdache & Uribe-Etxebarria 199x, 200x; Ritter & Wiltschko 2005, 2007)22. Tense relations are generally taken to be relations anchored to the speech time, and aspectual relations those anchored to the situation time. In this thesis, since we are interested in clause-typing, I am primarily concerned with the former type of relation. More recently, many analyses (e.g., Gennari 2003, among others) also make reference to an evaluation time, as in (61). (61) Evaluation Time (Teval)def: the time with respect to which the truth of the sentence (i.e., proposition) is evaluated The status of this fourth time with respect to the others is often vague. In this thesis, I integrate the evaluation time into the other times in a specific way, such that, for any clause, there are only 21 In order to relate multiple times, multiple relations are often necessary. For example, the specification of an English past perfect specifies a precedence relation between the reference time and the utterance time, and a precedence relation between the reference time and the situation time (cf. Klein 1994). (i) - COIN (Tref, T0) - COIN (Tref, Tsit) Because these relations can be established (at least semi-)independently, we expect that they could be established via multiple parts of the grammar: e.g., there is no need to assume that a single morpheme would give both the relation between Tref and Tsit and the relation between Tref and T0. 22 The [\u00C2\u00B1 coincidence] feature is in fact an over-simplified analysis, since the precedence relation could also in theory be reversed, where the topic time must precedent the situation time, yielding a future tense (Kamp & Rohrer 1983). To distinguish the future from the past, we therefore need some additional specification. While there is on independent grounds substantial agreement that the future needs a modal component (Jespersen 1924, Comrie 1985, Hornstein 1990, Abusch 1998, Copley 2002, Matthewson 2006, among others), the modal component is again not sufficient to derive the temporal properties of the future, and the ordering component of the future must be specified as [FOLLOW]. Thus an alternative analysis fully compatible with the data here is to specify the relations as [OVERLAP] and [PRECEDENCE]. 92 three relevant times. Specifically, I model the relation often characterized as a relation between reference time and speech time as a relation between reference time and evaluation time. (62) Tense relations: + COIN (Tref, Teval) - COIN (Tref, Teval) What the indexical/non-indexical split in clauses does is give the value of the evaluation time Teval. Within the situation semantics framework, all of these times will be derived from situations. Here I focus on the \u00E2\u0080\u0098times\u00E2\u0080\u0099 aspect, and use the \u00E2\u0080\u0098times\u00E2\u0080\u0099 notation. But if indexical INDEPENDENT clauses are evaluated with respect to a speech situation (s0), then the time they will be evaluated with respect to (i.e., the evaluation time) is the speech time T0. (63) s0 \u00EF\u0083\u00A0 T0 By transitivity, in indexical clauses the tense relation will always be between the reference time and the speech time. (64) Indexical: Evaluation time is speech time (Teval = T0) + COIN (Tref, T0) - COIN (Tref, T0) In anaphoric CONJUNCT clauses, by contrast, the proposition is evaluated with respect to some (unspecified) situation; thus it will be evaluated with respect to some (unspecified) time (in chapter 4, I argue that this time is established according to general principles of anaphora). In anaphoric CONJUNCT clauses, the tense relations will therefore be as in (65). (65) Anaphoric: Evaluation time is anaphorically give (Teval = T) + COIN (Tref, T) - COIN (Tref, T) The definition in (64) captures the fixed reference to speech time that indexical clauses have; in \u00C2\u00A73.2.2.4 I use the distinction between (64) and (65) to capture the different interpretations of indexical (Plains Cree INDEPENDENT) and anaphoric (Plains Cree CONJUNCT) clauses. 93 The second part of the claim is that a bare indexical clause in Plains Cree has a [+ coincidence] value;23 in order to get a [- coincidence] value, the temporal shifting preverb k\u00C3\u00AE- must be added to the verbal complex. Putting everything together, the temporal value of a bare INDEPENDENT order clause in Plains Cree is thus as in (66). (66) + COIN (Tref, T0) Thus, while bare INDEPENDENT clauses do not map directly onto either of the English past/present tense distinctions in terms of distribution, they are more like a present tense in that the reference time coincides with the speech time. This captures the generalization that Plains Cree\u00E2\u0080\u0099s INDEPENDENT clauses always have \u00E2\u0080\u0098present relevance\u00E2\u0080\u0099, a notion that is also recurrent in the literature for other languages (cf. Huddleston 1969, Lakoff 1970, Bennett & Partee 1972, Comrie 1976, Dowty 1979, Klein 1992, 1994, among others). There are at least two possible interpretations that the [+ COIN (Tref, T0)] relation is compatible with, depending on what the relation between Tevent and Tref is. First, it is compatible with a \u00E2\u0080\u0098present tense\u00E2\u0080\u0099 interpretation, where Tevent also coincides with Tref. (67) present: \u00E2\u0080\u0098I am happy.\u00E2\u0080\u0099 + COIN (Tref, T0) + COIN (Tref, Tevent) Second, the coincidence relation between Tref and Teval could be conjoined with a non- coincidence relation between Tevent and Tref. This is approximately the model for the English present perfect (Klein 1994): the reference time must be in the posttime (i.e., after) the situation time (i.e., expressed by the predicate), and must also include the utterance time. (68) present perfect \u00E2\u0080\u0098I have eaten three apples.\u00E2\u0080\u0099 + COIN (Tref, T0) - COIN (Tref, Tevent) Since Tevent is associated with the predicate domain and Tref with the propositional domain, and these are both below the clause-typing domain, I will not be concerned with the representation of 23 I leave as an open question whether the unmarked value of [+ coincidence] should be treated as an inherent (i.e., universal) value of indexical clauses, or whether this is specific to Plains Cree\u00E2\u0080\u0099s indexical clause. 94 aspectual value (cf. Klein 1992, Kratzer 1998, Bohnemeyer & Swift 2004). I show that while there is variation between a present and present perfect interpretation (which is indirectly related to aspectual value), there is always a [+ COIN] relation between the reference time Tref and the speech time T0. If the predicate is stative, the state holds at speech time; if the predicate is an activity (i.e. a-telic), the event or resultant state holds at speech time; if the predicate is an accomplishment (i.e., telic), then the resultant state holds at speech time (i.e., like a present perfect). This is summarized in table 3.9. ASPECT CLASS INTERPRETATION Stative predicates present Activity predicates present Telic predicates present perfect Table 3.9. Interpretation of INDEPENDENT clauses by aspectual class 3.3.1.1 Contrasting temporal interpretations of indexical and non-indexical clauses The temporal difference between indexical and non-indexical clauses is represented in (69): while the [+coincidence] relation is indexically given as the speech time in INDEPENDENT order clauses, it is unspecified in anaphoric CONJUNCT clauses, and must be given by context (specifically, it must have an antecedent). (69) a. + COIN (Tref, T0) INDEPENDENT b. + COIN (Tref, T) CONJUNCT Probably the most striking illustration of the anaphoric nature of non-indexical CONJUNCT clauses is that they may be embedded with respect to another clause, in which case they are dependent on the higher clause for temporal interpretation (cf. Chapter 4). In (70) the crying is interpreted relative to when Kim spoke to me, not relative to speech time. (The baby could still be crying, but the sentence says nothing about that.) 95 (70) Kim nik\u00C3\u00AE-itik b\u00C3\u00AAb\u00C3\u00AEsis \u00C3\u00AA-m\u00C3\u00A2tot K ni- k\u00C3\u00AE- it -ikw b\u00C3\u00AAbi -sis \u00C3\u00AA- m\u00C3\u00A2to -t K 1- PREV-tell.VTA-INV baby-DIM C1-cry.VAI-3 \u00E2\u0080\u0098Kim had told me that the baby was crying = + COIN (crying, T; T = TELL) The sentence in (70) tells us that the temporal interpretation of embedded anaphoric (CONJUNCT) clauses can be set with respect to some time that is not the speech time (i.e., it is not fixed). However, even in unembedded contexts, the temporal interpretation of CONJUNCT clauses is not fixed. In (71), both utterances are translated exactly the same way, but they are nevertheless temporally distinguished: the CONJUNCT form can be used to report something you heard earlier, or something cotemporaneous with speech time; the indexical INDEPENDENT has only the cotemporaneous interpretation. (71) a. \u00C3\u00AA-p\u00C3\u00AAhtawak nis\u00C3\u00AEmis wayawihtamihk \u00C3\u00AA-m\u00C3\u00AAtaw\u00C3\u00AAt CONJUNCT \u00C3\u00AA- p\u00C3\u00AAhtaw -ak ni- s\u00C3\u00AEmis wayawihtam -ihk \u00C3\u00AA- m\u00C3\u00AAtaw\u00C3\u00AA -t C1-hear.VTA-1>3 1- sibling outside -LOC C1-play.VAI -3 \u00E2\u0080\u0098\u00E2\u0080\u00A6I heard my little brother playing outside.\u00E2\u0080\u0099 comment: This one could mean any time; it could mean before, or it could mean I\u00E2\u0080\u0099m hearing him now b. nip\u00C3\u00AAhtaw\u00C3\u00A2w nis\u00C3\u00AEmis wayawihtamihk \u00C3\u00AA-m\u00C3\u00AAtaw\u00C3\u00AAt INDEPENDENT ni- p\u00C3\u00AAhtaw -\u00C3\u00A2 -w ni- s\u00C3\u00AEmis wayawihtam-ihk \u00C3\u00AA- m\u00C3\u00AAtaw\u00C3\u00AA -t C1-hear.VTA-DIR-3 1- sibling outside -LOC C1-play.VAI-3 \u00E2\u0080\u0098I heard my little brother playing outside.\u00E2\u0080\u0099 comment: \u00E2\u0080\u00A6like I\u00E2\u0080\u0099m on the phone with you, and he\u00E2\u0080\u0099s making noise, and I\u00E2\u0080\u0099m telling you about it Second, in places where INDEPENDENT clauses must be marked with k\u00C3\u00AE- (e.g., states that hold at some time other than utterance time), CONJUNCT clauses can be bare. The temporal value of the indexical clause is interpreted with respect to speech time, so an overt [- coincidence] element like k\u00C3\u00AE- is needed (72a, 73a). The temporal value of the anaphoric CONJUNCT clause (72b, (72b) is interpreted with respect to the time established by the previous clause: the times of the events coincide so k\u00C3\u00AE- is not needed. 96 (72) a. k\u00C3\u00A2-w\u00C3\u00A2pamak Jeff, k\u00C3\u00AE-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAw INDEPENDENT k\u00C3\u00A2-w\u00C3\u00A2pam -ak J k\u00C3\u00AE- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -w C2-see.VTA-1>3 J PREV-hungry.VAI -3 \u00E2\u0080\u0098When I saw Jeff, he was hungry.\u00E2\u0080\u0099 b. k\u00C3\u00A2-w\u00C3\u00A2pamak Jeff, \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAt CONJUUNCT24 k\u00C3\u00A2-w\u00C3\u00A2pam -ak J \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -t C2-see.VTA-1>3 J C1-hungry.VAI -3 \u00E2\u0080\u0098When I saw Jeff, he was hungry.\u00E2\u0080\u0099 (73) a. k\u00C3\u00A2-w\u00C3\u00A2pamak Jeff, k\u00C3\u00AE-\u00C3\u00A2hkosiwpayiw INDEPENDENT k\u00C3\u00A2- w\u00C3\u00A2pam -ak J k\u00C3\u00AE- \u00C3\u00A2hkosiwpayi-w C2-see.VTA -1>3 J PREV-sick.VAI -3 \u00E2\u0080\u0098When I saw Jeff he got sick (suddenly).\u00E2\u0080\u0099 b. k\u00C3\u00A2-k\u00C3\u00AE-w\u00C3\u00A2pamak Jeff, \u00C3\u00AA-\u00C3\u00A2hkosiwpayit CONJUNCT k\u00C3\u00A2- k\u00C3\u00AE- w\u00C3\u00A2pam -ak J \u00C3\u00AA- \u00C3\u00A2hkosiwpayi -t C2-PREV-see.VTA -1>3 J C1-sick.VAI -3 \u00E2\u0080\u0098When I saw Jeff, he got sick (suddenly).\u00E2\u0080\u0099 Third, in sequences of clauses, if k\u00C3\u00AE- is added to an indexical INDEPENDENT clause, it fixes the relation relative to the speech time, as in (74a, 75a); however if k\u00C3\u00AE- is added to an anaphoric CONJUNCT clause (74b, 75b), it shifts the event time with respect to the preceding clause, rather than with respect to speech time. See chapter 4 for more details. (74) a. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff nik\u00C3\u00AE-m\u00C3\u00AEcison\u00C3\u00A2n. INDEPENDENT \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J ni- k\u00C3\u00AE- m\u00C3\u00AEciso -n\u00C3\u00A2n C1-come-go.home.VAI-3 J 1- PREV-eat.VAI-1PL \u00E2\u0080\u0098When Jeff came, then we ate.\u00E2\u0080\u0099 = come home < eat b. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff, \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEcisoy\u00C3\u00A2hk CONJUNCT \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00AEciso -y\u00C3\u00A2n -k C1-come-go.home.VAI-3 J C1-PREV-eat.vai -1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6Jeff came home, we had eaten.\u00E2\u0080\u0099 (we = speaker & someone else) = eat < come home 24 Stative predicates in unembedded clauses still strongly prefer to be interpreted as holding at utterance time, but as the example shows, this is only a preference when it comes to CONJUNCT clauses. 97 (75) a. k\u00C3\u00A2-w\u00C3\u00A2pamak Jeff, k\u00C3\u00AE-ahkosiwp\u00C3\u00A2yiw INDEPENDENT k\u00C3\u00A2- w\u00C3\u00A2pam -ak J k\u00C3\u00AE- ahkosiwpayi -w C2- see.VTA-1>3 J PREV-sick.VAI -3 \u00E2\u0080\u0098When I saw Jeff he got sick (suddenly) (at that time).\u00E2\u0080\u0099 b. k\u00C3\u00A2-w\u00C3\u00A2pamak Jeff, tontoni \u00C3\u00AA-k\u00C3\u00AE-ahkosiwpayit CONJUNCT k\u00C3\u00A2- w\u00C3\u00A2pam -ak J \u00C3\u00AA- k\u00C3\u00AE- ahkosiwpayi -t C2-see.VTA -1>3 J C1-PREV-sick.VAI -3 \u00E2\u0080\u0098When I saw Jeff, he had gotten really very sick.\u00E2\u0080\u0099 Finally, the asymmetry between INDEPENDENT and CONJUNCT clauses can be seen in structures where the temporal adverbial t\u00C3\u00A2kosih \u00E2\u0080\u0098yesterday\u00E2\u0080\u0099 is in final position (74a). Indexical INDEPENDENT clauses are bad, but non-indexical CONJUNCT clauses are fine. (76) a. ?* kacis p\u00C3\u00AAyakw\u00C3\u00A2w nim\u00C3\u00AEcison ot\u00C3\u00A2kosihk INDEPENDENT kacis p\u00C3\u00AAyakw\u00C3\u00A2w ni- m\u00C3\u00AEciso -n ot\u00C3\u00A2kosihk only once 1- eat.VAI-SAP1 yesterday intended: I only ate once yesterday b. kacis p\u00C3\u00AAy\u00C3\u00A2kw\u00C3\u00A2w \u00C3\u00AA-m\u00C3\u00AEcisoy\u00C3\u00A2n ot\u00C3\u00A2kosihk CONJUNCT kacis p\u00C3\u00AAyakw\u00C3\u00A2w \u00C3\u00AA- m\u00C3\u00AEciso -y\u00C3\u00A2n ot\u00C3\u00A2kosihk only once C1-eat.VAI -1 yesterday \u00E2\u0080\u0098I only ate once yesterday.\u00E2\u0080\u0099 As we will see in the following sections, there is a general prohibition on final past time adverbials with indexical clauses lacking the temporal shifter k\u00C3\u00AE-, across all aspectual classes; this is expected if there is a fixed coincidence relation between reference time and speech time. However, there is no such prohibition on non-indexical CONJUNCT clauses; this is expected if the coincidence relation is between reference time and some unspecified time (i.e., that might be in the past). In this last section I have exemplified the difference between indexical INDEPENDENT clauses and non-indexical CONJUNCT clauses. In the next section, I focus on the temporal interpretations available for indexical clauses in stative and (atelic and telic) eventive predicates. These sections demonstrate the [+coincidence] relation between the reference time Tref and speech time T0. Because the relation is always evaluated relative to speech time T0, these sections also provide additional indirect evidence that the evaluation time for indexical clauses is always T0. 98 3.3.1.2 Indexical clauses present statives that hold at T0 Stative predicates are interpreted as holding at speech time in indexical (INDEPENDENT) clauses, as evidenced by the following characteristics: (i) they cannot be translated with a true past \u00E2\u0080\u0098used to (be)\u00E2\u0080\u0099 constructions; (ii) they cannot be used in explicitly past contexts; (iii) they are obligatory if the time span referred to (either by context or by a modifier) includes the utterance time; and (iv) they cannot co-occur with final past-time adverbials unless there is an overt temporal shifter. First, if Plains Cree was simply vague or ambiguous with respect to temporal relations, we would expect that it could be mapped onto either tense form in English25. However, when translations which unambiguously identify the predicate as not holding at speech time, such as the periphrastic \u00E2\u0080\u0098used to be\u00E2\u0080\u0099 (termed English true past constructions in Lakoff 1970), they are systematically rejected. The preverb k\u00C3\u00AE-, which is often called a \u00E2\u0080\u0098past tense\u00E2\u0080\u0099 (see chapter 7 for discussion), is used if the predicate does not hold at speech time. (77) Permanent stative predicates must have k\u00C3\u00AE- for unambiguous past translation a. Jeff kinosiw J kinosi -w J tall.VAI-3 = \u00E2\u0080\u0098Jeff is tall\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098Jeff used to be tall.\u00E2\u0080\u0099 b. Jeff k\u00C3\u00AE-kinosiw J k\u00C3\u00AE- kinosi -w J PREV-tall.VAI-3 \u00E2\u0089\u00A0 \u00E2\u0080\u0098Jeff is tall.\u00E2\u0080\u0099 = \u00E2\u0080\u0098Jeff used to be tall.\u00E2\u0080\u0099 25 For example, the Plains Cree diminutive -sis is ambiguous with respect to indicating size or quality (i). Notice that both English translations, which are non-ambiguous, are valid. (i) acimosis atimw -sis dog -DIM = \u00E2\u0080\u0098small dog (e.g., Pomeranian)\u00E2\u0080\u0099 = \u00E2\u0080\u0098puppy (e.g., baby German Shepherd)\u00E2\u0080\u0099 99 (78) Temporary stative predicates must have k\u00C3\u00AE- for unambiguous past translation a. n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAw Tomio n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -w T hungry.VAI-3 T = \u00E2\u0080\u0098Tomio is hungry.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098Tomio was/used to be hungry.\u00E2\u0080\u0099 b. k\u00C3\u00AE-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAw Tomio k\u00C3\u00AE- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -w T PREV- hungry.VAI-3 T \u00E2\u0089\u00A0 \u00E2\u0080\u0098Tomio is hungry.\u00E2\u0080\u0099 = \u00E2\u0080\u0098Tomio was/used to be hungry.\u00E2\u0080\u0099 Adding overt past contexts which disambiguate the interpretations (criteria (ii)) confirms these translations. For example, when talking about a person who\u00E2\u0080\u0099s no longer living, a stative predicate must be marked with k\u00C3\u00AE- (cf. Wolfart 1990, 2000:170). The absence of k\u00C3\u00AE- codes that the dead person can still have this state attributed to them, as nicely summarized by one consultant. (79) context: talking about a person who\u00E2\u0080\u0099s no longer living b. # n\u00C3\u00B4hkomip\u00C3\u00A2n kahkiyaw kisk\u00C3\u00AAyihtam n- \u00C3\u00B4hkom -p\u00C3\u00A2n kahkiyaw kisk\u00C3\u00AAyihtam -w 1-grandmother-former all know.VTI -3 \u00E2\u0080\u0098My grandmother (no longer here) knows everything.\u00E2\u0080\u0099 comment: OK if you believe you still have some communication with grandmother\u00E2\u0080\u00A6spiritual [communication], she\u00E2\u0080\u0099s dead, but you still have a relationship with her \u00E2\u0080\u0093 for example, a dream where she speaks to you b. n\u00C3\u00B4hkomip\u00C3\u00A2n kahkiyaw k\u00C3\u00AE-kisk\u00C3\u00AAyihtam n- \u00C3\u00B4hkom -p\u00C3\u00A2n kahkiyaw k\u00C3\u00AE- kisk\u00C3\u00AAyihtam -w 1- grandmother-former all PREV-know.VTI -3 \u00E2\u0080\u0098My grandmother (no longer here) knew everything.\u00E2\u0080\u0099 100 (80) context: talking about a friend who recently died a. # John kinosiw J kinosi -w J tall.VAI-3 \u00E2\u0080\u0098John is tall.\u00E2\u0080\u0099 b. John k\u00C3\u00AE-kinosiw J k\u00C3\u00AE- kinosi -w J PREV-tall.VAI-3 \u00E2\u0080\u0098John was tall.\u00E2\u0080\u0099 Likewise, in (81), the state being talked about (living in a particular house) held for some previous time span that does not include the time of speech. Again the bare indexical clause cannot be used. (81) context: walking by a house that speaker used to live in. Speaker points it out a. # niw\u00C3\u00AEkin \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwaskiy ni- w\u00C3\u00AEki -n \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwo askiy 1- live.VAI-SAP here DEM.INAN house four year \u00E2\u0080\u0098I\u00E2\u0080\u0099ve lived in this house for four years.\u00E2\u0080\u0099 b. nik\u00C3\u00AE-w\u00C3\u00AEkin \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwiskiy ni- k\u00C3\u00AE- w\u00C3\u00AEki -n \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwo askiy 1- PREV-live.VAI-SAP here DEM.INAN house four year \u00E2\u0080\u0098I lived at this house for four years.\u00E2\u0080\u0099 The past context for the INDEPENDENT clause in (82) is disambiguated by the initial clause: the speaker is talking about a state that held many years in the past. Again, the clause must be marked with k\u00C3\u00AE-. (82) context: old person talking about when they were young a. * k\u00C3\u00A2-(k\u00C3\u00AE)-oskin\u00C3\u00AEkiskw\u00C3\u00AAwiy\u00C3\u00A2n nikatawasisin k\u00C3\u00A2- k\u00C3\u00AE- oskin\u00C3\u00AEkiskw\u00C3\u00AAwi -y\u00C3\u00A2n ni- katawasisi -n C2-PREV-young.woman.VAI-1 1- beautiful.VAI-SAP --- b. k\u00C3\u00A2-(k\u00C3\u00AE)-oskin\u00C3\u00AEkiskw\u00C3\u00AAwiy\u00C3\u00A2n, nik\u00C3\u00AE-katawasisin k\u00C3\u00A2- k\u00C3\u00AE- oskin\u00C3\u00AEkiskw\u00C3\u00AAwi -y\u00C3\u00A2n ni- k\u00C3\u00AE- katawasisi -n C2-PREV-young.woman.VAI-1 1- PREV-beautiful.VAI-SAP \u00E2\u0080\u0098When I was a young woman, I was beautiful.\u00E2\u0080\u0099 101 Conversely, if the state does hold at present, a indexical (INDEPENDENT) clause must be bare (not marked with k\u00C3\u00AE-). In (83) the speaker is talking about a person recently met. In order for the state to be interpretable as holding at utterance time, the unmarked clause must be used. (If the clause is marked with k\u00C3\u00AE-, then the state must hold at a time previous to speech time). (83) context: talking about tall guy named Bernie after meeting him at a party a. iyikohk kinosiw Bernie iyikohk kinosi -w B so tall.VAI-3 B \u00E2\u0080\u0098Bernie is/was really tall.\u00E2\u0080\u0099 b. # iyikohk k\u00C3\u00AE-kinosiw Bernie iyikohk k\u00C3\u00AE- kinosi -w B so PREV-tall.VAI-3 B --- comment: *laughter* how did he get short? is it because he\u00E2\u0080\u0099s older and shrunk? Likewise, the contrast in (84) shows that with the adverbial phrase isp\u00C3\u00AE ot\u00C3\u00A2kosihk \u00E2\u0080\u0098since yesterday\u00E2\u0080\u0099, only the bare form is good. (84) a. nicihk\u00C3\u00AAyiht\u00C3\u00AAn ssp\u00C3\u00AE ot\u00C3\u00A2kosihk ni- c\u00C3\u00AEhk\u00C3\u00AAyiht\u00C3\u00AA -n isp\u00C3\u00AE ot\u00C3\u00A2kosin -k 1- happy.VTI -SAP TIME be.evening.VII-0 \u00E2\u0080\u0098I\u00E2\u0080\u0099ve been happy since yesterday.\u00E2\u0080\u0099 b. * nik\u00C3\u00AE-cihk\u00C3\u00AAyiht\u00C3\u00AAn ssp\u00C3\u00AE ot\u00C3\u00A2kosihk ni- c\u00C3\u00AEhk\u00C3\u00AAyiht\u00C3\u00AA -n isp\u00C3\u00AE ot\u00C3\u00A2kosin -k 1- happy.VTI -SAP TIME be.evening.VII-0 --- (intended: \u00E2\u0080\u0098I\u00E2\u0080\u0099ve been happy since yesterday.\u00E2\u0080\u0099) A final piece of evidence that predicates in an indexical clause must hold at speech time is their co-occurrence restrictions with past time adverbials. Distant past-time adverbials like kay\u00C3\u00A2s \u00E2\u0080\u0098long ago\u00E2\u0080\u0099 cannot co-occur at all. (85) * kay\u00C3\u00A2s nimiyosin kay\u00C3\u00A2s ni- miyosi -n long.ago. 1- pretty.VAI-SAP --- (intended: \u00E2\u0080\u0098A long time ago I was pretty.\u00E2\u0080\u0099) 102 More recent past time adverbials can co-occur with unmarked statives. If an unmarked INDEPENDENT clause is contrasted with a k\u00C3\u00AE-marked INDEPENDENT, the difference seems to be that in the latter case, the state of happiness no longer holds. (86) a. ot\u00C3\u00A2kosihk nicihk\u00C3\u00AAyiht\u00C3\u00AAn ot\u00C3\u00A2kosin -k ni- cihk\u00C3\u00AAyiht\u00C3\u00AA -n be.evening-0 1- happy.VTI -SAP \u00E2\u0080\u0098I was happy yesterday. b. ot\u00C3\u00A2kosihk nk\u00C3\u00AE-cihk\u00C3\u00AAyiht\u00C3\u00AAn ot\u00C3\u00A2kosin -k ni- k\u00C3\u00AE- cihk\u00C3\u00AAyiht\u00C3\u00AA -n be.evening.VII-0 1- PREV happy.VTI -SAP \u00E2\u0080\u0098I was happy yesterday.\u00E2\u0080\u0099 comment: there\u00E2\u0080\u0099s sort of an implied BUT\u00E2\u0080\u00A6 she\u00E2\u0080\u0099s not happy now However, there is a restriction on this co-occurrence: the adverbial must be in initial position. In final position, they are bad, as in (87). (87) a. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn m\u00C3\u00AEcim\u00C3\u00A2poy ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n m\u00C3\u00AEcim\u00C3\u00A2poy 1- want.VTI -SAP soup = \u00E2\u0080\u0098I want soup.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098I wanted soup.\u00E2\u0080\u0099 b. * ninitaw\u00C3\u00AAyiht\u00C3\u00AAn m\u00C3\u00AEcim\u00C3\u00A2poy ot\u00C3\u00A2kosihk ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n m\u00C3\u00AEcim\u00C3\u00A2poy ot\u00C3\u00A2kosin -k 1- want.VTI -SAP soup be.evening.VII-0 --- At this point, I am not sure what accounts for the distribution of the adverbials, although one possible line of analysis to pursue is that initial adverbials relate to the event time, while final adverbials relate to reference time (cf. Currie 1995 on time adverbials in Salish; also Klein 1992 on English adverbials with the perfect). 3.3.1.3 Indexical clauses present activities that coincide with T0 Like stative predicates, unmarked activity predicates are interpreted as present in indexical clauses, with either a habitual or imperfective reading, as shown in (88). 103 (88) Activity predicates must be marked with k\u00C3\u00AE- for unambiguous past reading a. nikisip\u00C3\u00AAkin\u00C3\u00AAn w\u00C3\u00AEy\u00C3\u00A2kana ni- kisip\u00C3\u00AAkin\u00C3\u00AA -n w\u00C3\u00AEy\u00C3\u00A2kan -a 1- wash.VTI -SAP dish -PL = \u00E2\u0080\u0098I am washing dishes (right now)\u00E2\u0080\u0099 = \u00E2\u0080\u0098I wash dishes in general\u00E2\u0080\u0099 b. nik\u00C3\u00AE-kisip\u00C3\u00AAkin\u00C3\u00AAn w\u00C3\u00AEy\u00C3\u00A2kana ni-k\u00C3\u00AE-kisip\u00C3\u00AAkin\u00C3\u00AA-n w\u00C3\u00AEy\u00C3\u00A2kan-a 1-PREV-wash.VTI-SAP dish-PL = \u00E2\u0080\u0098I had washed the dishes\u00E2\u0080\u0099 Activity predicates in bare indexical clauses act like stative predicates with respect to periphrastic past constructions; speakers reject periphrastic \u00E2\u0080\u0098used to\u00E2\u0080\u0099 past translations as in (89), even with the habitual element m\u00C3\u00A2na \u00E2\u0080\u0098usually/at times\u00E2\u0080\u0099. (89) a. nim\u00C3\u00A2ton ni- m\u00C3\u00A2to -n 1- cry.VAI-SAP = \u00E2\u0080\u0098I am crying / I cry.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098I used to cry.\u00E2\u0080\u0099 b. nim\u00C3\u00A2ton m\u00C3\u00A2na ni- m\u00C3\u00A2to -n m\u00C3\u00A2na 1- cry.VAI-SAP usually = \u00E2\u0080\u0098I cry at times.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098I used to cry (at times).\u00E2\u0080\u0099 Activity predicates also behave like stative predicates in that speakers reject the bare indexical clause in past contexts. Notice that in many cases (e.g., (90)), speakers will volunteer an English past translation, but reject the past context. (90) nim\u00C3\u00AEcison ni- m\u00C3\u00AEciso -n 1- eat.VAI-SAP = \u00E2\u0080\u0098I\u00E2\u0080\u0099m eating right now.\u00E2\u0080\u0099 = \u00E2\u0080\u0098I eat.\u00E2\u0080\u0099 (Like say you lost weight, and someone asks \u00E2\u0080\u0098aren\u00E2\u0080\u0099t you eating anymore?\u00E2\u0080\u0099) \u00E2\u0089\u00A0 \u00E2\u0080\u0098I ate\u00E2\u0080\u0099 (consultant allows translation \u00E2\u0080\u0098I ate\u00E2\u0080\u0099 but says it is bad in a past context, e.g., talking about the big meal you had earlier in the day) One possible explanation for this translation/interpretation discrepancy is that it stems from the convergence of tense (past) and aspect (completive) in English past tense constructions: in these 104 contexts, the speaker is attending to the fact that event time can precede and be completed by reference time, rather than attending to the relation between reference time and speech time. At any rate, when a past relation is established (e.g., by using a context), the speaker systematically rejects the bare indexical clause forms. Also like stative predicates, when the context is set up such that the reference time includes the speech time, an indexical clause is obligatorily bare. This includes present perfect contexts, as in (91). (91) context: walking across British Columbia, reach a friend\u00E2\u0080\u0099s house after three weeks, but still have a long ways to go a. mistahi nipimoht\u00C3\u00A2n m\u00C3\u00AEhc\u00C3\u00AAt k\u00C3\u00AEsik\u00C3\u00A2k mistahi ni- pim\u00C3\u00B4ht\u00C3\u00AA -n m\u00C3\u00AEhc\u00C3\u00AAt k\u00C3\u00AEsik\u00C3\u00A2 -k much 1- walk.VAI-SAP many be.day.VII-0 \u00E2\u0080\u0098I\u00E2\u0080\u0099ve walked a lot/many miles in many days.\u00E2\u0080\u0099 b. # mistahi nik\u00C3\u00AE-pimoht\u00C3\u00A2n mistahi ni- k\u00C3\u00AE- pimoht\u00C3\u00AA -n much 1- PREV-walk.VAI -SAP \u00E2\u0080\u0098I had walked a lot/ many miles\u00E2\u0080\u0099 comment: I wouldn\u00E2\u0080\u0099t guess that you\u00E2\u0080\u0099re going to walk more In terms of past time adverbials, unmarked activities can be modified by t\u00C3\u00A2kosihk yesterday, but like with statives, the adverbial must be in initial position. (92) a. ot\u00C3\u00A2kosihk, mistahi nim\u00C3\u00A2ton ot\u00C3\u00A2kosin -k mistahi ni- m\u00C3\u00A2to -n be.evening.VII-0 much 1- cry.VAI-SAP \u00E2\u0080\u0098I cried yesterday.\u00E2\u0080\u0099 b. ?? mistahi nim\u00C3\u00A2ton ot\u00C3\u00A2kosihk mistahi ni- m\u00C3\u00A2to -n ot\u00C3\u00A2kosin -k much 1- cry.VAI-SAP be.evening.VII-0 --- (intended: \u00E2\u0080\u0098I cried yesterday.\u00E2\u0080\u0099) 105 Also like statives, the utterance is much worse if the temporal adverbial refers to a distant past. In (93), the consultant fixed the unmarked form by adding the temporal sequencer k\u00C3\u00AE-, and a locative demonstrative.26 (93) a. ?? kay\u00C3\u00A2s kimiwan k\u00C3\u00A2y\u00C3\u00A2s kimiwan long.ago rain.VII --- b. kay\u00C3\u00A2s k\u00C3\u00AE-kimiwan \u00C3\u00B4ta kay\u00C3\u00A2s k\u00C3\u00AE- kimiwan \u00C3\u00B4ta long.ago PREV-rain.VII here \u00E2\u0080\u0098It rained here a long time ago.\u00E2\u0080\u0099 3.3.1.4 Indexical clauses present telic predicates whose result state coincides with T0 Unmarked telic predicates can only be interpreted as completed in bare indexical clauses. In terms of the coincidence relations, the result holds at speech time. Concurrent with their completion status, unmarked telic predicates can only translate into the English temporal system as a past tense. In (94) English present translations are rejected; however, like stative and activity predicates, the periphrastic past \u00E2\u0080\u0098used to\u00E2\u0080\u0099 is also rejected. 26 The addition of the locative in tandem with k\u00C3\u00AE- is suggestive of the well-known link between spatial and temporal deixis. In particular, while unmarked predicates are always taken to be spatially centered around the speaker (consistent with their indexical status), the marked predicate is clearly not; thus the overt proximal locative \u00C3\u00B4ta \u00E2\u0080\u0098here\u00E2\u0080\u0099 specifies that although the event is temporally distant it is still spatially coincident with the speech situation. 106 (94) Telic predicates can only have completed translation a. W\u00C3\u00A2pastim p\u00C3\u00AEkonam w\u00C3\u00A2pamowina27 W p\u00C3\u00AEkonam -w w\u00C3\u00A2pam -win -a W break.VTI-3 see.VTA-NOM-OBV \u00E2\u0080\u0098W. broke a mirror.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098W. is breaking a mirror.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098W. breaks mirrors.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098W. used to break mirrors.\u00E2\u0080\u0099 comment: The mirror is broken. b. Jeff paskisw\u00C3\u00AAw atimwa J paskisw -\u00C3\u00AA -w atimw -a J shoot.VTA-DIR-3 dog -OBV \u00E2\u0080\u0098Jeff shot the dog.\u00E2\u0080\u0099 Incompleted events are marked by preverbs or adverbials; in (93) the addition of m\u00C3\u00AAkw\u00C3\u00A2 corresponds with a progressive. (95) a. atim nipah\u00C3\u00AAw min\u00C3\u00B4sa atimw nipah -\u00C3\u00AA -w min\u00C3\u00B4s -a dog kill.VTA-DIR-3 cat -OBV \u00E2\u0080\u0098A/some dog killed a cat.\u00E2\u0080\u0099 b. atim m\u00C3\u00AAkw\u00C3\u00A2-nipah\u00C3\u00AAw min\u00C3\u00B4sa atimw m\u00C3\u00AAkw\u00C3\u00A2- nipah -\u00C3\u00AA -w min\u00C3\u00B4s -a dog IMP- kill.VTA-DIR-3 cat -OBV \u00E2\u0080\u0098A dog is killing a cat right now.\u00E2\u0080\u0099 comment: Maybe they\u00E2\u0080\u0099re looking out the window and see it happening. 27 It appears that the aspectual properties may not be hard-coded in these predicates; the other interpretations are made available by adding modifiers: m\u00C3\u00A2na \u00E2\u0080\u0098usually\u00E2\u0080\u0099 to get the habitual, m\u00C3\u00AAkwac- \u00E2\u0080\u0098midst\u00E2\u0080\u0099 to get an imperfective: (i) a. W\u00C3\u00A2pastim p\u00C3\u00AEkonam w\u00C3\u00A2pamwona m\u00C3\u00A2na W p\u00C3\u00AEkonam -w w\u00C3\u00A2pam -win -a m\u00C3\u00A2na W break.VTI-3 see.VTA-NOM-PL usually \u00E2\u0080\u0098W\u00C3\u00A2pastim breaks mirrors.\u00E2\u0080\u0099 (Lit.: \u00E2\u0080\u0098W\u00C3\u00A2pastim has broken a mirror repeatedly.\u00E2\u0080\u0099?) b. W\u00C3\u00A2pastim m\u00C3\u00AAkw\u00C3\u00A2c-p\u00C3\u00AEkonam w\u00C3\u00A2pamwona W m\u00C3\u00AAkw\u00C3\u00A2c p\u00C3\u00AEkonam -w w\u00C3\u00A2pam -win -a W now break.VTI-3 see.VTA-NOM-PL \u00E2\u0080\u0098W. is breaking mirrors right now.' This does not impact the analysis of clause-typing, which only requires an evaluation time of T0 (compatible with both (ia) and (iib)), but it suggests further work is need on aspectual classes in Plains Cree. 107 The present perfect contexts help to tease apart the aspectual interpretation (i.e., the result holds at T0) from the temporal interpretation (i.e., Tref precedes T0). If the former is the relevant interpretation of the clause, then we predict that present perfect contexts will be fine with bare indexical INDEPENDENT clauses, whereas if the latter analysis is more accurate, then we would expect some additional marking to be necessary in these contexts (just as m\u00C3\u00AAkw\u00C3\u00A2, which is used for imperfectives, is necessary in present contexts). In present perfect contexts, the bare indexical clause is used. For example, the \u00E2\u0080\u0098news\u00E2\u0080\u0099 context of the present perfect (Comrie 1976, Fenn 1987, Klein 1994) necessitates an unmarked INDEPENDENT clause. Notice that trying to use a k\u00C3\u00AE- marked clause elicits the kind of response we expect where reference time does not coincide with the speech time: the present relevance of the father\u00E2\u0080\u0099s fall and leg-breaking is no longer apparent. (96) context: a child\u00E2\u0080\u0099s father has just broken his leg; child runs to tell the news and get help a. nip\u00C3\u00A2pa w\u00C3\u00AEsakisin, p\u00C3\u00AEkonam osk\u00C3\u00A2t, p\u00C3\u00AA-w\u00C3\u00AEcihin\u00C3\u00A2n ni- p\u00C3\u00A2pa w\u00C3\u00AEsakisin -w p\u00C3\u00AEkonam -w o- sk\u00C3\u00A2t p\u00C3\u00AA- w\u00C3\u00AEcihi -n\u00C3\u00A2n 1- papa fall.VAI -3 break.VTI -3 3- leg come-help.VTA-1.PL \u00E2\u0080\u0098My dad got hurt, he broke his leg, come and help us!\u00E2\u0080\u0099 b. # nip\u00C3\u00A2pa k\u00C3\u00AE-w\u00C3\u00AEsaksin, k\u00C3\u00AE-p\u00C3\u00AEkonam osk\u00C3\u00A2t, p\u00C3\u00AA-w\u00C3\u00AEcihin\u00C3\u00A2n ni- p\u00C3\u00A2pa k\u00C3\u00AE- w\u00C3\u00AEsakisin-w k\u00C3\u00AE- p\u00C3\u00AEkonam -w o- sk\u00C3\u00A2t p\u00C3\u00AA- w\u00C3\u00AEcihi -n\u00C3\u00A2n 1- papa PREV-fall.VAI-3 PREV-break.VTI-3 3- leg come-help.VTA-1.PL --- comment: \u00E2\u0080\u00A6wouldn\u00E2\u0080\u0099t say it that way; sounds like the son is heartless/ungrateful (97) context: I found a hat that belonged to you several months ago, but didn\u00E2\u0080\u0099t tell you until just now (months later). I want to pretend I just found it a. kitastotin nimisk\u00C3\u00AAn ki(t)- astotin ni- misk\u00C3\u00AA -n 2- hat 1- find.VTI-SAP \u00E2\u0080\u0098I found your hat.\u00E2\u0080\u0099 (Last spring) comment: He will think you just found it. 108 b. # kitastotin nik\u00C3\u00AE-misk\u00C3\u00AAn ki(t)- astotin ni- k\u00C3\u00AE- misk\u00C3\u00AA -n 2- hat 1- PREV-find.VTI-SAP \u00E2\u0080\u0098I had found your hat.\u00E2\u0080\u0099 comment: if you wanted to pretend that you just found it, you wouldn\u00E2\u0080\u0099t use the k\u00C3\u00AE- Using the adverbial phrase anohc piko \u00E2\u0080\u0098just now\u00E2\u0080\u0099 also obligatorily requires a bare indexical clause, as in (98). (98) context: on a walk; snake has just slithered across path and into bushes a. anohc piko niw\u00C3\u00A2pam\u00C3\u00A2w kin\u00C3\u00AApik anohc piko ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w kin\u00C3\u00AApikw today QUANT 1- see.VTA-DIR-3 snake \u00E2\u0080\u0098I saw a snake just now.\u00E2\u0080\u0099 b. * anohc piko nik\u00C3\u00AE-w\u00C3\u00A2pam\u00C3\u00A2w kin\u00C3\u00AApik anohc piko ni- k\u00C3\u00AE- w\u00C3\u00A2pam -\u00C3\u00A2 -w kin\u00C3\u00AApikw today QUANT 1- PREV-see.VTA-DIR-3 snake --- comment: They\u00E2\u0080\u0099re sort of cancelling each other out\u00E2\u0080\u0093one is now, the other is before Finally, a bare predicate is used for a \u00E2\u0080\u0098perfect of result\u00E2\u0080\u0099 context, as shown in (99) and (100). (99) context: explaining to someone why you can\u00E2\u0080\u0099t see a. niwan\u00C3\u00AEht\u00C3\u00A2n nisk\u00C3\u00AEsik\u00C3\u00B4hk\u00C3\u00A2na ni- wan\u00C3\u00AEht\u00C3\u00A2 -n ni- sk\u00C3\u00AEsikwihk\u00C3\u00A2n -a 1- lose.VAI -SAP 1- glasses -PL \u00E2\u0080\u0098I lost my glasses.\u00E2\u0080\u0099 b. ?# nik\u00C3\u00AE-wan\u00C3\u00AEht\u00C3\u00A2n nisk\u00C3\u00AEsik\u00C3\u00B4hk\u00C3\u00A2na ni- k\u00C3\u00AE- wan\u00C3\u00AEht\u00C3\u00A2 -n ni- sk\u00C3\u00AEsikwihk\u00C3\u00A2n-a 1- PREV-lose.VAI-SAP 1- glasses -PL \u00E2\u0080\u0098I lost my glasses.\u00E2\u0080\u0099 109 (100) context: we are waiting for Tom to come so we can start dancing A: Tom \u00C3\u00B4ta ay\u00C3\u00A2w T \u00C3\u00B4ta ay\u00C3\u00A2 -w T here be.VAI-3 \u00E2\u0080\u0098Is Tom here?\u00E2\u0080\u0099 B: p\u00C3\u00AA-takosin p\u00C3\u00AA- takosin -w DIR-arrive.VAI-3 \u00E2\u0080\u0098(Yes) he\u00E2\u0080\u0099s arrived.\u00E2\u0080\u0099 comment: this one is implying that he\u00E2\u0080\u0099s still here, that we\u00E2\u0080\u0099re ready to dance What all of these contexts have in common is that they are contexts where we want to model that the result holds at speech time. Thus, just like stative and activity predicates, telic predicates in indexical clauses have a fixed temporal relation to the speech time. And in all of these contexts, bare indexical INDEPENDENT clauses are felicitous, indicating that we are tracking the coincidence of the reference time with speech time (T0). Bare indexical clauses with telic predicates can co-occur with otahkosihk \u00E2\u0080\u0098yesterday\u00E2\u0080\u0099, as in (101). Sometimes examples with past time adverbials are augmented by another clause with a stative predicate. In these examples, the telic predicate is overtly related to some non-telic predicate that holds at utterance time.28 (101) a. ot\u00C3\u00A2kosihk n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan sipw\u00C3\u00AAhtew, \u00C3\u00AAkwa anohc nikask\u00C3\u00AAyiht\u00C3\u00AAn. ot\u00C3\u00A2kosin -k ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan sipw\u00C3\u00AAht\u00C3\u00AA -w \u00C3\u00AAkwa anohc ni- kask\u00C3\u00AAyiht\u00C3\u00AA -n be.evening.VII-0 1- friend leave.VAI-3 and today 1- lonely.VTI -SAP \u00E2\u0080\u0098Yesterday my friend left, and now I\u00E2\u0080\u0099m lonely.\u00E2\u0080\u0099 b. nikis\u00C3\u00AEp\u00C3\u00AAkin\u00C3\u00AAn \u00C3\u00B4ma w\u00C3\u00AEy\u00C3\u00A2kan, kan\u00C3\u00A2tan ni- kis\u00C3\u00AEp\u00C3\u00AAkin\u00C3\u00AA -n \u00C3\u00B4ma w\u00C3\u00AEy\u00C3\u00A2kan kan\u00C3\u00A2tan 1- wash.VTI-SAP DEM.INAN dish clean.VII \u00E2\u0080\u0098I washed this dish, it\u00E2\u0080\u0099s clean.\u00E2\u0080\u0099 The data in (101) is consistent with the data we saw earlier, where the past time adverbial was acceptable as long as it was in initial position. 28 This parallels observations made by Lakoff (1970) regarding \u00E2\u0080\u0098present relevance\u00E2\u0080\u0099 in the English tense system. 110 3.3.1.5 Interim summary Given the times (situation time, reference time, evaluation time, and speech time) and relations [\u00C2\u00B1coincidence] that seem to be relevant to linguistic structure, we would expect indexical clauses to always make reference to the speech time. More specifically within a situation semantics analysis, the speech time is the temporal component of the speech situation by which the proposition in an indexical clause is evaluated. Consistent with our expectations, in this section we have seen data to show that there is a [+coincidence] relation between reference time and speech time in indexical clauses. Crucially, as we saw in \u00C2\u00A73.3.1.1, this interpretation is different from non-indexical clauses, where speech time is not necessarily taken into account. In the next section I argue that we can make the same claims for referents that we do for times, to derive the evidential force of indexical INDEPENDENT clauses. 3.3.2 Referential deixis: The role of the speaker in indexical clauses In this section, I show that indexical INDEPENDENT clauses always make reference to a speaker, a property which has not been previously discussed in the Algonquianist literature. If we think about which referents are necessary for a speech act, we see that first-person is crucial. Without a speaker, there is no speech act, and as long as there is a speaker, a speech act can occur29. Within a speech act, then, only first person is a referential constant; all other referents can come or go. If an indexical clause contains reference to a speech act (or, more formally, the speech situation), and every speech act contains a speaker, then that logically means that an indexical clause always contains reference to a speaker. (102) s0 \u00EF\u0083\u00A0 Speaker 29 It may be possible for reference to s0 to also yield second-person effects (i.e., common ground effects). On independent grounds, Plains Cree forms seem not to have a lot of common ground sensitive forms, so it is not surprising that I have found no common ground effects in the clause-typing domain either. 111 By contrast, non-indexical clauses are not deictic on the speech act; the proposition in a non-indexical clause is evaluated relative to a contextually-given situation. This means that non- indexical clauses will not necessarily contain reference to a speaker. In this section I provide evidence that indexical clauses, do asymmetrically divide referents between speakers (first person) and non-speakers (second and third persons) \u00E2\u0080\u0093 and that non-indexical clauses do not. Just as propositions in indexical clauses are temporally evaluated with respect to the speech time, so they are referentially evaluated with respect to the speaker. This is manifest in a number of domains. One way for the relation between the proposition and the speaker to be established is if the speaker is one of the participants in the event. Thus, if speaker is identified with one of the arguments of the predicate, an indexical clause will be preferred. Thus, in Plains Cree, indexical INDEPENDENT clauses are preferred when talking about the self. If the speaker is not identified with one of these arguments, then the reference must be established in some other way. I suggest that one way reference can be established is via the speaker\u00E2\u0080\u0099s perception of the event: if the event is directly perceived, then the speaker has a privileged relation to the proposition. This would mean that indexical clauses would have a \u00E2\u0080\u0098direct\u00E2\u0080\u0099 evidential force in the sense of Willett (1988) and Aikhenvald (2004) among others. Finally reference to the speaker can be given as the speaker\u00E2\u0080\u0099s epistemic state (i.e., certainty) about the proposition being expressed, or in subjective predicates, by the speaker\u00E2\u0080\u0099s attitude towards the proposition. All of these patterns converge to show that there is a specified relation between the proposition and the speaker in indexical clauses. Thus, despite the lack of dedicated morphological evidential marking, Plains Cree\u00E2\u0080\u0099s independent indicative mode has a privileged relationship with the speaker, and so has evidential force. 112 3.3.2.1 Person-effects: Preference for indexical clauses when talking about the self In this section I show that, across multiple classes of predicates, if the speaker is associated with one of the arguments of the verb, the clause is indexical. If we think about there being a coded relation between the proposition and the speaker, one way this relation could be realized is by the speaker being a participant in the event. When a non-indexical CONJUNCT order clause is used instead, a variety of distancing interpretive effects arise. If we think about the difference between indexical and non-indexical clauses within situation semantics, in the former case the proposition is evaluated with respect to the speech situation, while in the latter case the proposition is evaluated relative to some (unspecified) situation. Thus, in this analysis, indexical clauses (Plains Cree\u00E2\u0080\u0099s INDEPENDENT order) are specified, and non-indexical (i.e., anaphoric; Plains Cree\u00E2\u0080\u0099s CONJUNCT order) clauses are not specified. Via principles of Blocking (Trubetzkoy 1939, Jakobsen 1929, among many others), the choice of an unspecified form (here anaphoric clause-typing) implies that the specified form (here indexical clause-typing) is infelicitous; thus the situation in the non- indexical clause could not be the speech situation. For example, in (103), the INDEPENDENT is the only felicitous way for a speaker to express the current state of feeling cold; the \u00C3\u00AA-CONJUNCT, in this context as a whole utterance, has a distancing effect, and the consultant comments on how the non-indexical clause fails to convey the experience of being cold. 113 (103) a. nikawacin INDEPENDENT ni- kawaci -n 1- cold.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m cold.\u00E2\u0080\u0099 b. #? \u00C3\u00AA-kawaciy\u00C3\u00A2n CONJUNCT \u00C3\u00AA- kawaci -y\u00C3\u00A2n c1-cold.VAI-1 \u00E2\u0080\u0098I was cold.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098I\u00E2\u0080\u0099m cold.\u00E2\u0080\u0099 comment: it\u00E2\u0080\u0099s referring to when I was cold\u00E2\u0080\u00A6I wouldn\u00E2\u0080\u0099t say this to you. I would say nikawacin \u00E2\u0080\u0098I\u00E2\u0080\u0099m cold\u00E2\u0080\u0099\u00E2\u0080\u00A6 for \u00E2\u0080\u0098I\u00E2\u0080\u0099m cold\u00E2\u0080\u0099 I wouldn\u00E2\u0080\u0099t use [this form] Similarly, in (104), the indexical INDEPENDENT conveys the internal state of the speaker, while the \u00C3\u00AA-conjunct does not. Here the anaphoric CONJUNCT form codes a first person argument, and the first person is not bound, so it is associated with the speaker. Thus, when asked about the interpretation of (104b) the consultant disassociated the experience from speaking along temporal lines (even though, as we saw earlier, and will see in chapter 4, CONJUNCT clauses are unspecified for their temporal value). (104) a. nikisiw\u00C3\u00A2sin INDEPENDENT ni- kisiw\u00C3\u00A2si -n 1- angry.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m angry.\u00E2\u0080\u0099 comment: like, right now. It\u00E2\u0080\u0099s referring to I\u00E2\u0080\u0099m angry or I feel angry right now b. \u00C3\u00AA-kisiw\u00C3\u00A2siy\u00C3\u00A2n CONJUNCT \u00C3\u00AA- kisiw\u00C3\u00A2si -y\u00C3\u00A2n c1-angry.VAI-1 \u00E2\u0080\u0098I got angry.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098I\u00E2\u0080\u0099m angry right now.\u00E2\u0080\u0099 comment: like yesterday, or something In both these cases, the clause-typing conveys the internal state of self; however, in the non-indexical case, the internal state is disconnected from the current state (i.e., the state at the speech situation). In indexical clauses, by contrast, the internal state coded by the predicate is connected to the state at the speech situation. 114 The distancing effect is also seen in naming predicates (such as, but not limited to isiy\u00C3\u00AEhk\u00C3\u00A2so-/isiy\u00C3\u00AEhk\u00C3\u00A2t\u00C3\u00AA- \u00E2\u0080\u0098s/he is called\u00E2\u0080\u0099/ \u00E2\u0080\u0098it is called\u00E2\u0080\u0099): we find that predicates of naming almost universally utilize indexical clauses; anaphoric CONJUNCT clauses are only used with naming predicates in very specific contexts. In particular, when introducing oneself (i.e., when the participant of the event is the speaker), it is necessary to use the indexical (i.e., INDEPENDENT) clause-type. If the CONJUNCT is used, consultants react as though the speaker has no knowledge of their own name. (105) context: speaker naming self a. Clare nitsiy\u00C3\u00AEhk\u00C3\u00A2son INDEPENDENT C ni(t)- isiy\u00C3\u00AEhk\u00C3\u00A2so -n C 1- be.called.VAI-SAP \u00E2\u0080\u0098My name is Clare.\u00E2\u0080\u0099 b. # Clare \u00C3\u00AA-isiy\u00C3\u00AEhk\u00C3\u00A2soy\u00C3\u00A2n CONJUNCT C \u00C3\u00AA- isiy\u00C3\u00AEhk\u00C3\u00A2so -y\u00C3\u00A2n C C1-be.called.VAI-1 \u00E2\u0080\u0098My name is Clare.\u00E2\u0080\u0099 comment: *laughing* that\u00E2\u0080\u0099s just what they tell me\u00E2\u0080\u00A6. You don\u00E2\u0080\u0099t know yourself The only time where a naming predicate referring to self is found in the CONJUNCT order is when it is embedded under some other predicate \u00E2\u0080\u0093 where, as we have seen, an indexical clause is excluded. (106) ana iskw\u00C3\u00AAw kisk\u00C3\u00AAyihtam Clare \u00C3\u00AA-isiyihk\u00C3\u00A2soy\u00C3\u00A2n ana iskw\u00C3\u00AAw kisk\u00C3\u00AAyihtam -w C \u00C3\u00AA- isiyihk\u00C3\u00A2so -y\u00C3\u00A2n DEM.AN woman know.VTI -3 C C1-THUS.be.called.VAI-1 \u00E2\u0080\u0098That woman knows my name is Clare.\u00E2\u0080\u0099 The felicitousness of the non-indexical CONJUNCT clause in (106) has to do with its unspecified nature and will be discussed further in chapter 6. 3.3.2.2 Indexical clauses are infelicitous in contexts of unconsciousness Consciousness is a condition on being a speaker (cf. Searle 1965, Banfield 1982). So far we have looked at contexts where the speaker also happens to be one of the participants in the event 115 expressed by the proposition. However, if the speaker is unconscious during the event, then the speaker loses their privileged relation to the event. A lack of consciousness means by definition a lack of experience; something that happens while I am unconscious is experientially equivalent to something happening to someone else outside of my perceptual field (cf. Chung 2005 for Korean). In addition, because one undergoing unconsciousness does not have awareness during the event, the reporting of the event must always occur at a time that is distinct from the occurrence of the event. This means that when someone is reporting a proposition for which they were unconscious, this proposition cannot be connected to the speech situation. We expect indexical clauses to be infelicitous in such contexts, and they are. For example, in (107), the consultant immediately offered a clause in the \u00C3\u00AA-conjunct to express lack of consciousness. When presented with the independent indicative, the consultant accepted the form, but when asked to repeat it, always repeated the clause in the \u00C3\u00AA-conjunct. (107) a. # niwanitipskin\u00C3\u00AAn INDEPENDENT ni- wanitipiskin\u00C3\u00AA -n (presented) 1- lose.consciousness.VTI-LP \u00E2\u0080\u0098I (have) lost consciousness.\u00E2\u0080\u0099 b. \u00C3\u00AA-wanitipskinam\u00C3\u00A2n CONJUNCT \u00C3\u00AA- wanitipskiniam -\u00C3\u00A2n (offered) C1-lose.consciousness.VTI-1 \u00E2\u0080\u0098I lost consciousness.\u00E2\u0080\u0099 Clause-typing is also sensitive to contextually-specified unconsciousness \u00E2\u0080\u0093 we can take a predicate that has no particular intentional or consciousness properties, and the choice of clause- typing will indicate the experiential knowledge of the speaker. Thus, in (108), the predicate of falling (pahksini-) would usually be interpreted as happening while the participant is conscious, and when a context of consciousness is provided, the speaker prefers the indexical INDEPENDENT clause. 116 (108) context: speaker tripped over a chair, and fell to the floor a. nipahkisinin INDEPENDENT ni- pahkisin -n 1- fall.VAI -SAP \u00E2\u0080\u0098I fell.\u00E2\u0080\u0099 b. # \u00C3\u00AA-pahkisiniy\u00C3\u00A2n CONJUNCT \u00C3\u00AA- pahkisin -y\u00C3\u00A2n C1-fall.VAI -1 \u00E2\u0080\u0098I fell.\u00E2\u0080\u0099 By contrast, when provided with a context of unconsciousness (for example, falling during a faint), the consultant switches to the anaphoric \u00C3\u00AA-conjunct and rules the indexical INDEPENDENT infelicitous. (109) context: speaker blacked out and fell, woke up on the floor with a cut a. # nipahkisinin INDEPENDENT ni- pahkisin -n 1- fall.VAI -SAP \u00E2\u0080\u0098I fell.\u00E2\u0080\u0099 comment: no, you would say \u00C3\u00AA-pahksiniy\u00C3\u00A2n b. \u00C3\u00AA-pahkisiniy\u00C3\u00A2n CONJUNCT \u00C3\u00AA- pahkisin -y\u00C3\u00A2n C1-fall.VAI -1 \u00E2\u0080\u0098I fell.\u00E2\u0080\u0099 So far we have seen confirmation that if the speaker (i.e., the referent entailed by a speech act) is also one of the referents in the proposition, an indexical clause is preferred; an anaphoric clause gives rise to temporal/consciousness distancing effects. Given that indexical clauses always have reference to the speaker, we expect that indexical clauses that do not otherwise have a first person referent will (a) be restricted; and (b) have a special meaning where the speaker is invoked. When we look at passages of Plains Cree discourse, we find that there are strong tendencies on when a speaker uses an INDEPENDENT clause to talk about others. This leads to a split in the way first-person vs. non-first-person forms coincide with clause-typing. For example, if we compare the use of indexical INDEPENDENT and anaphoric CONJUNCT clauses in the passage 117 given below (108), we see that both of the predicates in which the speaker is a participant (kit\u00C3\u00A2cimost\u00C3\u00A2ton\u00C3\u00A2naw \u00E2\u0080\u0098we tell one another\u00E2\u0080\u0099 and nip\u00C3\u00AAhtaw\u00C3\u00A2wak \u00E2\u0080\u0098I have heard\u00E2\u0080\u0099) are in the INDEPENDENT order. However, all of the clauses referring solely to others (underlined) are in the \u00C3\u00AA-conjunct, except for one quotative (itw\u00C3\u00AAwak \u00E2\u0080\u0098they say\u00E2\u0080\u0099): (110) misak\u00C3\u00A2m\u00C3\u00AA ayisk \u00C3\u00B4ma, k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-pim\u00C3\u00A2tisiyahk, kit\u00C3\u00A2cimost\u00C3\u00A2ton\u00C3\u00A2naw m\u00C3\u00A2na t\u00C3\u00A2nis \u00C3\u00AA-ispayik aya, \u00C3\u00B4m \u00C3\u00A2ya, m-~ pim\u00C3\u00A2tisiwin \u00C3\u00B4ma k\u00C3\u00A2-pim\u00C3\u00A2tis\u00C3\u00AEtotamahk m\u00C3\u00AAkw\u00C3\u00A2c. p\u00C3\u00AAci-n\u00C3\u00A2way \u00C3\u00B4t\u00C3\u00AA nawac ayisiyiniwak (t\u00C3\u00A2nitahto nip\u00C3\u00AAhtaw\u00C3\u00A2wak) \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-miyaw\u00C3\u00A2tahkik, nawac ahp\u00C3\u00B4, \u00C3\u00AA-itw\u00C3\u00AAcik, \u00C3\u00AA-m\u00C3\u00AAkw\u00C3\u00A2-kitim\u00C3\u00A2kisicik, nawac \u00C3\u00AA-k\u00C3\u00AE-miyaw\u00C3\u00A2tahkik, os\u00C3\u00A2m \u00C3\u00AA-k\u00C3\u00AE-s\u00C3\u00A2kihitocik, nan\u00C3\u00A2tohk is \u00C3\u00AA-k\u00C3\u00AE-aya-w\u00C3\u00AEcihitocik m\u00C3\u00AEn \u00C3\u00A2ya, \u00C3\u00AA-k\u00C3\u00AE-kiyok\u00C3\u00A2tocik, miy\u00C3\u00AAkw\u00C3\u00A2-w\u00C3\u00A2skamis\u00C3\u00AEtw\u00C3\u00A2wi; \u00C3\u00AAkosi m\u00C3\u00A2n \u00C3\u00AEtw\u00C3\u00AAwak; \u00C3\u00AAkwa w\u00C3\u00AAtinahk \u00C3\u00AA-k\u00C3\u00AE-\u00C3\u00A2cimost\u00C3\u00A2tocik, \u00C3\u00AA-miyw\u00C3\u00A2sik k\u00C3\u00AEkway \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2miton-~-m\u00C3\u00A2misk\u00C3\u00B4tahkik. (Minde 1998:6) For all along, throughout our life, we tell one another about what is happening, about this life we are in the midst of living. In the past, people had been happier (I have heard many say that), they had been happier even when they were poor, because they used to love one another, they used to help one another in various ways, and they also visited one another when they were settled down; that is what they say; and they used to take time to tell stories to one another and to talk about good things. Since all indexical clauses have reference to a speaker, the use of the indexical INDEPENDENT clause for the quotative in this passage indicates that there is some other relation between the proposition and the speaker30. I now turn to the conditions under which an indexical (INDEPENDENT order) clause can felicitously be used to talk about others, and think about how the interpretation of these clauses can be thought of as invoking reference to a speaker. 3.3.2.3 Events in indexical clauses must be directly perceived by the speaker Indexical clauses which do not have any first person arguments are nevertheless predicted to make reference to the speaker. In this section I consider the evidential force of indexical clauses and suggest that this force is a result of the obligatory reference to the speaker. Following the evidential literature, I will take evidentiality to be concerned with the speaker\u00E2\u0080\u0099s source of information for the proposition being expressed; for example, direct experience, hearsay, or a dream (Aikhenvald 2004). The extent to which source of information is 30 Specifically, the INDEPENDENT here corresponds to something that the speaker has heard directly. See the following discussion for details. 118 a separate notion from epistemic certainty of the proposition (e.g., certain, probable, possible, impossible) is separate from evidentiality or whether one category can be derived from the other has remained a matter of debate (cf. Bybee 1985, Willett 1988); for the Plains Cree evidential system, the relevant notion seems to be source of information (Wolfart 1973, Blain & D\u00C3\u00A9chaine 2006a, b; 2007; Blain et al. 2006). However, as we will see, both source of information and certainty are defined in terms of the speaker\u00E2\u0080\u0099s relation to the proposition. When different sources of information are considered, the primary distinction that is cross-linguistically relevant is the distinction between direct and indirect sources of information. Direct evidentiality, also termed experiential or direct-perceptual evidentiality, includes information that the speaker has gained from personally perceiving (auditory, tactile, visual, etc. Aikhenvald 2004) an event or through self-knowledge (Garrett 2001). Indirect evidential force includes hearsay, reported, or inferred information (Willett 1988, Garrett 2001; Faller 2002, 2004; Aikhenvald 2004). Direct evidentiality by definition makes reference to the speaker, indirect evidentiality marks information coming from some other source. Under the indexical / non-indexical analysis of clauses, we thus expect that the direct/indirect evidential split could only map onto the clause-typing in one way. Indexical clauses, which always make reference to the speaker, should correspond to direct evidential force; non-indexical clauses, which do not have any reference to the speaker, should correspond to indirect evidential force. Further, the direct evidential force provides evidence that, even when the speaker is not one of the participants of the event, there is still reference to the speaker in the clause. Turning back to Plains Cree\u00E2\u0080\u0099s clause-typing system, INDEPENDENT clauses behave as if they have a direct evidential force: the INDEPENDENT indicates that the speaker has direct experiential evidence for the proposition being uttered. This is what happens with a clause like (111): the predicate is an attitude verb (miyw\u00C3\u00AAyihtam literally, \u00E2\u0080\u0098act.on.something.by.mind.in.a.good.way\u00E2\u0080\u0099; freely, \u00E2\u0080\u0098enjoy/like something\u00E2\u0080\u0099), and has a third- person subject (Anna); in this case, an indexical clause (the INDEPENDENT) is infelicitous in normal conversation. 119 (111) context: I see Anna cooking a lot; she smiles and laughs when she\u00E2\u0080\u0099s in the middle of cooking, so I infer that Anna must enjoy cooking a. # Anna miyw\u00C3\u00AAyihtam ka-k\u00C3\u00AEst\u00C3\u00AApot INDEPENDENT A miyw\u00C3\u00AAyihtam -w ka- k\u00C3\u00AEstepo -t A like.VTI -3 IRR-cook.VAI-3 \u00E2\u0080\u0098Anna likes to cook.\u00E2\u0080\u0099 b. Anna \u00C3\u00AA-m\u00C3\u00AEyw\u00C3\u00AAyihtahk ka-k\u00C3\u00AEst\u00C3\u00AApot CONJUNCT A \u00C3\u00AA- miyw\u00C3\u00AAyihtam -k ka- k\u00C3\u00AEstepo -t A C1-like.VTI -0 IRR-cook.VAI-3 \u00E2\u0080\u0098Anna likes to cook.\u00E2\u0080\u0099 comment: this is better (than independent) if I don\u00E2\u0080\u0099t actually know if Anna likes to cook, I\u00E2\u0080\u0099m just guessing because she cooks all the time In elicitation contexts, Plains Cree speakers will often allow intentional (i.e., attitude) predicates with third-person referents to have indexical INDEPENDENT clause-typing, but in this case, they often provide commentary on its felicity conditions. An example of this is given in (112), which has a third-person subject of a clause marked with w\u00C3\u00AE- \u00E2\u0080\u0098intend\u00E2\u0080\u0099. The consultant accepts the sentence, but comments about the necessary knowledge state of the speaker in order for this to be a felicitous statement. (112) Shujun w\u00C3\u00AEpac w\u00C3\u00AE-k\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2w otatosk\u00C3\u00AAwin INDEPENDENT S w\u00C3\u00AEpac w\u00C3\u00AE-k\u00C3\u00AEsiht\u00C3\u00A2-w o-atosk\u00C3\u00AA-win S soon.IPC INT-finish.VAI-3 3-work.VAI-NOM \u00E2\u0080\u0098Shujun\u00E2\u0080\u0099s going to finish her work soon.\u00E2\u0080\u0099 comment: it\u00E2\u0080\u0099s a fact that she\u00E2\u0080\u0099s going to finish her work soon\u00E2\u0080\u00A6 [you could say this] if you know where she\u00E2\u0080\u0099s going to be in her work In fact, the indexical clauses are only felicitous in contexts where the speaker has experienced (observed, heard, felt directly) some part of the event. Take for example the verb of saying itw\u00C3\u00AA- \u00E2\u0080\u0098say thus\u00E2\u0080\u0099 (used in the form itw\u00C3\u00AAwak in the passage in (110) above). As a predicate that overtly codes the flow of information, itw\u00C3\u00AA- has been analyzed as a predicate-level evidential in Plains Cree (Wolfart 1973, Blain & D\u00C3\u00A9chaine 2006, 2007), and is thus a good case to investigate in this section. As in the cases above, an indexical INDEPENDENT clause is used when the speaker heard the speech firsthand, while a non-indexical (CONJUNCT) clause does not have any such restriction. One consultant used a religious context, where source of information becomes especially important, and volunteered a minimal pair in (113) to highlight the distinction between 120 the two clause-types31: the indexical INDEPENDENT clause may only be used when reporting something told by a spirit in the speaker\u00E2\u0080\u0099s presence (cf. Cook & M\u00C3\u00BChlbauer 2007): (113) context: speaker hears the spirit a. \u00C3\u00AAkos \u00C3\u00AEtw\u00C3\u00AAw INDEPENDENT \u00C3\u00AAkosi itw\u00C3\u00AA -w thus thus.say.VAI-3 \u00E2\u0080\u0098that\u00E2\u0080\u0099s what he said\u00E2\u0080\u0099 b. # \u00C3\u00AAkos \u00C3\u00AA-itw\u00C3\u00AAt CONJUNCT \u00C3\u00AAkosi \u00C3\u00AA- itw\u00C3\u00AA -t thus C1-thus.say.VAI-3 \u00E2\u0080\u0098that\u00E2\u0080\u0099s what he said\u00E2\u0080\u0099 (114) context: someone else heard the spirit and reported the event to the speaker a. # \u00C3\u00AAkos \u00C3\u00AEtw\u00C3\u00AAw INDEPENDENT \u00C3\u00AAkosi itw\u00C3\u00AA -w thus thus.say.VAI-3 \u00E2\u0080\u0098that\u00E2\u0080\u0099s what he said\u00E2\u0080\u0099 b. \u00C3\u00AAkos \u00C3\u00AA-itw\u00C3\u00AAt CONJUNCT \u00C3\u00AAkosi \u00C3\u00AA- itw\u00C3\u00AA -t thus C1-thus.say.VAI-3 \u00E2\u0080\u0098that\u00E2\u0080\u0099s what he said\u00E2\u0080\u0099 Just like for first persons, naming predicates provide an important test case for third persons because naming is a performative action (Searle 1965, Ross 1970, Rutherford 1973, among 31 A further distinction in information flow may be made by use of the obviative, as (i) illustrates (see also M\u00C3\u00BChlbauer 2007 for discussion). The distinction in flow of information exhibited here by the clause-typing contrast and obviation is reminiscent of the distinctions discussed in Drapeau (1996), allowing for an interesting way to begin comparing how information flow is coded across the Cree dialect continuum. (i) a. \u00C3\u00AAkos \u00C3\u00AA-itw\u00C3\u00AAt CONJUNCT \u00C3\u00AAkosi \u00C3\u00AA- itw\u00C3\u00AA -t thus C1-thus.say.VAI-3 \u00E2\u0080\u0098that\u00E2\u0080\u0099s what he said\u00E2\u0080\u0099 Explanation: that\u00E2\u0080\u0099s what you\u00E2\u0080\u0099d say if someone told you that heard it b. \u00C3\u00AAkos \u00C3\u00AA-itw\u00C3\u00AAyit CONJUNCT W/ OBVIATIVE \u00C3\u00AAkosi \u00C3\u00AA- itw\u00C3\u00AA -yi -t thus C1-thus.say.VAI-DS-3 \u00E2\u0080\u0098that\u00E2\u0080\u0099s what he said\u00E2\u0080\u0099 Explanation: that\u00E2\u0080\u0099s what you\u00E2\u0080\u0099d say if it was third- or fourth-hand. 121 others): by using a particular name for someone, I as the speaker have direct (experiential) evidence for the name of that person. We thus expect naming predicates referring to third persons to be an example where the indexical clause is the neutral clause-type even for third- persons, and this expectation is fulfilled. Indexical clauses are always used with naming predicates unless the speaker is referring to the name of an individual that the speaker does not personally use (e.g., when talking about a person the speaker does not know). For example, in (115), the speaker first gives, in an INDEPENDENT clause, her kin-term for the man she is talking about (kin terms being used extensively and consistently in Plains Cree as forms of address (Mandelbaum 1940; Wolfart 2000)), and then switches to an anaphoric \u00C3\u00AA- CONJUNCT when she provides the nickname that other people used. Note that there is a speech hiccup in this example (also bolded): the speaker starts to use an indexical clause, stops, and then restarts and uses an anaphoric CONJUNCT clause. (115) \u00E2\u0080\u00A6, \u00E2\u0080\u0098nitaw\u00C3\u00AAm\u00C3\u00A2w\u00E2\u0080\u0099 nik\u00C3\u00AE-it\u00C3\u00A2hk\u00C3\u00B4m\u00C3\u00A2w m\u00C3\u00A2n \u00C3\u00A2na, n\u00C3\u00A2p\u00C3\u00AAw, INDEPENDENT ni(t)- aw\u00C3\u00AAm\u00C3\u00A2w ni- k\u00C3\u00AE- it\u00C3\u00A2hk\u00C3\u00B4m -\u00C3\u00A2 -w m\u00C3\u00A2na ana n\u00C3\u00A2p\u00C3\u00AAw 1- brother.in.law 1- PREV-use.VTA -DIR-3 usually DEM.AN man \u00E2\u0080\u0098..., I used to use the kin-term \u00E2\u0080\u0098my brother-in-law\u00E2\u0080\u0099 for him, for this man, \u00E2\u0080\u0098Black\u00E2\u0080\u0099 k\u00C3\u00AE-isiy\u00C3\u00AEhk-~, nickname anim \u00C3\u00AAkos \u00C3\u00AA-k\u00C3\u00AE-isiy\u00C3\u00AEhk\u00C3\u00A2sot. CONJUNCT B k\u00C3\u00AE- isiy\u00C3\u00AEhk-~ nickname anima \u00C3\u00AAkosi \u00C3\u00AA- k\u00C3\u00AE- isiy\u00C3\u00AEhkaso -t B PREV-thus.be.called.VAI -~ nickname DEM.INAN TOPIC C1-PREV-thus.be.called.VAI-3 his name was Black, that was his nickname.\u00E2\u0080\u0099 (AA 8.1) In fact, naming predicates in the \u00C3\u00AA-conjunct may be accompanied by the indirect evidential \u00C3\u00AAsa, which overtly identifies the source of information as indirect; this is illustrated in (116), where the speaker is talking about her husband\u00E2\u0080\u0099s younger half-brother, whom she had never met. (116) \u00E2\u0080\u00A6, \u00C3\u00AAkw \u00C3\u00AAsa \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00B4ni-pim\u00C3\u00A2tisit; \u00E2\u0080\u0098Paul\u00E2\u0080\u0099 \u00C3\u00AAs \u00C3\u00AAwako \u00C3\u00AA-k\u00C3\u00AE-isiy\u00C3\u00AEhk\u00C3\u00A2sot. \u00C3\u00AAkwa \u00C3\u00AAsa \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00B4ni-pim\u00C3\u00A2tisi-t P \u00C3\u00AAsa \u00C3\u00AAwako \u00C3\u00AA-k\u00C3\u00AE-isiy\u00C3\u00AEhk\u00C3\u00A2so-t and REPORT C1-PREV-stop-live.VAI-3 P REPORT RESUM C1-PREV-thus.be.called.VAI-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6, that one had died; Paul had been his name.\u00E2\u0080\u0099 (EM 31) The contrast between the INDEPENDENT and \u00C3\u00AA-CONJUNCT modes when talking about nonfirst- person referents is sometimes even described by consultants as a difference in information source. In (117a), the source corresponds to direct, sensory input to the speaker; in (117b), the source is indirect \u00E2\u0080\u0093 the information is coming from someone else. 122 (117) a. miyom\u00C3\u00A2cihow Anna INDEPENDENT miyom\u00C3\u00A2ciho -w A feel.well.VAI -3 A \u00E2\u0080\u0098She\u00E2\u0080\u0099s feeling well.\u00E2\u0080\u0099 comment: seems like you\u00E2\u0080\u0099re getting that from seeing her and looking at her b. \u00C3\u00AA-miyom\u00C3\u00A2cihot Anna CONJUNCT \u00C3\u00AA- miyom\u00C3\u00A2ciho -t A C1-feel.well.VAI-3 A \u00E2\u0080\u0098\u00E2\u0080\u00A6she\u00E2\u0080\u0099s feeling well.\u00E2\u0080\u0099 comment: \u00C3\u00AA-miyom\u00C3\u00A2cihot is more like you\u00E2\u0080\u0099re hearing about it 3.3.2.4 \u00C3\u00AAsa has mirative force in indexical clauses So far I have suggested that the evidential interpretation in indexical clauses could be seen as arising from the necessity of evaluating the proposition with respect to the speech situation. In particular, if the speech situation referent (the speaker) was not a participant in the event of the proposition, one way to connect the proposition to the speech situation is if the speaker has some spatio-temporal overlap with the event described in the proposition. Further evidence that indexical clauses must make reference to the speaker comes from the way they interact with other evidentials in Plains Cree. The evidential \u00C3\u00AAsa makes a good test case in that it is usually described as a reportative (Wolfart & Ahenakew 2000:34 gloss it as \u00E2\u0080\u0098reportedly\u00E2\u0080\u0099; Blain & D\u00C3\u00A9chaine 2007:265 explicitly claim it is a reportative). This is based on its use in narratives, such as (118a), where the speaker is retelling a story that had been told to her about events that happened about a hundred years earlier, \u00C3\u00AAsa is marked in almost every clause. (118) \u00E2\u0080\u00A6, \u00C3\u00AAkot\u00C3\u00AA \u00C3\u00AA-sa-s\u00C3\u00A2sakitisihk \u00C3\u00AAsa, ... \u00C3\u00AAkot\u00C3\u00AA \u00C3\u00AA- sa- s\u00C3\u00A2sakitisin -k \u00C3\u00AAsa there C1-RED-lie.on.back.VAI-0 EVID \u00E2\u0080\u0098and he was lying there on his back \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 9.8) When speakers are asked about these clauses in elicitation contexts, the indirect evidential interpretation is confirmed (119), although it is used in contexts broader than just a reportative. As we will see, with anaphoric \u00C3\u00AA-CONJUNCT clauses, it behaves like a general purpose indirect evidential. 123 (119) \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt \u00C3\u00AAsa Miranda \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t \u00C3\u00AAsa M C1-COME-go.VAI-3 EVID M \u00E2\u0080\u0098Apparently Miranda came.\u00E2\u0080\u0099 comment: \u00C3\u00AAsa you use if you didn\u00E2\u0080\u0099t see her yourself as a fact, if you don\u00E2\u0080\u0099t know The evidential \u00C3\u00AAsa expresses indirect evidentiality, and I am trying to show that the indexical INDEPENDENT clause-type requires a connection between the proposition and the speaker. Depending on what the specific semantics of \u00C3\u00AAsa are, the current analysis of indexical INDEPENDENT clauses predicts that the use of \u00C3\u00AAsa should either be excluded from co-occurring with them, or should give rise to a \u00E2\u0080\u0098stacked evidentiality\u00E2\u0080\u0099 effect where the interpretation of \u00C3\u00AAsa is additive to the interpretation of INDEPENDENT clauses (cf. LaPolla 2003). In this section, I show that \u00C3\u00AAsa is, in fact, sensitive to clause-typing.32 The particular interpretation of \u00C3\u00AAsa with indexical INDEPENDENT order clauses is reminiscent of \u00E2\u0080\u0098stacked evidentiality\u00E2\u0080\u0099, providing additional evidence that indexical clauses always make reference to the speaker. 3.3.2.4.1 The interaction of clause-typing and \u00C3\u00AAsa While \u00C3\u00AAsa has traditionally been thought of as a dedicated indirect evidential which always and only conveys that the speaker has no direct knowledge of the state of affairs expressed by the proposition, one finding of the present work is that its interpretation in fact varies depending on clause-typing. The table in 3.10 summarizes the different interpretations of \u00C3\u00AAsa that are available. CLAUSE-TYPE INTERPRETATION OF \u00C3\u008ASA Independent Mirative \u00C3\u00AA-conjunct Indirect Simple conjunct retrospective Table 3.10. Interpretations of \u00C3\u00AAsa33 32 For independent reasons, Blain and D\u00C3\u00A9chaine (2006, 2007) propose that \u00C3\u00AAsa is a CP-level evidential, which predicts on syntactic grounds that \u00C3\u00AAsa should interact with clause-typing. The present findings are consistent with their analysis. 33 \u00C3\u00AAsa also has a distinct interpretation with k\u00C3\u00A2- CONJUNCT clauses, although I have not been able to define it so far. Syntactically, the presence of \u00C3\u00AAsa in a k\u00C3\u00A2- CONJUNCT clause allows the clause to be a matrix clause. 124 In INDEPENDENT order clauses, \u00C3\u00AAsa is used to convey the speaker\u00E2\u0080\u0099s surprise at the state of affairs expressed by the proposition. Cross-linguistically, constructions which convey this surprise can be said to have a mirative function and are found in many languages (cf. DeLancey 1997, 2001). The state of affairs is directly experienced by the speaker, but is contrary to their previous expectations; miratives are variously summarized as \u00E2\u0080\u009Cunexpected information\u00E2\u0080\u009D (DeLancey 1997), or \u00E2\u0080\u009Cunprepared mind\u00E2\u0080\u009D (Aikhenvald 2004). It should be noted that it is very common (though not universal) for the expression of mirativity to be the result of an \u00E2\u0080\u0098indirect\u00E2\u0080\u0099 evidential occurring with some other element (Aikhenvald 2004). (120) mirativity: expressed via INDEPENDENT clause and \u00C3\u00AAsa a. nis\u00C3\u00B4niyamin ni-s\u00C3\u00B4niyami -n 1- have.money.VAI-SAP \u00E2\u0080\u0098I have money.\u00E2\u0080\u0099 b. nis\u00C3\u00B4niyamin \u00C3\u00AAsa ni-s\u00C3\u00B4niyami -n \u00C3\u00AAsa 1-have.money.VAI-SAP EVID \u00E2\u0080\u0098I have money!?\u00E2\u0080\u0099 comment: if you have money in your wallet that you didn\u00E2\u0080\u0099t know you had I will return to the discussion of \u00C3\u00AAsa and INDEPENDENT clauses below in \u00C2\u00A73.3.2.4.2. If \u00C3\u00AAsa co-occurs with a non-indexical \u00C3\u00AA-CONJUNCT, the interpretation is that the speaker has only indirect knowledge of the state of affairs expressed by the proposition. This is the interpretation that is well-known and usually cited in the Algonquian literature. (121) indirect evidence: expressed by \u00C3\u00AA-CONJUNCT and \u00C3\u00AAsa a. \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt Miranda \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t M C1-COME-go.VAI-3 M \u00E2\u0080\u0098Miranda came.\u00E2\u0080\u0099 b. \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt \u00C3\u00AAsa Miranda \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t \u00C3\u00AAsa M C1-COME-go.VAI-3 EVID M \u00E2\u0080\u0098Apparently Miranda came.\u00E2\u0080\u0099 comment: \u00C3\u00AAsa you use if you didn\u00E2\u0080\u0099t see her yourself as a fact, if you don\u00E2\u0080\u0099t know 125 Finally, \u00C3\u00AAsa may be used with simple CONJUNCT clauses to express past in a modal construction (see chapter 6 for discussion). (122) retroactive: expressed by simple CONJUNCT and \u00C3\u00AAsa a. m\u00C3\u00B4y ka-ki-fonahwak nim\u00C3\u00A2ma m\u00C3\u00B4y ka- k\u00C3\u00AE- fonahw -ak ni- m\u00C3\u00A2ma NEG IRR-PREV-call.VTA -1>3 1- mother \u00E2\u0080\u0098I shouldn\u00E2\u0080\u0099t call my mom.\u00E2\u0080\u0099 (trying to decide what to do) b. m\u00C3\u00B4y ka-ki-fonahwak nim\u00C3\u00A2ma \u00C3\u00AAsa m\u00C3\u00B4y ka- k\u00C3\u00AE- fonahw -ak ni- m\u00C3\u00A2ma \u00C3\u00AAsa NEG IRR-PREV-call.VTA -1>3 1- mother EVID \u00E2\u0080\u0098I shouldn\u00E2\u0080\u0099t have called my mother.\u00E2\u0080\u0099 (after having a bad conversation) Here \u00C3\u00AAsa has a temporal shifting function (cf. Wolvengrey 2001), and explicitly highlights what Blain & D\u00C3\u00A9chaine call \u00E2\u0080\u0098the retrospective component\u00E2\u0080\u0099 of indirect evidentiality (Blain & D\u00C3\u00A9chaine 2007). These three contexts do not contradict the characterization of \u00C3\u00AAsa as an indirect evidential in that many indirect evidentials are reported to have a retrospective aspect and many also are reported to be part of a mirative construction, including Turkish (Turkic; Aksu-Koc & Slobin 1986), Hare (Athabaskan; DeLancey 1990, 1997), Kham (Tibetan; DeLancey 1992, 1997), and Tsafiki (Barbacoan; Dickinson 2000). The contexts do provide evidence that the semantic characterization of \u00C3\u00AAsa is more abstract than the convenient label \u00E2\u0080\u0098indirect evidential\u00E2\u0080\u0099 conveys. It may be more accurate to think of \u00C3\u00AAsa as coding distance between the speaker and the proposition, whether it be perceptual distance (yielding the indirect evidential interpretation), psychological distance (as in the mirative interpretation), or temporal distance (yielding the temporal shifting interpretation) (cf. the discussion of distantive force in East Cree (James et al. 2001), and in Korean (Chung 2005)). Finally, the prediction made by the current analysis of INDEPENDENT clauses, that \u00C3\u00AAsa should interact with them, is upheld. 3.3.2.4.2 Mirativity as incongruent experience Since mirativity conveys the speaker\u00E2\u0080\u0099s surprise, it is important to show that the mirative interpretation is available regardless of whether or not the speaker is coded in the clause (e.g., as an argument of the predicate) in order to demonstrate that it is the indexical clause-type which 126 introduces this meaning. In fact, we see that the mirative interpretation is present in first-person (123), second-person (124), and third-person (including inanimate) predicates (125), provided they are INDEPENDENT clauses. In (123), the speaker\u00E2\u0080\u0099s state of tiredness is unexpected given the speaker\u00E2\u0080\u0099s knowledge state up until that point. The combination of the clause-typing and the evidential conveys information that is incongruent with the speakers \u00E2\u0080\u0098premonitory awareness\u00E2\u0080\u0099 (Aksu-Koc & Slobin 1986, 1988; Dickinson 2000). (123) a. nin\u00C3\u00AAstosin ni- n\u00C3\u00AAstosi -n 1- tired.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m tired.\u00E2\u0080\u0099 b. nin\u00C3\u00AAstosin \u00C3\u00AAsa ni- n\u00C3\u00AAstosi -n \u00C3\u00AAsa 1- tired.VAI-SAP EVID \u00E2\u0080\u0098I\u00E2\u0080\u0099m tired?!\u00E2\u0080\u0099 comment: you didn\u00E2\u0080\u0099t know that you were tired until after you stopped or went out for a breath of fresh air. Maybe you only worked for a short time and suddenly you were tired Such a description is also consistent with the context in (122), where the speaker is expecting exactly the opposite of what actually happens. In this context, \u00C3\u00AAsa must be used with the indexical INDEPENDENT order, rather the CONJUNCT, showing that it is the combination of the evidential marker and the clause-typing which conveys the mirativity. (124) context: speaker believes hearer isn\u00E2\u0080\u0099t coming, but hearer unexpectedly shows up a. (Oh) ki-p\u00C3\u00AA-itoht\u00C3\u00A2n \u00C3\u00AAsa! INDEPENDENT Oh ki- p\u00C3\u00AA- itoht\u00C3\u00A2 -n \u00C3\u00AAsa Oh 2- DIR-go.VAI-SAP EVID \u00E2\u0080\u0098Oh, you came!\u00E2\u0080\u0099 b. # Oh, \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAyan \u00C3\u00AAsa CONJUNCT Oh \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -yan \u00C3\u00AAsa Oh C1-DIR-go.VAI-2 EVID \u00E2\u0080\u0098Oh, you came!\u00E2\u0080\u0099 127 Even with predicates that have no overt arguments at all, such as weather verbs, \u00C3\u00AAsa can be used with the INDEPENDENT order to convey surprise on the speaker\u00E2\u0080\u0099s part. Insofar as speaking to oneself involves treating oneself as another, the consultant\u00E2\u0080\u0099s comment on this piece of data, \u00E2\u0080\u0098it\u00E2\u0080\u0099s more something I would think or say to myself\u00E2\u0080\u0099 also suggests the speaker\u00E2\u0080\u0099s psychological distance (conveyed by \u00C3\u00AAsa) from the state of affairs being experienced. (125) context: didn\u00E2\u0080\u0099t know it was raining, step outside; OR wake up in the morning, look out the window kimiwan \u00C3\u00AAsa kimiwan \u00C3\u00AAsa rain.VII-0 EVID \u00E2\u0080\u0098It\u00E2\u0080\u0099s raining.\u00E2\u0080\u0099 comment: this sounds kind of funny in conversation; it\u00E2\u0080\u0099s more something I would think or say to myself Turning next to the distribution of \u00C3\u00AAsa in running speech, we see confirmation of these judgments. First, in the narratives I have worked with, \u00C3\u00AAsa occurs much more freely with anaphoric \u00C3\u00AA-conjunct clauses than with the indexical INDEPENDENT clauses; given that reported narratives will have a lot of indirect evidentiality, this is expected. For example, in the following passage, \u00C3\u00AAsa is marked on virtually every clause. The striking exception to this pattern is when an INDEPENDENT clause is used: in all three cases (bracketed with the clause bolded), \u00C3\u00AAsa is missing. These three examples instantiate the only INDEPEDENT clauses in this span. The first independent clause is when the speaker is explaining background information about the structure of the lodge. The other two independent clauses mark the crucial point of the story: while the speaker may not know exactly all the details leading up to the shooting of the bird, the shooting of the bird did happen \u00E2\u0080\u0093 without that, there is no story. Hence, these clauses appear in the INDEPENDENT order, and \u00C3\u00AAsa is no longer used. 128 (126) \u00C3\u00AAkwa, kis\u00C3\u00AAyiniw aw \u00C3\u00AAkwa \u00C3\u00AAsa \u00C3\u00AA-s\u00C3\u00A2sakitisihk, m\u00C3\u00AAtoni mistah \u00C3\u00AA-m\u00C3\u00A2miton\u00C3\u00AAyihtahk \u00C3\u00B4ma m\u00C3\u00A2ka p\u00C3\u00A2skisikan wiy \u00C3\u00AAsa \u00C3\u00AA-k\u00C3\u00AE-nakatam\u00C3\u00A2ht \u00C3\u00AAsa. [\u00C3\u00AAkwa \u00C3\u00B4hi m\u00C3\u00AEkiw\u00C3\u00A2hpa ayis taw\u00C3\u00A2wa \u00C3\u00B4t\u00C3\u00AA ispimihk,] \u00C3\u00AAkwa March anim \u00C3\u00AA-ispayik \u00C3\u00AAsa, \u00C3\u00AAkot\u00C3\u00AA \u00C3\u00AA-sa-s\u00C3\u00A2sakitisihk \u00C3\u00AAsa, k\u00C3\u00AAtahtaw \u00C3\u00AAsa k\u00C3\u00A2- w\u00C3\u00A2pam\u00C3\u00A2t \u00C3\u00AAkot\u00C3\u00AA \u00C3\u00AA-pimih~, ispimihk \u00C3\u00AA-pimihy\u00C3\u00A2yit niska. m\u00C3\u00AAton \u00C3\u00AAsa k\u00C3\u00A2-\u00C3\u00AAkicik\u00C3\u00A2wit \u00C3\u00AAs \u00C3\u00AA- isit\u00C3\u00A2cimot iskw\u00C3\u00A2ht\u00C3\u00AAmihk, \u00C3\u00AAkota \u00C3\u00AA-pimakociniyit \u00C3\u00AAsa k\u00C3\u00A2p\u00C3\u00A2skisw\u00C3\u00A2t \u00C3\u00AAkoni anihi niska; [nipah\u00C3\u00AAw, p\u00C3\u00AA-n\u00C3\u00AEhtakociniyiwa.] (AA 9.8, brackets added) Now the old man was lying on his back, thinking about things a very great deal \u00E2\u0080\u0093 but he had been left with a gun. [Now, these lodges are open at the top, of course,] and it was March at the time, and he was lying there on his back when suddenly he saw some geese flying overhead. He was very slow in crawling to the door, and as the geese flew over he shot at them; [he killed one and it came falling down.] Second, in at least some cases where \u00C3\u00AAsa occurs with the indexical INDEPENDENT in textual sources, it does not have the same interpretation as when it occurs with the \u00C3\u00AA-conjunct34. For example, in (127), taken from a little later in the same story as above, \u00C3\u00AAsa is used twice. The first clause is the CONJUNCT \u00C3\u00AA-p\u00C3\u00AA-takoht\u00C3\u00AAcik \u00E2\u0080\u0098\u00E2\u0080\u00A6they arrived\u00E2\u0080\u0099, and \u00C3\u00AAsa has an (untranslated) reportative function. In the second clause, \u00C3\u00AAsa occurs with an INDEPENDENT clause k\u00C3\u00AE- pap\u00C3\u00A2moht\u00C3\u00AAyiwa \u00E2\u0080\u0098they were walking about\u00E2\u0080\u0099, and the event is surprising; this appears to be a mirative interpretation of \u00C3\u00AAsa (cf. Bloomfield 1962, Macaulay 2004 on the mirative function of - \u00C3\u00AAsa in Menominee)35. (127) \u00C3\u00AA-p\u00C3\u00AA-takoht\u00C3\u00AAcik \u00C3\u00AAsa, CONJUNCT \u00C3\u00AA-p\u00C3\u00AA-takoht\u00C3\u00AA-t-k \u00C3\u00AAsa C1-COME-arrive.VAI-3-PL EVID p\u00C3\u00B4t \u00C3\u00B4hi k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AAyiwa \u00C3\u00AAsa [laughter], ... INDEPENDENT p\u00C3\u00B4ti ohi k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AA-yi-w-a \u00C3\u00AAsa behold DEM PREV-walk.about.VAI-DEP-3-OBV EVID \u00E2\u0080\u0098When they arrived there, behold, the old people were walking about, ...\u00E2\u0080\u0099 (AA 9.8) In this example, there is also the particle of surprise p\u00C3\u00B4ti \u00E2\u0080\u0098behold\u00E2\u0080\u0099, which is cross-linguistically a common type of element to occur with mirative clauses (cf. Aksu-Koc & Slobin 1986, 34 The examples cited below offer cases where the non-reportative reading is clear from the linguistic context. In some other cases, this reading is not clear from the linguistic context; in such instances, more work with fluent speakers is needed to understand the interpretation of \u00C3\u00AAsa. 35 The mirative reading of \u00C3\u00AAsa also appears in nominal clauses, such as (i), whereby the speaker expresses surprise at the situation she finds herself in. (i) isp\u00C3\u00AE \u00C3\u00AAkwa \u00C3\u00AA-w\u00C3\u00A2saskot\u00C3\u00AAnik\u00C3\u00AAhk aya, c\u00C3\u00AEk \u00C3\u00AAs \u00C3\u00B4ma nip\u00C3\u00AAwinihk \u00C3\u00AA-n\u00C3\u00AEpawiy\u00C3\u00A2n; \u00E2\u0080\u0098Then, when they lit the lamp, here I was standing close to the bed;\u00E2\u0080\u0099 (Minde 1998:\u00C2\u00A741) 129 Dickenson 2000; Aikhenvald 2004), and in Plains Cree has been independently observed to occur with \u00C3\u00AAsa when surprise is being expressed (Wolfart & Ahenakew 1998:165). Similarly, in another case (128), \u00C3\u00AAsa and an INDEPENDENT clause co-occur when the speaker is reporting a speech that she seems to believe will surprise her audience. In fact, she goes on to state that the event was one she personally witnessed. Thus the presence of \u00C3\u00AAsa does not preclude that the proposition was directly witnessed by the speaker. (128) k\u00C3\u00AE-w\u00C3\u00AEhtamaw\u00C3\u00AAw \u00C3\u00AAsa ayisiyiniwa, \u00E2\u0080\u009Cs\u00C3\u00B4skw\u00C3\u00A2c nama k\u00C3\u00AEkway \u00C3\u00AAwakw \u00C3\u00A2nima ninitaw\u00C3\u00AAyiht\u00C3\u00AAn niy \u00C3\u00A2ya, k-\u00C3\u00A2tosk\u00C3\u00A2tam\u00C3\u00A2n, niw\u00C3\u00AE-\u00C3\u00B4m-\u00C3\u00A2ya-~ niw\u00C3\u00AE-t\u00C3\u00B4t\u00C3\u00AAn \u00C3\u00B4ma, niw\u00C3\u00AE-okistik\u00C3\u00AAwiyin\u00C3\u00AEwin, \u00C3\u00AAkos \u00C3\u00AAwako niwi-~ niw\u00C3\u00AE-kis\u00C3\u00A2t\u00C3\u00AAn, os\u00C3\u00A2m mistahi nika-waniht\u00C3\u00A2n nitatosk\u00C3\u00AAwin, \u00C3\u00B4ma okim\u00C3\u00A2hk\u00C3\u00A2niwiy\u00C3\u00A2ni,\u00E2\u0080\u009D \u00C3\u00AA-k\u00C3\u00AE-itw\u00C3\u00AAt, n\u00C3\u00AEsta nik\u00C3\u00AE-p\u00C3\u00AAhtaw\u00C3\u00A2w \u00C3\u00AAkos \u00C3\u00AA-itw\u00C3\u00AAt. (EM 27) He had told the people, \u00E2\u0080\u009CI simply do not want that kind of work at all; I am going to do this, I am going to farm, and so I am going to stay with it, because I will lose too much of my working time if I am a chief,\u00E2\u0080\u009D he had said, and I myself had heard him say that. To summarize, the interpretation of \u00C3\u00AAsa is consistently distinguished on the basis of clause- typing. In particular, with INDEPENDENT clauses, \u00C3\u00AAsa conveys surprise at the state of affairs experienced by the speaker; it does not convey lack of speaker\u00E2\u0080\u0099s experience. If the reference to speaker were merely a default value \u00E2\u0080\u0093 an implication that could be cancelled \u00E2\u0080\u0093 we could expect \u00C3\u00AAsa to have its regular \u00E2\u0080\u0098reportative\u00E2\u0080\u0099 function. The fact that \u00C3\u00AAsa has a mirative function is thus evidence that indexical clauses are specified: have a fixed referential deixis such that even when there is no first-person marking in the clause, the speaker\u00E2\u0080\u0099s perspective on the proposition is always present. 3.3.2.5 Speaker commitment to the proposition A common property of direct evidentials is that they can be licensed in restricted contexts where the speaker has integrated information into their knowledge base, even if the information is not part of their personal experience (Dickenson 2001, Aikhenvald 2004, among others). This must be information from a trusted source, and there is often a time lapse between the time when the speaker learned the information and the time when the speaker conveys the information to someone else. For example, Dickenson (2001) reports that in Tsafiki, one speaker used an 130 indirect evidential immediately after finding out from his mother the city where he was born, but a day later used the direct evidential. In terms of the current discussion, the relevant part of the phenomena is who has an epistemic commitment to the proposition \u00E2\u0080\u0093 it is the speaker. If indexical clauses always contain reference to the speech act, than the person epistemically committed to the proposition should be the speaker. This expectation is also fulfilled. For example, (129) was uttered in the context of the speaker visiting with the subject \u00E2\u0080\u0093 a clear case of direct experience, and a place where we would expect a direct evidential to be licensed. The speaker then comments that this utterance could also imply that you know about Betty\u00E2\u0080\u0099s illness from someone else. (129) context: Speaker visited with Betty \u00C3\u00A2hkosiw Betty an\u00C3\u00B4s INDEPENDENT \u00C3\u00A2hkosi -w B anohc sick.VAI-3 B today \u00E2\u0080\u0098Betty\u00E2\u0080\u0099s sick today.\u00E2\u0080\u0099 comment: you\u00E2\u0080\u0099re just stressing that that\u00E2\u0080\u0099s a fact. It\u00E2\u0080\u0099s for sure that she\u00E2\u0080\u0099s sick. It seems like you know that or you believe it from someone else When an explicit context of reporting was set up (as in the following example), the speaker lays out the restrictions necessary for the indexical INDEPENDENT to be felicitous, and contrasts it with a second example with a non-indexical CONJUNCT clause. This latter clause does not make reference to the speaker\u00E2\u0080\u0099s commitment to the proposition. 131 (130) context: someone told speaker that Betty was sick today a. ahkosiw Betty an\u00C3\u00B4s INDEPENDENT ahkosi -w B anohc sick.VAI-3 B today \u00E2\u0080\u0098Betty\u00E2\u0080\u0099s sick today.\u00E2\u0080\u0099 comment: if you know this person, and you know this person is pretty honest and reliable, and that they\u00E2\u0080\u0099re not going to lie. [Otherwise], you would have to add \u00E2\u0080\u0098I heard\u00E2\u0080\u0099\u00E2\u0080\u00A6you\u00E2\u0080\u0099re not going make such an active statement b. \u00C3\u00AA-ahkosit Betty an\u00C3\u00B4s CONJUNCT: NEUTRAL \u00C3\u00AA- ahkosi -t B anohc C1-sick.VAI-3 B today \u00E2\u0080\u0098Betty\u00E2\u0080\u0099s sick today.\u00E2\u0080\u0099 comment: this covers your tracks a little more. It doesn\u00E2\u0080\u0099t feel as strong as [the INDEPENDENT]. this one has wiggle room. Usually that one is you heard it. What it means is it\u00E2\u0080\u0099s not first-hand knowledge Likewise, in narrative, indexical clauses can be used to convey emphatic certainty of an event: there is a contrast between INDEPENENT and CONJUNCT clauses in terms of whether the speaker is committed to the proposition. As one speaker commented on listening to this passage: \u00E2\u0080\u009CShe starts off questioning: \u00E2\u0080\u0098I\u00E2\u0080\u0099m guessing why it happened, but it did happen.\u00E2\u0080\u0099\u00E2\u0080\u009D Notice that the English translation of the indexical k\u00C3\u00AE-nipah\u00C3\u00AAw \u00E2\u0080\u0098s/he killed someone\u00E2\u0080\u0099 has the emphatic did, used for emphatic affirmatives. (131) k\u00C3\u00A2-p\u00C3\u00AEhtam\u00C3\u00A2n ana \u00C3\u00AEskw\u00C3\u00AAw \u00C3\u00AA-nipah\u00C3\u00A2t on\u00C3\u00A2p\u00C3\u00AAma, \u00C3\u00AA-p\u00C3\u00A2skisw\u00C3\u00A2t. \u00C3\u00AAkosi k\u00C3\u00AE-~ nik\u00C3\u00AE-koskw\u00C3\u00AAyiht\u00C3\u00AAn, mistah \u00C3\u00A2yis \u00C3\u00AA-k\u00C3\u00AE-miyohtw\u00C3\u00A2t mistah \u00C3\u00A2na \u00C3\u00AEskw\u00C3\u00AAw, miton \u00C3\u00AAtikw\u00C3\u00AA kwayask \u00C3\u00AA-k\u00C3\u00AE-kisiw\u00C3\u00A2hikot anihi k\u00C3\u00A2-k\u00C3\u00AE-p\u00C3\u00A2skisw\u00C3\u00A2t on\u00C3\u00A2p\u00C3\u00AAma, k\u00C3\u00AE-nipah\u00C3\u00AAw. (AA 5.6) \u00E2\u0080\u00A6 later I heard that this woman killed her husband, she shot him. So I was greatly shocked, for that woman had been very good-natured, she must have been angered exceeding by her husband when she shot him, and she did kill him. 3.3.2.6 Subjective predicates convey speaker\u00E2\u0080\u0099s attitude The last piece of evidence that indexical clauses have deictic reference to the speaker comes from the fact that predicates which lack any first person argument may be used in the indexical clause-type to convey the speaker\u00E2\u0080\u0099s opinion. For example, the predicate miyw\u00C3\u00A2si- \u00E2\u0080\u0098it (inan.) is 132 good\u00E2\u0080\u0099 in (130) is interpreted as good with respect to the speaker. Thus, (130) was offered as a translation for the English \u00E2\u0080\u0098I like this chair\u00E2\u0080\u0099; significantly, the anaphoric CONJUNCT clause-type was not judged an appropriate translation. (132) context: translation task for \u00E2\u0080\u0098I like this chair\u00E2\u0080\u0099 a. m\u00C3\u00AEyw\u00C3\u00A2sin \u00C3\u00B4ma t\u00C3\u00AAhtapiwin INDEPENDENT m\u00C3\u00AEyw\u00C3\u00A2sin \u00C3\u00B4ma t\u00C3\u00AAhtapi -win (offered) good.VII DEM.INAN sit -NOM \u00E2\u0080\u0098This is a nice chair.\u00E2\u0080\u0099 comment: by saying that, it implies that you like it comment: if you use m\u00C3\u00AEyw\u00C3\u00A2sin about something that someone else has, then the other person has to give it to you. It\u00E2\u0080\u0099s very powerful b. # \u00C3\u00AA-miyw\u00C3\u00A2sik \u00C3\u00B4ma t\u00C3\u00AAhtapiwin CONJUNCT \u00C3\u00AA-miyw\u00C3\u00A2si-k \u00C3\u00B4ma t\u00C3\u00AAhtapi-win (presented) C1-good.VII-0 DEM sit-NOM \u00E2\u0080\u0098This is a nice chair.\u00E2\u0080\u0099 comment: you\u00E2\u0080\u0099re saying it\u00E2\u0080\u0099s nice so it could mean you like it\u00E2\u0080\u00A6 I would never say this if I wanted to be taken as liking this chair It is significant that the type of predicate where this effect shows up most strongly is in \u00E2\u0080\u0098subjective\u00E2\u0080\u0099 predicates (cf. Lasersohn 2005, Stephenson 2007). In some languages, where the indexical status of the clause is not marked, these predicates introduce an oblique phrase such as the German mir \u00E2\u0080\u0098to/for me\u00E2\u0080\u0099. In Plains Cree, where the indexical status is morpho-syntactically marked by the clause-typing, the relation to the speaker is already given by the presence of the speech situation variable. 3.3.2.7 Interim summary In this section I have reported on a number of previously undocumented facts about the interpretation of indexical INDEPENDENT clauses relating to the presence of a speaker coded in the speech situation (s0) variable. Indexical clauses pair the proposition with the speech situation; since a speech situation always entails a speaker, by transitivity indexical clauses pair the proposition with a speaker. Clause-typing of intentional predicates thus shows marked 133 person effects. When the subject of the predicate is identical with the speaker (first person), there is a match between one of the individuals in the proposition and the individual that the proposition is paired with; there is a strong tendency to use indexical clauses. When the subject of the predicate is distinct from the speaker (e.g., a third person), then use of an indexical clause means that the speaker has some other relation to the proposition: the speaker may have experienced the event coded by the proposition, be epistemically committed to the proposition, or be providing an evaluation of the proposition. 3.4 Summary: Structural and semantic conditions on indexical clauses This chapter put forward the claim that indexical clauses have both structural and semantic conditions on them that separate them from non-indexical clauses. Structurally, indexical clauses are subject to anti-c-command: they can never be embedded, and they cannot be preceded. Semantically, indexical clauses are temporally and referentially indexical: temporal relations are always and only calculated with respect to speech time, and reference always includes reference to a speaker. In the next chapter I examine non-indexical clauses \u00E2\u0080\u0093 Plains Cree\u00E2\u0080\u0099s anaphoric CONJUNCT order. 134 CHAPTER 4 ANAPHORIC CLAUSES: PLAINS CREE\u00E2\u0080\u0099S CONJUNCT ORDER 4.1 Proposal: Anaphoric clauses In chapter 3, we saw that indexical clauses are anchored in the discourse in a particular way. We looked at the syntax and semantics of indexical clauses as two sides of the same coin: in terms of their syntax, we saw they could not be c-commanded, and in terms of their semantics, we saw that the proposition in an indexical clause is evaluated with respect to the (indexical) speech situation. A clause that is not anchored in this way correspondingly lacks the restrictions of indexical clauses. In Plains Cree, for example, the CONJUNCT order of clauses can occur in both matrix and embedded environments (1a-b; cf. Wolfart 1973, 1996; Dahlstrom 1991; Blain 1997; Cook & M\u00C3\u00BChlbauer 2006; Cook 2007); (1) a. \u00C3\u00AA-w\u00C3\u00A2pamak atim MATRIX \u00C3\u00AA- w\u00C3\u00A2pam -ak atim C1-see.VTA-1>3 dog \u00E2\u0080\u0098\u00E2\u0080\u00A6I see a dog.\u00E2\u0080\u0099 b. nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-w\u00C3\u00A2pamak atim EMBEDDED ni- kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- w\u00C3\u00A2pam -ak atim 1- know.VTI -SAP C1-see.VTA -1>3 dog \u00E2\u0080\u0098I know I saw a dog.\u00E2\u0080\u0099 In Plains Cree, then, we have a system which morpho-syntactically distinguishes between an indexical and a non-indexical clause. In this chapter, I provide an account for the non-indexical clauses. I take the distinction between indexical and anaphoric pronominal forms (Bar-Hillel 1954, Kaplan 1989, among many others) and extend it to clauses: I claim that a non-indexical clause, lacking the specification of an indexical clause, must have an antecedent \u00E2\u0080\u0093 in the same way that a pronominal anaphor does. 135 I then argue that just as the anti-c-command and anti-precedence conditions syntactically model the requirement that dependencies be resolved clause-internally in indexical clauses, the absence of these restriction syntactically models the possibility of dependency relations being established cross-clausally. The behaviour of the variables within a clause directly reflects how the clause is syntactically introduced into the discourse. If the clause itself is subject to the principles of anaphora (i.e., precedence and/or c-command, to be made more explicit below), the variables within the clause may be bound by an antecedent that obeys those same principles. Following Williams (1997), who takes anaphora to be an 'elsewhere' case, I claim that the anaphoric properties we see in anaphoric clauses are what arise in the absence of other restrictions. 4.2 From pronominal to clausal anaphora In the rich literature on anaphoric argument expressions, an element is said to be anaphoric if its reference is not fixed, but is rather determined by some other expression (cf. Hockett 1958, Ross 1969, Langacker 1969, McCawley 1988, Reinhart 1983, Safir 2004, among others). This other expression is called the antecedent, and an anaphoric element is coreferential (since it \u00E2\u0080\u0098co- refers\u00E2\u0080\u0099) with the antecedent. For example, in (2a), we cannot tell who him refers to, since Mary is female and therefore not a possible antecedent; in (2b), we understand that him refers to the same person as Tom (indicated by the matching indices), (although it could also \u00E2\u0080\u0093 infelicitously in this context \u00E2\u0080\u0093 refer to some other individual not mentioned in the sentence). (2) a. # Mary is angry at him. b. Tomi doesn\u00E2\u0080\u0099t know that Mary is angry at himi. In (2), then, him is anaphoric, and in (2b) if him corefers with Tom, then Tom is the antecedent. There are at least two kinds of questions that we must address in order to understand what an anaphoric clause is. First, what are the possible forms of an anaphoric element? Second, what is the nature of the antecedent (i.e., its form and relation to the anaphoric element)? 136 4.2.1 The forms of anaphoric elements If DPs (i.e., argument expressions) are the anaphoric elements we are talking about, we observe two different forms: zero-anaphora and proform anaphora. A zero anaphor has no phonological content at all (e.g., the phonologically-empty subject in 3a). A proform has phonological content that stands in for the antecedent (e.g., him is anaphoric on John in 3b). (3) a. [John]i wants [\u00EF\u0081\u00B8]i to eat breakfast. b. Mary called [John]i and invited [him]i over. These two forms can be represented as in (4). (4) a. [antecedent] [ \u00EF\u0081\u00B8 ] ZERO-ANAPHOR b. [antecedent] [ anaphor ] PROFORM ANAPHOR Notice that these forms are not necessarily restricted to DPs. For example, both zero-anaphora and proform anaphora can be found in English for much larger constituents such as predicates (VPs) and entire clauses (CPs) (cf. McCawley 1988, Williams 1997). (5) Predicate anaphora a. John [broke an arm]i on Tuesday and Mary did [\u00EF\u0081\u00B8]i on Wednesday. b. I [climbed on the house-roof]i, and John did [it]i too. (6) Clausal anaphora a. Sam kept arguing [that Lenin was Jewish]i , but he couldn\u00E2\u0080\u0099t convince us [\u00EF\u0081\u00B8]i. b. Mary said [that Roger was an idiot]i , but I\u00E2\u0080\u0099m sure she doesn\u00E2\u0080\u0099t really think [so]i. (from McCawley 1988:319) Anaphora can also be embedded within a larger structure. For example, in (7) the pronominal form his is embedded within the object DP, and has as its antecedent the subject DP John. 137 (7) [John]i loves [ [his]i dog]. Any time we have a constituent with subconstituents, it is possible that an anaphoric element will be one of those subconsitituents. Here I claim that Plains Cree has a type of clause that hosts embedded anaphoric elements: the CONJUNCT order. In particular, whereas indexical clauses are evaluated with respect to a speech situation (8a), anaphoric CONJUNCT clauses are evaluated with respect to an anaphoric situation: there must always be an antecedent for it. (8) a. [CP s0 [C \u00E2\u0080\u00A6 ] ] INDEXICAL: situation is indexical b. [antecedent]i [CP si [C \u00E2\u0080\u00A6 ] ] ANAPHORIC: situation is anaphoric Therefore, even when an anaphoric clause is a matrix clause, it must have an antecedent in order for the proposition to be evaluated. Just as the sense of a DP with an embedded anaphor \u00E2\u0080\u0093 such as his dog \u00E2\u0080\u0093 cannot be resolved without an antecedent1, so the sense of a CP with an anaphorically given situation cannot be resolved without an appropriate antecedent. This leads us to the next question: what can be a possible antecedent for an anaphor? 4.2.2 The relation between anaphor and antecedent The second set of properties about anaphora that we must understand in order to evaluate anaphoric clauses is with respect to their relation to the antecedent. This has to do with how an anaphoric element is licensed: with what may an anaphoric element corefer, and what relation must hold between the anaphoric element and its antecedent? Antecedent licensing of anaphoric elements has received a huge amount of study in formal linguistics. Williams (1997) observes that the antecedent licensing of English anaphoric elements is sensitive both to c-command and precedence: the anaphor must either follow or be in a subordinate relation to the antecedent. If the anaphor it follows its antecedent term paper, it may occur in either a matrix clause (10a) or a subordinate clause (10b); but if the anaphor 1 Actually, English pronominal forms like him are not inherently anaphoric \u00E2\u0080\u0093 they are simply proforms. As such, their reference may be resolved either anaphorically (the part of the puzzle relevant here) or deictically. The latter case covers gestural pointing, as well as salient (Heim & Kratzer 1998, Kratzer 2007) or symbolic (Fillmore 1975) reference. 138 precedes its antecedent, it must be in a subordinate clause2 as in (10c); otherwise, the utterance becomes ungrammatical (10d). (9) Generalized pattern of anaphoric dependence a. Anyone [who has written their term paperi ] can turn iti in to me now. b. Anyone can turn their term paperi in to me now [who has written iti ]. c. Anyone [who has written iti ] can turn their term paperi in to me now. d. * Anyone can turn iti in to me now [who has written their TERM PAPERi]. (Williams 1997, (22)) The pattern in (10a-d) is termed the Generalized Pattern of Anaphoric Dependence (GPAD) and is summarized by Williams as \u00E2\u0080\u009CDependence can be forward; or it can be \u00E2\u0080\u0098backward and down\u00E2\u0080\u0099.\u00E2\u0080\u009D (Williams 1997:588; cf. also Lakoff 1967, Ross 199, McCawley 1988). According to this pattern, there are two conditions in which antecedent licensing of an anaphoric element may take place: in a c-command condition, where the antecedent is in a clause that c-commands the clause containing the anaphor; and in a precedence condition, where the antecedent precedes the anaphoric element. It is important to realize that the precedence condition does not specify how far the antecedent may be from the co-referent anaphor. Consider for example the following actual situation. My husband and I were reading a story about a girl in England who befriends a robin redbreast. Being from North America, we weren\u00E2\u0080\u0099t familiar with what a robin redbreast was, and were particularly puzzled since we kept thinking of North American robins \u00E2\u0080\u0093 which this didn\u00E2\u0080\u0099t seem to be. Three days later, while on a walk and talking about something unrelated (probably linguistics), we walked by a North American robin on someone\u00E2\u0080\u0099s lawn. My husband looked at it for a moment, then abruptly changed the topic, saying (9). (10) It can\u00E2\u0080\u0099t be like the robins around here. In this utterance, the pronoun it refers back to the robin redbreast in the story we had read three days earlier. In order to understand the reference of this pronoun (and I did), the hearer needs to have a context previously established with the speaker. The utterance in (9) is crucially not in a discourse initial position, and this is what defines anaphoric (from Greek \u00CE\u00B1\u00CE\u00BD\u00CE\u00B1\u00CF\u0086\u00CF\u0089\u00CF\u0081\u00CE\u00B5\u00CE\u00B9\u00CE\u00BD \u00E2\u0080\u0098to carry 2 Notice that the anaphoric element does not need to be directly c-commanded by the antecedent; thus a relative clause as in the examples is a sufficient condition for an antecedence relation to be established. 139 forward\u00E2\u0080\u0099) elements; they always look back to something else, on which they are dependent for reference. Thus we expect anaphoric clauses to have a \u00E2\u0080\u0098common-ground\u00E2\u0080\u0099 effect, whereas indexical clauses will not. This I cover in \u00C2\u00A74.3.2. As evidence for my claim that anaphoric clauses correspond to Plains Cree\u00E2\u0080\u0099s CONJUNCT order, I show that the precedence and c-command conditions of the Generalized Pattern of Anaphoric Dependency govern the distribution of the anaphoric clauses themselves as well as the possible antecedents for the anaphoric links within those anaphoric clauses. 4.2.3 Conditions on antecedent-licensing: C-command and precedence As with pronominal anaphora, I show that only one of the two conditions (c-command or precedence) need hold in order for antecedent licensing of anaphoric clauses to take place. Before turning to the Plains Cree patterns that are attested, let us look in more detail at what patterns each condition predicts. The first condition is the c-command condition, given in (11). (11) C-command condition on anaphoric elements: An anaphoric element is licensed by an antecedent if that antecedent is in a clause c-commanding the clause containing the anaphoric element. Under the c-command condition, there are only two structures in which an anaphoric element may be licensed: if the anaphor is in a subordinate clause relative to and follows the antecedent, or if the anaphor is in a subordinate clause relative to and precedes the antecedent. These are shown in Table 4.1. C-command condition met? \u00E2\u009C\u0094 Structures CP 5 anteced. CP 5 anaphor CP 5 CP anteced. 5 anaphor Table 4.1. Structures which meet the c-command condition 140 According to the c-command condition, all of the following structures are undefined for anaphora, including structures where the antecedent is in a subordinate clause relative to the anaphor, and structures where there is no c-command relation between the clauses in which the anaphor and antecedent occur. All of the logical possibilities are given in Table 4.2. C-command condition met? \u00E2\u009C\u0096 (antecedent in a subordinate clause relative to anaphor) \u00E2\u009C\u0096 (no c-command relation between clauses) Structures CP 5 CP anaphor 5 anteced. CP 5 anaphor CP 5 anteced. CP CP 5 5 anaphor anteced. CP CP 5 5 anteced. anaphor Table 4.2. Structures which do not meet the c-command condition The other condition on antecedent licensing of anaphoric elements is the precedence condition, given in (12). (12) Precedence condition on anaphoric elements: An anaphoric element is licensed by an antecedent if the antecedent precedes it. If we restrict our attention to just the precedence condition, we see that there three syntactic relations between clauses that will respect precedence: the clause which contains the anaphor may be subordinate to, superordinate to, or non-subordinate to the clause which contains the antecedent. Precedence condition met? \u00E2\u009C\u0094 Structures CP 5 anteced. CP 5 anaphor CP 5 CP anaphor 5 anteced. CP CP 5 5 anteced. anaphor Table 4.3. Structures which meet the precedence condition The precedence condition will also leave some configurations undefined for anaphoric elements. If the potential antecedent follows the anaphoric element, according to the precedence condition the antecedent cannot be licensed in any syntactic configuration. The structures in table 4.4 141 below are exactly the same as those in table 4.3 above, but the precedence relation has been reversed, and the configuration is undefined. Precedence condition met? \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Structure CP CP 5 5 anaphor anteced. CP 5 anaphor CP 5 anteced. CP 5 CP anteced. 5 anaphor Table 4.4. Structures which do not meet the precedence condition Remember, however, that only one of the two conditions is necessary for the licensing relation between the anaphor and antecedent to be established. Thus, there are two configurations which are undefined once both conditions have been applied, as in table 4.5. C-command condition met? \u00E2\u009C\u0096 Precedence condition met? \u00E2\u009C\u0096 Structures CP 5 anaphor CP 5 anteced. CP CP 5 5 anaphor anteced. Table 4.5. Structures which are undefined for antecedent licensing of anaphoric elements The ill-formed dependency between it and term paper given in (10d), here repeated as (13), is thus ruled out in that the anaphor element it is not preceded by its potential antecedent term paper, nor is it in a subordinate clause with respect to the clause which contains the potential antecedent. (13) a. * Anyone can turn iti in to me now [who has written their TERM PAPERi]. b. * CP 5 iti CP 5 [term paper]i 142 Likewise, two coordinated (14a) or two independent, adjacent (14b) clauses do not permit a backwards dependency to be established. (14) a. * Hei won the race and we welcomed home JOHNi. Williams (1997: ex. 23b) b. * HeI knocked. I opened the door and greeted JOHNI. Here the anaphor he cannot be dependent on John, since John neither precedes nor is in a c- commanding clause relative to he. (15) CP CP 5 5 hei Johni Notice that the potential antecedent \u00E2\u0080\u0098term paper\u00E2\u0080\u0099 in (13) is capitalized to reflect its status as new information, which in turn means that it is not itself anaphoric on some preceding antecedent in the discourse. The distinction is important because the string in (13) is not in and of itself ungrammatical: it is the particular antecedent-anaphor relationship which causes the problem. Within a larger discourse, where there is a previous mention of \u00E2\u0080\u0098term paper\u00E2\u0080\u0099 available to serve as the antecedent for \u00E2\u0080\u0098it\u00E2\u0080\u0099 the string becomes grammatical (notice crucially that the final instance of \u00E2\u0080\u0098term paper\u00E2\u0080\u0099 cannot be stressed). (16) [I assume you recall that this course requires a term paperi.] Anyone can turn iti in to me now [who has WRITTEN their term paperi]. (adapted from Williams 1997: ex. 27) In this example, the second instance of \u00E2\u0080\u0098term paper\u00E2\u0080\u0099 is itself anaphoric on the first instance, and the formal anaphor it is likewise anaphoric on the preceding instance. The anaphoric dependency can be determined in English by the placement of stress: the fact that the second instance of \u00E2\u0080\u0098term paper\u00E2\u0080\u0099 is destressed indicates that it is not new information (and thus not the antecedent). The relevant dependencies could be represented as a kind of \u00E2\u0080\u0098many-to-one\u00E2\u0080\u0099 linking dependency a la Higginbotham (1983): both \u00E2\u0080\u0098it\u00E2\u0080\u0099 and the final instance of \u00E2\u0080\u0098term paper\u00E2\u0080\u0099 are anaphorically dependent on the first instance of \u00E2\u0080\u0098term paper\u00E2\u0080\u0099, which is the antecedent. 143 (17) [\u00E2\u0080\u00A6 term paper \u00E2\u0080\u00A6 ] [\u00E2\u0080\u00A6 it \u00E2\u0080\u00A6] [ \u00E2\u0080\u00A6 term paper \u00E2\u0080\u00A6 ] ANTECEDENT ANAPHOR ANAPHOR Crucially, the relation between the antecedent/anaphoric elements in (17) does not have to be a chain, where the antecedent / anaphor relation must be calculated with respect to each local pair of elements. The chain analysis would require that the first instance of term paper be that antecedent for it, and that it would itself serve as an antecedent to the second instance of term paper, as in (18). (18) [\u00E2\u0080\u00A6 term paper \u00E2\u0080\u00A6 ] [\u00E2\u0080\u00A6 it \u00E2\u0080\u00A6] [ \u00E2\u0080\u00A6 term paper \u00E2\u0080\u00A6 ] ANTECEDENT ANAPHOR ANAPHOR Just as English anaphora require (something like) a linking analysis to capture the kinds of dependencies that they establish, we will see that the anaphoric links in anaphoric clauses are best represented by a linking-type analysis rather than a chaining-type analysis. 4.2.4 Summary: The properties of anaphoric clauses Indexical clauses are evaluated with respect to the speech situation. Anaphoric clauses, on the other hand, are evaluated with respect to an unspecified situation; I claim that the properties of this situation are determined by an antecedent, just as the reference of an anaphoric pronominal are determined by an antecedent. I show that the distribution of anaphoric clauses, which are morpho-syntactically distinguished from their indexical counterparts in Plains Cree, is accounted for by the c-command and precedence conditions in \u00C2\u00A74.3. I further claim that variables within an anaphoric clause are subject to the antecedent- licensing conditions of anaphora: the antecedent must be in a superordinate or preceding clause. In \u00C2\u00A74.4 I show that this correctly accounts for the range of interpretations within anaphoric clauses. Dependent elements are anaphoric in anaphoric clauses, but deictic in indexical clauses. 144 4.3 The distribution of anaphoric clauses In this section I show how the distribution of anaphoric clauses can be accounted for by the general licensing mechanisms proposed for pronominal anaphora. 4.3.1 Anaphoric clauses are subject to precedence and/or c-command The discussion of antecedent licensing of anaphoric clauses is broken into constructions for which c-command holds; constructions for which precedence holds, and constructions which satisfy neither c-command nor precedence. 4.3.1.1 Anaphoric clauses that must be c-commanded are not subject to precedence For three of the four subtypes of anaphoric CONJUNCT clauses, the morpho-syntactic marking on the clause corresponds with obligatory embedding. This includes clauses introduced by the complementizer k\u00C3\u00A2-, the subjunctive CONJUNCT (with a null complementizer and plural suffix -i), and the simple CONJUNCT (with a null complementizer and modal prefix ka-). The simplest way to demonstrate the embeddedness of these clauses is to test their ability to be uttered on their own (i.e., as a complete proposition). As shown in (4-6), none of these clauses pass this test. They thus have a very local dependency \u00E2\u0080\u0093 they must be part of a larger constituent in order to be grammatical. This is exemplified for k\u00C3\u00A2-clauses in (19), subjunctive clauses in (20), and simple CONJUNCT clauses in (21). (19) a. * k\u00C3\u00A2-k\u00C3\u00AEsit\u00C3\u00AApoy\u00C3\u00A2n K\u00C3\u0082-CONJUNCT k\u00C3\u00A2- k\u00C3\u00AEsit\u00C3\u00AApo -t C2-cook.VAI-3 -- b. kisit\u00C3\u00AAw k\u00C3\u00A2-k\u00C3\u00AEsit\u00C3\u00AApoy\u00C3\u00A2n kisit\u00C3\u00AA -w k\u00C3\u00A2- k\u00C3\u00AEsit\u00C3\u00AApo -y\u00C3\u00A2n be.hot.VII-3 C2- cook.VAI -1 \u00E2\u0080\u0098It\u00E2\u0080\u0099s hot when I cook.\u00E2\u0080\u0099 145 (20) a. * Jeff nikamoci SUBJUNCTIVE J nikamo -t -i J sing.VAI-3 -SUBJ -- b. Jeff nikamoci w\u00C3\u00A2pakaniy ka-w\u00C3\u00A2pahtam J nikamo -t -i w\u00C3\u00A2pakaniy ka- w\u00C3\u00A2pahtam-w J sing.VAI-3-SUBJ flower IRR-see.VTI -3 \u00E2\u0080\u0098Should Jeff sing, he will see a flower.\u00E2\u0080\u0099 (21) a. * ka-kawsimoyan SIMPLE ka- kawsimo -yan IRR-go.to.bed.VAI-2 --- b. piko ka-kawsimoyan \u00C3\u00AAkwa piko ka- kawsimo -yan \u00C3\u00AAkwa it.is.necessary IRR-go.to.bed.VAI -2 now \u00E2\u0080\u0098You have to go to bed now!\u00E2\u0080\u0099 (father to child) Since these clauses are obligatorily embedded, they fulfill the c-command condition on anaphora3. We expect that they will be insensitive to the precedence condition: they should be able to either precede or follow their antecedent, as the structures in (22) represent. (22) a. CP 5 antecedent CP 5 anaphor b. CP 5 CP antecedent 5 anaphor This prediction is borne out for these CONJUNCT clauses. For example, we see that k\u00C3\u00A2-clauses can either precede or follow the clause they modify (i.e., are dependent on). The relevant structures are repeated below each example. 3 It is also theoretically possible that an embedded clause could be anteceded by some clause other than the immediately superordinate clause. The point is that there must always be an immediately superordinate clause. 146 (23) k\u00C3\u00A2-CONJUNCT clause as anaphor: following a. \u00E2\u0080\u00A6, \u00C3\u00AAkos \u00C3\u00A2niki m\u00C3\u00A2n \u00C3\u00AA-t\u00C3\u00B4tahkik, tourists k\u00C3\u00A2-takoht\u00C3\u00AAyit,\u00E2\u0080\u009D ... \u00C3\u00AAkosi aniki m\u00C3\u00A2na \u00C3\u00AA-t\u00C3\u00B4tam -k-k tourists k\u00C3\u00A2-takoht\u00C3\u00AA -yi -t TOPIC DEM.AN usually C1-do.VTI-0-PL tourists C2-arrive.VAI-DS -3 \u00E2\u0080\u0098\u00E2\u0080\u00A6, that is what they do when tourists arrive,\u00E2\u0080\u009D ... \u00E2\u0080\u0099 (AA 3.2) b. CP 5 \u00C3\u00AA-t\u00C3\u00B4tahkik CP 5 k\u00C3\u00A2-takoht\u00C3\u00AAyit (24) k\u00C3\u00A2-CONJUNCT clause as anaphor: preceding a. k\u00C3\u00A2-minahot, kahkiyaw awiya \u00C3\u00AA-asam\u00C3\u00A2t; k\u00C3\u00A2-minaho -t kahkiyaw awiya \u00C3\u00AA-asam-\u00C3\u00A2-t C2-kill.animal.VAI-3 all someone C1-feed.VTA-DIR-3 \u00E2\u0080\u0098when he killed an animal, he fed everyone;\u00E2\u0080\u0099 (AA 1.7) b. CP 5 CP \u00C3\u00AA-asam\u00C3\u00A2t 5 k\u00C3\u00A2-minahot Similarly, subjunctive clauses \u00E2\u0080\u0093 identified by the lack of a proclitic complementizer and the suffix -i \u00E2\u0080\u0093 may either precede or follow the clause they are dependent on. (25) SUBJUNCTIVE clause as anaphor: following a. \u00E2\u0080\u00A6 , \u00E2\u0080\u0098\u00C3\u00AAkos \u00C3\u00B4ma t-\u00C3\u00AAsin\u00C3\u00A2kwan \u00C3\u00B4ma, maskihkiy os\u00C3\u00AEht\u00C3\u00A2yani,\u00E2\u0080\u0099 nititik,\u00E2\u0080\u009D \u00E2\u0080\u00A6 \u00C3\u00AAkosi \u00C3\u00B4ma ta-isin\u00C3\u00A2kwan \u00C3\u00B4ma maskihkiy os\u00C3\u00AEht\u00C3\u00A2 -yan-i ni(t)-it -ik -w TOPIC DEM IRR-thus.look.VII DEM medicine make.VAI-2 -SUBJ 1- say.VTA-INV -3 \u00E2\u0080\u0098 \u00E2\u0080\u009C\u00E2\u0080\u00A6, \u00E2\u0080\u0098It will look like that when you make the medicine,\u00E2\u0080\u0099 he said to me,\u00E2\u0080\u009D \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 4.3) b. CP 5 t-\u00C3\u00AAsin\u00C3\u00A2kwan CP 5 os\u00C3\u00AEht\u00C3\u00A2yani 147 (26) SUBJUNCTIVE clause as anaphor: preceding a. ..., \u00E2\u0080\u009C \u00E2\u0080\u0098miskahkw\u00C3\u00A2wi, nika-misihon, maskihk\u00C3\u00AEwiyiniwak,\u00E2\u0080\u0099 ...\u00E2\u0080\u009D miskam -k-w\u00C3\u00A2w-i ni-ka-misiho -n maskihk\u00C3\u00AEwiyiniw-ak find.VTI-0-3PL -SUBJ 1- IRR-trouble.VAI-SAP doctor -PL \u00E2\u0080\u0098..., \u00E2\u0080\u009C \u00E2\u0080\u0098If the doctors find it, I\u00E2\u0080\u0099ll be in trouble,\u00E2\u0080\u0099 ...\u00E2\u0080\u009D \u00E2\u0080\u0099 (AA 4.8) b. CP 5 CP nika-misihon 5 miskahkw\u00C3\u00A2wi Finally, simple CONJUNCT clauses are also able to precede or follow their antecedent, although the case of precedence is very rare and seems to be highly restricted. (27) simple CONJUNCT clause as anaphor: following a. \u00E2\u0080\u00A6, awa \u00C3\u00AA-p\u00C3\u00AA-kakw\u00C3\u00AAcimak ka-\u00C3\u00A2h-\u00C3\u00A2cimostawit, \u00E2\u0080\u00A6 awa \u00C3\u00AA- p\u00C3\u00AA- kakw\u00C3\u00AAcim-ak ka- \u00C3\u00A2h- \u00C3\u00A2cimostaw-it DEM.AN C1- COME-ask.VTA -1>3 IRR-RED-tell.VTA -3>1 \u00E2\u0080\u0098\u00E2\u0080\u00A6 I have come to ask [her] \u00E2\u0080\u00A6to tell me stories.\u00E2\u0080\u0099 (FA in AA 1) b. CP 6 \u00C3\u00AA-p\u00C3\u00AA-kakw\u00C3\u00AAcimak CP 5 ka-\u00C3\u00A2h-\u00C3\u00A2cimostawit 148 (28) simple CONJUNCT clause as anaphor: preceding a. ..., \u00C3\u00AA-titipikwanahahkik \u00C3\u00AAkoni anih \u00C3\u00A2ya, oc\u00C3\u00AEhkw\u00C3\u00AAhikana, \u00C3\u00AA- titipikwanaham-k-k \u00C3\u00AAkoni anihi aya oc\u00C3\u00AEkw\u00C3\u00AAhikan-a C1-sew.VTI -0-PL TOPIC DEM.INAN CONN moccasin -PL \u00E2\u0080\u0098..., sewing it around the vamp of the gathered moccasins\u00E2\u0080\u00A6 ka-miyon\u00C3\u00A2kwaniyiki \u00C3\u00AA-k\u00C3\u00AE-is\u00C3\u00AEht\u00C3\u00A2cik m\u00C3\u00A2na. ka- miyon\u00C3\u00A2kwan-yi -k-i \u00C3\u00AA- k\u00C3\u00AE- is\u00C3\u00AEht\u00C3\u00A2 -t -k m\u00C3\u00A2na IRR-good.look.VII -DS-0-PL C1-PREV-make.VAI-3-PL usually so as to make them look nice.\u00E2\u0080\u0099 (EM 20) b. CP 5 CP \u00C3\u00AA-k\u00C3\u00AE-is\u00C3\u00AEht\u00C3\u00A2cik m\u00C3\u00A2na 5 ka-miyon\u00C3\u00A2kwaniyiki This means that for all clause-types whose embeddedness can be identified from the morpho- syntactic marking of the clause (i.e., CONJUNCT agreement with k\u00C3\u00A2-, ka-, or the suffix -i), there is a c-command condition. Corresponding with the satisfaction of c-command, there are no absolute linear restrictions between the embedded clause and the superordinate clause. This data is enough to demonstrate that these anaphoric clauses are licensed where the antecedent-anaphor relation satisfies c-command; I will return to a more detailed analysis of the syntax and semantics of these clauses in chapters 5 and 6. 4.3.1.2 Anaphoric clauses with subordinating particles are not subject to precedence There are also many cases where a clause\u00E2\u0080\u0099s embeddedness in Plains Cree is determined by a particle that introduces a particular kind of subordinate clause, such as the degree-marker iyikohk, concessives like kiy\u00C3\u00A2m \u00E2\u0080\u0098although\u00E2\u0080\u0099 and \u00C3\u00A2ta \u00E2\u0080\u0098even\u00E2\u0080\u0099, and non-interrogative locatives like ita \u00E2\u0080\u0098where\u00E2\u0080\u0099. I take these particles to be a complementizer in C, since they interact with the clause- typing (i.e., choice of k\u00C3\u00A2- vs. \u00C3\u00AA-) and the presence of the particle affects the distributional properties of the clause. 149 Since these clauses are subordinate, we again expect that the precedence condition need not hold. This prediction also holds. For example, degree clauses introduced by the particle iyikohk \u00E2\u0080\u0098so\u00E2\u0080\u0099 are also insensitive to precedence, as demonstrated in (29) and (30). (29) iyikohk-clause as anaphor: following a. ..., \u00C3\u00A2skaw m\u00C3\u00A2na nik\u00C3\u00AE-n\u00C3\u00AAp\u00C3\u00AAwihik, \u00C3\u00A2skaw m\u00C3\u00A2na ni-k\u00C3\u00AE- n\u00C3\u00AAp\u00C3\u00AAwih -ik -w sometimes usually 1- PREV-shame.VTA-INV-3 \u00E2\u0080\u0098..., my husband used to put me to shame at times iyikohk \u00C3\u00AA-k\u00C3\u00AE-miyohtw\u00C3\u00A2t niw\u00C3\u00AEkim\u00C3\u00A2kan, ... iyikohk \u00C3\u00AA-k\u00C3\u00AE-miyohtw\u00C3\u00A2-t ni-w\u00C3\u00AEkim\u00C3\u00A2kan DEGREE C1-PREV-good.natured.VAI-3 1-spouse because he was so good-natured...\u00E2\u0080\u0099 (EM 29) b. CP 6 nik\u00C3\u00AE-n\u00C3\u00AAp\u00C3\u00AAwihik CP 5 iyikohk \u00C3\u00AA-k\u00C3\u00AE-miyohtw\u00C3\u00A2t (30) iyikohk-clause as anaphor: preceding a. \u00E2\u0080\u00A6 iyikohk \u00C3\u00AA-k\u00C3\u00AE-miyokiht\u00C3\u00A2y\u00C3\u00A2hk askipw\u00C3\u00A2wa, iyikohk \u00C3\u00AA- k\u00C3\u00AE- miyokiht\u00C3\u00A2-y\u00C3\u00A2n-k askipw\u00C3\u00A2w-a DEG C1-PREV-grow.VAI -1 -PL potato -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6, when we grew such a good crop of potatoes, \u00C3\u00AAkosi m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-isi-tipaham\u00C3\u00A2hk, \u00E2\u0080\u00A6 \u00C3\u00AAkosi m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- isi-tipaham -\u00C3\u00A2n-k TOPIC usually C1-PREV-SO-measure.VTI-1 -PL that is how we measured them, ...\u00E2\u0080\u0099 (EM 54) (lit: \u00E2\u0080\u0098we grew such a good crop of potatoes that we measured them that way\u00E2\u0080\u0099) b. CP 5 CP \u00C3\u00AA-k\u00C3\u00AE-isi-tipaham\u00C3\u00A2hk 5 iyikohk \u00C3\u00AA-k\u00C3\u00AE-miyokiht\u00C3\u00A2y\u00C3\u00A2hk 150 Similarly, ita \u00E2\u0080\u0098where\u00E2\u0080\u0099 clauses can occur in both linear relations to the superordinate clause4. In (31), the ita \u00E2\u0080\u0098where\u00E2\u0080\u0099 clause follows the superordinate clause. (31) ita-clause as anaphor: following a. nitakopayin\u00C3\u00A2n \u00C3\u00AAkwa anita \u00C3\u00AAkwa \u00C3\u00AA-wi-~, ita \u00C3\u00AA-w\u00C3\u00AEkicik. ni-takopayi -n\u00C3\u00A2n \u00C3\u00AAkwa anita \u00C3\u00AAkwa ita \u00C3\u00AA- w\u00C3\u00AEki -t -k 1- drive.up.VAI-1PL then there then where C1-live.VAI-3-PL \u00E2\u0080\u0098Then we drove up there where they lived. (AA 2.2) b. CP 6 nitakopayin\u00C3\u00A2n CP 5 ita \u00C3\u00AA-w\u00C3\u00AEkicik In (32), the ita \u00E2\u0080\u0098where\u00E2\u0080\u0099 clause ita \u00C3\u00AA-n\u00C3\u00AEpawiy\u00C3\u00A2n \u00E2\u0080\u0098where I stood\u00E2\u0080\u0099 precedes the superordinate clause it is modifying. (Notice that the corresponding clause in the English follows the superordinate clause.) (32) ita-clause as anaphor: preceding a. \u00E2\u0080\u00A6, ita \u00C3\u00AA-n\u00C3\u00AEpawiy\u00C3\u00A2n, ot\u00C3\u00A2hk \u00C3\u00B4t\u00C3\u00AA \u00C3\u00AA-p\u00C3\u00AA-n\u00C3\u00AEpawit, \u00E2\u0080\u00A6 ita \u00C3\u00AA- n\u00C3\u00AEpawi -y\u00C3\u00A2n ot\u00C3\u00A2hk \u00C3\u00B4t\u00C3\u00AA \u00C3\u00AA- p\u00C3\u00AA- n\u00C3\u00AEpawi -t where C1-stand.VAI-1 behind there C1-COME-stand.VAI-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6 [he] came and stood over there behind me, just where I stood, \u00E2\u0080\u00A6 (AA 12.10) b. CP 5 CP \u00C3\u00AA-p\u00C3\u00AA-n\u00C3\u00AEpawit 5 ita \u00C3\u00AA-n\u00C3\u00AEpawiy\u00C3\u00A2n Thus in contexts where a CONJUNCT clause is unambiguously embedded with respect to some higher clause, they work like anaphors do: they may either precede or follow their antecedent. I turn now to a more complicated case: the distribution of \u00C3\u00AA-clauses with no overt subordinating particles. 4 Many of these clauses do have a more common position relative to the higher clause (some more frequently precede the matrix clause, others more frequently follow it). Dahlstrom (2006) discusses some of these patterns for Fox; at the moment it is not clear what determines which order occurs (but see M\u00C3\u00BChlbauer 2003, 2008; Wolvengrey 2007, and D\u00C3\u00A9chaine 2007 for discussion of principles of ordering in Plains Cree). 151 4.3.1.3 Anaphoric clauses that are subject to precedence: Unembedded \u00C3\u00AA-clauses In addition to clauses which are unambiguously embedded (as shown by their inability to stand on their own), we expect that some anaphoric clauses will be licensed under the precedence condition, and thus not need to be embedded. In Plains Cree, we see this behaviour exhibited in CONJUNCT clauses with the \u00C3\u00AA- complementizer (henceforth, \u00C3\u00AA-clauses). Anaphoric \u00C3\u00AA-clauses are often found in relatively long chains introduced by an initial indexical clause. For example, in (33), the first clause is indexical, and is marked with the temporal-shifting element k\u00C3\u00AE- (as is obligatory in this context); the following anaphoric clauses, while referring to successive events within the same episode, do not have any temporal marking at all. (33) [ Indexical ] [anaphoric] [anaphoric] . . . (i) \u00C3\u00AAkwa n\u00C3\u00AAwos\u00C3\u00A2p-~ nistos\u00C3\u00A2p-k\u00C3\u00AEsik\u00C3\u00A2w nik\u00C3\u00AE-pap\u00C3\u00A2m\u00C3\u00A2cihon\u00C3\u00A2n \u00C3\u00AAkot\u00C3\u00AA, INDEXICAL \u00C3\u00AAkwa n\u00C3\u00AAwos\u00C3\u00A2p-~ nistos\u00C3\u00A2p-k\u00C3\u00AEsik\u00C3\u00A2 -w ni-k\u00C3\u00AE- pap\u00C3\u00A2m\u00C3\u00A2ciho-n\u00C3\u00A2n \u00C3\u00AAkot\u00C3\u00AA and fourteen-~ thirteen- day.VII-3 1- PREV-go.about.VAI -1PL there \u00E2\u0080\u0098Then we toured about over there for fourteen-~ for thirteen days, (ii) a tour \u00C3\u00AA-otinam\u00C3\u00A2hk oti, ANAPHORIC a tour \u00C3\u00AA- otinam -\u00C3\u00A2n-k oti a tour C1-take.VAI-1 -PL especially we took a tour, (iii) bus \u00C3\u00AA-p\u00C3\u00B4siy\u00C3\u00A2hk, aya, ANAPHORIC bus \u00C3\u00AA- p\u00C3\u00B4si -y\u00C3\u00A2n-k aya bus C1-travel.VAI-1 -PL CONN we travelled on a bus, (iv) thirteen-day tour \u00C3\u00AA-otinam\u00C3\u00A2hk, ANAPHORIC thirteen-day tour \u00C3\u00AA- otinam -\u00C3\u00A2n-k thirteen-day tour C1-take.VTI-1 -PL we took a thirteen-day tour (v) thirty-six \u00C3\u00AA-ihtasiy\u00C3\u00A2hk, ANAPHORIC thirty-six \u00C3\u00AA- ihtasi -y\u00C3\u00A2n-k thirty-six C1-be.VAI-1 -PL with thirty-six of us (vi) bus an[a] \u00C3\u00AA-p\u00C3\u00B4siy\u00C3\u00A2hk. ANAPHORIC bus ana \u00C3\u00AA- p\u00C3\u00B4si -y\u00C3\u00A2n-k bus DEM.AN C1-travel.VAI-1 -PL travelling on the bus.\u00E2\u0080\u0099 (AA 3.2) 152 Notice that line (ii) and line (iv) are repetitions of the same clause with the same verbal predicate inflected for the same participants (\u00C3\u00AA-otinam\u00C3\u00A2hk \u00E2\u0080\u0098we took it\u00E2\u0080\u0099), and lines (iii) and (vi) are also alternates of each other with the same verbal complex (\u00C3\u00AA-p\u00C3\u00B4siy\u00C3\u00A2hk \u00E2\u0080\u0098we travelled\u00E2\u0080\u0099). Discoursally, these clauses seem to be restrictions of the main clause. Thus, we can analyze this sequence of clauses as in (34) (here I represent each CP only with the verbal complexes for simplicity\u00E2\u0080\u0099s sake): the clauses in lines (ii-iii) restate and restrict line (i); those in lines (iv) and (vi) repeat (ii- iii), with (v) further restricting (vi). The arrows in the tree flow from anaphor to antecedent. (34) CP1 / i 6 CPii CPiii CPiv CPvi nik\u00C3\u00AE-pap\u00C3\u00A2m\u00C3\u00A2cihon\u00C3\u00A2n 5 5 5 5 \u00C3\u00AA-otinam\u00C3\u00A2hk \u00C3\u00AA-p\u00C3\u00B4siy\u00C3\u00A2hk \u00C3\u00AA-otinam\u00C3\u00A2hk CPv \u00C3\u00AA-p\u00C3\u00B4siy\u00C3\u00A2hk 5 \u00C3\u00AA-ihtasiy\u00C3\u00A2hk The only clause in this sequence that might be said to be embedded in this example is \u00C3\u00AA- ihtasiy\u00C3\u00A2hk.5 None of the other clauses are embedded within any of the other clauses, and in particular none of the anaphoric clauses are embedded relative to the initial indexical clause. However, despite the lack of embedding, there is still a dependence between the anaphoric clauses and the indexical clause, most clearly seen in the temporal interpretation they receive. Specifically, the anaphoric clauses carry the same temporal value as the indexical clause, even though they are not marked in any way other than being typed as anaphoric by the \u00C3\u00AA- complementizer in the CONJUNCT order. The licensing conditions on anaphora predict that, since the c-command condition does not hold, the precedence condition must. If the anaphor is in a non-initial position, precedence is satisfied (35a), but if the anaphor is in an initial position, precedence is not satisfied (35b). 5 In this example it is also entirely possible that the \u00C3\u00AA-ihtasiy\u00C3\u00A2hk clause is not embedded either, since it is formally identical to the other clauses; instead it could be second in a three-part chain with lines (iv) and (vi). I chose the embedding analysis on the basis of the discourse structure. 153 (35) a. CP CP ANAPHORIC 5 5 antecedent anaphor b. * CP CP UNDEFINED 5 5 anaphor antecedent This turns out to capture the distribution of the unembedded \u00C3\u00AA-conjunct clauses. For example, in (36), we have an initial clause marked with the preverb k\u00C3\u00AE- and the temporal particle m\u00C3\u00A2na \u00E2\u0080\u0098usually\u00E2\u0080\u0099, followed by a second clause which has no marking. There is thus an analogous temporal dependency between these clauses as we saw above. The overt nominal aw\u00C3\u00A2sisak \u00E2\u0080\u0098children\u00E2\u0080\u0099 is also in the initial clause, and is interpreted as the subject of both clauses. Crucially, the anaphoric clause, which lacks both the temporal specification and the overt nominal, cannot be placed before the antecedent clause (36b). (36) a. [antecedent] [anaphor] [ \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-itoht\u00C3\u00AAcik m\u00C3\u00A2na aw\u00C3\u00A2sisak ] [ \u00C3\u00AA-nikamocik ] \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00AA- itoht\u00C3\u00AA -t -k m\u00C3\u00A2na aw\u00C3\u00A2sis-ak \u00C3\u00AA- nikamo -t -k C1-PREV-COME-go.VAI-3-PL usually child -PL C1-sing.VAI-3-PL \u00E2\u0080\u0098The children used to come and they used to sing.\u00E2\u0080\u0099 comment: this could mean either they were singing while they came, or that they sang when they got there b. [anaphor] [antecedent] ! [ \u00C3\u00AA-nikamocik ] [ \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-itoht\u00C3\u00AAcik m\u00C3\u00A2na aw\u00C3\u00A2sisak ] \u00C3\u00AA- nikamo -t -k \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00AA- itoht\u00C3\u00AA -t -k m\u00C3\u00A2na aw\u00C3\u00A2sis-ak C1-sing.VAI-3-PL C1-PREV-COME-go.VAI-3-PL usually child -PL --- (intended: \u00E2\u0080\u0098The children used to sing and come.\u00E2\u0080\u0099) Some might object that the absence of the overt nominal in the initial clause of (36b) accounts for the utterance\u00E2\u0080\u0099s ungrammaticality, but if the initial clause is subordinated to the second clause, then precedence should not be a problem. The following example confirms this: when the \u00C3\u00AA- in the initial clause is replaced with the unambiguously subordinating complementizer k\u00C3\u00A2-, the utterance is grammatical, as in (37). 154 (37) [anaphor] [antecedent] k\u00C3\u00A2-nikamocik, \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-itoht\u00C3\u00AAcik m\u00C3\u00A2na aw\u00C3\u00A2sisak k\u00C3\u00A2-nikamo -t-k \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00AA- itoht\u00C3\u00AA -t -k m\u00C3\u00A2na aw\u00C3\u00A2sis-ak C2-sing.VAI-3-PL C1-PREV-COME-go.VAI-3-PL usually child -PL \u00E2\u0080\u0098Only when the children sang did they come.\u00E2\u0080\u0099 Within Plains Cree, the sensitivity to precedence is striking; as we have seen, in general, subordinate clauses are not sensitive to it. For example, a subordinate clause introduced by the complementizer k\u00C3\u00A2- may either precede or follow the superordinate clause, as in (38). (38) a. Jane k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2c-atosk\u00C3\u00AAt \u00C3\u00A2hkosiwipayiw. J k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-\u00C3\u00A2tosk\u00C3\u00AA -t \u00C3\u00A2hkosiw-ipayi -w J C2-NOW- work.VAI-3 sick.VAI- INCH.VAI-3 \u00E2\u0080\u0098When Jane was working, she got ill / sick\u00E2\u0080\u0099 b. Jane \u00C3\u00A2hkosiwipayiw k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2c-atosk\u00C3\u00AAt J \u00C3\u00A2hkosiw-ipayi -w k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-\u00C3\u00A2tosk\u00C3\u00AA -t J sick.VAI-INCH.VAI-3 C2-NOW- work.VAI-3 \u00E2\u0080\u0098When Jane was working, she got ill / sick\u00E2\u0080\u0099 Cross-linguistically, a sensitivity to precedence is striking because it is reminiscent of the patterns seen for \u00E2\u0080\u0098clause-chains\u00E2\u0080\u0099. Most of the languages claimed to have clause-chaining have the same strict observance of precedence relations: a dependent clause that is part of a chain may only be in one position relative to the non-dependent clause6. If CONJUNCT clauses are anaphoric clauses, this is exactly the pattern predicted: where anaphoric clauses are not embedded with respect to their antecedent, the antecedent must precede them in order for the dependency to be established. 4.3.2 Long-distance precedence of antecedent for anaphoric clauses We have seen that \u00C3\u00AA-clauses differ from other CONJUNCT clauses in that they need not be c- commanded with respect to their antecedent; i.e., \u00C3\u00AA-clauses are unembedded. The unembedding property of \u00C3\u00AA-clauses is in fact quite general: they are also grammatical as stand-alone sentences. Thus, in elicitation translation tasks, consultants will often offer anaphoric \u00C3\u00AA-clauses as stand- 6 See \u00C2\u00A74.5.1.2 for discussion on this point. 155 alone clauses, and claim that they are interchangeable with their indexical INDEPENDENT counterparts, as in (39) below. (39) elicitation task: translation of \u00E2\u0080\u0098Sol is crying.\u00E2\u0080\u0099 a. Sol m\u00C3\u00A2tow INDEPENDENT S m\u00C3\u00A2to -w S cry.VAI-3 \u00E2\u0080\u0098Sol is crying / cries.\u00E2\u0080\u0099 b. Sol \u00C3\u00AA-m\u00C3\u00A2tot CONJUNCT S \u00C3\u00AA- m\u00C3\u00A2to -t S C1-cry.VAI-3 \u00E2\u0080\u0098\u00E2\u0080\u00A6Sol is crying / cries.\u00E2\u0080\u0099 comment: \u00E2\u0080\u009CThey mean the same thing.\u00E2\u0080\u009D However, when we look at the contexts in which these utterances occur, stand-alone indexical clauses and stand-alone anaphoric clauses become distinct. In particular, an indexical clause (INDEPENDENT) is felicitous in an out-of-the-blue context, while an anaphoric clause (CONJUNCT) requires an established context \u00E2\u0080\u0093 which both the speaker and hearer share \u00E2\u0080\u0093 for its felicity7. Thus, when speakers are asked to identify how they would use an INDEPENDENT clause as opposed to an unembedded CONJUNCT clause, their answers indicate that the two clause-types are quite different. In this section I consider the contexts that license anaphoric clauses, with reference to the discussion at the beginning of the chapter, where I posited that anaphora are found in established contexts. The analysis of CONJUNCT clauses as anaphoric allows us to understand their behaviour when they occur as stand-alone utterances. This expectation provides a way to relate some previously undiscussed and puzzling data to familiar principles. In the current section I consider the distribution of anaphoric CONJUNCT clauses in out-of-the-blue contexts vs. established contexts (\u00C2\u00A74.3.2.1), in focus contexts (\u00C2\u00A74.3.2.2), and in different speaking modes (narrative, conversation, and elicitation) (\u00C2\u00A74.3.2.3). 7 The nature of a translation task is such that the relevant context-dependency does not emerge in this example; see Cook & M\u00C3\u00BChlbauer (2006) for further discussion. 156 4.3.2.1 Out-of-the-blue contexts vs. established contexts When asked to think about the contexts in which an anaphoric CONJUNCT clause is used, as opposed to an indexical INDEPENDENT clause, speakers often contrast an out-of-the-blue context (for INDEPENDENT clauses) with an established context (for CONJUNCT clauses). This is what happens in (40-41): the indexical INDEPENDENT clause is felicitous where there has been no cue that the hearer is hungry, and there is no reason on the part of the speaker to suppose that the speaker is hungry. By contrast, the anaphoric \u00C3\u00AA-conjunct clause is used in a context where the actions of the addressee create a presupposition for asking the question. (40) context: someone is coming to visit; just arrived a. kin\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00A2n c\u00C3\u00AE INDEPENDENT ki- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00A2-n c\u00C3\u00AE 2- hungry.VAI -SAP Q \u00E2\u0080\u0098Are you hungry?\u00E2\u0080\u0099 b. # \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAyan c\u00C3\u00AE CONJUNCT \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -yan c\u00C3\u00AE C1-hungry.VAI -2 Q \u00E2\u0080\u0098\u00E2\u0080\u00A6are you hungry?\u00E2\u0080\u0099 (cf. \u00E2\u0080\u0098Is it that you\u00E2\u0080\u0099re hungry?\u00E2\u0080\u0099) (41) context: see someone rummaging in the refrigerator a. # kin\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00A2n c\u00C3\u00AE INDEPENDENT ki- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00A2-n c\u00C3\u00AE 2- hungry.VAI -SAP Q \u00E2\u0080\u0098Are you hungry?\u00E2\u0080\u0099 b. \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAyan c\u00C3\u00AE CONJUNCT \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -yan c\u00C3\u00AE C1-hungry.VAI -2 Q \u00E2\u0080\u0098\u00E2\u0080\u00A6are you hungry?\u00E2\u0080\u0099 (cf. \u00E2\u0080\u0098Is it that you\u00E2\u0080\u0099re hungry?\u00E2\u0080\u0099) This anaphoric form is similar to the way in which, in English, context licenses a clefted form of the question.8 8 In fact, Plains Cree speakers who use Cree as their primary language will often translate \u00C3\u00AA-clauses into clefts, even when the English cleft has an extra presupposition that makes the English infelicitous. For example, during one elicitation session, someone excused himself to use the washroom, and a Cree speaker teased him: \u00E2\u0080\u009CIt\u00E2\u0080\u0099s you who has the bladder problem.\u00E2\u0080\u009D 157 When consultants are asked about using anaphoric CONJUNCT clauses in a situation where the context has explicitly not been given, the anaphoric clause is rejected as infelicitous. For example, in (42), the INDEPENDENT form can be felicitously uttered, but the anaphoric CONJUNCT form cannot. (42) context: out-of-the-blue (used to express the concept \u00E2\u0080\u0098Life\u00E2\u0080\u0099s great!\u00E2\u0080\u0099 or \u00E2\u0080\u0098It\u00E2\u0080\u0099s all good!\u00E2\u0080\u0099) a. miyw\u00C3\u00A2sin INDEPENDENT miyw\u00C3\u00A2si -n good.VII -0 \u00E2\u0080\u0098[it\u00E2\u0080\u0099s] good.\u00E2\u0080\u0099 b. # \u00C3\u00AA-miyw\u00C3\u00A2sik CONJUNCT \u00C3\u00AA- miyw\u00C3\u00A2si -k C1-good.VII -0 \u00E2\u0080\u0098\u00E2\u0080\u00A6it\u00E2\u0080\u0099s good.\u00E2\u0080\u0099 comment: it\u00E2\u0080\u0099s understandable, but I\u00E2\u0080\u0099d add something with this one9 An anaphoric clause in an initial position \u00E2\u0080\u0093 e.g., at the beginning of a discourse \u00E2\u0080\u0093 is quite restricted, if indeed it happens at all. One potential textual example of this is given in (43), which is the initial clause in Ahenakew (2000), a text based on a recorded interview. Here the interviewer is identifying herself, her interviewee, and the purpose of the conversation into the tape recorder. (43) \u00C3\u00AA-w\u00C3\u00AE-\u00C3\u00A2h-\u00C3\u00A2cimoy\u00C3\u00A2hk \u00C3\u00B4m \u00C3\u00B4t[a] \u00C3\u00A2nohc, \u00E2\u0080\u00A6 \u00C3\u00AA- w\u00C3\u00AE- \u00C3\u00A2h- \u00C3\u00A2cimo -y\u00C3\u00A2n-k \u00C3\u00B4ma \u00C3\u00B4ta anohc C1-INT-RED-story.VAI-1 -PL DEM.INAN here today \u00E2\u0080\u0098We are going to tell stories here today, \u00E2\u0080\u00A6 \u00E2\u0080\u0099 (FA in AA 1) Notice that this involves a complex context: the speaker is talking into a tape-recorder, rather than to the interviewer. The intended addressee will only be able to receive this message via the tape recorder, and so the presence of the tape recorder must be assumed by both the speaker and 9 The consultant added a preverbal resumptive topic marker to the anaphoric clause (i) to provide the relevant context: (i) \u00C3\u00AAwakw \u00C3\u00A2nima \u00C3\u00AA-miyw\u00C3\u00A2sik \u00C3\u00AAwakw anima \u00C3\u00AA- miyw\u00C3\u00A2si -k TOPIC DEM.INAN C1-good.VII -0 \u00E2\u0080\u0098It\u00E2\u0080\u0099s good.\u00E2\u0080\u0099 158 the hearer in this discourse \u00E2\u0080\u0093 it is a quintessential \u00E2\u0080\u009Cestablished context\u00E2\u0080\u009D. Within this context, the speaker is simply identifying the relevant activity as story-telling. 4.3.2.2 Contrastive focus Another environment where we can see the difference between indexical matrix clauses and anaphoric matrix clauses is when some part of the proposition has contrastive focus. Both the syntax and semantics of contrastive focus are relevant for the analysis of clause-typing proposed here. Semantically, contrastive focus is characterized as the choice of one alternative out of a (discoursally-given) set of alternatives (Rooth 1996, Hagstrom 1998, among others). Thus in a contrastive focus context we need a context: the speaker and hearer need to share the set of alternatives. Semantically then, contrastive focus should satisfy the anaphoric nature of CONJUNCT clauses. Syntactically, contrastive focus is treated as a (minimally) bi-partite structure (Rooth 1996): the focus part, and whatever is left over. In terms of the syntax, contrastive focus will invoke a clause external dependency between the open proposition and the focus element associated with it. Thus, syntactically, contrastive focus of any dependent in a clause should be syntactically incompatible with indexical INDEPENDENT clauses. Taken together, we expect that contrastive focus constructions will use anaphoric clauses, not indexical clauses. Of particular note in this example is the fact that the \u00C3\u00AA-clause is immediately followed by the demonstrative \u00C3\u00B4ma, which is not associated with any noun. This is significant because post-positional \u00C3\u00B4ma serves to introduce the element it follows as a predicate, and is interpreted as a subject of that predicate, rather than a modifying demonstrative. Consider the pair in (44): when the demonstrative precedes the nominal, the phrase is interpreted as a deictic DP \u00E2\u0080\u0098this knife\u00E2\u0080\u0099, but when the demonstrative follows the nominal, the phrase is interpreted as an instance of predication \u00E2\u0080\u0098this is a knife\u00E2\u0080\u0099 with m\u00C3\u00B4hkom\u00C3\u00A2n \u00E2\u0080\u0098knife\u00E2\u0080\u0099 as the predicate, and \u00C3\u00B4ma \u00E2\u0080\u0098this\u00E2\u0080\u0099 as the subject: 159 (44) a. \u00C3\u00B4ma m\u00C3\u00B4hkom\u00C3\u00A2n \u00C3\u00B4ma m\u00C3\u00B4hkom\u00C3\u00A2n DEM.INAN knife = \u00E2\u0080\u0098this knife \u00E2\u0080\u00A6\u00E2\u0080\u0099 \u00E2\u0089\u00A0 \u00E2\u0080\u0098this is a knife\u00E2\u0080\u0099 b. m\u00C3\u00B4hkom\u00C3\u00A2n \u00C3\u00B4ma m\u00C3\u00B4hkom\u00C3\u00A2n \u00C3\u00B4ma knife DEM.INAN \u00E2\u0089\u00A0 \u00E2\u0080\u0098this knife \u00E2\u0080\u00A6\u00E2\u0080\u0099 = \u00E2\u0080\u0098this is a knife\u00E2\u0080\u0099 (D\u00C3\u00A9chaine 1997) Analyzing the predicative example in (44b) as a kind of predicate inversion where the predicate has undergone raising (Moro 1999, D\u00C3\u00A9chaine 1997), is supported by the semantic properties associated with [X DEM ] strings: contrastive focus (Ahenakew 1987, Reinholtz 1995, 1999, Blain 1997, D\u00C3\u00A9chaine 1997, Okimasis & Ratt 1999, Wolvengray 2003). For example , in (45a), the initial locative element \u00C3\u00B4ta \u00E2\u0080\u0098here\u00E2\u0080\u0099 is focussed, and in (45b) the clause \u00C3\u00AA-p\u00C3\u00AAhoy\u00C3\u00A2n \u00E2\u0080\u0098I\u00E2\u0080\u0099m waiting\u00E2\u0080\u0099 in initial position corresponds to predicate-focus. (45) a. \u00C3\u00B4t \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AAhoy\u00C3\u00A2n \u00C3\u00B4ta \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AAho -y\u00C3\u00A2n here DEM.INAN C1-wait.VAI-1 \u00E2\u0080\u0098I\u00E2\u0080\u0099m waiting here. / It\u00E2\u0080\u0099s here that I\u00E2\u0080\u0099m waiting.\u00E2\u0080\u0099 (from Wolvengrey 2003, ex. 5b) b. \u00C3\u00AA-p\u00C3\u00AAhoy\u00C3\u00A2n \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AAho -y\u00C3\u00A2n \u00C3\u00B4ma C1-wait.VAI-1 DEM.INAN \u00E2\u0080\u0098I\u00E2\u0080\u0099m waiting.\u00E2\u0080\u0099 / \u00E2\u0080\u0098It\u00E2\u0080\u0099s that I\u00E2\u0080\u0099m waiting.\u00E2\u0080\u0099 (from Wolvengrey 2003, ex. 6a) Since [XP \u00C3\u00B4ma] structures are associated with contrastive focus of the XP, we expect an asymmetry between anaphoric CONJUNCT clauses and indexical INDEPENDENT ones. With anaphoric clauses the [XP \u00C3\u00B4ma] structure introduces contrastive predicate-focus. We expect the indexical INDEPENDENT clause to not be able to have this interpretation, and it doesn\u00E2\u0080\u0099t. 160 (46) context: argument over whether the speaker is angry or tired a. \u00C3\u00AA-n\u00C3\u00AAstosiy\u00C3\u00A2n \u00C3\u00B4ma, m\u00C3\u00B4y \u00C3\u00AA-kisiw\u00C3\u00A2siy\u00C3\u00A2n CONJUNCT \u00C3\u00AA-n\u00C3\u00AAstosi-y\u00C3\u00A2n \u00C3\u00B4ma m\u00C3\u00B4y \u00C3\u00AA-kisiw\u00C3\u00A2si-y\u00C3\u00A2n C1-tired.VAI-1 DEM.INAN NEG C1-angry.VAI-1 \u00E2\u0080\u0098I\u00E2\u0080\u0099m TIRED, not ANGRY.\u00E2\u0080\u0099 b. # nin\u00C3\u00AAstosin \u00C3\u00B4ma, m\u00C3\u00B4y nikisiw\u00C3\u00A2sin INDEPENDENT ni-n\u00C3\u00AAstosi -n \u00C3\u00B4ma m\u00C3\u00B4y ni- kisiw\u00C3\u00A2si -n 1-tired.VAI-SAP DEM.INAN NEG 1- angry.VAI-SAP \u00E2\u0080\u0098I AM tired, I am NOT angry.\u00E2\u0080\u0099 If the CP is an indexical clause, the only available interpretation is verum-focus: i.e., over the polar value of the entire proposition. This is exemplified in (47), where the two speakers are arguing over whether the proposition has a positive or negative value. The anaphoric CONJUNCT clause is infelicitous here, but the INDEPENDENT is appropriate. (47) context: argument over whether speaker is angry or not angry a. # \u00C3\u00AAha, \u00C3\u00AA-kisiw\u00C3\u00A2siy\u00C3\u00A2n \u00C3\u00B4ma CONJUNCT \u00C3\u00AAha \u00C3\u00AA- kisiw\u00C3\u00A2si -y\u00C3\u00A2n \u00C3\u00B4ma yes C1-angry.VAI-1 DEM.INAN \u00E2\u0080\u0098Yes, I\u00E2\u0080\u0099m angry, but \u00E2\u0080\u00A6\u00E2\u0080\u0099 b. \u00C3\u00AAha, nikisiw\u00C3\u00A2sin \u00C3\u00B4ma INDEPENDENT \u00C3\u00AAha ni- kisiw\u00C3\u00A2si -n \u00C3\u00B4ma yes 1- angry.VAI-SAP DEM.INAN \u00E2\u0080\u0098Yes, I AM angry.\u00E2\u0080\u0099 Verum-focus means that only the polarity of the proposition is available for focus in an indexical clause \u00E2\u0080\u0093 there can be no variables within the proposition. The association of polarity with the C domain is also independently argued for on the basis of Swampy Cree, a closely related language to Plains Cree (cf. Reinholtz 2007). 4.3.2.3 The distribution of clause-typing in elicitation: A discourse effect Elicitation offers a somewhat peculiar discourse context as opposed to regular language-use: utterances are devoid of their usual context. If we consider elicitation as a kind of (constructed) discourse, we can compare it to narratives and conversation, which are themselves quite distinct 161 discourse modes (Hockett 1958, Smith 2003, among others). In all three, we expect the initial utterance of the discourse to be indexical, as summarized in table 4.6. Table 4.6. Clause-typing in discourse-initial position We then expect that subsequent anaphoric clauses will be licensed \u00E2\u0080\u0093 since they are in a non- initial position and therefore satisfy the precedence condition \u00E2\u0080\u0093 until a new scene needs to be established. In elicitation, however, the scene is often re-established from one utterance to the next; for example, if I am testing aspectual classes of predicates, then each utterance will be unrelated to the next, and every utterance effectively begins a new discourse. In other words, such utterances are all in out-of-the-blue contexts. Table 4.7. Sequencing of clause-types by discourse-type This helps explain the preponderance of INDEPENDENT clauses as opposed to CONJUNCT clauses in Plains Cree. In sharp contrast to longer conversations and narratives \u00E2\u0080\u0093 where anaphoric clauses are by far more common \u00E2\u0080\u0093 anaphoric clauses are treated as \u00E2\u0080\u0098special\u00E2\u0080\u0099 in elicitation10. For example, in (48) the speaker proffered an INDEPENDENT clause, and when presented with its CONJUNCT counterpart, translated it with the nominal dislocated (i.e., as non- neutral). 10 Anecdotally, the preponderance of indexical INDEPENDENT clauses in elicitation contexts as opposed to narrative contexts is so striking that linguists more familiar with Plains Cree discourse will raise questions about the validity of the data. Discourse-type Initial utterance Elicitation indexical (PC: INDEPENDENT) Conversation indexical (PC: INDEPENDENT) Narrative indexical (PC: INDEPENDENT) Discourse-type Sequencing of clauses Elicitation indexical, indexical, indexical (PC: INDEPENDENT, INDEPENDENT, INDEPENDENT) Conversation Narrative indexical, anaphoric, anaphoric (PC: INDEPENDENT, CONJUNCT, CONJUNCT) 162 (48) a. \u00C3\u00A2hkosiw Tomio INDEPENDENT ahksoi -w T (offered) sick.VAI-3 T \u00E2\u0080\u0098Tomio is sick\u00E2\u0080\u0099 b. \u00C3\u00AA-\u00C3\u00A2hksosit Tomio CONJUNCT \u00C3\u00AA- ahkosi -t T (presented) C1-sick.VAI-3 T \u00E2\u0080\u0098He is sick, Tomio.\u00E2\u0080\u0099 When asked about anaphoric CONJUNCT clauses, consultants will in fact often refer to other modes of discourse, saying things like \u00E2\u0080\u009Cthis is something you would say in the middle of a story\u00E2\u0080\u009D or \u00E2\u0080\u009Cmaybe if you were talking to someone\u00E2\u0080\u009D. More specifically, consultants will often provide a conversational context for the CONJUNCT clause in question. This is exemplified in (49); here the speaker had been presented with the form \u00C3\u00AA-k\u00C3\u00AE-n\u00C3\u00AAstosiy\u00C3\u00A2n \u00E2\u0080\u0098\u00E2\u0080\u00A6I was tired\u00E2\u0080\u0099, had ruled it a good utterance, and had then been asked to provide a situation in which the utterance would be felicitous. The consultant responded by providing a preceding question (which is itself presuppositional; see chapter 6), and then used the targeted form as the response11. (49) context: constructed context for \u00C3\u00AA-k\u00C3\u00AE-n\u00C3\u00AAstosiy\u00C3\u00A2n A: T\u00C3\u00A2n\u00C3\u00AAhki \u00C3\u00B4ma \u00C3\u00AAk\u00C3\u00A2 k\u00C3\u00A2-p\u00C3\u00AA-itoht\u00C3\u00AAyan t\u00C3\u00A2n-\u00C3\u00AAhki \u00C3\u00B4ma \u00C3\u00AAk\u00C3\u00A2 k\u00C3\u00A2-p\u00C3\u00AA- it\u00C3\u00B4ht\u00C3\u00AA -yan WH-REASON DEM.INAN NEG C2-COME-go.VAI-2 \u00E2\u0080\u0098Why didn\u00E2\u0080\u0099t you come?\u00E2\u0080\u0099 [Lit: why is it that you didn\u00E2\u0080\u0099t come?] B: \u00C3\u00AA-k\u00C3\u00AE-os\u00C3\u00A2m-n\u00C3\u00AAstosiy\u00C3\u00A2n \u00C3\u00AA- k\u00C3\u00AE- os\u00C3\u00A2m-n\u00C3\u00AAstosi -y\u00C3\u00A2n C1-PREV-DEG- tired.VAI-1 \u00E2\u0080\u0098\u00E2\u0080\u00A6I was too tired.\u00E2\u0080\u0099 Here the context on which the clause is anaphoric has been linguistically defined by a preceding clause; by virtue of its having been uttered within the same conversation, it is part of the context that has been established between the speaker and hearer. Anaphoric clauses are expected to emerge in elicitation contexts where the targeted utterance is non-initial in the discourse. This is also borne out: there is a strong tendency, particularly when doing English-Cree translation tasks, for a consultant to start with a Cree 11 The consultant has also added the quantifier os\u00C3\u00A2m to the target form. This doesn\u00E2\u0080\u0099t change the context, and it indicates that the speaker has \u00E2\u0080\u0098taken the utterance on as their own\u00E2\u0080\u0099: she added some element that made the utterance feel appropriate, rather than simply repeating a citation form (Cook & M\u00C3\u00BChlbauer 2006). 163 INDEPENDENT clause, and then, if asked to repeat the utterance, to switch to an anaphoric CONJUNCT clause. In (50), the initial translation task uses an indexical INDEPENDENT clause. Here I was interested in the presence of the -a suffix on the noun iskw\u00C3\u00AAw \u00E2\u0080\u0098woman\u00E2\u0080\u0099 and the -yi suffix in the embedded clause, and was trying to remove them \u00E2\u0080\u0093 a process which makes the utterance ungrammatical (see Wolfart 1973, 1996; Cook & M\u00C3\u00BChlbauer 2006, M\u00C3\u00BChlbauer 2007, chapter 5 of this thesis). Thus I was using the same utterance and changing only one element. In this non- initial utterance, the consultant switched to the anaphoric CONJUNCT clause12. (50) a. context: translation of \u00E2\u0080\u0098that man wants that woman to dance .\u00E2\u0080\u0099 ana nap\u00C3\u00AAw nitaw\u00C3\u00AAyim\u00C3\u00AAw anihi iskw\u00C3\u00AAwa ka-nimihitoyit INDEPENDENT ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim -\u00C3\u00AA -w anihi iskw\u00C3\u00AAw -a ka- nimihito -yi -t DEM.AN man want.VTA -DIR-3 DEM.INAN woman-OBV IRR-dance.VAI-DS-3 \u00E2\u0080\u0098That man wants that woman to dance.\u00E2\u0080\u0099 b. context: consultant repeating (a) with the -yi suffix missing (*) ana n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA-nitaw\u00C3\u00AAyim\u00C3\u00A2t iskw\u00C3\u00AAwa ka-nimihitot CONJUNCT ana n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA- nitaw\u00C3\u00AAyim -\u00C3\u00A2 -t iskw\u00C3\u00AAw -a DEM.AN man C1-want.VTA -DIR-3 woman-OBV --- (intended: \u00E2\u0080\u0098That man wants the woman to dance.\u00E2\u0080\u0099) Here the consultant takes the initial utterance as a context for the repetition of the utterance. This is parallel to what consultants do with pronominal anaphora, where they will initially insist on an overt nominal to identify a referent, but will, on repeating the utterance, revert to a pronominal form (51). 12 One can easily imagine the trouble this creates for collecting minimal pairs. 164 (51) a. context: initial utterance atim nip\u00C3\u00A2w INDEPENDENT W/ OVERT NOMINAL atimw nip\u00C3\u00A2 -w dog sleep.VAI-3 \u00E2\u0080\u0098The dog is sleeping\u00E2\u0080\u0099 b. context: non-initial utterance tanisp\u00C3\u00AE \u00C3\u00AA-k\u00C3\u00AE-nip\u00C3\u00A2t CONJUNCT W/OUT OVERT NOMINAL t\u00C3\u00A2n -isp\u00C3\u00AE \u00C3\u00AA- k\u00C3\u00AE- nip\u00C3\u00A2 -t Q -time C1-PREV-sleep.VAI-3 \u00E2\u0080\u0098When did he [the dog] go to sleep?\u00E2\u0080\u0099 To sum up, anaphoric CONJUNCT clauses appear less frequently in elicitation contexts than in narrative contexts or conversational contexts, and the places where anaphoric CONJUNCT clauses emerge in elicitation are similar to the places where pronominal anaphora occur, such as non- initial utterances that are related to some previous utterance (i.e., have the same propositional content). 4.3.3 Interim summary I have shown that anaphoric CONJUNCT clauses behave like nominal anaphora in that they need an antecedent and that the antecedent-anaphor relation is satisfied by either a c-command condition or a precedence condition. I have also shown that they can be licensed by an established context, which is consistent with the behaviour of pronominal anaphora and can be thought of as satisfying precedence over a larger domain. I now turn to temporal and referential dependent elements contained within anaphoric CONJUNCT clauses and show that the relation between these dependent elements and their antecedents also obey the c-command and precedence conditions on anaphora. 165 4.4 Anaphoric dependencies in anaphoric clauses So far we have seen that anaphoric clauses are licensed by the general principles of anaphora: they may be licensed by a linguistic antecedent given that either the precedence or c-command condition holds. Recall that during our first look at anaphoric clauses back in chapter 3, we saw that the temporal interpretation of an anaphoric clause could be set relative to the temporal interpretation of its antecedent. For example, in (52a) we have a chain of clauses with k\u00C3\u00AE- marked only on the initial clause, but all within the same temporal setting. Here the temporal shifter k\u00C3\u00AE-, which is obligatory in indexical clauses, cannot be used on the non-initial anaphoric clauses in the sequence (52b-c), nor can all the anaphoric clauses be marked (52d). (52) a. \u00C3\u00AAkwa mistahi m\u00C3\u00A2n \u00C3\u00A2ya, \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, ANTECEDENT \u00C3\u00AAkwa mistahi m\u00C3\u00A2na aya \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n -k and much usually CONN C1-PREV-go.about.VAI-1 -PL \u00E2\u0080\u0098And we used to go around a lot, \u00C3\u00AA-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw ANAPHOR \u00C3\u00AA- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n -k \u00C3\u00A2skaw C1-go.with.VTA-DIR-1 -PL sometimes sometimes going along with her \u00C3\u00AA-~ \u00C3\u00AA-pap\u00C3\u00A2mi-mawisot, ... ANAPHOR \u00C3\u00AA- pap\u00C3\u00A2mi- mawiso -t C1-about.PV-pick.berries.VAI-3 as she went about berry-picking, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 17) b. * \u00E2\u0080\u00A6, \u00C3\u00AA-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw \u00C3\u00AA-~ \u00C3\u00AA-pap\u00C3\u00A2mi-mawisot, \u00C3\u00AA- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n-k \u00C3\u00AA- k\u00C3\u00AE- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n -k \u00C3\u00A2skaw C1-go.about.VAI-1 -PL C1-PREV-go.with.VTA-DIR-1 -PL times \u00C3\u00AA- pap\u00C3\u00A2mi- mawiso -t C1-go- pick.berries.VAI-3 --- 166 c. * \u00E2\u0080\u00A6, \u00C3\u00AA-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, \u00C3\u00AA-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw \u00C3\u00AA-~ \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2mi-mawisot, \u00C3\u00AA- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n-k \u00C3\u00AA- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n -k \u00C3\u00A2skaw C1-go.about.VAI-1 -PL C1-go.with.VTA-DIR-1 -PL times \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2mi- mawiso -t C1-PREV-go- pick.berries.VAI-3 --- d. * \u00E2\u0080\u00A6, \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw \u00C3\u00AA-~ \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2mi-mawisot, \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n-k \u00C3\u00AA- k\u00C3\u00AE- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n -k \u00C3\u00A2skaw C1-PREV-go.about.VAI-1 -PL C1-PREV-help.VTA-DIR-1 -PL times \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2mi- mawiso -t C1-PREV-go- pick.berries.VAI-3 --- The anaphoric properties of clauses, then, affect not only the distribution of the clause itself, but the distribution and interpretation of dependent expressions contained within the anaphoric clause. In principle, since I take anaphora to be an elsewhere (i.e., generalized) phenomena, any variable in the clause should be able to establish clause-external relations with an antecedent, and multiple anaphor-antecedent relations can be established at once. For example, we expect temporal reference (which is in the situation variable s), referent tracking (i.e., argument variables like the different subject marker -yi), and modifier variables (e.g., isi \u00E2\u0080\u0098thus\u00E2\u0080\u0099 and ohci \u00E2\u0080\u0098for\u00E2\u0080\u0099) to be able to have clause-external antecedents, regardless of where they are introduced in the clause. (53) antecedenti CP 3 si 3IP 3 ohci-i 3VP -yii 3 isi-i 3 ohci-i 167 The table in 4.8 summarizes the patterns we expect: in indexical clauses these dependent elements must either be deictic, or they are ungrammatical; in anaphoric clauses, they are anaphoric.13 VARIABLE INDEXICAL ANAPHORIC Temporal k\u00C3\u00AE- \u00E2\u009C\u0094(Deictic) \u00E2\u009C\u0094 (Anaphoric) Referential -yi \u00E2\u009C\u0094(Deictic) \u00E2\u009C\u0094 (Anaphoric) isi- \u00E2\u009C\u0094(Deictic) \u00E2\u009C\u0094 (Anaphoric) ohci- \u00E2\u009C\u0094(Deictic) \u00E2\u009C\u0094 (Anaphoric) isp\u00C3\u00AE \u00E2\u0080\u0098time\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (Anaphoric) Modifier variables ita \u00E2\u0080\u0098place\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0094 (Anaphoric) Table 4.8. Deictic vs. anaphoric dependents In this section, I show that the distribution and interpretation of both the temporal shifting element k\u00C3\u00AE-, and the different subject marker -yi are subject to the c-command and precedence conditions on anaphora.14 4.4.1 k\u00C3\u00AE- is subject to c-command and precedence in CONJUNCT clauses In chapter 3 we saw that k\u00C3\u00AE- \u00E2\u0080\u0098PREVIOUS\u00E2\u0080\u0099 was obligatory in indexical clauses to shift the reference time relative to the evaluation time. Here I look at its function in anaphoric clauses, where we shall see that instead of shifting relative to utterance time, k\u00C3\u00AE- shifts relative to the time in the antecedent clause. Consider the following set of contrasts. In both examples of the first set (54a-b), we have an anaphoric clause followed by an indexical clause. They differ minimally in that (54b) has k\u00C3\u00AE- added to the indexical clause. Notice that the temporal sequencing of the events (the time of 13 See the discussion in chapter 3 for the discussion of modifier variables, which I showed could have clause- external antecedents in anaphoric CONJUNCT clauses, but not in indexical INDEPENDENT clauses. 14 Note that the current analysis predicts that it is possible for an anaphoric clause to host multiple different anaphoric relations (for example, an anaphoric temporal element relating to one clause, and an anaphoric referential element relating to another clause). Such cases are attested in Plains Cree. 168 eating relative to the time of coming home) is unaffected by this difference; in both cases the coming home precedes the eating15. (54) a. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff nim\u00C3\u00AEcison\u00C3\u00A2n. INDEPENDENT \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J ni-m\u00C3\u00AEciso-n\u00C3\u00A2n C1-come-go.home.VAI-3 J 1-eat.VAI-1PL \u00E2\u0080\u0098Jeff is home, we ate.\u00E2\u0080\u0099 = PRECEDE (come home, eat) b. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff nik\u00C3\u00AE-m\u00C3\u00AEcison\u00C3\u00A2n. INDEPENDENT W/ k\u00C3\u00AE- \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J ni- k\u00C3\u00AE- m\u00C3\u00AEciso -n\u00C3\u00A2n C1-come-go.home.VAI-3 J 1- PREV-eat.VAI-1PL \u00E2\u0080\u0098When Jeff came, then we ate.\u00E2\u0080\u0099 = PRECEDE (come home, eat) In the second set of examples (55a-b), the indexical clause has been replaced by an anaphoric clause. When k\u00C3\u00AE- is added in (55b), it acts anaphorically, it reverses the temporal relation between the two events: in this example the eating occurs before the coming home. With the change in temporal ordering is a concomitant change in the participants of the eating: since the eating happened before Jeff came home, Jeff could not be one of the eaters. (55) a. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff, \u00C3\u00AA-m\u00C3\u00AEcisoy\u00C3\u00A2hk CONJUNCT \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J \u00C3\u00AA- m\u00C3\u00AEciso -y\u00C3\u00A2n -k C1-come-go.home.VAI-3 J C1-eat.VAI -1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6Jeff came home and we ate.\u00E2\u0080\u0099 (we = speaker & Jeff) = PRECEDE (come home, eat) b. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff, \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEcisoy\u00C3\u00A2hk CONJUNCT W/ k\u00C3\u00AE- \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00AEciso -y\u00C3\u00A2n -k C1-come-go.home.VAI-3 J C1-PREV-eat.VAI -1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6Jeff came home, we had eaten.\u00E2\u0080\u0099 (we = speaker & someone else) = PRECEDE (eat, come home) The next sections show that the precedence and/or c-command conditions must hold in order for k\u00C3\u00AE- to behave anaphorically in an anaphoric clause. 15 The only difference in the translation is the presence of the overt temporal connective then in the example containing k\u00C3\u00AE-. I suspect that this relates to the emphatic reading discussed below; at any rate, the presence of k\u00C3\u00AE- clearly does not reverse the relative temporal sequencing of the two events in Plains Cree. 169 4.4.1.1 Precedence without c-command: k\u00C3\u00AE- is anaphoric The first context I examine is one where the anaphoric element (here k\u00C3\u00AE-) follows its antecedent, but is not subordinate to it: precedence holds, but c-command does not. (56) CP CP 5 5 antecedent anaphor This is essentially the case we saw above. In the second clause, the presence of k\u00C3\u00AE- yields an interpretation where the event in the second clause temporally precedes the event in the second clause. There are many instances of this kind of dependency. One case is in conversational sequences like (57). Here the k\u00C3\u00AE- in the second persons\u00E2\u0080\u0099 response provides the necessary sequencing of the crying before the laughing. (57) A. \u00C3\u00AA-p\u00C3\u00A2hpit ana iskw\u00C3\u00AAw \u00C3\u00AA- p\u00C3\u00A2hpi -t ana iskw\u00C3\u00AAw C1-laugh.VAI-3 DEM.AN woman \u00E2\u0080\u0098That woman is/was laughing.\u00E2\u0080\u0099 B. \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2tot \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00A2to -t C1-PREV-cry.VAI-3 \u00E2\u0080\u0098She was crying (earlier).\u00E2\u0080\u0099 Here, the first speaker\u00E2\u0080\u0099s utterance provides the reference time which k\u00C3\u00AE- is evaluated with respect to. This latter example confirms that an \u00C3\u00AA-conjunct clause\u00E2\u0080\u0099s dependency may be to some other speaker\u00E2\u0080\u0099s utterance in a dialogue, as we saw earlier in the discussion of elicitation and conversation (cf. \u00C2\u00A74.3.2.3). If the anaphoric reading is excluded, the utterance becomes ungrammatical: the k\u00C3\u00AE- in the second-clause must have an antecedent in order for the asymmetry between the two clauses to be resolved. This accounts for the ungrammatical examples we saw earlier when we attempted to place k\u00C3\u00AE- on some non-initial CONJUNCT clause, repeated here as (58). In such a case k\u00C3\u00AE- does not have an antecedent; and under the targeted interpretation must itself be the antecedent; since this violates the precedence and c-command conditions, it is ungrammatical. 170 (58) * [ anaphoric ] [ antecedent ] ! [ \u00C3\u00AA-nikamocik ] [ \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-itoht\u00C3\u00AAcik m\u00C3\u00A2na aw\u00C3\u00A2sisak ] \u00C3\u00AA- nikamo -t-k \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00AA- itoht\u00C3\u00AA -t -k m\u00C3\u00A2na aw\u00C3\u00A2sis-ak C1-sing.VAI-3-PL C1-PREV-COME-go.VAI-3-PL usually child -PL --- (intended: \u00E2\u0080\u0098The children sang and came.\u00E2\u0080\u0099) The anaphoric behaviour of k\u00C3\u00AE- is also exhibited in the clause chains in Plains Cree narratives, where, in a way that is remarkable from an English perspective, temporal sequencing is almost universally done without adverbials like before and after. In (59), we have a sequence of three clauses. The first two clauses are unmarked, and the temporal interpretation follows from the linear ordering of the events (\u00C3\u00AA-m\u00C3\u00A2h-manipitahk \u00E2\u0080\u0098s/he pulled them up\u00E2\u0080\u0099 linearly and temporally precedes \u00C3\u00AA-m\u00C3\u00A2h-m\u00C3\u00AEcit \u00E2\u0080\u0098s/he ate them\u00E2\u0080\u0099). The final clause lacks any of the temporal connectives (such as \u00C3\u00AAkwa \u00E2\u0080\u0098and\u00E2\u0080\u0099 or m\u00C3\u00AEna \u00E2\u0080\u0098also\u00E2\u0080\u0099) that we saw earlier, but is marked with k\u00C3\u00AE-. Temporally, the event in this clause (\u00C3\u00AA-k\u00C3\u00AE-kan\u00C3\u00A2ciht\u00C3\u00A2t \u00E2\u0080\u0098s/he cleaned them\u00E2\u0080\u0099) is interpreted as preceding the eating: k\u00C3\u00AE- is shifting the temporal reference. (59) a. s\u00C3\u00B4skw\u00C3\u00A2c \u00C3\u00AA-m\u00C3\u00A2h-manipitahk anihi oc\u00C3\u00AApihkisa, s\u00C3\u00B4skw\u00C3\u00A2c \u00C3\u00AA- m\u00C3\u00A2h-manipitam -k anihi oc\u00C3\u00AApihkis-a \u00C3\u00AA -m\u00C3\u00A2h-m\u00C3\u00AEci -t \u00C3\u00AAkoni simply C1-RED- pull.up.VTI-3 DEM.OBV root -OBV C1-RED-eat.VTI-3 RESUMP \u00E2\u0080\u0098\u00E2\u0080\u00A6she simply pulled up those little roots \u00C3\u00AA-m\u00C3\u00A2h-m\u00C3\u00AEcit \u00C3\u00AAkoni ... \u00C3\u00AA-k\u00C3\u00AE-kan\u00C3\u00A2ciht\u00C3\u00A2t \u00C3\u00AAkoni. \u00C3\u00AA- m\u00C3\u00A2h-m\u00C3\u00AEci -t \u00C3\u00AAkoni \u00C3\u00AA- k\u00C3\u00AE- kanaciht\u00C3\u00A2 -t \u00C3\u00AAkoni C1-RED- eat.VTI-3 RESUMP C1-PREV-clean.VAI -3 RESUMP \u00E2\u0080\u0098\u00E2\u0080\u00A6and cleaned and ate them.\u00E2\u0080\u0099 (EM 71) Similarly, in (60), the first two clauses are unmarked, while the third clause is marked with k\u00C3\u00AE- and reports on an event that had happened much earlier in the story (several months). Again, the only indication of this sequencing is this preverb, which corresponds with the past perfect in the English. 171 (60) \u00C3\u00AAkwa, kis\u00C3\u00AAyiniw aw \u00C3\u00AAkwa \u00C3\u00AAsa \u00C3\u00AA-s\u00C3\u00A2sakitisihk, \u00C3\u00AAkwa kis\u00C3\u00AAyiniw awa \u00C3\u00AAkwa \u00C3\u00AAsa \u00C3\u00AA- s\u00C3\u00A2skitisin -k and old.man this.AN and EVID C1-lay.on.back.VAI-3 \u00E2\u0080\u0098Now the old man was lying on his back, m\u00C3\u00AAtoni mistah \u00C3\u00AA-m\u00C3\u00A2miton\u00C3\u00AAyihtahk \u00C3\u00B4ma \u00E2\u0080\u0093 mitoni mistahi \u00C3\u00AA- m\u00C3\u00A2miton\u00C3\u00AAyihtam-k \u00C3\u00B4ma intense much C1-think.VTI -3 this.INAN thinking about things a very great deal \u00E2\u0080\u0093 m\u00C3\u00A2ka p\u00C3\u00A2skisikan wiy \u00C3\u00AAsa \u00C3\u00AA-k\u00C3\u00AE-nakatam\u00C3\u00A2ht \u00C3\u00AAsa. m\u00C3\u00A2ka p\u00C3\u00A2skisikan wiya \u00C3\u00AAsa \u00C3\u00AA- k\u00C3\u00AE- nakatam\u00C3\u00A2-h -t \u00C3\u00AAsa but gun emphatic EVID C1-PREV-left.VTI -US-3 EVID but he had been left with a gun.\u00E2\u0080\u0099 (AA 9.8) All of the cases we have seen so far have no temporal reference point other than the event time of the preceding clause, and no temporal connectives such as \u00C3\u00AAkwa \u00E2\u0080\u0098and, now/then\u00E2\u0080\u0099 (cf. Ogg 1991). In the following example, both an overt temporal adverb (ot\u00C3\u00A2kosihk \u00E2\u0080\u0098yesterday\u00E2\u0080\u0099), and \u00C3\u00AAkwa are present, and we see that it opens up additional temporal sequencing possibilities. The interpretation of \u00C3\u00AA-k\u00C3\u00AE-ahkosit \u00E2\u0080\u0098s/he was sick\u00E2\u0080\u0099) is ambiguous: it can either mean that nim\u00C3\u00A2ma \u00E2\u0080\u0098my mom\u00E2\u0080\u0099 was sick at the time she came to visit, or that she had been sick previous to her coming to visit16. (61) a. ot\u00C3\u00A2kosihk nim\u00C3\u00A2ma p\u00C3\u00AA-kiyok\u00C3\u00AAw, \u00C3\u00AAkwa \u00C3\u00AA-ahkosit ot\u00C3\u00A2kosin -k ni-m\u00C3\u00A2ma p\u00C3\u00AA- kiyok\u00C3\u00AA -w \u00C3\u00AAkwa \u00C3\u00AA- ahkosi -t be.evening.VII-0 1-mama come-visit.VAI-3 and C1-sick.AI-3 \u00E2\u0080\u0098My mother came to visit yesterday and she was sick at the time.\u00E2\u0080\u0099 b. ot\u00C3\u00A2kosihk nim\u00C3\u00A2ma p\u00C3\u00AA-kiyok\u00C3\u00AAw, \u00C3\u00AAkwa \u00C3\u00AA-k\u00C3\u00AE-ahkosit ot\u00C3\u00A2kosin -k ni- m\u00C3\u00A2ma p\u00C3\u00AA- kiyok\u00C3\u00AA -w \u00C3\u00AAkwa \u00C3\u00AA- ahkosi -t be.evening.VII-0 1- mama COME-visit.VAI-3 and C1-sick.AI -3 = \u00E2\u0080\u0098Yesterday my mother came to visit and she had been sick\u00E2\u0080\u0099 = \u00E2\u0080\u0098Yesterday my mother came to visit and she was sick at the time.\u00E2\u0080\u0099 16 Given the fact that the predicate ahkosi- \u00E2\u0080\u0098sick.VAI\u00E2\u0080\u0099 seems to vary between whether it is a stative or inchoative predicate (Wolfart, p.c.), some of the ambiguity in interpretation may be attributed to that. The interpretation may also depend on the temporal sequencing properties of \u00C3\u00AAkwa \u00E2\u0080\u0098and, now/then\u00E2\u0080\u0099, which I do not have a full account of, but discuss in \u00C2\u00A74.5.1. 172 While I do not have a full account of this, crucially an anaphoric relation between the two clauses can still be established. This means that we can maintain the precedence condition for k\u00C3\u00AE- in clause sequences. 4.4.1.2 Precedence and c-command: k\u00C3\u00AE- is anaphoric If k\u00C3\u00AE- appears in an anaphoric clause that is subordinate to the preceding clause, we predict that it should be anaphoric. Both the precedence and c-command conditions hold. (62) CP 5 antecedent CP 5 anaphor This prediction is borne out when we look at the interpretation of k\u00C3\u00AE- in subordinate clauses. In (63), we have a perceptual predicate where the proposition is the thing perceived: i.e., the \u00C3\u00AA- conjunct clause is behaving as a part of the larger clause. If the subordinate clause is not marked with k\u00C3\u00AE-, it may be interpreted as simultaneous or sequenced (63a). If k\u00C3\u00AE- is added to the subordinate clause, the interpretation is restricted so that only the sequenced interpretation is available (63b). (63) a. w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-mispohk CONJUNCT w\u00C3\u00A2pahtam-w J \u00C3\u00AA- mispon -k see.VTI -3 J C1-snow.VII-0 \u00E2\u0080\u0098Jeff saw that it (had) snowed.\u00E2\u0080\u0099 = (i) PRECEDE (snow, see) = (ii) OVERLAP (snow, see) b. w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-k\u00C3\u00AE-mispohk CONJUNCT W/ k\u00C3\u00AE- w\u00C3\u00A2pahtam-3 J \u00C3\u00AA- k\u00C3\u00AE- mispon -k see.VTI -3 J C1-PREV-snow.VII-0 \u00E2\u0080\u0098Jeff saw that it (had) snowed.\u00E2\u0080\u0099 = (i) PRECEDE (snow, see) \u00E2\u0089\u00A0 (ii) OVERLAP (snow, see) 173 While double-marking of k\u00C3\u00AE- (i.e., in both clauses) in this context was ruled inappropriate, the consultant noted that, insofar as the utterance meant anything, only the sequenced interpretation was valid. (64) ? k\u00C3\u00AE-w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-k\u00C3\u00AE-mispohk k\u00C3\u00AE- w\u00C3\u00A2pahtam-3 J \u00C3\u00AA- k\u00C3\u00AE- mispon -k prev-see.VTI -3 J C1-PREV-snow.VII-0 --- = (i) PRECEDE (snow, see) \u00E2\u0089\u00A0 (ii) OVERLAP (snow, see) comment: don\u00E2\u0080\u0099t need both k\u00C3\u00AE-\u00E2\u0080\u0099s \u00E2\u0080\u0093 it\u00E2\u0080\u0099s too much The interpretation of k\u00C3\u00AE- in the dependent clauses in example (65 and 66) is also obligatorily calculated with respect to the higher clause. Notice that the sequencing effect is insensitive to the aspectual class of the predicate in the dependent clause: in (65) it is eventive, while in (66) it is stative. (65) a. nip\u00C3\u00AAht\u00C3\u00AAn Jeff \u00C3\u00AA-nikamot CONJUNCT ni-p\u00C3\u00AAht\u00C3\u00AA -n J \u00C3\u00AA- nikamo -t 1- hear.VTI-SAP J C1-sing.VAI-3 \u00E2\u0080\u0098I heard that Jeff was singing\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (sing, hear)17 = (ii) OVERLAP (sing, hear) b. nip\u00C3\u00AAht\u00C3\u00AAn Jeff \u00C3\u00AA-k\u00C3\u00AE-nikamot CONJUNCT W/ k\u00C3\u00AE- ni- p\u00C3\u00AAht\u00C3\u00AA -n J \u00C3\u00AA- k\u00C3\u00AE- nikamo -t 1- hear.VTI-SAP J C1-PREV-sing.VAI-3 \u00E2\u0080\u0098I heard Jeff had sang.\u00E2\u0080\u0099 = (i) PRECEDE (sing, hear) \u00E2\u0089\u00A0 (ii) OVERLAP (sing, hear) 17 I am not sure why the PRECEDE interpretation is unavailable in this example, but acceptable in, e.g., (63a) and (67a). If this represents a consistent pattern, it is important for understanding the temporal behaviour of unmarked \u00C3\u00AA-CONJUNCT clauses. Since it is not of direct import to understanding the anaphoric behaviour of k\u00C3\u00AE- here, I leave it for further research. 174 (66) a. niw\u00C3\u00A2paht\u00C3\u00AAn Joe \u00C3\u00AA-kisiw\u00C3\u00A2sit CONJUNCT ni-w\u00C3\u00A2paht\u00C3\u00AA-n J \u00C3\u00AA- kisiw\u00C3\u00A2si -t 1- see.VTI -SAP J C1-angry.VAI-3 \u00E2\u0080\u0098I see Joe is/was mad.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (angry, see) = (ii) OVERLAP (angry, see) b. niw\u00C3\u00A2paht\u00C3\u00AAn Joe (\u00C3\u00AAsa) \u00C3\u00AA-k\u00C3\u00AE-kisiw\u00C3\u00A2sit CONJUNCT W/ k\u00C3\u00AE- ni-w\u00C3\u00A2paht\u00C3\u00AA-n J \u00C3\u00AAsa \u00C3\u00AA- k\u00C3\u00AE- kisiw\u00C3\u00A2si -t 1- see.VTI -SAP J EVID C1-PREV-angry.VAI-3 \u00E2\u0080\u0098I see Joe was mad.\u00E2\u0080\u0099 = (i) PRECEDE (angry, see) \u00E2\u0089\u00A0 (ii) OVERLAP (angry, see) Dependent clauses introduced by a verb of speaking behave the same way: the presence of k\u00C3\u00AE- in the dependent clause obligatorily sequences it with respect to the matrix verb. This is illustrated in (67) by the infelicity of using k\u00C3\u00AE- in a dependent clause in a context where the event time should not be sequenced with respect to the higher predicate. (67) context: you are going to meet someone you don\u00E2\u0080\u0099t know at the airport, and he gives you information that you can identify him by. Later, you are relaying these to a friend a. n\u00C3\u00A2p\u00C3\u00AAw nitik \u00C3\u00AA-kinosit CONJUNCT n\u00C3\u00A2p\u00C3\u00AAw n- it -ik \u00C3\u00AA- kinosi -t man 1- tell.VTA-INV C1-tall.VAI-3 \u00E2\u0080\u0098He told me he is/was tall.\u00E2\u0080\u0099 b. # n\u00C3\u00A2p\u00C3\u00AAw nitik \u00C3\u00AA-k\u00C3\u00AE-kinosit CONJUNCT W/ k\u00C3\u00AE- n\u00C3\u00A2p\u00C3\u00AAw n-it -ik \u00C3\u00AA- k\u00C3\u00AE- kinosi -t man 1-tell.VTA-INV C1-PREV-tall.VAI-3 \u00E2\u0080\u0098This man told me that he used to be tall.\u00E2\u0080\u0099 comment: *laughter* how are you gonna know he used to be tall?? Other kinds of subordinate clauses also exhibit the expected pattern. The pair in (68) shows the contrast in interpretation when k\u00C3\u00AE- is used in a relative clause; in (69), we see the same contrast in a temporal modification clause; and in (70) the contrast again in a reason clause introduced by ayis \u00E2\u0080\u0098for/because\u00E2\u0080\u0099. 175 (68) a. nim\u00C3\u00B4sahkin\u00C3\u00A2w b\u00C3\u00AAb\u00C3\u00AEsis ana k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-m\u00C3\u00A2tot CONJUNCT ni- m\u00C3\u00B4sahkin -\u00C3\u00A2 -w b\u00C3\u00AAb\u00C3\u00AE -sis ana k\u00C3\u00A2- m\u00C3\u00AAkw\u00C3\u00A2- m\u00C3\u00A2to -t 1- pick.up.VTA-DIR-3 baby-DIM DEM.AN C2-WHILE- cry.VAI-3 \u00E2\u0080\u0098I picked up the baby, the one that was crying.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (cry, pick up) = (ii) OVERLAP (cry, pick up) b. nim\u00C3\u00B4sahkin\u00C3\u00A2w ana b\u00C3\u00AAb\u00C3\u00AEsis ana k\u00C3\u00A2-k\u00C3\u00AE-m\u00C3\u00A2tot CONJUNCT W/ k\u00C3\u00AE- ni- m\u00C3\u00B4sahkin -\u00C3\u00A2 -w ana b\u00C3\u00AAbi -sis ana k\u00C3\u00A2- k\u00C3\u00AE- m\u00C3\u00A2to -t 1- pick.up.VTA-DIR-3 DEM.AN baby-DIM DEM.AN C2-PREV-cry.VAI-3 \u00E2\u0080\u0098I picked up the baby that had been crying.\u00E2\u0080\u0099 = (i) PRECEDE (cry, pick up) \u00E2\u0089\u00A0 (ii) OVERLAP (cry, pick up) (69) a. Clare k\u00C3\u00A2kik\u00C3\u00AA m\u00C3\u00A2tow, nikamoci CONJUNCT C k\u00C3\u00A2kike m\u00C3\u00A2to -w nikamo -t -i C always cry.VAI-3 sing.VAI-3-SUBJ \u00E2\u0080\u0098Clare always cries when she sings.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (sing, cry) = (ii) OVERLAP (sing, cry) b. Clare k\u00C3\u00A2kik\u00C3\u00AA m\u00C3\u00A2tow, k\u00C3\u00AE-nikamoci18 CONJUNCT W/ k\u00C3\u00AE- C k\u00C3\u00A2kike m\u00C3\u00A2to -w k\u00C3\u00AE- nikamo -t -i C always cry.VAI-3 PREV-sing.VAI-3-SUBJ \u00E2\u0080\u0098Clare always cries when she\u00E2\u0080\u0099s through/done singing.\u00E2\u0080\u0099 = (i) PRECEDE (sing, cry) \u00E2\u0089\u00A0 (ii) OVERLAP (sing, cry) 18 In the corpus data I\u00E2\u0080\u0099ve looked at, the constructions parallel to the data in (69) are always also marked with initial change, which ablauts the initial vowel (cf. Hockett 1966, Wolfart 1973). In this case, since the initial vowel is the [i] in k\u00C3\u00AE-, the changed vowel should be [a]. A relevant example from Minde (1997) is given in (i): here we get the same sequencing effect as in the elicitation data (i.e., the resting and sleeping always took place before the working in the fields). (i) ... \u00C3\u00AAkwa k\u00C3\u00A2-ayiw\u00C3\u00AApici k\u00C3\u00A2-nip\u00C3\u00A2ci, \u00C3\u00AAkwa IC. k\u00C3\u00AE- ayiw\u00C3\u00AApi -t -i IC.k\u00C3\u00AE- nip\u00C3\u00A2 -t -i and IC-PREV-rest.VAI -3-SUBJ IC-PREV-sleep.VAI-3-SUBJ k\u00C3\u00AE-wayaw\u00C3\u00AEw m\u00C3\u00A2n \u00C3\u00AA-nitaw-\u00C3\u00A2tosk\u00C3\u00AAt kistik\u00C3\u00A2nihk. k\u00C3\u00AE- wayaw\u00C3\u00AE -w m\u00C3\u00A2na \u00C3\u00AA- nitaw-\u00C3\u00A2tosk\u00C3\u00AA -t kistik\u00C3\u00A2n -ihk PREV-go.out.VAI-3 HAB C1-DIR- work.VAI-3 field -LOC \u00E2\u0080\u0098..., then when he had rested and slept, he would still go out to go and work in the fields.\u00E2\u0080\u0099 (EM 28) None of the 6 speakers I have worked with control this process in elicitation contexts when eliciting these particular constructions: some use periphrastic constructions, while others, including the one whose data is cited above, simply use the unchanged form. 176 (70) a. \u00C3\u00AA-m\u00C3\u00A2toy\u00C3\u00A2n ayis \u00C3\u00AA-sipw\u00C3\u00AAht\u00C3\u00AAt nim\u00C3\u00A2ma CONJUNCT \u00C3\u00AA- m\u00C3\u00A2to -y\u00C3\u00A2n ayis(k) \u00C3\u00AA- sipw\u00C3\u00AAht\u00C3\u00AA -t ni- m\u00C3\u00A2ma C1-cry.VAI-1 for C1-leave.VAI -3 1- mom \u00E2\u0080\u0098I cried because my mom left.\u00E2\u0080\u0099 = (i) PRECEDE (leave, cry) = (ii) OVERLAP (leave, cry) b. \u00C3\u00AA-m\u00C3\u00A2toy\u00C3\u00A2n ayis \u00C3\u00AA-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAt nim\u00C3\u00A2ma CONJUNCT W/ k\u00C3\u00AE- \u00C3\u00AA- m\u00C3\u00A2to -y\u00C3\u00A2n ayis \u00C3\u00AA- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -t ni -m\u00C3\u00A2ma C1-cry.VAI-1 for C1-PREV-leave.VAI -3 1 -mom \u00E2\u0080\u0098I cried because my mom (had) left.\u00E2\u0080\u0099 = (i) PRECEDE (leave, cry) \u00E2\u0089\u00A0 (ii) OVERLAP (leave, cry) 4.4.1.3 C-command without precedence: k\u00C3\u00AE- is anaphoric The third configuration in which a relation between an anaphor and an antecedent can be established is when the antecedent follows the anaphor (precedence does not hold), but the antecedent is in a higher clause than the antecedent (c-command holds). This is represented in (71). (71) CP 5 CP antecedent 5 anaphor When k\u00C3\u00AE- is in an anaphoric clause, we then expect that, for any of the utterances we saw in the last section (where the subordinate clause followed its matrix clause), the order of the two clauses can be switched without altering the temporal relation between the two events. Again, this prediction is borne out in the data. For clauses that are introduced by propositional predicates (verbs of thinking, speaking and perception), placing the dependent clause before the matrix clause yields an utterance that is not particularly natural (cf. chapter 5), but the judgments on the temporal relation between the events are quite clear. I here give examples with w\u00C3\u00A2paht- \u00E2\u0080\u0098see.VTI\u00E2\u0080\u0099 (72a) and wihtamaw- \u00E2\u0080\u0098tell.VTA\u00E2\u0080\u0099 (72b). 177 (72) a. ? \u00C3\u00AA-k\u00C3\u00AE-mispohk w\u00C3\u00A2pahtam Jeff CONJUNCT W/ k\u00C3\u00AE- \u00C3\u00AA- k\u00C3\u00AE- mispon -k w\u00C3\u00A2paht-am J C1-PREV-snow.VII-0 see.VTI -3 J \u00E2\u0080\u0098Jeff saw that it had snowed.\u00E2\u0080\u0099 = (i) PRECEDE (snow, see) \u00E2\u0089\u00A0 (ii) OVERLAP (snow, see) b. ? \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEcisot niw\u00C3\u00AEhtam\u00C3\u00A2k Jeff \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00AEciso -t ni- wihtam -aw -ikw J C1-PREV-eat.VAI -3 1- tell.VTA -BEN-INV J \u00E2\u0080\u0098Jeff told me he had eaten.\u00E2\u0080\u0099 = (i) PRECEDE (eat, tell) \u00E2\u0089\u00A0 (ii) OVERLAP (eat, tell) The other subordinate clauses are found more commonly preceding the clause they are dependent on; the interpretation of k\u00C3\u00AE- is again insensitive to the change in ordering of the two clauses. A contrast between a relative clause that is unmarked vs. one that is marked with k\u00C3\u00AE- is given in (73): the insertion of k\u00C3\u00AE- reverses the available interpretations. (73) a. aw\u00C3\u00A2sis k\u00C3\u00A2-mow\u00C3\u00A2t cookie \u00C3\u00AA-m\u00C3\u00A2ci-m\u00C3\u00A2tot CONJUNCT aw\u00C3\u00A2sis k\u00C3\u00A2- mow -\u00C3\u00A2 -t cookie \u00C3\u00AA- m\u00C3\u00A2ci-m\u00C3\u00A2to -t child C2-eat.VTA-DIR-3 cookie C1-start-cry.VAI-3 \u00E2\u0080\u0098The child eating the cookie is starting to cry\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (cry, eat) = (ii) OVERLAP (cry, eat) b. aw\u00C3\u00A2sis k\u00C3\u00A2-k\u00C3\u00AE-mow\u00C3\u00A2t cookie \u00C3\u00AA-maci-m\u00C3\u00A2tot CONJUNCT W/ k\u00C3\u00AE- aw\u00C3\u00A2sis k\u00C3\u00A2- k\u00C3\u00AE- mow -\u00C3\u00A2 -t cookie \u00C3\u00AA- m\u00C3\u00A2ci-m\u00C3\u00A2to -t child C2-PREV-eat.VTA-DIR-3 cookie C1-start-cry.VAI-3 \u00E2\u0080\u0098The child that had been/was eating the cookie is starting to cry now\u00E2\u0080\u0099 = (i) PRECEDE (cry, eat) \u00E2\u0089\u00A0 (ii) OVERLAP (cry, eat) Likewise, we see that initial subjunctive clauses (74a) and initial reason clauses (74b) behave just like their non-initial counterparts: the superordinate clause acts as an antecedent. 178 (74) a. k\u00C3\u00AE-nikamoci, Clare k\u00C3\u00A2kik\u00C3\u00AA m\u00C3\u00A2tow CONJUNCT W/ k\u00C3\u00AE- k\u00C3\u00AE- nikamo -t -i C k\u00C3\u00A2kike m\u00C3\u00A2to -w PREV-sing.VAI-3-SUBJ C always cry.VAI-3 \u00E2\u0080\u0098When she\u00E2\u0080\u0099s done singing, Clare always cries.\u00E2\u0080\u0099 = (i) PRECEDE (sing, cry) \u00E2\u0089\u00A0 (ii) OVERLAP (sing, cry) b. ayis \u00C3\u00AA-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAt nim\u00C3\u00A2ma, \u00C3\u00AA-m\u00C3\u00A2toy\u00C3\u00A2n CONJUNCT W/ k\u00C3\u00AE- ayis \u00C3\u00AA- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -t ni- m\u00C3\u00A2ma \u00C3\u00AA- m\u00C3\u00A2to -y\u00C3\u00A2n for C1-PREV-leave.VAI-3 1- mom C1-cry.VAI-1 \u00E2\u0080\u0098Because my mother left I cried.\u00E2\u0080\u0099 = (i) PRECEDE (leave, cry) \u00E2\u0089\u00A0 (ii) OVERLAP (leave, cry) I have now established that if either precedence or c-command holds between a clause marked with k\u00C3\u00AE- and its potential antecedent, k\u00C3\u00AE- is anaphoric on a reference time given by the clause which serves as the antecedent. 4.4.1.4 No precedence, no c-command: k\u00C3\u00AE- is not anaphoric The configuration where an anaphor may not establish a relationship to a potential antecedent is when the antecedent follows the anaphor (i.e., precedence does not hold) and the anaphor is not in a subordinate clause relative to the antecedent (i.e., c-command does not hold either). This would be the case if the antecedent were in an embedded clause (75a), or if the anaphor and antecedent are in separate coordinated clauses (75b), or if there is simply a sequence of two independent clauses (75c). (75) a. * CP 6 anaphor CP 5 antecedent b. * CP 9 CP and CP 5 5 anaphor antecedent 179 c. * CP CP 5 5 anaphor antecedent The structures in (75) violate both the precedence condition and the c-command condition. The prediction is that the potential anaphor will fail to be dependent on the subsequent clause as the antecedent. The prediction for k\u00C3\u00AE- is that in an initial CONJUNCT clause, it cannot shift the event time of that clause with respect to a following clause it is not subordinated under. Recall that it is specifically the dependency between the potential anaphoric element and antecedent pair that is prohibited by these structures, and that we could have structures such as (77), where both elements are (co-)dependent on some other antecedent. (76) [I assume you recall that this course requires a term paperi.] Anyone can turn iti in to me now [who has WRITTEN their term paperi]. (adapted from Williams 1997: ex. 27) (77) [\u00E2\u0080\u00A6 term paper \u00E2\u0080\u00A6 ] [\u00E2\u0080\u00A6 it \u00E2\u0080\u00A6] [ \u00E2\u0080\u00A6 term paper \u00E2\u0080\u00A6 ] ANTECEDENT ANAPHOR ANAPHOR This is important for the current discussion because k\u00C3\u00AE- can always sequence the event time relative to some other topic time: k\u00C3\u00AE- anchors to utterance time in indexical INDEPENDENT clauses, and can anchor to a time of some other clause in anaphoric CONJUNCT clauses. There is also the complication of shifting reference time in narratives (cf. Hinrich 1984, Kamp & Rohrer 1983, Bittner 2008). Although the contexts in which shifting reference time is possible and/or obligatory have not thoroughly worked out even for English (e.g., the extent of the role of aspectual class of the two predicates), it seems that a similar sort of pattern is seen in Plains Cree, where the temporal order of two events, will, all else being equal, reflect the order in which they are presented. For example, in (78) the order of presentation is k\u00C3\u00AEw\u00C3\u00AA- \u00E2\u0080\u0098come.home\u00E2\u0080\u0099, m\u00C3\u00AEciso- \u00E2\u0080\u0098eat\u00E2\u0080\u0099, and the temporal interpretation is that the coming home preceded the eating. 180 (78) a. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff, \u00C3\u00AA-m\u00C3\u00AEcisoy\u00C3\u00A2hk CONJUNCT \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J \u00C3\u00AA- m\u00C3\u00AEciso -y\u00C3\u00A2n -k C1-come-go.home.VAI-3 J C1-eat.VAI -1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6Jeff came home and we ate.\u00E2\u0080\u0099 (we = speaker & Jeff) = PRECEDE (come home, eat) Taking these two factors into account, the GPAD predicts that in any of the configurations given above in (75), the k\u00C3\u00AE- will not affect the temporal relation between the clause it is in and the following clause at all. Precedence is not entirely ruled out (since it can be established via shifting reference time), but should be equally available regardless of whether k\u00C3\u00AE- is there or not. Further, an overlap reading should be possible, whereas it was not possible when the conditions on anaphora were met. This is exactly what we see. In the superordinate clause (79), the sequenced reading should be strange on semantic grounds, so if k\u00C3\u00AE- was coding temporal precedence, we would expect it to be infelicitous. But it is not: the utterance is fine, and codes two contemporaneous events. (79) Superordinate clauses: k\u00C3\u00AE- fails to be anaphoric k\u00C3\u00AE-w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-mispohk k\u00C3\u00AE- w\u00C3\u00A2pahtam-w J \u00C3\u00AA- mispon -k PREV-see.VTI -3 J C1-snow.VII-0 \u00E2\u0080\u0098Jeff had seen it snowed.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 PRECEDE (see, snow) = OVERLAP (see, snow) context: snow was falling when Jeff looked out the window With two coordinated clauses, the first coordinate and second coordinate are interpreted as overlapping in (80a); the addition of k\u00C3\u00AE- in (80b) does not allow the sequenced interpretation. (80) Coordinated clauses: k\u00C3\u00AE- fails to be anaphoric a. \u00C3\u00AA-kinosit Jack, \u00C3\u00AAkwa \u00C3\u00AA-tak\u00C3\u00A2hk\u00C3\u00A2p\u00C3\u00AAwit \u00C3\u00AA- kinosi -t J \u00C3\u00AAkwa \u00C3\u00AA- tak\u00C3\u00A2hk\u00C3\u00A2p\u00C3\u00AAwi -t C1-tall.VAI-3 J and C1-good.looking.man.VAI-3 \u00E2\u0080\u0098Jack is tall, and a good-looking man.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (tall, good looking) = (ii) OVERLAP (tall, good looking) 181 b.. \u00C3\u00AA-k\u00C3\u00AE-kinosit Jack, \u00C3\u00AAkwa \u00C3\u00AA-tak\u00C3\u00A2hk\u00C3\u00A2p\u00C3\u00AAwit.19 \u00C3\u00AA- k\u00C3\u00AE- kinosi -t J \u00C3\u00AAkwa \u00C3\u00AA- tak\u00C3\u00A2hk\u00C3\u00A2p\u00C3\u00AAwi -t C1-PREV-tall.VAI-3 J and C1-good.looking.man.VAI-3 \u00E2\u0080\u0098Jeff used to be tall, and he was handsome / a good-looking man.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (tall, good looking) = (ii) OVERLAP (tall, good looking) Finally, in (81) we have a sequence of two eventive predicates. Both the sequenced and overlap interpretations are available in (81a), and the addition of k\u00C3\u00AE- again does not eliminate the overlap interpretation. (81) Chained clauses: k\u00C3\u00AE- fails to be anaphoric a. \u00C3\u00AA-sipw\u00C3\u00AAht\u00C3\u00AAt nim\u00C3\u00A2ma, \u00C3\u00AA-m\u00C3\u00A2toy\u00C3\u00A2n \u00C3\u00AA- sipw\u00C3\u00AAht\u00C3\u00AA -t ni- m\u00C3\u00A2ma \u00C3\u00AA- m\u00C3\u00A2to -y\u00C3\u00A2n C1-leave.VAI -3 1- mom C1-cry.VAI -1 \u00E2\u0080\u0098My mother left, I cried.\u00E2\u0080\u0099 = (i) PRECEDE (leave, cry) = (ii) OVERLAP (leave, cry) b. \u00C3\u00AA-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAt nim\u00C3\u00A2ma, \u00C3\u00AA-m\u00C3\u00A2toy\u00C3\u00A2n \u00C3\u00AA- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -t ni- m\u00C3\u00A2ma \u00C3\u00AA- m\u00C3\u00A2to -y\u00C3\u00A2n C1-PREV-leave.VAI-3 1- mama C1-cry.VAI-1 \u00E2\u0080\u0098My mother left, I cried.\u00E2\u0080\u0099 = (i) PRECEDE (leave, cry) = (ii) OVERLAP (leave, cry) Here the temporal relations are being established apart from k\u00C3\u00AE-, and, importantly, they are not changed by adding k\u00C3\u00AE- to the initial clause. In all cases k\u00C3\u00AE- fails to be anaphoric, as predicted. 4.4.2 -yi is subject to c-command and precedence in CONJUNCT clauses The suffix -yi in Plains Cree is a subject-oriented reference-tracking marker which codes disjoint reference between the subject of its clause and some other clause (M\u00C3\u00BChlbauer 2007, 2008). For 19 See also 4.2.1.2 for parallel examples with non-stative predicates. 182 example, in (82), the subject of the singing clause (Solveiga) is different from the subject from the knowing clause (Jeff), and is marked with suffix -yi.20 (82) Jeff kisk\u00C3\u00AAyihtam Solveiga \u00C3\u00AA-nikamoyit. CONJUNCT W/ -yi J. kisk\u00C3\u00AAyihtam -w S \u00C3\u00AA- nikamo -yi -t J know.VTI -3 S C1-sing.VAI-DS -3 \u00E2\u0080\u0098Jeff knows that Solveiga is singing.\u00E2\u0080\u0099 The syntax of -yi can be represented as in (83).21 The subject associated with -yi is disjoint from some other element (cf. also Saxon 1986 on disjoint anaphora). (83) CP 2 2IP 2 (Subjx, \u00E2\u0089\u00A0 y) 2vP -yi 5 (M\u00C3\u00BChlbauer 2008) The presence of -yi thus always requires that the clause be interpreted with respect to another clause: it indirectly establishes a cross-clausal dependency22. The current analysis of clause-typing predicts that the cross-clausal dependency will be subject to c-command and/or precedence in anaphoric clauses (\u00C2\u00A74.3.2.1). The dependency between -yi and its antecedent is claimed by M\u00C3\u00BChlbauer (2007, 2008) to be sensitive to precedence and c-command, and the current analysis provides a systematic consideration of this claim, and situates it within a larger picture of anaphoric dependencies in Plains Cree clauses. The following table presents, for each condition on anaphora, the percentages and total number of occurrences of a -yi marked clause in a two-and-a-half hour narrative (Minde 1997). 20 Note that \u00E2\u0080\u0098Solveiga\u00E2\u0080\u0099 is not marked with obviation (i.e., the suffix -a) in this sentence; obviation is commonly dropped on common names in elicitation contexts. See Cook & M\u00C3\u00BChlbauer (2006) for discussion. 21 The traditional analysis of -yi is that it marks obviative agreement. For discussion and evidence as to why this analysis cannot be correct, see M\u00C3\u00BChlbauer (2007); since it is not of direct import here, I do not replicate the arguments. However, it is important for readers to know that -yi only marks disjoint reference over third persons: it will never, for example, be marked if the subject of the antecedent clause is a speech act participant. 22 In fact, cross-linguistically, languages which exhibit switch-reference marking also exhibit extensive cosubordination (i.e., clause-chaining), and the switch-reference marking is restricted to cosubordinated clauses (see, e.g., Stirling 1993). 183 Conditions on anaphora -yi (Minde 1997) Attested? % (tokens) C-command and precedence \u00E2\u009C\u0094 55 (59) Precedence \u00E2\u009C\u0094 27 (29) C-command \u00E2\u009C\u0094 14 (15) No c-command, no precedence \u00E2\u009C\u0096 0 (0) antecedent [anaphor [ant] ] \u00E2\u009C\u0094 4 (4) Table 4.9. Distribution of -yi by anaphoric configuration While each of the three possible antecedent-anaphor relations are instantiated in this narrative, the relation that is predicted to be impossible is unattested. I here consider each condition in turn, supplementing the textual data with elicitation data. 4.4.2.1 C-command and precedence: -yi is licensed If the anaphor both follows and is subordinated to its antecedent, it fulfills both conditions on anaphora. This is the most common configuration for a -yi clause and its antecedent clause. (84) CP 5 antecedent CP 5 anaphor (-yi) Examples of this pattern with three different anaphoric CONJUNCT clauses are given: (85a) is a temporal modification clause introduced by the k\u00C3\u00A2- complementizer (which can only occur when it is embedded with respect to a higher clause). (85b) is a relative clause introduced by the \u00C3\u00AA- complementizer, it is an inanimate intransitive verb modifying \u00C3\u00AAkotowahk \u00E2\u0080\u0098that kind\u00E2\u0080\u0099 (which is associated with the verb\u00E2\u0080\u0099s subject position). (85c) is a purpose clause introduced by a clause that lacks an overt complementizer. (85) a. [\u00C3\u00AA-k\u00C3\u00AE-CONJ [k\u00C3\u00A2-CONJ-yi ] ] ..., \u00C3\u00AA-k\u00C3\u00AE-miciminamaw\u00C3\u00A2t misatimwa aya k\u00C3\u00A2-nakay\u00C3\u00A2h\u00C3\u00A2yit, ... \u00C3\u00AA- k\u00C3\u00AE- miciminamaw -\u00C3\u00A2 -t misatimw -a aya k\u00C3\u00A2- nakay\u00C3\u00A2h -\u00C3\u00A2 -yi -t C1-PREV-hold.for.VTA -DIR-3 horse -OBV CONN C2- break.VTA-DIR-DS-3 \u00E2\u0080\u0098She even held the horses for him, she told me, when he broke them, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 66) 184 b. [\u00C3\u00AA-k\u00C3\u00AE-CONJ [\u00C3\u00AA-CONJ-yi] ] ... \u00C3\u00AA-k\u00C3\u00AE-kikiskahk m\u00C3\u00A2n \u00C3\u00AAkotowahk \u00C3\u00AA-m\u00C3\u00AEkisiwiyiki. \u00C3\u00AA- k\u00C3\u00AE- kikiskam -k m\u00C3\u00A2na \u00C3\u00AAkotowahk \u00C3\u00AA- m\u00C3\u00AEkisiwi -yi -k -i C1-PREV-wear.VTI -0 usually that.kind C1-beaded.VII-DS-0-PL \u00E2\u0080\u0098\u00E2\u0080\u00A6 he used to wear beaded ones.\u00E2\u0080\u0099 (EM 68) (lit: \u00E2\u0080\u0098...he used to wear [shoes] that were beaded.\u00E2\u0080\u0099) c. [k\u00C3\u00A2-CONJ [CONJ-yi] ] ..., k\u00C3\u00A2-misipot\u00C3\u00A2t aya pahk\u00C3\u00AAkin, ka-y\u00C3\u00B4sk\u00C3\u00A2yik, ... k\u00C3\u00A2- misipot -\u00C3\u00A2 -t aya pahk\u00C3\u00AAkin ka- y\u00C3\u00B4sk\u00C3\u00A2 -yi -k C2-roll.VTA-DIR-3 CONN hide IRR-soft.VII-DS -0 \u00E2\u0080\u0098..., when she rolled a hide over the blade so that it would be soft, ...\u00E2\u0080\u0099 (EM 18) In all three cases the non-initial, subordinate clause is marked with -yi. 4.4.2.2 Precedence without c-command: -yi is licensed If precedence holds, an antecedent-anaphor relation may be established regardless of the subordinate relation between the two clauses. There are thus a number of precedence-governed clausal relations where we expect -yi to occur. First, we expect -yi to be possible in the second conjunct of a coordinated clause23. (86) CP 9 CP AND CP 5 5 antecedent anaphor (-yi) The overtly coordinated clauses in (87) fit into this class. Here the subject in the first clause (\u00C3\u00A2y\u00C3\u00A2w- \u00E2\u0080\u0098have\u00E2\u0080\u0099) is Dan Minde, and the subject in the second clause (\u00C3\u00A2y\u00C3\u00A2w- \u00E2\u0080\u0098have\u00E2\u0080\u0099) is Sam Minde. The two clauses are conjoined with m\u00C3\u00A2ka \u00E2\u0080\u0098but\u00E2\u0080\u0099, and the disjoint subject marker \u00E2\u0080\u0093yi is in the second conjunct. 23 I do not at present have a full analysis of coordination. On the one hand, in anaphoric clauses, the second coordinate can host anaphoric elements like \u00E2\u0080\u0093yi. On the other hand, indexical clauses can also be coordinated, although there are restrictions on coordination (e.g., an overt coordinator such as \u00C3\u00AAkwa \u00E2\u0080\u0098and/then\u00E2\u0080\u0099 is necessary, and such coordinators may function at a discourse level, rather than simply inter-clausally; cf. Ogg 1991). Further, even when indexical clauses are coordinated, they cannot host anaphoric elements. 185 (87) [CP IND ] m\u00C3\u00A2ka [CP \u00C3\u00AA-CONJ-yi ] ..., m\u00C3\u00B4ya mihc\u00C3\u00AAt oht-\u00C3\u00A2-~ ohc-\u00C3\u00A2y\u00C3\u00A2w\u00C3\u00AAw wiya nisis awa Dan Minde, m\u00C3\u00B4ya mihc\u00C3\u00AAt ohc- \u00C3\u00A2y\u00C3\u00A2w -\u00C3\u00AA -w wiya nisis awa DM NEG many PREV-have.VTA-DIR-3 EMPH father-in-law DEM.AN DM \u00E2\u0080\u0098..., my father-in-law Dan Minde, he did not have many, m\u00C3\u00A2ka wiy \u00C3\u00B4-~ os\u00C3\u00AEma, mihc\u00C3\u00AAt aya pisiskiwa, mostoswa \u00C3\u00AAkwa misatimwa m\u00C3\u00A2ka wiya o-s\u00C3\u00AEm -a mihc\u00C3\u00AAt aya pisiskiw -a mostosw -a \u00C3\u00AAkwa misatimw -a BUT EMPH 3-SIBLING-OBV many CONN animal -OBV cow -OBV and horse -OBV \u00C3\u00AA-k\u00C3\u00AE-ay\u00C3\u00A2w\u00C3\u00A2yit. \u00C3\u00AA- k\u00C3\u00AE- ay\u00C3\u00A2w -\u00C3\u00A2 -yi -t C1-PREV-have.VTA-DIR-DS-3 but his younger brother [Sam Minde] had many animals, cattle and horses.\u00E2\u0080\u0099 (EM 45) Two clauses which are coordinate with respect to each other and both subordinate to some higher predicate show the same pattern. In (88) the subject of the first conjunct (waw\u00C3\u00A2n\u00C3\u00AAyiht- \u00E2\u0080\u0098worry\u00E2\u0080\u0099) is k\u00C3\u00AAht\u00C3\u00AA-ayak \u00E2\u0080\u0098the old people\u00E2\u0080\u0099; and subject of the second conjunct (n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA- \u00E2\u0080\u0098hungry\u00E2\u0080\u0099) is \u00C3\u00B4taw\u00C3\u00A2simisiw\u00C3\u00A2wa \u00E2\u0080\u0098their children\u00E2\u0080\u0099. The subjects are disjoint, and the second conjunct contains -yi. (88) [\u00C3\u00AA-CONJ [ka-CONJ \u00C3\u00AAkwa ka-CONJ-yi ] ] ... k\u00C3\u00AAht\u00C3\u00AA-ayak \u00C3\u00AA-~ \u00C3\u00AA-kway\u00C3\u00A2tastam\u00C3\u00A2socik m\u00C3\u00AEciwin, k\u00C3\u00AAht\u00C3\u00AA-aya -k \u00C3\u00AA- kway\u00C3\u00A2tastamaw -iso -t -k m\u00C3\u00AEci -win old-people-PL C1-get.food.VTA -REFLX-3-PL eat.VTI-NOM \u00E2\u0080\u0098... [the old people would get] food ready for themselves \u00C3\u00AAk\u00C3\u00A2 ka-waw\u00C3\u00A2n\u00C3\u00AAyihtahkik; \u00C3\u00AAk\u00C3\u00A2 ka- waw\u00C3\u00A2n\u00C3\u00AAyihtam -k -k NEG IRR-worry.VTI -0 -PL so that they would not have to worry about it; \u00C3\u00AAkwa m\u00C3\u00AEn \u00C3\u00B4taw\u00C3\u00A2simisiw\u00C3\u00A2wa \u00C3\u00AAk\u00C3\u00A2 ka-waw-~ ka-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAyit. \u00C3\u00AAkwa m\u00C3\u00AEna o(t)- aw\u00C3\u00A2s -im-is-w\u00C3\u00A2w-a \u00C3\u00AAk\u00C3\u00A2 ka-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA-yi-t and also 3- child -DJ-DIM-3.PL-OBV NEG IRR-hungry.VAI-DEP-3 and so that their children would not have to go hungry.\u00E2\u0080\u0099 (EM 17) Two clauses do not have to be overtly coordinated (e.g., with \u00C3\u00AAkwa) in order for one to be in a non-subordinative relation to the other: they may simply occur next to each other (what in this thesis I am calling clause-chains). 186 (89) CP CP 5 5 antecedent anaphor If the two clauses have different subjects, the second one may be marked with -yi (as in 90).24 (90) [IND \u00C3\u00AA-CONJ-yi ] \u00C3\u00AAkwa k\u00C3\u00AE-omisiw, \u00E2\u0080\u0098Sophie\u00E2\u0080\u0099 \u00C3\u00AA-isiy\u00C3\u00AEhk\u00C3\u00A2soyit; \u00C3\u00AAkwa k\u00C3\u00AE- omisi -w S \u00C3\u00AA- isiy\u00C3\u00AEhk\u00C3\u00A2so -yi -t and PREV-have.older.sister.VAI-3 S C1-thus.be.called.VAI-DS -3 \u00E2\u0080\u0098And he [my husband] had an older sister, Sophie was her name;\u00E2\u0080\u0099 (EM 31) Finally, we predict that the anaphoric link may be superordinate to its antecedent, as long as the antecedent precedes it. This particular configuration was not attested in the textual data, but was confirmed to be grammatical in elicitation. (91) CP 5 CP anaphor 5 antecedent (92) [k\u00C3\u00A2-CONJ ] \u00C3\u00AA-CONJ-yi ] k\u00C3\u00A2-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAt aw\u00C3\u00A2sis, \u00C3\u00AA-at\u00C3\u00A2w\u00C3\u00AAyit iskw\u00C3\u00AAwa m\u00C3\u00AEcisowin k\u00C3\u00A2- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -t aw\u00C3\u00A2sis \u00C3\u00AA- at\u00C3\u00A2w\u00C3\u00AA -yi -t iskw\u00C3\u00AAw -a m\u00C3\u00AEciso -win C2-hungry.VAI -3 child C1-buy.VAI-DS -3 woman-OBV eat.VAI-NOM \u00E2\u0080\u0098When the child is hungry, the woman buys food.\u00E2\u0080\u0099 4.4.2.3 C-command without precedence: -yi is licensed The third configuration where we predict that an anaphoric relation may be established, is one which violates precedence, but respects c-command, as in (93). 24 Note that -yi marking is not obligatory under the same conditions that k\u00C3\u00AE- is. This is discussed in more detail in chapter 5. For the present discussion, I am concerned with where -yi is possible and where it is impossible. 187 (93) CP 5 CP antecedent 5 anaphor These structures are also attested for -yi, as illustrated in (94). In (94) we see overt subordinators such as iyikohk \u00E2\u0080\u0098instead\u00E2\u0080\u0099 and \u00C3\u00A2ta \u00E2\u0080\u0098although\u00E2\u0080\u0099 in clauses that are morpho- syntactically coded as subordinate (e.g., simple CONJUNCT in 94a; k\u00C3\u00A2- complementizer in 94b). The antecedent clause follows the subordinate clause; since c-command is respected, the anaphoric link can be established. (94) a. [CP [CP iyikohk ka-CONJ-yi ] \u00C3\u00AA-k\u00C3\u00AE-t\u00C3\u00B4tahkik ] (i) iyikohk ka-misiwan\u00C3\u00A2taniyik anima wiy\u00C3\u00A2s, iyikohk ka- misiwan\u00C3\u00A2tan -yi -k anima wiy\u00C3\u00A2s DEG IRR-be.destroyed.VII-DS-0 DEM.INAN meat \u00E2\u0080\u0098Instead of the meat being destroyed, (ii) \u00C3\u00AAkos \u00C3\u00A2nima m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-t\u00C3\u00B4tahkik, \u00E2\u0080\u00A6 \u00C3\u00AAkosi anima m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- t\u00C3\u00B4tam -k -k TOP DEM.INAN usually C1-PREV-do.VTI -0 -PL they used to do [these things], ...\u00E2\u0080\u0099 (EM 57) b. [CP [CP \u00C3\u00A2ta k\u00C3\u00A2-CONJ-yi ] k\u00C3\u00AE-w\u00C3\u00AEcih\u00C3\u00AAwak ] (i) \u00C3\u00A2ta wiy \u00C3\u00AAtokw\u00C3\u00AA m\u00C3\u00A2-~, ita k-\u00C3\u00A2yimaniyik, \u00C3\u00A2ta wiya \u00C3\u00AAtokw\u00C3\u00AA ita k\u00C3\u00A2- ayiman -yi -k although EMPH EVID LOC C2- be.hard.VII-DS -0 \u00E2\u0080\u0098Where it was hard, though, (ii) n\u00C3\u00A2p\u00C3\u00AAwak m\u00C3\u00AEna m\u00C3\u00A2na k\u00C3\u00AE-w\u00C3\u00AEcih\u00C3\u00AAwak w\u00C3\u00AEwiw\u00C3\u00A2wa, \u00E2\u0080\u00A6 n\u00C3\u00A2p\u00C3\u00AAw -ak m\u00C3\u00AEna m\u00C3\u00A2na k\u00C3\u00AE- w\u00C3\u00AEcih -\u00C3\u00AA -w -ak w\u00C3\u00AEwi -w\u00C3\u00A2w -a man -PL also usually PREV-help.VTA-DIR-3 -PL wife -3.PL -OBV I guess the men used to help their wives, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 17) Likewise, in (95) the subordinate k\u00C3\u00A2-clause modifies the following \u00C3\u00AA-clause. The -yi marks that the subject of the modifying clause is disjoint from the subject of the following, unmarked clause. 188 (95) [CP [CP k\u00C3\u00A2-CONJ-yi ] \u00C3\u00AA-CONJ ] k\u00C3\u00A2-p\u00C3\u00AEhtikw\u00C3\u00AAyit awiya, \u00C3\u00AAkos \u00C3\u00AA-asam\u00C3\u00A2t k\u00C3\u00AEkway, ... k\u00C3\u00A2- p\u00C3\u00AEhtikw\u00C3\u00AA -yi -t awiy -a, \u00C3\u00AAkosi \u00C3\u00AA- asam -\u00C3\u00A2 -t k\u00C3\u00AEkway C2-enter.VAI-DS-3 someone-OBV, thus C1-feed.VTA-DIR-3 something \u00E2\u0080\u0098When someone came to visit he fed them something right away, ...\u00E2\u0080\u0099 (AA 1.7) 4.4.2.4 No c-command, no precedence: -yi is not licensed When neither the precedence condition nor the c-command condition hold, we expect that an anaphoric-antecedent relation cannot be licensed. For example, if the clause with the potential anaphoric link is superordinated to the other clause, and is not preceded by it, we predict that the anaphoric-antecedent relation is undefined; thus -yi will be ungrammatical. (96) a. * CP 6 anaphor (-yi) CP 5 antecedent This prediction is accurate; in (97), the superordinate \u00C3\u00AA-clause has -yi, and the subordinate k\u00C3\u00A2- clause follows it. The utterance is ungrammatical. (97) a. * [CP \u00C3\u00AA-CONJ-yi [CP k\u00C3\u00A2-CONJ ] ] b. * \u00C3\u00AA-at\u00C3\u00A2w\u00C3\u00AAyit iskw\u00C3\u00AAwa m\u00C3\u00AEcisowin, k\u00C3\u00A2-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAt aw\u00C3\u00A2sis \u00C3\u00AA- \u00C3\u00A2t\u00C3\u00A2w\u00C3\u00AA -yi -t iskw\u00C3\u00AAw -a m\u00C3\u00AEciso -win k\u00C3\u00A2- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -t aw\u00C3\u00A2sis C1-buy.VAI-DS-3 woman-OBV eat.VAI-NOM C2- hungry.VAI -3 child --- (intended: \u00E2\u0080\u0098...the woman bought food when her child was hungry\u00E2\u0080\u0099) We also predict that the first conjunct in a coordinated structure will not be able to host -yi. 189 (98) * CP 9 CP and CP 5 5 anaphor antecedent This prediction is also borne out. In (99), two clauses are overtly coordinated with \u00C3\u00AAkwa. We observe that the second coordinate may bear -yi marking; However, the first coordinate, crucially, is not allowed to have -yi marking. (99) a. Jeff \u00C3\u00AA-nikamot \u00C3\u00AAkwa Clarewa \u00C3\u00AA-nimihitoyit J \u00C3\u00AA- nikamo -t \u00C3\u00AAkwa C -wa \u00C3\u00AA- nimihito -yi -t J C1-sing.VAI-3 and C-OBV C1-dance.VAI -DS -3 \u00E2\u0080\u0098Jeff was singing and Clare was dancing.\u00E2\u0080\u0099 b. * Jeffa \u00C3\u00AA-nikamoyit \u00C3\u00AAkwa Clare \u00C3\u00AA-nimihitot J -a \u00C3\u00AA- nikamo -yi -t \u00C3\u00AAkwa C \u00C3\u00AA- nimihito -t J-OBV C1-sing.VAI-DS-3 and C C1-dance.VAI-3 --- We also find that the configurations which violate both precedence and c-command are unattested in running speech. In the final row of table 4.9 above, I showed that there were four examples where a configuration similar to that in (97) arose: the superordinate clause is marked with -yi, and the next clause it introduces (either subordinated or coordinated) is unmarked. If the superordinate clause were anaphoric on the subordinate clause, these examples would run counter to the precedence and c-command conditions on antecedent-licensing. However, a closer inspection shows that these are all cases where the anaphoric (i.e., the - yi marked) clause is anaphoric on a previous clause, and the unmarked subordinate clause is also dependent on that previous clause: they have the structure in (100): (100) ANTECEDENT [ ANAPHOR [ ANAPHOR ] ] For example, (101b) contains two clauses that are simply sequenced with respect to each other: \u00C3\u00AA-ahkosiyit (\u00E2\u0080\u0098[his father] was sick, with -yi) and \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEcih\u00C3\u00A2t (\u00E2\u0080\u0098he was helping him\u00E2\u0080\u0099, without -yi). At first glance, this looks problematic for the anaphoric account. 190 (101) a. * CP CP 5 5 anaphor antecedent b. \u00C3\u00B4ht\u00C3\u00A2wiya \u00C3\u00AA-\u00C3\u00A2hkosiyit \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEcih\u00C3\u00A2t. o- oht\u00C3\u00A2wiy -a \u00C3\u00AA- \u00C3\u00A2hkosi -yi -t \u00C3\u00AA- k\u00C3\u00AE- w\u00C3\u00AEcih -\u00C3\u00A2 -t 3- father -OBV C1-sick.VAI-DS-3 C1-PREV-help.VTA-DIR-3 \u00E2\u0080\u0098his father was ill and he was helping him.\u00E2\u0080\u0099 (EM 27) However, if we look at the immediately preceding sequence of clauses, we see that the speaker has been talking about her husband, and the things her husband did, and the way he used to work. In particular, the initial clause given in (103) is \u00C3\u00AA-\u00C3\u00A2cimostawit \u00E2\u0080\u0098he told me\u00E2\u0080\u0099, where her husband is the subject of the clause (the overt noun occurred earlier in the discourse). The second and third clauses nistos\u00C3\u00A2p \u00C3\u00AA-itahtopiponw\u00C3\u00AAt \u00E2\u0080\u0098he was 13 years old\u00E2\u0080\u0099 and k\u00C3\u00A2-k\u00C3\u00AE-\u00C3\u00A2tosk\u00C3\u00AAt \u00E2\u0080\u0098he worked\u00E2\u0080\u0099 have the same subject. Crucially, the \u00C3\u00AA-ahkosiyit \u00E2\u0080\u0098he was sick\u00E2\u0080\u0099 clause is in a non-initial position: it is the fourth clause and -yi is marking disjoint subjecthood relative to the first three clauses. The final clause \u00C3\u00AA-w\u00C3\u00AEcih\u00C3\u00A2t \u00E2\u0080\u0098he helped him\u00E2\u0080\u0099 is unmarked, since it is simply a return to the subject of the initial clauses. (102) (i) p\u00C3\u00AAyakw\u00C3\u00A2w \u00C3\u00AA-\u00C3\u00A2cimostawit, p\u00C3\u00AAyakw\u00C3\u00A2w \u00C3\u00AA- \u00C3\u00A2cimostaw -it once C1-tell.VTA -1>3 Once he told me the story of when he had begun, (ii) nistos\u00C3\u00A2p \u00C3\u00AA-itahtopiponw\u00C3\u00AAt, nistos\u00C3\u00A2p \u00C3\u00AA- itahtopiponw\u00C3\u00AA -t thirteen C1-be.thus.many.winters.VAI-3 at the age of thirteen, (iii) k\u00C3\u00A2-k\u00C3\u00AE-m\u00C3\u00A2c-\u00C3\u00A2tosk\u00C3\u00AAt kistik\u00C3\u00A2nihk; k\u00C3\u00A2- k\u00C3\u00AE- m\u00C3\u00A2ci- \u00C3\u00A2tosk\u00C3\u00AA -t kistik\u00C3\u00A2n -ihk C2-PREV-start- work.VAI-3 field -LOC to work in the fields; (iv) \u00C3\u00B4ht\u00C3\u00A2wiya \u00C3\u00AA-\u00C3\u00A2hkosiyit o- oht\u00C3\u00A2wiy -a \u00C3\u00AA- \u00C3\u00A2hkosi -yi -t 3- father -OBV C1-sick.VAI-DS-3 his father was ill (v) \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEcih\u00C3\u00A2t. \u00C3\u00AA- k\u00C3\u00AE- w\u00C3\u00AEcih -\u00C3\u00A2 -t C1-PREV-help.VTA-DIR-3 and he was helping him.\u00E2\u0080\u0099 (EM 27) 191 The structure of an utterance like (103) would thus be something like (104). In particular, CPi functions as the antecedent for both CPiv and CPv, rather than there being an anaphor-antecedent relation between CPiv and CPv. (103) CPi Civ CPv 6 6 6 \u00C3\u00AA-\u00C3\u00A2cimostawit CPiii \u00C3\u00AA-ahkosiyit \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEcih\u00C3\u00A2t 6 CPii k\u00C3\u00A2-k\u00C3\u00AE-m\u00C3\u00A2ci-\u00C3\u00A2tosk\u00C3\u00AAt 6 \u00C3\u00AA-itahtopiponw\u00C3\u00AAt The other cases where at first glance the conditions on antecedent-licensing appear to be contradicted all turn out to be like the one just discussed: the subject of the putative antecedent clause is the same as the subject in a clause that precedes the anaphoric clause, and in all cases there is no overt nominal on the putative antecedent clause. Thus the potentially contradictory examples turn out to be further confirmation that -yi in CONJUNCT clauses is behaving like anaphors in other languages. 4.5 The cross-linguistic typology of anaphoric clauses Just like pronominal anaphora can occur in matrix and embedded clauses, anaphoric clauses crosscut the traditional syntactic division between matrix (unembedded) and subordinate (embedded) clauses. As we have seen, an anaphoric clause may be subordinated to another clause, but it may also be a matrix clause and licensed by precedence or context. Indexical clauses, on the other hand, cannot be embedded. This gives us the typology for clauses in table 4.10. Matrix Embedded Indexical \u00E2\u009C\u0094 \u00E2\u009C\u0096 Anaphoric \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 4.10. Anaphoricity vs. embedding 192 We have already seen that in Plains Cree, where I have claimed that INDEPENDENT order clauses are indexical, and CONJUNCT clauses are anaphoric, the former are restricted to matrix clauses, but the latter occur in both contexts, as expected. In this section, I compare Plains Cree\u00E2\u0080\u0099s clause- typing system to two other systems. First, I look at how anaphoric clauses relate to \u00E2\u0080\u0098clause-chaining\u00E2\u0080\u0099, a phenomena that is pervasive in the areal region of New Guinea, and also present in many North American languages. This shows that the clause-typing split in Plains Cree is not specific to the language: in order to talk about clause-chains, the same distinction between \u00E2\u0080\u009Cdependency\u00E2\u0080\u009D (cf. Foley & Van Valin 1984, what I am calling anaphoricity) and embedding must be made. Second, I look at how the anaphoric/indexical split maps onto English clause-typing. In particular, since an English matrix clause is not morpho-syntactically marked as indexical or anaphoric, we expect it to occur in both contexts. I show that we see cases of English matrix anaphoric clauses in the phenomena known as \u00E2\u0080\u0098modal subordination\u00E2\u0080\u0099 (cf. Roberts 1989). 4.5.1 Chained clauses are anaphoric clauses In their work on the typology of inter-clausal relations, Foley & Van Valin (1984) recognize that: \u00E2\u0080\u00A6dependence is not equivalent to embeddedness. That is, whether a clause is dependent in some way upon another clause is independent of whether it is embedded as an argument of another clause. As we will see, many languages possess constructions in which one unit is dependent upon another and yet is clearly not embedded in it. (Foley & Van Valin 1984:243) Foley & Van Valin use the term cosubordinate for clauses which are dependent but not embedded. Cosubordination is often referred to as clause-chaining; it is a pervasive feature of many languages of the Pacific and North America (cf. Stirling 1993). Plains Cree\u00E2\u0080\u0099s unembedded anaphoric clauses share a number of properties with cosubordinate clauses. In particular, they share the same kind of dependencies (here I focus on temporal dependencies); they pattern together with respect to the division between coordination and subordination; and they both have a fixed order relative to the antecedent clause. 193 4.5.1.1 The significance of asymmetric marking One of the tests by which coordination is distinguished from subordination is the morpho- syntactic locus test (Zwicky 1985, Kazenin & Testelets 2004). According to this test, the morpho-syntactic realization of a dependency between two constituents and the superordinate structure will depend on the relation between these two constituents. If the constituents are coordinated, then the dependency must be marked on both coordinates; but if they are not (traditionally, if one is subordinate to the other), the dependency will only be marked once. This is used, for example, to explain the double-marking of the possessive marker in (104a), which is coordinated, contrasted with the obligatory single-marking in (104b) which is subordinate. (104) a. the king\u00E2\u0080\u0099s and the queen\u00E2\u0080\u0099s palace b. the king(*\u00E2\u0080\u0099s) of England\u00E2\u0080\u0099s palace (from Kazenin & Testelets 2004) In languages which are considered to have clause-chaining, this diagnostic differentiates clause- chains from a sequence of coordinated clauses. Clause-chains are headed by a clause which is marked for a number of features relating the proposition to the speech act (e.g., evidential value, illocutionary force, polarity, and/or tense). The other clauses in the chain are not marked for these features, but only for same-subject or different-subject morphology relative to the marked clause. For example, in Amele (Papuan; Papua New Guinea), the marked clause is marked for what Stirling (1993) calls \u00E2\u0080\u009CRemote Past\u00E2\u0080\u009D temporal marking, but other clauses in a clause chain are not so marked. (105) Ho bu-busal-en dana age qo-in pig Sim-run_out-3sg_DS man 3Pl hit-3Pl-RemP \u00E2\u0080\u0098As the pig ran out the men killed it.\u00E2\u0080\u0099 (Stirling 1993: 203) For Amele, this asymmetry in marking is one of the diagnostics for identifying clause-chains, since it indicates, according to the morpho-syntactic locus test, that the two clauses cannot be analyzed as coordinate clauses. This means that the asymmetric marking can be used as a diagnostic for the relation between two clauses. If the dependent clauses are acting like coordinated clauses, they should require symmetric marking; conversely, if it does not require symmetric marking, then it is not a coordinated clause. 194 This is the asymmetry we observe for unembedded anaphoric CONJUNCT clauses. In the sequence in (106), the preverb k\u00C3\u00AE- is a temporal sequencer which sequences the reference time relative to some evaluation time. Here, k\u00C3\u00AE- is only marked on the initial clause; adding k\u00C3\u00AE- to the other clauses makes the utterance ungrammatical. (106) a. \u00C3\u00AAkwa mistahi m\u00C3\u00A2n \u00C3\u00A2ya, \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, ANTECEDENT \u00C3\u00AAkwa mistahi m\u00C3\u00A2na aya \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n-k and much usually CONN C1-PREV-go.about.VAI-1 -PL \u00E2\u0080\u0098And we used to go around a lot, \u00C3\u00AA-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw ANAPHORIC \u00C3\u00AA- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n-k \u00C3\u00A2skaw C1-go.VTA-DIR-1 -PL sometimes sometimes going along with her \u00C3\u00AA-~ \u00C3\u00AA-pap\u00C3\u00A2mi-mawisot, ANAPHORIC \u00C3\u00AA- pap\u00C3\u00A2mi-mawiso -t C1-go- pick.berries.VAI-3 as she went about berry-picking, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 17) b. * \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw \u00C3\u00AA-~ \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2mi-mawisot \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n-k \u00C3\u00AA- k\u00C3\u00AE- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n-k \u00C3\u00A2skaw C1-PREV-go.about.VAI-1 -PL C1-PREV-go.VTA-DIR-1 -PL sometimes \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2mi-mawiso -t C1-PREV-go- pickberry.VAI-3 --- comment: there\u00E2\u0080\u0099s too many k\u00C3\u00AE-\u00E2\u0080\u0099s \u00E2\u0080\u00A6 It\u00E2\u0080\u0099s worse to put them in, and I wouldn\u00E2\u0080\u0099t The asymmetric marking in clause chains parallels the asymmatric marking we see in subordinate clauses \u00E2\u0080\u0093 only one marking is necessary, as in (107). 195 (107) a. k\u00C3\u00AE-w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-mispohk k\u00C3\u00AE- w\u00C3\u00A2pahtam -w J \u00C3\u00AA- mispon -k PREV-see.VTI -3 J C1-snow.VII-0 \u00E2\u0080\u0098Jeff saw that it snowed.\u00E2\u0080\u0099 b. * k\u00C3\u00AE-w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-k\u00C3\u00AE-mispohk k\u00C3\u00AE- w\u00C3\u00A2pahtam -w J \u00C3\u00AA- k\u00C3\u00AE- mispon -k PREV-see.VTI -3 J C1-PREV-snow.VII-0 --- (intended: \u00E2\u0080\u0098Jeff saw that it snowed.\u00E2\u0080\u0099) By contrast, as expected, two overtly coordinated clauses require matching marking. For example, in Tonkawa, one clause can get its illocutionary force from a following clause (the dependent clause\u00E2\u0080\u0099s dependency showing up in the switch-reference marking). If the illocutionary force (morphologically realized as the suffix -w) is marked on both clauses, there is a concomitant addition of an overt coordinator \u00CA\u0094e:-ta (108). (108) a. Asymmetric marking = no coordinator Tekeke\u00CA\u0094e:k \u00C5\u00A1\u00CA\u0094a:pa-ta ke-ya\u00C5\u00A1e-w. in.that.bush hide-SAME 1sgU-watch-IMP \u00E2\u0080\u0098Hide in that bush and watch me.\u00E2\u0080\u0099 b. Symmetric marking = coordinator (\u00CA\u0094e:-ta) Tekeke\u00CA\u0094e:k \u00C5\u00A1\u00CA\u0094a:pa-w \u00CA\u0094e:-ta ke-ya\u00C5\u00A1e-w. in.that.bush hide-IMP and-SAME 1sgU-watch-IMP \u00E2\u0080\u0098Hide in that bush and watch me.\u00E2\u0080\u0099 (Hoijer 1949, in Foley & Van Valin 1984:258) Plains Cree behaves like Tonkawa. If two clauses are coordinated with \u00C3\u00AAkwa \u00E2\u0080\u0098and/then\u00E2\u0080\u0099 and both clauses have the same (presence or absence of) marking, the coordination is fine (as evidenced by the fact that the simultaneous reading can occur) (109a-b). 196 (109) a. Jane \u00C3\u00AA-atosk\u00C3\u00AAt \u00C3\u00AAkwa \u00C3\u00AA-m\u00C3\u00A2tot J \u00C3\u00AA- atosk\u00C3\u00AA -t \u00C3\u00AAkwa \u00C3\u00AA- m\u00C3\u00A2to -t J C1-work.VAI-3 and C1-cry.VAI-3 \u00E2\u0080\u0098Jane worked and cried.\u00E2\u0080\u0099 = (i) PRECEDE (work, cry) = (ii) SIMULTANEOUS (work, cry) b. Jane \u00C3\u00AA-k\u00C3\u00AE-atosk\u00C3\u00AAt \u00C3\u00AAkwa \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2tot J \u00C3\u00AA- k\u00C3\u00AE- atosk\u00C3\u00AA -t \u00C3\u00AAkwa \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00A2to -t J C1-PREV-work.VAI-3 and C1-PREV-cry.VAI-3 \u00E2\u0080\u0098Jane had worked and she had cried\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) PRECEDE (work, cry) = (ii) SIMULTANEOUS (work, cry) comment: this is better if you have \u00C3\u00AAkwa in it comment: you\u00E2\u0080\u0099re putting it in the same time If only the second clause is marked with k\u00C3\u00AE-, the utterance is acceptable, but the two events cannot be cotemporaneous (110a). Finally, the utterance where the first clause is marked with k\u00C3\u00AE- degrades: the temporal relations become unclear, and the consultant volunteered a form with \u00C3\u00AAkwa removed. (110) a. Jane \u00C3\u00AA-atosk\u00C3\u00AAt \u00C3\u00AAkwa \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00A2tot. J \u00C3\u00AA- atosk\u00C3\u00AA -t \u00C3\u00AAkwa \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00A2to -t J C1-work.VAI-3 and C1-PREV-cry.VAI-3 \u00E2\u0080\u0098Jane is working and she had cried\u00E2\u0080\u0099 = (i) PRECEDE (work, cry) \u00E2\u0089\u00A0 (ii) SIMULTANEOUS (work, cry) comment: when she was at work, she looked like she had cried b. ? Jane \u00C3\u00AA-k\u00C3\u00AE-atosk\u00C3\u00AAt \u00C3\u00AAkwa \u00C3\u00AA-m\u00C3\u00A2tot J \u00C3\u00AA- k\u00C3\u00AE- atosk\u00C3\u00AA -t \u00C3\u00AAkwa \u00C3\u00AA- m\u00C3\u00A2to -t J C1-PREV-work.VAI-3 and C1-cry.VAI-3 \u00E2\u0080\u0098Jane had worked and she was crying\u00E2\u0080\u0099 comment: you could say \u00E2\u0080\u0098J \u00C3\u00AA-k\u00C3\u00AE-atosk\u00C3\u00AAt \u00C3\u00AA-m\u00C3\u00A2tot\u00E2\u0080\u0099 Plains Cree unembedded anaphoric clauses thus have the same behaviour as clause- chaining with respect to the morpho-syntactic locus test: they pattern with subordinate clauses, rather than coordinate clauses. The sensitivity of unembedded anaphoric clauses to the overt coordinator \u00C3\u00AAkwa \u00E2\u0080\u0098and/then\u00E2\u0080\u0099 means that we can use its presence as a diagnostic for the relative structure of two 197 clauses. If there is an overt coordinator like \u00C3\u00AAkwa, then the two clauses are coordinated and both coordinates must be equally marked. The data below shows that \u00C3\u00AAkwa \u00E2\u0080\u0098and/then\u00E2\u0080\u0099 is connecting only clauses which have an overt temporal operator: the dependent clauses are not targeted by \u00C3\u00AAkwa (cf. Ogg 1991, who claims that \u00C3\u00AAkwa is a \u00E2\u0080\u0098sentential\u00E2\u0080\u0099 connective that is often interpreted as connecting a clause it is not adjacent to). More generally, \u00C3\u00AAkwa \u00E2\u0080\u0098and/then\u00E2\u0080\u0099 only coordinates like constituents, and in order for two constituents to be \u00E2\u0080\u0098alike\u00E2\u0080\u0099 they must both be marked for temporal force as well as clause-typing. We observe a sequence of nine \u00C3\u00AA-conjunct clauses, four of which are marked with k\u00C3\u00AE- and five of which lack it. Relevant to the current discussion is that all of the clauses which are marked with the temporal sequencer k\u00C3\u00AE- are also marked with an overt clausal sequencer \u00C3\u00AAkwa (and, in all cases but one, with the habitual m\u00C3\u00A2na as well). (111) kiy\u00C3\u00A2m \u00C3\u00A2ta k\u00C3\u00A2-pipok, \u00C3\u00A2hci piko m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-y\u00C3\u00AEkinik\u00C3\u00AAt nik\u00C3\u00A2w\u00C3\u00AEn\u00C3\u00A2n. kiy\u00C3\u00A2m \u00C3\u00A2ta k\u00C3\u00A2-pipon -k \u00C3\u00A2hci piko m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- y\u00C3\u00AEkinik\u00C3\u00AA -t ni- k\u00C3\u00A2w\u00C3\u00AE -n\u00C3\u00A2n even though C2-winter.VII-0 still QUANT usually C1-PREV-milk.VAI -3 1- mom -1PL \u00E2\u0080\u0098Even during the winter our mother would still milk the cows. \u00C3\u00AAkwa mistahi m\u00C3\u00A2n \u00C3\u00A2ya, \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2moht\u00C3\u00AAy\u00C3\u00A2hk, \u00C3\u00AAkwa mistahi m\u00C3\u00A2na aya \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2moht\u00C3\u00AA -y\u00C3\u00A2n -k and much usually CONN C1-PREV-go.around.VAI-1 -PL And we used to go around a lot, \u00C3\u00AA-w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2y\u00C3\u00A2hk \u00C3\u00A2skaw \u00C3\u00AA-~ \u00C3\u00AA-pap\u00C3\u00A2mi-mawisot, \u00C3\u00AA- w\u00C3\u00AEc\u00C3\u00AAw -\u00C3\u00A2 -y\u00C3\u00A2n -k \u00C3\u00A2skaw \u00C3\u00AA- pap\u00C3\u00A2m -i- mawiso -t C1-go.along.VTA-DIR-1 -PL sometimes C1-around-PV-pick.berries.VAI-3 sometimes going along with her as she went about berry-picking, \u00C3\u00AAkwa \u00C3\u00AA-k\u00C3\u00AE-nayahtahk m\u00C3\u00A2na m\u00C3\u00AEnisa aya, \u00C3\u00AAkwa \u00C3\u00AA- k\u00C3\u00AE- nayahtam -k m\u00C3\u00A2na m\u00C3\u00AEnisa aya and C1-PREV-carry.on.back.VTI-0 usually berries CONN and she used to carry the berries on her back \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAhtat\u00C3\u00A2t \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AAhtat -\u00C3\u00A2 -t C1-COME-go.home.VTA-DIR-3 and bring them back home, 198 \u00C3\u00AAkwa \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00A2sahk mis\u00C3\u00A2skwat\u00C3\u00B4mina. \u00C3\u00AAkwa \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00A2sam -k mis\u00C3\u00A2skwat\u00C3\u00B4min -a and C1-PREV-dry.VTI-0 berry -PL and she used to dry saskatoons. \u00C3\u00AAkwa m\u00C3\u00AEna takwahimin\u00C3\u00A2na m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-takwahahk, \u00C3\u00AAkwa m\u00C3\u00AEna takwahimin\u00C3\u00A2n -a m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- takwaham -k and also choke.cherry -PL usually C1-PREV-pound.VTI -0 And she also used to pound chokecherries \u00C3\u00AA-p\u00C3\u00A2sahk \u00C3\u00AAkoni; k\u00C3\u00A2-pipok \u00C3\u00AAkoni \u00C3\u00AA-m\u00C3\u00AEciy\u00C3\u00A2hk. \u00C3\u00AA- p\u00C3\u00A2sam -k \u00C3\u00AAkoni k\u00C3\u00A2- pipon -k \u00C3\u00AAkoni \u00C3\u00AA- m\u00C3\u00AEci -y\u00C3\u00A2n -k C1-dry.VTI-0 TOPIC C2-winter.VII-0 TOPIC C1-eat.VTI-1 -PL and dry them; these we ate during the winter.\u00E2\u0080\u0099 (EM 17) 4.5.1.2 Fixed relative order of the anaphoric clause and antecedent clause Another classic feature of clause-chains is that the order of the dependent clause relative to the main clause is fixed. In many of the languages discussed (Longacre 1983, Stirling 1993), the order is [ dependent main ]; i.e., they are head-final. (112) [dependent ] [main] Tonkawa [Tekeke\u00CA\u0094e:k \u00C5\u00A1\u00CA\u0094a:pa-ta ] [ ke-ya\u00C5\u00A1e-w ]. in.that.bush hide-SAME 1sgU-watch-IMP \u00E2\u0080\u0098Hide in that bush and watch me.\u00E2\u0080\u0099 (Hoijer 1949, in Foley & Van Valin 1984:258) In Plains Cree, the main (antecedent) clause always precedes the dependent (anaphoric) clause. Under the current analysis, the ordering facts predict that in head-final clause-chains the dependent clauses must be subject to c-command. This is a question I am not prepared to answer since I am not familiar enough with the relevant languages. However, in further research, I suspect this would be relevant to Giv\u00C3\u00B3n\u00E2\u0080\u0099s (2001) discussion about head-initial vs. head-final \u00E2\u0080\u009Cclause-chains.\u00E2\u0080\u009D A supporting piece of evidence that this prediction is on the right track is that anaphoric clauses can be matrix clauses (i.e., are not subject to c-command); to my knowledge, dependent clauses in a clause-chain cannot (Stirling 1993) (i.e., indicating they are subject to c- command). 199 In each of the following examples, the initial antecedent clause is marked with k\u00C3\u00AE-, while the anaphoric clauses follow them. The antecedent clauses also have additional temporal particles, including habitual m\u00C3\u00A2na \u00E2\u0080\u0098usually\u00E2\u0080\u0099, the connective \u00C3\u00AAkwa \u00E2\u0080\u0098and\u00E2\u0080\u0099 and/or the connective m\u00C3\u00AEna \u00E2\u0080\u0098and/also\u00E2\u0080\u0099. Here the point I wish to make is that all of the sequencing elements are showing up on the same clauses, rather than being distributed across the clauses. (113) \u00C3\u00AA-k\u00C3\u00AE-sipw\u00C3\u00AApicicik m\u00C3\u00A2n ANTECEDENT \u00C3\u00AA- k\u00C3\u00AE- sipw\u00C3\u00AApici -t -k m\u00C3\u00A2na C1-PREV-move.VAI -3-PL usually \u00E2\u0080\u0098they would move their camps out \u00C3\u00AA-nitawi-w\u00C3\u00AEkicik \u00C3\u00AAkot\u00C3\u00AA, \u00E2\u0080\u00A6 ANAPHORIC \u00C3\u00AA- nitawi- w\u00C3\u00AEki -t -k \u00C3\u00AAkot\u00C3\u00AA C1-go- live.VAI-3-PL there and go to live out there, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 12) (114) \u00C3\u00AAkwa m\u00C3\u00AEn \u00C3\u00AA-k\u00C3\u00AE-nit\u00C3\u00A2misohk ANTECEDENT \u00C3\u00AAkwa m\u00C3\u00AEna \u00C3\u00AA-k\u00C3\u00AE- nit\u00C3\u00A2miso -hk and also C1-PREV-look.for.berries.VAI-USC \u00E2\u0080\u0098And people used to look for berries \u00C3\u00AA-mawasohk, \u00E2\u0080\u00A6 ANAPHORIC \u00C3\u00AA- mawaso -hk C1-pick.berries.VAI-USC and pick berries, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 9.2) (115) ANTECEDENT m\u00C3\u00A2ka m\u00C3\u00A2n \u00C3\u00AAkot\u00C3\u00AA m\u00C3\u00AEna, iyikohk m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2mipiciy\u00C3\u00A2hk misiw \u00C3\u00AEt\u00C3\u00AA north, m\u00C3\u00A2ka m\u00C3\u00A2na \u00C3\u00AAkot\u00C3\u00AA m\u00C3\u00AEna iyikohk m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2mipici -y\u00C3\u00A2n -k misiw\u00C3\u00AA it\u00C3\u00AA north but usually there also DEG usually C1-PREV-move.VAI -1 -PL all there north \u00E2\u0080\u0098But we also used to move our camp about so much, all over the north, \u00C3\u00AA-minahocik, ANAPHORIC \u00C3\u00AA- minaho -t -k C1-hunt.VAI-3-PL they killed animals \u00C3\u00AA-mawasoy\u00C3\u00A2hk k\u00C3\u00A2-n\u00C3\u00AEpihk, \u00E2\u0080\u00A6 ANAPHORIC \u00C3\u00AA- mawaso -y\u00C3\u00A2n -k k\u00C3\u00A2- n\u00C3\u00AEpin -k C1-pick.berries.VAI-1 -PL C2-summer.VII-0 and we picked berries in the summer, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 1.4) As we saw earlier in the chapter, an unembedded anaphoric clause cannot precede its antecedent; an example of this is given in (116). 200 (116) a. n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA-p\u00C3\u00AA-w\u00C3\u00AEtatosk\u00C3\u00AAmit, [\u00C3\u00AA-ahkosiyit w\u00C3\u00AEwa] ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA- p\u00C3\u00AA- w\u00C3\u00AEtatosk\u00C3\u00AAm -it \u00C3\u00AA- ahkosi -yi -t w- \u00C3\u00AEw -a 1- friend C1-COME-work.with.VTA-3>1 C1-sick.VAI-DS-3 3- wife-OBV \u00E2\u0080\u0098\u00E2\u0080\u00A6my friend came to work with me, his wife was sick.\u00E2\u0080\u0099 b. ! [\u00C3\u00AA-ahkosiyit w\u00C3\u00AEwa, ] n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA-p\u00C3\u00AA-w\u00C3\u00AEtatosk\u00C3\u00AAmit \u00C3\u00AA- ahkosi -yi -t w- \u00C3\u00AEw -a ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA- p\u00C3\u00AA- w\u00C3\u00AEtatosk\u00C3\u00AAm -it C1-sick.VAI-DS-3 3- wife-OBV 1- friend C1-COME-work.with.VTA-3>1 --- (intended: \u00E2\u0080\u0098...his wife was sick, my friend came to visit me.\u00E2\u0080\u0099) 4.5.2 English modally subordinated clauses are anaphoric clauses I have claimed that the indexical/anaphoric distinction in Plains Cree clauses is directly mapped into the morpho-syntactic distinction between INDEPENDENT and CONJUNCT order clauses. However, the analysis predicts that even in a language which does not morphologically distinguish between these two clause-types, such as English, there should still be semantic evidence for the distinction between indexical and anaphoric clauses In English, the indexical/anaphoric distinction is most clearly seen by contrasting a matrix clause and an embedded one. Without any other context, the matrix clause (117a) is taken as presenting a proposition that the speaker believes (i.e., an indexical clause), and the embedded clause (117b) as presenting a proposition that the islanders (i.e., the subject of the higher clause) believe (i.e., an anaphoric clause). (117) a. Death is never natural. b. The islanders believe that death is never natural. However, a proposition in a matrix clause can also be interpreted relative to a preceding clause, in a phenomena identified by Roberts (1989) as \u00E2\u0080\u0098generalized modal subordination\u00E2\u0080\u0099. In modal subordination, unembedded clauses are in the (discourse) scope of some operator in a previous clause: no overt modal operator, nor any other type of morpho-syntactic distinction is necessary (although the cases where an overt modal is present are by far the most common cases discussed in the literature). 201 Consider the following example, a paragraph consisting of three \u00E2\u0080\u0098sentences\u00E2\u0080\u0099.25 (118) The islanders believe a lot of strange things about the world. Death is never natural and each death must be avenged. The gods punish those who do not avenge the death of their family members. (from Farkas 1992:88) The first sentence makes a claim about the islanders and their beliefs. The next two sentences are interpreted as elaborations on this claim: as English readers, we understand the propositions \u00E2\u0080\u0098death is never natural\u00E2\u0080\u0099 \u00E2\u0080\u0098each death must be avenged\u00E2\u0080\u0099 and \u00E2\u0080\u0098the gods punish those who do not avenge the death of their family members\u00E2\u0080\u0099 to be things that the islanders believe (and not, for example, something that the writer believes). (119) The islanders believe p1, p2, p3 INDEXICAL p1: Death is never natural ANAPHORIC p2: Each death must be avenged ANAPHORIC p3: The gods punish those who do not avenge the death ANAPHORIC of their family members. In terms of their interpretation, then, these propositions behave like an embedded proposition rather than an (indexical) matrix clause, even though they are morphologically indistinguishable from the latter. (120) CP CP CP 6 9 6 the islanders 5and 5 gods punish CP believe \u00E2\u0080\u00A6 death never each death 6 natural avenged those who don\u00E2\u0080\u0099t avenge Roberts (1989) characterizes clauses that are modally subordinated having a \u00E2\u0080\u0098telescoping\u00E2\u0080\u0099 function, where the modally subordinated clauses are in a part-whole or subset-set relation to the antecedent clause. 25 When one starts looking at English discourse through this perspective, examples appear in abundance: in newspapers, in letters, in stories, in conversations; it is an important issue to investigate more thoroughly in future research. 202 Contexts like (118) are exactly where we see the indexical/anaphoric distinction marked in Plains Cree, as in the following example (121). The first line contains an indexical INDEPENDENT clause nikiskisin \u00E2\u0080\u0098I remember\u00E2\u0080\u0099. The following anaphoric CONJUNCT clauses provide the details of what she remembered \u00E2\u0080\u0093 they are all interpreted relative to the indexical clause. We also see internal structure of the sequence of anaphoric clauses: clauses (ii), (iv), (vi), (xi), and (xii) are marked with the temporal sequencer k\u00C3\u00AE- and correspond to either a change of subject or an overt nominal (i.e., the temporal anaphora and pronominal anaphora are working in tandem, as expected by discourse analysis done on other languages; cf. Giv\u00C3\u00B3n 2001, Smith 2003, among others). The first anaphoric clause introduces p\u00C3\u00AAyak \u00C3\u00AAkota m\u00C3\u00A2na kis\u00C3\u00AAyiniw \u00E2\u0080\u0098a certain old man\u00E2\u0080\u0099; the subject changes in (iv) and (vi), and an overt nominal is used in (xi) and (xii). In these and only these cases, the clause is marked with k\u00C3\u00AE-: the temporal anchoring and referential anchoring are working in tandem.26 (121) (i) m\u00C3\u00A2cik \u00C3\u00AAkosp\u00C3\u00AE anima nikiskisin k\u00C3\u00A2-m\u00C3\u00AEhcin\u00C3\u00AAhk anima, aya, INDEPENDENT m\u00C3\u00A2cik \u00C3\u00AAkosp\u00C3\u00AE anima ni- kiskisi -n wait.and.see then DEM.INAN 1- remember.VAI-SAP k\u00C3\u00A2- m\u00C3\u00AAhcin\u00C3\u00AAhk anima aya C2-die.out.VAI DEM.INAN CONN \u00E2\u0080\u0098For instance I remember how, at the time of the great epidemic (ii) p\u00C3\u00AAyak \u00C3\u00AAkota m\u00C3\u00A2na kis\u00C3\u00AAyiniw, c\u00C3\u00AEki nik\u00C3\u00AE-wa-w\u00C3\u00AEtapim\u00C3\u00A2kanin\u00C3\u00A2n, p\u00C3\u00AAyak \u00C3\u00AAkota m\u00C3\u00A2na kis\u00C3\u00AAyiniw c\u00C3\u00AEki ni-k\u00C3\u00AE- wa- w\u00C3\u00AEtapim\u00C3\u00A2kani -n\u00C3\u00A2n one there usually old.man close 1- PREV-RED-neighbor.VAI -1PL a certain old man, a close neighbour of ours, iyikohk \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2mi-pamiht\u00C3\u00A2sot, CONJUNCT iyikohk \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00A2p\u00C3\u00A2mi- pamiht\u00C3\u00A2so -t DEG C1-PREV-go.about-tend.VAI -3 went about looking after the sick (iii) \u00C3\u00AA-pap\u00C3\u00A2m-\u00C3\u00A2h-a-~-p\u00C3\u00A2h-p\u00C3\u00AEhtikw\u00C3\u00AAt w\u00C3\u00A2skahikana, CONJUNCT \u00C3\u00AA- pap\u00C3\u00A2m- p\u00C3\u00A2h- p\u00C3\u00AEhtikw\u00C3\u00AA -t w\u00C3\u00A2skahikan -a C1-go.about-RED-go.inside.VAI -3 house -PL going into each of the houses 26 In line (ii), there is an additional INDEPENDENT clause: c\u00C3\u00AEki nik\u00C3\u00AE-wa-w\u00C3\u00AEtapim\u00C3\u00A2kanin\u00C3\u00A2n \u00E2\u0080\u0098he was a close neighbor of ours\u00E2\u0080\u0099. Notice that this clause functions as a parenthetical \u00E2\u0080\u0093 the speaker\u00E2\u0080\u0099s comment to the side about the person she is talking about \u00E2\u0080\u0093 the indexical clause has the expected disconnect from the discourse of the main story line of the man\u00E2\u0080\u0099s actions during the epidemic. 203 (iv) mistah \u00C3\u00A2yis anim \u00C3\u00AA-k\u00C3\u00AE-\u00C3\u00A2hkosihk s\u00C3\u00B4skw\u00C3\u00A2c, CONJUNCT mistahi ayis anima \u00C3\u00AA- k\u00C3\u00AE- \u00C3\u00A2hkosi -hk s\u00C3\u00B4skw\u00C3\u00A2c very for DEM.INAN C1-PREV-sick.VAI -USC just for the people were extremely sick (v) iyikohk mihc\u00C3\u00AAt k\u00C3\u00A2-k\u00C3\u00AE-nipicik; CONJUNCT iyikohk mihc\u00C3\u00AAt k\u00C3\u00A2-k\u00C3\u00AE- nipi -t -k DEG many C2-PREV-die.VAI-3-PL and there were so many who died (vi) \u00C3\u00AA-k\u00C3\u00AE-pap\u00C3\u00A2mi-p\u00C3\u00AEhtikwat\u00C3\u00A2t mihta, CONJUNCT \u00C3\u00AA- k\u00C3\u00AE- pap\u00C3\u00A2mi -p\u00C3\u00AEhtikwat\u00C3\u00A2 -t mihta C1-PREV-go.about-carry.inside.VAI-3 firewood he went about hauling wood (vii) nipiy \u00C3\u00AA-ast\u00C3\u00A2t, CONJUNCT nipiy \u00C3\u00AA- ast\u00C3\u00A2 -t water C1-place.VAI-3 and filling up the water supply, (viii) \u00C3\u00AAkwa kahkiyaw k\u00C3\u00AEkway \u00C3\u00AA-t\u00C3\u00B4tahk, CONJUNCT \u00C3\u00AAkwa kahkiyaw k\u00C3\u00AEkway \u00C3\u00AA- t\u00C3\u00B4tam -k and all thing C1-do.VTI -0 doing everything, (ix) kiy\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2ci p\u00C3\u00AAyak w\u00C3\u00A2skahikan, CONJUNCT iy- k\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2 -t -i p\u00C3\u00AAyak w\u00C3\u00A2skahikan IC-finish.VAI-3-SUBJ one house and when he had finished one house, (x) kotakihk \u00C3\u00AA-itoht\u00C3\u00AAt; CONJUNCT kotak -ihk \u00C3\u00AA- itoht\u00C3\u00AA -t other -LOC C1-go.VAI -3 he went to the next; (xi) m\u00C3\u00B4y \u00C3\u00AA-ohc-\u00C3\u00A2hkosit \u00C3\u00AAwakw \u00C3\u00A2n[a] \u00C3\u00A2na kis\u00C3\u00AAyiniw, CONJUNCT m\u00C3\u00B4ya \u00C3\u00AA- ohci- \u00C3\u00A2hkosi -t \u00C3\u00AAwakw ana ana kis\u00C3\u00AAyiniw NEG C1-PREV-sick.VAI-3 TOP DEM.AN DEM.AN old.man that old man did not get sick, (xii) \u00C3\u00AA-k\u00C3\u00AE-ma-m\u00C3\u00B4s\u00C3\u00A2p\u00C3\u00AAwit ana kis\u00C3\u00AAyiniw. CONJUNCT \u00C3\u00AA- k\u00C3\u00AE- ma- m\u00C3\u00B4s\u00C3\u00A2p\u00C3\u00AAwi -t ana kis\u00C3\u00AAyiniw C1-PREV-RED-widower.VAI-3 DEM.AN old.man that old man was a widower.\u00E2\u0080\u0099 (AA 1.9) Taking the correlation of k\u00C3\u00AE-marking and reference-tracking to indicate structure, the utterance in (121) can be structurally represented as in (122). 204 (1 22 ) C P i C P i i C P i v C P v i C P x i C P x ii 5 6 C P i ii 6 5 C P v ii C P v iii C P x 5 5 n ik is ki si n \u00C3\u00AA -k \u00C3\u00AE-p am ih ta so t 5 \u00C3\u00AA- k\u00C3\u00AE -a hk os ih k C P v \u00C3\u00AA- k\u00C3\u00AE -p \u00C3\u00AEh tik w at \u00C3\u00A2t 5 5 5 \u00C3\u00AA- oh c- \u00C3\u00A2h ko si t \u00C3\u00AA -k \u00C3\u00AE-m a- m \u00C3\u00B4s \u00C3\u00A2p \u00C3\u00AAw it \u00E2\u0080\u0098I re m em be r\u00E2\u0080\u0099 \u00C3\u00AA- p\u00C3\u00AE ht ok w \u00C3\u00AAt 5 \u00C3\u00AA- as t\u00C3\u00A2 t \u00C3\u00AA- t\u00C3\u00B4 ta hk C P i x \u00C3\u00AA -it oh t\u00C3\u00AA t k\u00C3\u00A2 -k \u00C3\u00AE-n ip ic ik 5 ki y\u00C3\u00AE s\u00C3\u00AE ht \u00C3\u00A2c i In th is st ru ct ur e, th e ar ro w s f lo w fr om th e an ap ho ric c la us e to it s a nt ec ed en t. S ub or di na te c la us es a re e m be dd ed u nd er th ei r an te ce de nt c la us e, n on -s ub or di na te c la us es a re n ot (r eg ar dl es s o f t he ir an ap ho ric st at us ). Th er e is a n in iti al in de xi ca l c la us e ni ki sk isi n \u00E2\u0080\u0098I re m em be r\u00E2\u0080\u0099 a nd th en fi ve c la us es (i i, iv , v i, xi , a nd x ii) a na ph or ic al ly d ep en de nt on th e in de xi ca l c la us e. C la us es (i i) an d (v i) al so se rv e as a nc ho rs fo r t he a na ph or ic al ly d ep en de nt c la us es (i ii) , ( vi i), (v iii ) a nd (x ). Th er e ar e al so tw o su bo rd in at e cl au se s: c la us e (v ) i s s ub or di na te d to c la us e (iv ), an d cl au se (i x) is su bo rd in at ed to c la us e (x ). R ob er ts su gg es ts th at m od al su bo rd in at io n in E ng lis h co ul d be se m an tic al ly m od el le d us in g si tu at io n se m an tic s: a c la us e th at is m od al ly su bo rd in at ed w ith re sp ec t t o so m e ot he r c la us e m us t b e gi ve n a si tu at io n in w hi ch to b e ev al ua te d. T hi s i s f or m al ly v er y si m ila r t o th e an al ys is p ro po se d he re fo r a na ph or ic c la us es (i .e ., th e si tu at io n in w hi ch th e pr op os iti on o f a n an ap ho ric c la us es is an ap ho ric al ly g iv en ). T he c ur re nt a na ly si s t hu s o ff er s a w ay to u nd er st an d th e se m an tic si m ila rit ie s o f u ne m be dd ed -b ut -m od al ly - su bo rd in at ed a nd e m be dd ed c la us es in E ng lis h. 205 4.6 Summary This chapter has talked about anaphoric clauses, which take the form of CONJUNCT clauses in Plains Cree. Taking Williams\u00E2\u0080\u0099 (1997) observed patterns for licensing dependencies between anaphors and antecedents as a starting point, I showed that anaphoric CONJUNCT clauses have all of the familiar patterns of pronominal anaphors. If they are embedded, they are insensitive to precedence; if they are unembedded, precedence must be respected. This pattern directly parallels the distribution of pronominal anaphors. Second, anaphoric CONJUNCT clauses have the same kinds of discourse- dependencies that pronominal forms do: they are infelicitous in out-of-the-blue contexts. Finally, variables within anaphoric CONJUNCT clauses have relations to their antecedents that are subject to the c-command and precedence conditions seen for pronominal anaphora. In the next chapter I turn to the syntax of these anaphoric clauses. 206 CHAPTER 5 THE SYNTAX OF ANAPHORIC CLAUSES 5.1 Proposal: Chained, adjoined, and mediated argument clauses In the last chapter I established that anaphoric CONJUNCT clauses contain dependent elements whose antecedent must be either in a preceding or c-commanding clause. In this chapter, I turn to the nature of the structural relations that may hold between an anaphoric clause and the antecedent clause. Differentiating relations between clauses has typically been difficult in Plains Cree for multiple reasons. For one thing, there is not much inflectional clause-typing morphology to distinguish between clauses. Anaphoric clauses in Plains Cree come in five morpho-syntactic flavors, as shown in table 5.1. PLAINS CREE CONJUNCT \u00E2\u0080\u0098modes\u00E2\u0080\u0099 FORM GLOSS changed conjunct (1) \u00C3\u00AA- nip\u00C3\u00A2t \u00E2\u0080\u00A6s/he is sleeping changed conjunct (2) k\u00C3\u00A2-nip\u00C3\u00A2t when s/he sleeps (iterative) changed conjunct n\u00C3\u00AAp\u00C3\u00A2ci whenever s/he slept (subjunctive) simple conjunct nip\u00C3\u00A2ci if/when s/he sleeps (irrealis) simple conjunct ka-nip\u00C3\u00A2t him/her to sleep Table 5.1. Morpho-syntactic classification of anaphoric CONJUNCT clauses However, as we will see, the morpho-syntax of the verbal complex often cross-cuts the syntactic classes; for example, all five classes in table 5.1 seem to behave in some contexts as adjunct clauses. A second \u00E2\u0080\u0093 albeit related \u00E2\u0080\u0093 problem is that some of the diagnostics used for distinguishing different syntactic relations between clauses in languages like English are not directly applicable to Plains Cree (Blain 1997, Long 1999); language-internal diagnostics are needed in order to develop an accurate classification. In this chapter, I propose that anaphoric clauses can be divided into three classes: chained clauses as in (1), adjoined clauses as in (2), and mediated argument clauses (including both object- and subject-mediated clauses) as in (3). 207 (1) CHAINED CLAUSES CP1 CP2 CP3 5 5 5 ANTECEDENT ANAPHORIC ANAPHORIC (2) a. Right-adjoined CP1 3 CP1 CP2 5 5 ANTECEDENT ANAPHORIC a. Left-adjoined CP1 3 CP2 CP1 5 5 ANAPHORIC ANTECEDENT ADJOINED CLAUSES (3) a. Object-mediated CP 5 VP 3 VP CPi 2 5 V DPi ANAPHORIC b. Subject-mediated CP 5 IP 3 IP CPi 2 5 DPi 2ANAPHORIC VP MEDIATED ARGUMENT CLAUSES In terms of the relation between anaphoric clauses and their antecedent, chained clauses are excluded from the antecedent clause in the sense of May (1985), Chomsky (1986): no part of the antecedent clause dominates the anaphoric clause. Adjunct clauses are c-commanded by the clause they are adjoined to. Argument-mediated clauses are c-commanded by the clause they they are adjoined to, and in addition are licensed by an argument position. In \u00C2\u00A75.2, I lay out the two classes of diagnostics I use motivate my tripartite division: exclusion tests, and island tests. In \u00C2\u00A75.3, I walk through the exclusion diagnostics and show that they pick out chained clauses (those that are not c-commanded by any other clause). In \u00C2\u00A75.4, I walk through the c-command diagnostics and show that they pick out argument-mediated clause (those that are licensed by an argument position). Finally, in \u00C2\u00A75.5, I look at the implications of the analysis with respect to (i) 208 the non-existence of argument clauses; (ii) complementation; and (iii) the syntax of copy-to- object constructions. 5.2 The diagnostics I use two classes of tests to pick out the three syntactic classes of clauses. 5.2.1 Exclusion tests Exclusion tests are tests that pick out something which is excluded, as defined in (4). (4) Exclusiondef: element \u00CE\u00B1 is excluded iff there is no element \u00CE\u00B2 that dominates it I use three exclusion tests: (i) sensitive to precedence, (ii) necessity of prosodic breaks; and (iii) ability to be a matrix clause. Exclusion tests should uniformly isolate chained clauses as opposed to adjoined or mediated argument clauses. In the case of precedence and prosodification tests, the test shows that the relevant process is obligatory for chained clauses but optional for for adjoined and argument mediated clauses. In the case of the matrix clause tests, we observe that chained clauses may always be matrix clauses, but adjoined and mediated argument clauses sometiems cannot. 5.2.2 Island tests Island tests are so-called after Ross (1967), who discovered that adjuncts are \u00E2\u0080\u009Cislands\u00E2\u0080\u009D - they do not allow elements to escape from them \u00E2\u0080\u0093 but arguments are not. Following work arguing that 209 (mediated) argument clauses fail to be islands because of their relation to an argument position, I use these tests to isolate mediated argument clauses as opposed to chained or adjoined clauses. There are three island tests: (i) wh-fronting (ii) quantifier-fronting (iii) argument expression-fronting In all of these cases mediated argument clauses allow fronting of the relevant element, but adjoined and chained clauses do not. 5.3 Applying the exclusion tests The first set of diagnostics pick out chained clauses as opposed to adjoined and mediated argument clauses. 5.3.1 Linear precedence Recall that anaphora must be licensed either by the c-command condition or the precedence condition. If an anaphoric clause fails to be licensed by c-command, then we expect that it must be subject to the precedence condition: otherwise, it could not be anaphoric. Thus we expect the following pattern. CHAINED ADJOINED MEDIATED ARG. Always subject to precedence? \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Table 5.2. Diagnostic 1: Subjection to precedence This is what we find, as exemplified in each of the next three subsections. 210 5.3.1.1 Chained clauses must follow their antecedent Chained clauses must follow the antecedent clause that licenses them. As expected, reversing the order of a chained anaphoric clause with respect to the antecedent clause is not possible. (5) a. [antecedent] [chained clause] b. * [chained clause] [antecedent] In (6) the antecedent clause contains the temporal shifting preverb k\u00C3\u00AE-, the habitual m\u00C3\u00A2na \u00E2\u0080\u0098usually\u00E2\u0080\u0099 and the overt nominal aw\u00C3\u00A2sisak \u00E2\u0080\u0098children\u00E2\u0080\u0099; the anaphoric clause \u00C3\u00AA-nikamocik \u00E2\u0080\u0098\u00E2\u0080\u00A6they used to sing\u00E2\u0080\u0099 must follow it.1 (6) a. [ \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-itoht\u00C3\u00AAcik m\u00C3\u00A2na aw\u00C3\u00A2sisak, ] [ \u00C3\u00AA-nikamocik ] CHAIN \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00AA- it\u00C3\u00B4ht\u00C3\u00AA -t -k m\u00C3\u00A2na aw\u00C3\u00A2sis -ak \u00C3\u00AA- nikamo -t -k C1-PREV-COME-go.VAI-3-PL usually child -PL C1-sing.VAI-3-PL \u00E2\u0080\u0098\u00E2\u0080\u00A6the children used to come and they used to sing.\u00E2\u0080\u0099 b. ! [ \u00C3\u00AA-nikamocik, ] [ \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-itoht\u00C3\u00AAcik m\u00C3\u00A2na aw\u00C3\u00A2sisak ] \u00C3\u00AA- nikamo -t -k \u00C3\u00AA- k\u00C3\u00AE- p\u00C3\u00AA- it\u00C3\u00B4ht\u00C3\u00AA -t -k m\u00C3\u00A2na aw\u00C3\u00A2sis -ak C1-sing.VAI-3-PL C1-PREV-COME-go.VAI-3-PL usually child -PL --- (intended: \u00E2\u0080\u0098\u00E2\u0080\u00A6the children used to come and they used to sing.\u00E2\u0080\u0099) Similarly in (7), the anaphoric clause contains the different-subject marker -yi, and in order to get the chained interpretation, it must follow its antecedent clause. (7) a. n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA-p\u00C3\u00AA-w\u00C3\u00AEtatosk\u00C3\u00AAmit, [\u00C3\u00AA-\u00C3\u00A2hkosiyit w\u00C3\u00AEwa] CHAIN ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA- p\u00C3\u00AA- w\u00C3\u00AEtatosk\u00C3\u00AAm -it \u00C3\u00AA- \u00C3\u00A2hkosi -yi -t w- \u00C3\u00AEw -a 1- friend C1-COME-work.with.VTA-3>1 C1-sick.VAI-DS-3 3- wife-OBV \u00E2\u0080\u0098\u00E2\u0080\u00A6my friend came to work with me, his wife was sick.\u00E2\u0080\u0099 b. ! [\u00C3\u00AA-\u00C3\u00A2hkosiyit w\u00C3\u00AEwa, ] n\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA-p\u00C3\u00AA-w\u00C3\u00AEtatosk\u00C3\u00AAmit \u00C3\u00AA- \u00C3\u00A2hkosi -yi -t w- \u00C3\u00AEw -a ni- w\u00C3\u00AEc\u00C3\u00AAw\u00C3\u00A2kan \u00C3\u00AA- p\u00C3\u00AA- w\u00C3\u00AEtatosk\u00C3\u00AAm -it C1-sick.VAI-DS-3 3- wife-OBV 1- friend C1-COME-work.with.VTA-3>1 --- (intended: \u00E2\u0080\u0098\u00E2\u0080\u00A6his wife was sick, my friend came to visit me.\u00E2\u0080\u0099) 1 Recall that the exclamation point indciates a string that may be well-formed under some interpretation, but not well-formed with the particular interpretation being tested. See chapter 1 for details. 211 5.3.1.2 Adjoined clauses can precede their antecedent By contrast, adjoined clauses are c-commanded by their antecedent, so we expect that they will be able to either precede or follow the antecedent. This expectation is also borne out, as exemplified by adjunct clauses which are introduced by subordinating particles (e.g., subordinate negation \u00C3\u00AAk\u00C3\u00A2 in (8)), by the clause-typing (e.g., k\u00C3\u00A2- in (9)). (8) a. nisipw\u00C3\u00AAht\u00C3\u00A2n \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA-miyom\u00C3\u00A2cihoy\u00C3\u00A2n ADJUNCT ni-sipw\u00C3\u00AAht\u00C3\u00A2 -n \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- miyom\u00C3\u00A2ciho -y\u00C3\u00A2n 1-leave.VAI -SAP NEG C1-feel.well.VAI -1 \u00E2\u0080\u0098I left because I wasn\u00E2\u0080\u0099t feeling well.\u00E2\u0080\u0099 b. \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA-miyom\u00C3\u00A2cihoy\u00C3\u00A2n, nisipw\u00C3\u00AAht\u00C3\u00A2n \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- miyom\u00C3\u00A2ciho-y\u00C3\u00A2n ni-sipw\u00C3\u00AAht\u00C3\u00A2-n NEG C1-feel.well.VAI-1 1-leave.VAI-SAP \u00E2\u0080\u0098I left because I wasn\u00E2\u0080\u0099t feeling well.\u00E2\u0080\u0099 (9) a. Jane k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2c-atosk\u00C3\u00AAt \u00C3\u00A2hkosiwipayiw ADJUNCT J k\u00C3\u00A2- m\u00C3\u00AAkw\u00C3\u00A2- atosk\u00C3\u00AA -t \u00C3\u00A2hkosiwipayi -w J C2-MIDST- work.VAI-3 get.sick.VAI -3 \u00E2\u0080\u0098When Jane was working, she got ill / sick\u00E2\u0080\u0099 b. Jane \u00C3\u00A2hkosiwpayiw k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2c-atosk\u00C3\u00AAt J \u00C3\u00A2hkosiwpayi -w k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2- atosk\u00C3\u00AA -t J get.sick.VAI -3 C2-MIDST- work.VAI-3 \u00E2\u0080\u0098When Jane was working, she got ill / sick\u00E2\u0080\u0099 comment: it\u00E2\u0080\u0099s the same as [a] This data demonstrates that, although embedded clauses may prefer one of the two positions relative to the superordinate clause (cf. Dahlstrom 2006 for Fox), both positions are in principle available (see also the discussion in chapter 4). 5.3.1.3 Mediated argument clauses usually (but not always) follow their antecedent The ordering of mediated argument clauses relative to the main clause is more complicated than adjoined clauses. On the one hand, this class of clauses can be interpreted both when it precedes and follows the main clause; in this sense it does not look like chained clauses. 212 (10) a. w\u00C3\u00A2pahtam Jeff \u00C3\u00AA-k\u00C3\u00AE-mispohk VOLUNTEERED w\u00C3\u00A2pahtam-w J \u00C3\u00AA- k\u00C3\u00AE- mispon -k see.VTI -3 J C1-PREV-snow.VII-0 \u00E2\u0080\u0098Jeff saw that it had snowed.\u00E2\u0080\u0099 a. ? \u00C3\u00AA-k\u00C3\u00AE-mispohk Jeff w\u00C3\u00A2pahtam PRESENTED \u00C3\u00AA- k\u00C3\u00AE- mispon -k J w\u00C3\u00A2pahtam-w C1-PREV-snow.VII-0 J see.VTI -3 \u00E2\u0080\u0098Jeff saw that it had snowed.\u00E2\u0080\u0099 comment: it means \u00E2\u0080\u0098Jeff saw that it had snowed.\u00E2\u0080\u0099 But I would say it [the other way] However, as the consultant\u00E2\u0080\u0099s comment indicates, an order where the embedded clause precedes the higher clause is quite artificial. It is not an order that is seen in running speech, speakers never volunteer this order (in marked contrast to many adjoined clauses), and often comment that they would not use such utterances. More importantly, with some of these clauses, the embedded clause is judged completely bad if it precedes the c-commanding clause. This includes both irrealis object-mediated clauses in (11) and subject-mediated clauses (12). (11) a. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn ka-m\u00C3\u00AEcisoy\u00C3\u00A2n MEDIATED OBJECT ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n ka- m\u00C3\u00AEciso -y\u00C3\u00A2n 1- want.VTI -SAP IRR-eat.VAI -1 \u00E2\u0080\u0098I would like to eat.\u00E2\u0080\u0099 b. * ka-m\u00C3\u00AEcisoy\u00C3\u00A2n ninitaw\u00C3\u00AAyiht\u00C3\u00AAn ka- m\u00C3\u00AEciso -y\u00C3\u00A2n ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n IRR-eat.VAI -1 1- want.VTI -SAP --- (intended: \u00E2\u0080\u0098I would like to eat.\u00E2\u0080\u0099) 213 (12) a. miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt John MEDIATED SUBJECT miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J good.VII DEM.INAN C1-COME-go.VAI-3 J \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that John came.\u00E2\u0080\u0099 b. * \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt John, miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J miyw\u00C3\u00A2sin DEM.INAN C1-COME-go.VAI-3 J good.VII --- (intended: \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that John came.\u00E2\u0080\u0099) Thus, mediated argument clauses have a restriction on them that adjoined clauses do not. I will return to this issue in \u00C2\u00A75.5.1. 5.3.2 Ability to be a matrix clause based on morpho-syntax Since chained clauses are not embedded as a constituent of some other clause, we expect them to have more independence than adjoined and mediated argument clauses, which are constituents. If a clause cannot stand on its own, by part-whole logic, it must be a part of some constituent. Conversely, if a clause is not part of some constituent, it must be able to stand on its own. Therefore, we expect that clauses which participate in chaining structures should always be able to occur as matrix clauses, while those which are embedded under another clause will not have to, as summarized in table 5.3. CHAINED ADJOINED MEDIATED ARG. Always subject to precedence? \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Always has matrix capability? \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Table 5.3. Diagnostic 2: Matrix capability This accounts for the relation between the morpho-syntactic clause-type allowed in chained, adjoined, and mediated argument clauses, and the ability of the of the clause to be a matrix clause. There are four morpho-syntactic clause-types possible, as in (13); only anaphoric \u00C3\u00AA- clauses are possible in matrix environments. 214 (13) a. * k\u00C3\u00A2-w\u00C3\u00A2pamak atim k\u00C3\u00A2- w\u00C3\u00A2pam -ak atim C2- see.VTA-1>3 dog --- b. * ka-w\u00C3\u00A2pamak atim ka- w\u00C3\u00A2pam -ak atim IRR-see.VTA -1>3 dog --- c. * w\u00C3\u00A2pamaki atim w\u00C3\u00A2pam-ak -i atim see.VTA-1>3-SUBJ dog --- d. \u00C3\u00AA-w\u00C3\u00A2pamak atim \u00C3\u00AA- w\u00C3\u00A2pam -ak atim C1-see.VTA-1>3 dog \u00E2\u0080\u0098\u00E2\u0080\u00A6I see a dog.\u00E2\u0080\u0099 On the basis of the data in (13), in each of the next three subsections, I show the mapping between each syntactic class of clauses and the ability of a form to be a matrix clause. We will see that chained clauses are always capable of being matrix clauses, but adjoined and mediated argument clauses are not, as expected. 5.3.2.1 Chained clauses are always capable of being matrix clauses Only \u00C3\u00AA-CONJUNCT clauses can stand on their own (Cook 2007), and only \u00C3\u00AA-CONJUNCT clauses can occur in chains. There is a one-to-one mapping between participating in a clause chain and being a matrix clause. Clause-type Matrix clause? CHAINED k\u00C3\u00A2-CONJUNCT \u00E2\u009C\u0096 \u00E2\u009C\u0096 simple CONJUNCT (w/ ka-) \u00E2\u009C\u0096 \u00E2\u009C\u0096 subjunctive CONJUNCT (w/ -i) \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00C3\u00AA-CONJUNCT \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 5.4. Only potential matrix clauses can be chained clauses 215 5.3.2.2 Adjoined clauses are not always capable of being matrix clauses Unlike chained clauses, adjoined clauses can be any of the four clause-types; not all adjoined clauses can be matrix clauses. This is expected, since adjoined clauses are by part of a larger clause. Clause-type Matrix clause? ADJOINED k\u00C3\u00A2-CONJUNCT \u00E2\u009C\u0096 \u00E2\u009C\u0094 simple CONJUNCT (w/ ka-) \u00E2\u009C\u0096 \u00E2\u009C\u0094 subjunctive CONJUNCT (w/ -i) \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00C3\u00AA-CONJUNCT \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 5.5. Adjoined clauses need not have potential to be matrix clauses 5.3.2.3 Mediated argument clauses are not always capable of being matrix clauses Mediated argument clauses employ two different clauses types (\u00C3\u00AA- and simple CONJUNCT). Not all mediated argument clauses can be matrix clauses. Again, this is consistent with them being part of a larger clause. Clause-type Matrix clause? MEDIATED ARGUMENT k\u00C3\u00A2-CONJUNCT \u00E2\u009C\u0096 \u00E2\u009C\u0096 simple CONJUNCT (w/ ka-) \u00E2\u009C\u0096 \u00E2\u009C\u0094 subjunctive CONJUNCT (w/ -i) \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00C3\u00AA-CONJUNCT \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 5.6. Mediated argument clauses need not have potential to be matrix clauses 5.3.3 Prosodification In Plains Cree running speech, the prosodic break between two clauses can be marked by up to four properties (pitch shift, amplitude shift, final-syllable lengthening, and a pause), with an overall correlation between the degree of clausal-relatedness and the degree of intonational 216 (in)dependence (Cook 2006). These findings are consistent with cross-linguistic findings that syntactic integration of two clauses corresponds with prosodic integration (Chafe 1988). This means we expect chained clauses, which are not a constituent (i.e., part) of another clause, to not be prosodically integrated with the antecedent clause, whereas adjoined and mediated argument clauses may be prosodically integrated. CHAINED ADJOINED MEDIATED ARG. Always subject to precedence? \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Always has matrix capability? \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Requires intonational break? \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Table 5.7. Diagnostic 3: Intonational break Again the following subsections show the relevant data patterns as expected. 5.3.3.1 Chained clauses require an intonational break Consultants impose prosodic requirements chained clauses in elicitation contexts: chained clauses can only be constructed if there is minimally a pause (and often a corresponding pitch and amplitude shift) at the end of each clause (marked orthographically with a comma). (14) a. \u00C3\u00AA-k\u00C3\u00AE-k\u00C3\u00AEs-kimiwahk, Tom \u00C3\u00AA-pimoht\u00C3\u00AAt CHAIN \u00C3\u00AA- k\u00C3\u00AE- k\u00C3\u00AEs- kimiwan-k T \u00C3\u00AA- pim\u00C3\u00B4ht\u00C3\u00AA -t C1-PREV-FINISH-rain.VII -0 T C1-walk.VAI-3 \u00E2\u0080\u0098It had stopped raining, Tom was walking.\u00E2\u0080\u0099 b. ! \u00C3\u00AA-k\u00C3\u00AE-k\u00C3\u00AEs-kimiwahk Tom \u00C3\u00AA-pimoht\u00C3\u00AAt \u00C3\u00AA- k\u00C3\u00AE- k\u00C3\u00AEs- kimiwan-k T \u00C3\u00AA- pim\u00C3\u00B4ht\u00C3\u00AA -t C1-PREV-FINISH-rain.VII -0 T C1-walk.VAI-3 --- (intended: \u00E2\u0080\u0098It had stopped raining, Tom was walking.\u00E2\u0080\u0099) 5.3.3.2 Adjoined clauses do not require an intonational break Adjoined clauses do not require a prosodic break. When asked, consultants will often allow a prosodic break, but comment that \u00E2\u0080\u009Cit isn\u00E2\u0080\u0099t necessary\u00E2\u0080\u009D; they do not volunteer it. 217 (15) a. ? \u00C3\u00AA-k\u00C3\u00AE-k\u00C3\u00AEs-kimiwahk, Tom k\u00C3\u00A2-pimoht\u00C3\u00AAt ADJUNCT \u00C3\u00AA- k\u00C3\u00AE- k\u00C3\u00AEs- kimiwan-k T k\u00C3\u00A2-pim\u00C3\u00B4ht\u00C3\u00AA -t C1-PREV-FINISH-rain.VII -0 T C2-walk.VAI-3 \u00E2\u0080\u0098It had stopped raining when Tom went walking.\u00E2\u0080\u0099 b. \u00C3\u00AA-k\u00C3\u00AE-k\u00C3\u00AEs-kimiwahk Tom k\u00C3\u00A2-pimoht\u00C3\u00AAt \u00C3\u00AA- k\u00C3\u00AE- k\u00C3\u00AEs- kimiwan-k T k\u00C3\u00A2-pim\u00C3\u00B4ht\u00C3\u00AA -t C1-PREV-FINISH-rain.VII -0 T C2-walk.VAI-3 \u00E2\u0080\u0098It had stopped raining when Tom went walking.\u00E2\u0080\u0099 This is expected in that adjoined clauses are also syntactically integrated into a higher clause. 5.3.3.3 Mediated argument clauses do not require an intonational break Mediated argument clauses pattern with adjoined clauses: they do not require an intonational break between the two clauses. Thus the intonational break in (16a) is dispreferred; while an integrated prosodic contour (i.e., without a pause) is appropriate (16b). (16) a. ? ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim\u00C3\u00AAw, anihi iskw\u00C3\u00AAwa ka-nimihitoyit MEDIATED ARGUMENT ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim -\u00C3\u00AA -w anihi iskw\u00C3\u00AAw -a ka-nimihito -yi -t DEM.AN man want.VTA -DIR-3 DEM.OBV woman -OBV IRR-dance.VAI-DEP-3 \u00E2\u0080\u0098That man wants that woman to dance.\u00E2\u0080\u0099 b. ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim\u00C3\u00AAw anihi iskw\u00C3\u00AAwa ka-nimihitoyit ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim -\u00C3\u00AA -w anihi iskw\u00C3\u00AAw -a ka-nimihito -yi -t DEM.AN man want.VTA -DIR-3 DEM.OBV woman -OBV IRR-dance.VAI-DEP-3 \u00E2\u0080\u0098That man wants that woman to dance.\u00E2\u0080\u0099 5.3.4 Interim summary In this section I showed that chained clauses have a set of properties that can be derived from analyzing them as excluded clauses: (i) since they are not c-commanded by any other clause, they are sensitive to precedence; (ii) as excluded clauses, we have independent evidence that they can occur in matrix clause environments; 218 (iii) an intonational break is required between chained clauses, indicating that they are not as closely syntactically integrated. In the next section I apply the second set of diagnostics, those that pick out mediated argument clauses as opposed to any other clauses. 5.4 Applying the island tests A classic test for argument-adjunct (or, more neutrally, argument/non-argument) distinctions is island effects \u00E2\u0080\u0093 the (in)ability of an element to escape out a clause (CP) and occur in a non-local position relative to it, as illustrated in (17) (Ross 1967, Chomsky 1977, Huang 1982, Manzini 1992, among many others). (17) ... \u00CE\u00B1 \u00E2\u0080\u00A6 XP \u00E2\u0080\u00A6 [CP \u00E2\u0080\u00A6 Since at least Ross (1967) it has been noticed that CPs which are adjoined to a higher CPs, rather than being associated with an argument position, act as islands for various kinds of movement operations (e.g., wh-movement, focus-movement). (18) Extraction from clause associated with object position a. Nettie knows I like ice cream. [CP [IP Nettie knows [CP I like ice cream ] ] ] b. What does Nettie know I like? [CP What does [IP Nettie know [CP I like ti ] ] ] 219 (19) Extraction from adjoined clause a. Denver laughs when I eat ice cream. [CP [IP Denver laughs ] [CP when I eat ice cream ] ] b. * What does Denver laugh when I eat? [CP What does [IP Denver laugh ] [CP when I eat ti ] ] Thus there is said to be an argument-adjunct asymmetry with respect to island effects. In Plains Cree there is evidence that the asymmetry is broader. On the one hand, not only adjoined clauses but also chained clauses are islands. On the other hand, a clause that is associated with an argument position, rather than itself being an argument, is enough to allow an element to escape out of it. I consider three kinds of elements that can front in Plains Cree2. Wh-words, quantifiers, and argument expressions can all front across an intervening clause if the clause they are associated with is an mediated argument clause. 5.4.1 Long distance wh-construal must be across mediated arguments In a wh-construction, there is an operator in Spec, CP associated with a gap (20a). Cross- linguistically, this operator is sensitive to clause-boundaries: it cannot move out of an adjunct clause (cf. Huang\u00E2\u0080\u0099s Condition on Extraction Domains 1982). (20) a. [CP Whati [IP \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 ] ] b. * [CP Whati \u00E2\u0080\u00A6 [CP because \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 ] 2 The properties of quantifiers and wh-words have been discussed in the literature for Plains Cree (Blain 1997, Long 1999); the properties of argument expressions have not, to my knowledge. Neither have the fronting properties been discussed as a single class of properties. This discussion is thus more general than has been traditionally recognized for Plains Cree. 220 The invalidity of this structure accounts for the ungrammaticality of English utterances like (21), where the wh-word who is supposed to be associated with the object of saw. (21) * Who did John kiss Mary because he saw? In Plains Cree, Blain (1997) argues that wh-words are generated external to the clause, and the operator within the clause is null (22).3 In long-distance extraction, however, the null operator must obey the same conditions that the overt operator in English does; we expect adjunct clauses, e.g., a reason clause introduced by os\u00C3\u00A2m \u00E2\u0080\u0098because,\u00E2\u0080\u0099 to behave like adjunct clauses in English. (22) a. [ whi ] [CP Opi [IP \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6] b. * [whi ] [CP Opi \u00E2\u0080\u00A6 [CP os\u00C3\u00A2m \u00E2\u0080\u0098because\u00E2\u0080\u0099 [IP \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 ] ] ] Extending the logic of Huang (1982) and following work (e.g., Chomsky 1986, Manzini 1992), we expect chained clauses to behave like adjunct clauses. Since they are not licensed by an argument position, long-distance wh-construal should be impossible. We thus expect a bifurcation between chained and adjoined clauses on the one hand, and mediated arguments on the other. CHAINED ADJOINED MEDIATED ARG. Long distance wh-fronting \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 5.8. Only mediated argument clauses allow long-distance wh-words The following subsections give the data supporting this generalization. In this section I start with the mediated argument clauses to show that the construction is possible, and then move to adjoined and chained clauses. 3 Differences in behaviour of wh-questions between Plains Cree and English include lack of multiple wh-questions; presence of a dedicated yes-no question particle (cf. Cheng 1991), fixed ordering of the wh-word even in echo questions, and non-obligatory agreement (e.g., animacy, obviation) between the wh-word and the argument. See Blain (1997) for details and chapter 6 for more discussion. 221 5.4.1.1 Wh-words can be long distance with mediated argument clauses Mediated argument clauses allow the wh-word to be long distance from the clause which contains the gap that the wh-word is associated with. For example, in (23) the argument wh- word aw\u00C3\u00AEna \u00E2\u0080\u0098who\u00E2\u0080\u0099 is associated with the object position in John \u00C3\u00AA-oc\u00C3\u00AAm\u00C3\u00A2t \u00E2\u0080\u0098John kissed x\u00E2\u0080\u0099. (23) Argument extraction from a mediated object clause aw\u00C3\u00AEna \u00C3\u00AA-itw\u00C3\u00AAyan \u00C3\u00AA-it\u00C3\u00AAyihtaman John \u00C3\u00AA-oc\u00C3\u00AAm\u00C3\u00A2t aw\u00C3\u00AEna \u00C3\u00AA- itw\u00C3\u00AA -yan \u00C3\u00AA- it\u00C3\u00AAyihtam -an J \u00C3\u00AA- oc\u00C3\u00AAm -\u00C3\u00A2 -t who C1-say.VAI-2 C1-think.VTI -2 J C1-kiss.VTA-DIR-3 \u00E2\u0080\u0098who did you say you think John kissed?\u00E2\u0080\u0099 (Blain 1997:186) [ wh ]i [CP Opi \u00C3\u00AA-itw\u00C3\u00AAyan [CP \u00C3\u00AA-it\u00C3\u00AAyihtaman [CP John \u00C3\u00AA-oc\u00C3\u00AAm\u00C3\u00A2t ti ] ] ] The wh-word aw\u00C3\u00AEna \u00E2\u0080\u0098who\u00E2\u0080\u0099 in (24) is likewise associated with the subject position of the embedded clause. (24) aw\u00C3\u00AEna \u00C3\u00B4ma Tom \u00C3\u00AA-w\u00C3\u00AEhtam\u00C3\u00A2sk \u00C3\u00AA-cihk\u00C3\u00AAyihtamiyit ayis nim\u00C3\u00A2ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt aw\u00C3\u00AEna \u00C3\u00B4ma \u00C3\u00AA- w\u00C3\u00AEhtamaw-isk \u00C3\u00AA- cihk\u00C3\u00AAyihtam-yi-t ayis ni-mama \u00C3\u00AA-p\u00C3\u00AA- itoht\u00C3\u00AA -t who DEM.INAN C1-tell.VTA -3>2 C1-happy.VTI -DS-3 for 1-mom C1-COME-go.VAI-3 \u00E2\u0080\u0098Who is it that Tom told you is happy because my mother came to visit?\u00E2\u0080\u0099 Adjunct wh-words may also front from a mediated argument clause, as in (25), where the manner wh-word t\u00C3\u00A2nis\u00C3\u00AEsi \u00E2\u0080\u0098how\u00E2\u0080\u0099 is questioning the manner of leaving in the lower clause (as can be seen the by the extraction marker isi in that clause; cf. Wolfart 1973, Cook 2004 for details). (25) Adjunct extraction from a mediated object clause t\u00C3\u00A2ns\u00C3\u00AEsi \u00C3\u00AA-k\u00C3\u00AE-itw\u00C3\u00AAt Misti \u00C3\u00AA-isi-sipw\u00C3\u00AAht\u00C3\u00AAyit W\u00C3\u00A2pastimwa t\u00C3\u00A2nis\u00C3\u00AEs \u00C3\u00AA- k\u00C3\u00AE- itw\u00C3\u00AA -t M \u00C3\u00AA- isi- sipw\u00C3\u00AAht\u00C3\u00AA -yi -t W-a how C1-PREV-say.VAI-3 M C1-THUS-leave.VAI-DS-3 W-OBV \u00E2\u0080\u0098What did Misti say was the way that W\u00C3\u00A2pastim walked away?\u00E2\u0080\u0099 In Plains Cree, subject-oriented clauses and object-oriented clauses pattern together with respect to long-distance wh-movement (cf. Wiltschko 1995 on German extraposed clauses). In (26), we have an intransitive verb \u00C3\u00AA-miyw\u00C3\u00A2sik \u00E2\u0080\u0098it is good\u00E2\u0080\u0099 with an inanimate subject; the mediated subject clause John \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt \u00E2\u0080\u0098John came\u00E2\u0080\u0099 is the subject of the verb \u00C3\u00AA-miyw\u00C3\u00A2sik. 222 (26) Argument extraction from a mediated subject clause aw\u00C3\u00AEna \u00C3\u00B4ma \u00C3\u00AA-miyw\u00C3\u00A2sik \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt aw\u00C3\u00AEna \u00C3\u00B4ma \u00C3\u00AA- miyw\u00C3\u00A2si -k \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t who DEM.INAN C1-good.VII -0 C1-COME-go.VAI-3 \u00E2\u0080\u0098Who was it that came over that made it nice?\u00E2\u0080\u0099 Thus both mediated-subject and mediated-object clauses allow the wh-word to front across the higher clause. 5.4.1.2 Wh-words cannot be long distance with adjoined clauses As Blain (1997) shows, a wh-question cannot be construed with a position in an adjoined clause. In (27a) we see that the wh-word aw\u00C3\u00AEnihi \u00E2\u0080\u0098who.OBV\u00E2\u0080\u0099 cannot be associated with an object position of the adjoined clause t\u00C3\u00A2n\u00C3\u00AAhki k\u00C3\u00A2-p\u00C3\u00B4n-kiyokaw\u00C3\u00A2t \u00E2\u0080\u0098why she stopped visiting him/her\u00E2\u0080\u0099. And in (27b), the wh-word aw\u00C3\u00AEna \u00E2\u0080\u0098who\u00E2\u0080\u0099 cannot be associated with a subject position of the adjoined clause os\u00C3\u00A2m \u00C3\u00AA-p\u00C3\u00AEkon\u00C3\u00A2t \u00E2\u0080\u0098because s/he broke it\u00E2\u0080\u0099. (27) a. [aw\u00C3\u00AEnihii ] [CP Opi \u00E2\u0080\u00A6 [CP t\u00C3\u00A2n\u00C3\u00AAhki [IP \u00E2\u0080\u00A6 k\u00C3\u00A2-p\u00C3\u00B4n-kiyokaw\u00C3\u00A2t obji \u00E2\u0080\u00A6 ] ] ] * aw\u00C3\u00AEnihi k\u00C3\u00A2-kakw\u00C3\u00AAcimat Mary t\u00C3\u00A2n\u00C3\u00AAhki k\u00C3\u00A2-p\u00C3\u00B4n-kiyokaw\u00C3\u00A2t aw\u00C3\u00AEna -hi k\u00C3\u00A2- kakw\u00C3\u00AAcim -at M t\u00C3\u00A2n\u00C3\u00AAhki k\u00C3\u00A2- p\u00C3\u00B4n- kiyokaw -\u00C3\u00A2 -t who -OBV C2- ask.VTA -2>3 M Q.why C2- STOP-visit.VTA -DIR-3 --- (intended: \u00E2\u0080\u0098Who did you ask Mary why she stopped visiting?\u00E2\u0080\u0099) (Blain 1997) b. [aw\u00C3\u00AEnai ] [CP Opi \u00E2\u0080\u00A6 [CP os\u00C3\u00A2m [IP \u00E2\u0080\u00A6 subji \u00C3\u00AA-p\u00C3\u00AEkon\u00C3\u00A2t \u00E2\u0080\u00A6 ] ] ] * aw\u00C3\u00AEna k\u00C3\u00A2-m\u00C3\u00A2toy\u00C3\u00A2n os\u00C3\u00A2m \u00C3\u00AA-p\u00C3\u00AEkon\u00C3\u00A2t kitaw\u00C3\u00A2sisihk\u00C3\u00A2na aw\u00C3\u00AEna k\u00C3\u00A2-m\u00C3\u00A2to -y\u00C3\u00A2n os\u00C3\u00A2m \u00C3\u00AA-p\u00C3\u00AEkon -\u00C3\u00A2 -t ki(t)-aw\u00C3\u00A2sisihk\u00C3\u00A2n-a who C2-cry.VAI-1 because C1-break.VTA-DIR-3 2- doll -OBV --- (intended: \u00E2\u0080\u0098Who did you cry because \u00E2\u0080\u00A6 broke your doll?\u00E2\u0080\u0099) (Blain 1997:189) Similarly, (28) provides a minimal pair showing that while the wh-word aw\u00C3\u00AEna \u00E2\u0080\u0098who\u00E2\u0080\u0099 can be associated with a long-distance subject of a mediated argument clause, it cannot be associated with the long-distance subject of an adjoined clause. For example, \u00C3\u00AA-cik\u00C3\u00AAyihtamiyit \u00E2\u0080\u0098someone is happy\u00E2\u0080\u0099 is associated with an argument position of \u00C3\u00AA-wihtam\u00C3\u00A2sk \u00E2\u0080\u0098he told x to you\u00E2\u0080\u0099 in (28a). In (28b), we have a reason clause introduced by the subordinator ayis \u00E2\u0080\u0098for/because\u00E2\u0080\u0099 and extraction from this clause is impossible. 223 (28) a. Extraction from subject of mediated-argument clause aw\u00C3\u00AEna \u00C3\u00B4ma Tom \u00C3\u00AA-w\u00C3\u00AEhtam\u00C3\u00A2sk \u00C3\u00AA-cihk\u00C3\u00AAyihtamiyit ayis nim\u00C3\u00A2ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt aw\u00C3\u00AEna \u00C3\u00B4ma \u00C3\u00AA- w\u00C3\u00AEhtamaw-isk \u00C3\u00AA- cihk\u00C3\u00AAyihtam-yi-t ayis ni-m\u00C3\u00A2ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AA-t who DEM.INAN C1-tell.VTA -3>2 C1-happy.VTI-DS-3 for 1-mom C1-COME-go.VAI-3 \u00E2\u0080\u0098Who is that Tom told you is happy because my mother came to visit?\u00E2\u0080\u0099 b. Extraction from subject of adjoined clause * aw\u00C3\u00AEna \u00C3\u00B4ma Tom \u00C3\u00AA-w\u00C3\u00AEhtam\u00C3\u00A2sk \u00C3\u00AA-cihk\u00C3\u00AAyihtam\u00C3\u00A2n ayis \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt aw\u00C3\u00AEna \u00C3\u00B4ma \u00C3\u00AA- w\u00C3\u00AEhtamaw -isk \u00C3\u00AA- cihk\u00C3\u00AAyihtam-\u00C3\u00A2n ayis \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t who DEM.INAN C1-tell.VTA -3>2 C1-happy.VTI -1 for C1-COME-go.VAI-3 --- (intended: \u00E2\u0080\u0098Whoi did Tom tell you I was happy because ti came to visit?\u00E2\u0080\u0099) 5.4.1.3 Wh-words cannot be long distance with chained clauses Finally, wh-words cannot be construed with an argument in a chained clause across the antecedent clause. In (29a), kahkiyaw m\u00C3\u00AEnisa \u00E2\u0080\u0098all the berries\u00E2\u0080\u0099 is the object of the second clause; in (29b) I have attempted to construct a wh-word associated with this object position of the chained clause, but the result is ungrammaticality. (29) a. aw\u00C3\u00A2sisak kw\u00C3\u00AAsimocikiht\u00C3\u00A2wak, kahkiyaw m\u00C3\u00AEnisa \u00C3\u00AA-m\u00C3\u00AEcisocik CHAIN aw\u00C3\u00A2sis -ak kw\u00C3\u00AAsimocikiht\u00C3\u00A2 -w -ak kahkiyaw m\u00C3\u00AEnis -a \u00C3\u00AA- m\u00C3\u00AEciso -t -k child -PL have.fun.VAI -3 -PL ALL berry -XT C1-eat.VAI-3 -PL \u00E2\u0080\u0098The children were having a lot of fun, they ate all the berries.\u00E2\u0080\u0099 b. * [k\u00C3\u00AEkwayi ] [CP Opi \u00E2\u0080\u00A6 ] [CP [IP \u00E2\u0080\u00A6 \u00C3\u00AA-m\u00C3\u00AEcisocik obji \u00E2\u0080\u00A6 ] ] ] * k\u00C3\u00AEkway awasisak kw\u00C3\u00AAsimocikiht\u00C3\u00A2wak \u00C3\u00AA-m\u00C3\u00AEcisocik k\u00C3\u00AEkway aw\u00C3\u00A2sis -ak kw\u00C3\u00AAsimocikiht\u00C3\u00A2 -w -ak \u00C3\u00AA- m\u00C3\u00AEciso -t -k what child -PL have.fun.VAI -3 -PL C1-eat.VAI-3 -PL --- (intended: \u00E2\u0080\u0098Whati did the children have fun, they ate ti?\u00E2\u0080\u0099) I now turn to another domain in which we see long-distance phenomena: quantification. 224 5.4.2 Long distance quantifier-fronting must be across mediated arguments It has been noticed for some time that quantifiers in Plains Cree, like other languages of the Cree dialect continuum, may be discontinuous from the argument over which they quantify (Matthewson & Reinholtz 1996, Reinholtz 1995, 1999; Russell & Reinholtz 1996; Wolvengrey 2003, among others). An example of this is given in (30), where aw\u00C3\u00A2sisak \u00E2\u0080\u0098children\u00E2\u0080\u0099 can either immediately follow the quantifier n\u00C3\u00AEso \u00E2\u0080\u0098two\u00E2\u0080\u0099, or can occur at the end of the utterance, with the verbal complex k\u00C3\u00AE-p\u00C3\u00A2hpiwak \u00E2\u0080\u0098they laughed\u00E2\u0080\u0099 intervening. (30) a. \u00C3\u00B4ki n\u00C3\u00AEso aw\u00C3\u00A2sisak k\u00C3\u00AE-p\u00C3\u00A2hpiwak \u00C3\u00B4ki n\u00C3\u00AEso aw\u00C3\u00A2sis-ak k\u00C3\u00AE- p\u00C3\u00A2hpi -w-ak DEM two child -PL PREV-laugh.VAI-3-PL \u00E2\u0080\u0098These two children laughed.\u00E2\u0080\u0099 b. \u00C3\u00B4ki n\u00C3\u00AEso k\u00C3\u00AE-p\u00C3\u00A2hpiwak aw\u00C3\u00A2sisak \u00C3\u00B4ki n\u00C3\u00AEso k\u00C3\u00AE- p\u00C3\u00A2hpi -w-ak aw\u00C3\u00A2sis-ak DEM two PREV-laugh.VAI-3-PL child -PL \u00E2\u0080\u0098These two children laughed.\u00E2\u0080\u0099 (from Wolvengrey 2003: 9a-b) Here I show that an entire clause may intervene between the quantifier and the element being quantified over. With respect to the syntactic classification proposed in this chapter, we expect the same patterns as for wh-words, as summarized in table 5.9. CHAINED ADJOINED MEDIATED ARG. Long distance wh-fronting \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Long distance quantifier-fronting \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 5.9. Only mediated argument clauses allow long-distance quantifier-fronting The next subsections bear this out. 225 5.4.2.1 Mediated argument clauses permit long distance quantifier-fronting As expected, mediated argument clauses allow quantifiers to escape as schematized in (31); such examples are found in discourses, and judged acceptable in elicitation contexts. (31) [ Qj ] [CP ...argi ... [CP ... DPj \u00E2\u0080\u00A6]i ] MEDIATED ARGUMENT For example, in (32), the partitive quantificational phrase mihc\u00C3\u00AAt aniki \u00E2\u0080\u0098many of them\u00E2\u0080\u0099 is in initial position, and the nominal iskw\u00C3\u00AAwak \u00E2\u0080\u0098women\u00E2\u0080\u0099 is in final position. They are separated not only by the clause with which the quantification phrase is associated (m\u00C3\u00AEkiskihkahcik\u00C3\u00AA- \u00E2\u0080\u0098do.beadwork.VAI\u00E2\u0080\u0099), but by the matrix clause kaskiht\u00C3\u00A2- \u00E2\u0080\u0098succeed.at.VAI\u00E2\u0080\u0099. (32) mihc\u00C3\u00AAt aniki \u00C3\u00AA-k\u00C3\u00AE-kaskiht\u00C3\u00A2cik aya \u00C3\u00AA-m\u00C3\u00AEkisihkahcik\u00C3\u00AAcik \u00C3\u00AAkosp\u00C3\u00AE iskw\u00C3\u00AAwak, \u00E2\u0080\u00A6 mihc\u00C3\u00AAt aniki \u00C3\u00AA- k\u00C3\u00AE- kaskiht\u00C3\u00A2 -t -ik aya \u00C3\u00AA- m\u00C3\u00AEkisihkahcik\u00C3\u00AA -t-ik \u00C3\u00AAkosp\u00C3\u00AE iskw\u00C3\u00AAw-ak many DEM.AN C1-PREV -able.VAI-3-PL CONN C1-do.beading -3-PL then woman-PL \u00E2\u0080\u0098Many of the women used to be able to do beadwork then, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 48) This is also shown in (33) where kahkiyaw \u00E2\u0080\u0098all\u00E2\u0080\u0099 may occur either adjacent to the subject of the lower clause (33a)4 or in initial position with the matrix clause intervening (33b): in both cases it quantifies over iskw\u00C3\u00AAwa \u00E2\u0080\u0098women\u00E2\u0080\u0099. (33) a. J p\u00C3\u00AAhtaw\u00C3\u00AAw kahkiyaw iskw\u00C3\u00AAwa \u00C3\u00AA-nikamoyit J p\u00C3\u00AAhtaw -\u00C3\u00AA -w kahkiyaw iskw\u00C3\u00AAw -a \u00C3\u00AA- nikamo -yi -t J hear.VTA-DIR-3 all woman -OBV C1-sing.VAI-DS -3 \u00E2\u0080\u0098Jeff heard all the women singing.\u00E2\u0080\u0099 b. kahkiyaw J p\u00C3\u00AAhtaw\u00C3\u00AAw iskw\u00C3\u00AAwa \u00C3\u00AA-nikamoyit kahkiyaw J p\u00C3\u00AAhtaw -\u00C3\u00AA -w iskw\u00C3\u00AAw -a \u00C3\u00AA- nikamo -yi -t all J hear.VTA-DIR-3 woman -OBV C1-sing.VAI-DS -3 \u00E2\u0080\u0098Jeff heard all the women singing.\u00E2\u0080\u0099 Demonstratives, which pattern like quantifiers with respect to their discontinuous properties (Reinholtz 1999, Wolvengrey 2003), may also front long distance out of a mediated argument clause. 4 On independent grounds, quantifiers have been shown be restricted from occurring after the verbal complex of the clause they are associated with (see, for example, Reinholtz 1995, 1999, Matthewson & Reinholtz 1998); this rules out the possibility that kahkiyaw is just quantifying over the object of p\u00C3\u00AAhtaw\u00C3\u00AAw in (33a). 226 (34) Demonstratives front out of mediated argument clauses a. nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-n\u00C3\u00AEmihitocik \u00C3\u00B4ki n\u00C3\u00A2p\u00C3\u00AAwak ni- kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- nimihito -t -k \u00C3\u00B4ki n\u00C3\u00A2p\u00C3\u00AAw -ak 1- know.VTI -SAP C1-dance.VAI-3 -PL DEM.AN man -PL \u00E2\u0080\u0098I know these men are dancing.\u00E2\u0080\u0099 b. \u00C3\u00B4ki nikisk\u00C3\u00AAyihten \u00C3\u00AA-n\u00C3\u00AEmihtocik n\u00C3\u00A2p\u00C3\u00AAwak \u00C3\u00B4ki ni-kisk\u00C3\u00AAyiht\u00C3\u00AA-n \u00C3\u00AA- nimihito -t -k n\u00C3\u00A2p\u00C3\u00AAw-ak DEM.AN 1-know.VTI -SAP C1-dance.VAI-3-PL man -PL \u00E2\u0080\u0098These men here, I know that they\u00E2\u0080\u0099re dancing.\u00E2\u0080\u0099 5.4.2.2 Adjoined clauses do not permit long distance quantifier-fronting As expected, adjoined clauses do not permit long-distance quantifier fronting. This is schematized in (35). (35) * [Qi ] [CP [C ... ] [CP ...DPi ... ] ] This is exemplified by a clause with the causal subordinator ayis \u00E2\u0080\u0098for/because\u00E2\u0080\u0099 and the subordinator \u00C3\u00AA-. In (36a), the quantificational element is continuous with the nominal; in (36b), kahkiyaw \u00E2\u0080\u0098all\u00E2\u0080\u0099 has been fronted, but the result is ungrammatical. (36) Long distance quantifier-fronting not allowed in adjoined clauses a. nik\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00A2n ayis kahkiyaw n\u00C3\u00A2p\u00C3\u00AAwak \u00C3\u00AA-nimihitocik ni- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00A2 -n ayis kahkiyaw n\u00C3\u00A2p\u00C3\u00AAw -ak \u00C3\u00AA- nimihito -t -k 1- PREV-leave.VAI -SAP for all man -PL C1-dance.VAI-3 -PL \u00E2\u0080\u0098I left because all the men were dancing.\u00E2\u0080\u0099 b. * kahkiyaw nik\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00A2n ayis \u00C3\u00AA-nimihitocik n\u00C3\u00A2p\u00C3\u00AAwak kahkiyaw ni-k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00A2 -n ayis \u00C3\u00AA- nimihito -t -k n\u00C3\u00A2p\u00C3\u00AAw -ak all 1- PREV-leave.VAI -SAP for C1-dance.VAI-3-PL man -PL --- (intended: \u00E2\u0080\u0098I left because all the men were dancing.\u00E2\u0080\u0099) 227 5.4.2.3 Chained clauses do not permit long distance quantifier-fronting Chained clauses do not allow discontinuous quantifiers (as in the structure in (37)) either. (37) * [Qi ] [CP ... ] [CP ...DPi ... ] For example, in (38a) we have the quantificational phrase kahkiyaw m\u00C3\u00AEnisa \u00E2\u0080\u0098all the berries\u00E2\u0080\u0099; in (38b) the quantifier kahkiyaw \u00E2\u0080\u0098all\u00E2\u0080\u0099 has been displaced to the front, yielding ungrammaticality. (38) Long distance quantifier-fronting not allowed in chained clauses a. aw\u00C3\u00A2sisak kw\u00C3\u00AAsimocikihtawak, kahkiyaw m\u00C3\u00AEnisa \u00C3\u00AA-m\u00C3\u00AEcisocik aw\u00C3\u00A2sis -ak kw\u00C3\u00AAsimocikiht\u00C3\u00A2 -w -ak kahkiyaw m\u00C3\u00AEnisa \u00C3\u00AA- m\u00C3\u00AEciso -t -k child -PL have.fun.VAI -3 -PL ALL berries C1-eat.VAI-3 -PL \u00E2\u0080\u0098The children were having a lot of fun, they ate all the berries.\u00E2\u0080\u0099 b. * kahkiyaw n\u00C3\u00A2p\u00C3\u00AAsis kw\u00C3\u00AAsimohcikiht\u00C3\u00A2w m\u00C3\u00AEnisa \u00C3\u00AA-m\u00C3\u00AEcisot kahkiyaw n\u00C3\u00A2p\u00C3\u00AAsis kw\u00C3\u00AAsimohcikiht\u00C3\u00A2-w m\u00C3\u00AEnisa \u00C3\u00AA- m\u00C3\u00AEciso -t all boy have.fun.VAI -3 berries C1-eat.VAI -3 --- (intended: \u00E2\u0080\u0098The boy was having fun, he ate all the berries.\u00E2\u0080\u0099) 5.4.3 Long distance argument-expression fronting The same kinds of clausal dependencies that are transparent for discontinuous quantifiers also prove to be transparent for some other discontinuous constituents. We once again expect that this will only be possible with mediated argument clauses, as summarized in table 5.10. CHAINED ADJOINED MEDIATED ARG. Long distance wh-fronting \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Long distance quantifier-fronting \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Long distance argument-fronting \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 5.10. Only mediated argument clauses allow long-distance argument expressions 228 5.4.3.1 Mediated argument clauses permit long distance argument-fronting Non-quantificational arguments may also be discontinuous from the clause they are associated with. In (39), the argument n\u00C3\u00A2p\u00C3\u00AAwak \u00E2\u0080\u0098men\u00E2\u0080\u0099 is the subject of the clause \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAcik \u00E2\u0080\u0098they came\u00E2\u0080\u0099. (39a) shows n\u00C3\u00A2p\u00C3\u00AAwak \u00E2\u0080\u0098men\u00E2\u0080\u0099 adjacent to the clause with which it is associated; however in (39b) the matrix clause nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00E2\u0080\u0098I know it\u00E2\u0080\u0099 intervenes. Even though n\u00C3\u00A2p\u00C3\u00AAwak \u00E2\u0080\u0098men\u00E2\u0080\u0099 is not associated with any argument position in this matrix clause, the utterance is fine. (39) a. nikisk\u00C3\u00AAyiht\u00C3\u00AAn n\u00C3\u00A2p\u00C3\u00AAwak \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAcik ni-kisk\u00C3\u00AAyiht\u00C3\u00AA-n n\u00C3\u00A2p\u00C3\u00AAw-ak \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t-k 1-know.VTI -SAP man -PL C1-COME-go.VAI-3-PL \u00E2\u0080\u0098I know the men came.\u00E2\u0080\u0099 b. n\u00C3\u00A2p\u00C3\u00AAwak nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAcik n\u00C3\u00A2p\u00C3\u00AAw-ak ni-kisk\u00C3\u00AAyiht\u00C3\u00AA-n \u00C3\u00AA-p\u00C3\u00AA- itoht\u00C3\u00AA -t-k man -PL 1-know.VTI -SAP C1-COME-go.VAI-3-PL \u00E2\u0080\u0098I know the men came.\u00E2\u0080\u0099 Whole argument phrase (e.g., with quantifiers) may also occur in initial position, with an intervening clause between it and the clause with which it is syntactically associated. For example, in (40) kahkiyaw n\u00C3\u00A2p\u00C3\u00AAwak \u00E2\u0080\u0098all the men\u00E2\u0080\u0099 is the subject of \u00C3\u00AA-nimihitocik \u00E2\u0080\u0098they are dancing\u00E2\u0080\u0099, but occurs in initial position. (40) kahkiyaw n\u00C3\u00A2p\u00C3\u00AAwak nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-nimihitocik kahkiyaw n\u00C3\u00A2p\u00C3\u00AAw -ak ni- kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- nimihito -t -k all man -PL 1- know.VTI -SAP C1- dance.VAI-3-PL \u00E2\u0080\u0098I know all the men are dancing.\u00E2\u0080\u0099 Examples of fronted argument expressions also occur in corpora. For example, in (41), the quantified phrase kahkiyaw \u00C3\u00AAkoni \u00E2\u0080\u0098all these things\u00E2\u0080\u0099 is the object of the lower clause ka-t\u00C3\u00B4tam\u00C3\u00A2hk \u00E2\u0080\u0098us to do\u00E2\u0080\u0099. (41) kahkiyaw \u00C3\u00AAkoni \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEhtam\u00C3\u00A2kawiy\u00C3\u00A2hk aya ka-t\u00C3\u00B4tam\u00C3\u00A2hk, ... kahkiyaw \u00C3\u00AAkoni \u00C3\u00AA- k\u00C3\u00AE- [w\u00C3\u00AEht-amaw]-ikawi-y\u00C3\u00A2n-k aya ka- t\u00C3\u00B4tam-\u00C3\u00A2n-k all TOPIC C1-PREV-tell.VTA-APPL-USC -1 -PL CONN IRR-do.VTI-1 -PL \u00E2\u0080\u0098all these things she used to tell us to do, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 49) Here the upper clause is inflected for three arguments: the subject is suppressed via the unspecified subject -ikawi, the object is morphologically expressed via the 1st-person plural 229 object -y\u00C3\u00A2hk, and the applicative morpheme -amaw introduces an indirect object argument, with which the dependent clause is associated. While the presence of this applicative argument might at first glance appear to allow kahkiyaw \u00C3\u00AAkoni \u00E2\u0080\u0098all these things\u00E2\u0080\u0099 to be an argument of the upper clause, this analysis does not work on at least two counts. First, this analysis would have to posit that the lower clause is an adjunct clause. However, in the absence of an overt subordinating particle, adjunct ka-clauses are interpreted as purpose clauses (cf. chapter 7), which is not what is happening here. Consider the contrast between (41) above and (42), given here: (42) \u00E2\u0080\u009Ckis\u00C3\u00AA-manitow \u00C3\u00B4m \u00C3\u00AA-k\u00C3\u00AE-os\u00C3\u00AEh\u00C3\u00A2t ayisiyiniwa, ka-miyaw\u00C3\u00A2tamiyit, \u00E2\u0080\u00A6\u00E2\u0080\u009D kis\u00C3\u00AA-manitow \u00C3\u00B4ma \u00C3\u00AA- k\u00C3\u00AE- os\u00C3\u00AEh -\u00C3\u00A2 -t ayisiyiniw-a ka- miyaw\u00C3\u00A2tam -yi -t God DEM.INAN C1-PREV-make.VTA-DIR-3 person -OBV IRR-happy.VTI -DS -3 \u00E2\u0080\u0098 \u00E2\u0080\u009CGod has created man to be happy, \u00E2\u0080\u00A6\u00E2\u0080\u009D \u00E2\u0080\u0099 (EM 37) Second, an analysis where kahkiyaw \u00C3\u00AAkoni \u00E2\u0080\u0098everything\u00E2\u0080\u0099 is the indirect object of wihtamaw- \u00E2\u0080\u0098tell.VTA\u00E2\u0080\u0099 claims that the speaker is being told about something (i.e., kahkiyaw \u00C3\u00AAkoni \u00E2\u0080\u0098all these things\u00E2\u0080\u0099), rather than directed to do something (i.e., ka-totam\u00C3\u00A2hk \u00E2\u0080\u0098what we should do\u00E2\u0080\u0099). Even oblique arguments that are not subcategorized for by the verbal complex (and which generally have more restricted ordering properties; cf. Dahlstrom 1995, M\u00C3\u00BChlbauer 2003) can be discontinuous from the clause they are associated with. For example, in (43), the modifier nan\u00C3\u00A2tohk isi \u00E2\u0080\u0098in various ways\u00E2\u0080\u0099 is associated with the \u00C3\u00AA-w\u00C3\u00AEcihikot \u00E2\u0080\u0098she helped him\u00E2\u0080\u0099. (43) nan\u00C3\u00A2tohk isi m\u00C3\u00AEn \u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00A2pahtam\u00C3\u00A2n \u00C3\u00AA-w\u00C3\u00AEcihikot aya ow\u00C3\u00AEkim\u00C3\u00A2kana, \u00E2\u0080\u00A6 nan\u00C3\u00A2tohk isi m\u00C3\u00AEna \u00C3\u00AA- k\u00C3\u00AE- w\u00C3\u00A2pahtam-\u00C3\u00A2n \u00C3\u00AA- w\u00C3\u00AEcih -iko -t aya o-w\u00C3\u00AEkim\u00C3\u00A2kan-a various way also C1-PREV-see.VTI -1 C1-help.VTA-INV -3 CONN 3-wife -OBV \u00E2\u0080\u0098I also saw that his wife helped him in various ways, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 43 presented in elicitation) comment: the nan\u00C3\u00A2tohk isi is telling you about all the ways the wife helped her husband. Likewise, in (44), the topic marker \u00C3\u00AAwakw \u00C3\u00A2nima \u00E2\u0080\u0098that\u00E2\u0080\u0099 (referring back to the immediately previous discourse) is associating with the relative root isi \u00E2\u0080\u0098this way\u00E2\u0080\u0099; \u00C3\u00AAwakw\u00C3\u00A2nima describes the way the speaker saw her husband. 230 (44) \u00C3\u00AAwakw \u00C3\u00A2nima k\u00C3\u00AAhcin\u00C3\u00A2 aya \u00C3\u00AA-k\u00C3\u00AE-miyw\u00C3\u00AAyihtam\u00C3\u00A2n, \u00C3\u00AAwakw anima k\u00C3\u00AAhcina aya \u00C3\u00AA- k\u00C3\u00AE- miyw\u00C3\u00AAyihtam -\u00C3\u00A2n TOP DEM.INAN certainly CONN C1-PREV-happy.VTI -1 \u00C3\u00AA-k\u00C3\u00AE-oh-~ aya \u00C3\u00AA-k\u00C3\u00AE-isi-w\u00C3\u00A2pamak niw\u00C3\u00AEkim\u00C3\u00A2kan ... aya \u00C3\u00AA- k\u00C3\u00AE- isi- w\u00C3\u00A2pam -ak ni- w\u00C3\u00AEkim\u00C3\u00A2kan CONN C1-PREV-THUS-see.VTA -1>3 1- spouse \u00E2\u0080\u0098I certainly used to be happy that I could see my husband in this light ...\u00E2\u0080\u0099 (EM 28) 5.4.3.2 Adjoined clauses do not permit long distance argument-fronting Adjoined clauses also behave as expected: they do not permit long distance argument-fronting. Thus, for example, in (45) we have an adjoined clause introduced with the k\u00C3\u00A2- clause-typing proclitic. The argument expression aw\u00C3\u00A2sisak \u00E2\u0080\u0098children\u00E2\u0080\u0099 cannot occur to the left of the matrix clauses. (45) * aw\u00C3\u00A2sisak nik\u00C3\u00AE-m\u00C3\u00A2ton k\u00C3\u00A2-m\u00C3\u00AAtaw\u00C3\u00AAcik wayawihtamihk aw\u00C3\u00A2sis -ak ni- k\u00C3\u00AE- m\u00C3\u00A2to -n k\u00C3\u00A2- m\u00C3\u00AAtaw\u00C3\u00AA -t -k wayawihtamihk child -PL 1- PREV-cry.VAI-SAP C2- play.VAI-3-PL outside --- (intended: \u00E2\u0080\u0098I cried while the children were playing outside.\u00E2\u0080\u0099) When there is an argument phrase in an adjoined clause, it cannot occur to the left of the subordinating particle, as exemplified with ayis \u00E2\u0080\u0098for/because\u00E2\u0080\u0099 in (46). (46) a. nik\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00A2n ayis kahkiyaw n\u00C3\u00A2p\u00C3\u00AAwak \u00C3\u00AA-nimihitocik ni- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -n ayis kahkiyaw n\u00C3\u00A2p\u00C3\u00AAw-ak \u00C3\u00AA- nimihito -t -k 1- PREV-leave.VAI-SAP for all man -PL C1-dance.VAI-3-PL \u00E2\u0080\u0098I left because all the men were dancing.\u00E2\u0080\u0099 b. * nik\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00A2n kahkiyaw n\u00C3\u00A2p\u00C3\u00AAwak ayis \u00C3\u00AA-nimihitocik ni- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -n kahkiyaw n\u00C3\u00A2p\u00C3\u00AAw-ak ayis \u00C3\u00AA- nimihito -t -k 1- PREV-leave.VAI-SAP all man -PL for C1-dance.VAI-3-PL --- (intended: \u00E2\u0080\u0098I left because all the men were dancing.\u00E2\u0080\u0099) c. * kahkiyaw n\u00C3\u00A2p\u00C3\u00AAwak nik\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00A2n ayis \u00C3\u00AA-nimihitocik kahkiyaw n\u00C3\u00A2p\u00C3\u00AAw -ak ni- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00A2 -n ayis \u00C3\u00AA- nimihito -t -k all man -PL 1- PREV-leave.VAI-SAP for C1-dance.VAI-3-PL --- (intended: \u00E2\u0080\u0098I left because all the men were dancing.\u00E2\u0080\u0099) 231 5.4.3.3 Chained clauses do not permit long distance argument-fronting Chained clauses pattern with adjoined clauses: the argument cannot front out of them. For example, in (47), m\u00C3\u00AEnisa \u00E2\u0080\u0098berries\u00E2\u0080\u0099 are supposed to be associated with the object position of \u00C3\u00AA- m\u00C3\u00AEcisot \u00E2\u0080\u0098he is eating\u00E2\u0080\u0099, but it is impossible if m\u00C3\u00AEnisa \u00E2\u0080\u0098berries\u00E2\u0080\u0099 is in an initial position. (47) a. * m\u00C3\u00AEnisa n\u00C3\u00A2p\u00C3\u00AAsis kw\u00C3\u00AAsimocikiht\u00C3\u00A2w \u00C3\u00AA-m\u00C3\u00AEcisot CHAIN m\u00C3\u00AEnisa n\u00C3\u00A2p\u00C3\u00AAsis kw\u00C3\u00AAsimocik\u00C3\u00AEht\u00C3\u00A2 -w \u00C3\u00AA- m\u00C3\u00AEciso -t berries boy have.fun.VAI -3 C1-eat.VAI -3 --- (intended: \u00E2\u0080\u0098The boy was having fun, eating berries.\u00E2\u0080\u0099) 5.4.4 (Non-)obligatory switch-reference picks out object oriented clauses In chapter 4, we saw that the dependent reference marker -yi is an anaphoric different-subject marker: it marks that the subject of the predicate to which it is attached is disjoint from some clause-external argument (see also M\u00C3\u00BChlbauer 2008). (48) CP 2 2IP 2 (Subjx, \u00E2\u0089\u00A0 y) 2vP -yi 5 (M\u00C3\u00BChlbauer 2008) We saw that the relation between these two clauses must have the same configuration as the relation between an anaphor and its antecedent: either precedence or c-command. What I consider in more detail here are the conditions under which -yi marking is obligatory. Switch reference is obligatory in object-oriented clauses, but not in any other type (chained, adjoined, or subject-oriented). Chained clauses behave like coordinated clauses: in some contexts -yi is used, but it may also be omitted without changing the well-formedness of the construction. 232 (49) -yi can be omitted in chained clauses a. Jeff \u00C3\u00AA-nikamot, Clarewa \u00C3\u00AA-nimihitoyit Jeff \u00C3\u00AA- nikamo -t C-wa \u00C3\u00AA- nimihito -yi -t J C1-sing.VAI-3 C-OBV C1-dance.VAI-DS-3 \u00E2\u0080\u0098Jeff sang, Clare danced.\u00E2\u0080\u0099 b. Jeff \u00C3\u00AA-nikamot, Clare \u00C3\u00AA-nimihitot Jeff \u00C3\u00AA- nikamo -t C \u00C3\u00AA- nimihito -t J C1-sing.VAI-3 C C1-dance.VAI-3 \u00E2\u0080\u0098Jeff sang, Clare danced.\u00E2\u0080\u0099 Adjoined clauses that are introduced either by the complementizer k\u00C3\u00A2-, or by some subordinating particle also show behaviour that is parallel to coordinated clauses: -yi may occur, but it is not obligatory (cf. Long 1999). For example, a modificational clause introduced by the complementizer k\u00C3\u00A2- can have a different subject from the clause it is modifying it, and there is no need for dependent reference marking. The ordering of the two clauses has no effect (50a-b). (50) -yi can be omitted in adjoined clauses in either order a. Jeff nip\u00C3\u00A2w, Clare k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2c5-n\u00C3\u00B4tinik\u00C3\u00AAt J nip\u00C3\u00A2 -w C k\u00C3\u00A2- m\u00C3\u00AAkw\u00C3\u00A2c- n\u00C3\u00B4tinik\u00C3\u00AA -t J sleep.VAI-3 C C2-while- fight.VAI-3 \u00E2\u0080\u0098Jeff is sleeping while Clare is fighting.\u00E2\u0080\u0099 b. Jane m\u00C3\u00A2na k\u00C3\u00A2-miyopayit, Beth \u00C3\u00AA-m\u00C3\u00A2tot J m\u00C3\u00A2na k\u00C3\u00A2- miyopayi -t B \u00C3\u00AA- m\u00C3\u00A2to -t J usually C2- good.VAI -3 B C1-cry.VAI-3 \u00E2\u0080\u0098When Jane has good fortune, Beth cries.\u00E2\u0080\u0099 In fact, if the modificational clause contains both referents (one proximate, the other obviative) and the following main clause is unmarked, the subject cannot be determined: it may either be the same subject as in the modificational clause (in this case, Jeff), or it may be different (in this case atim \u00E2\u0080\u0098dog\u00E2\u0080\u0099). 5 The form m\u00C3\u00AAkw\u00C3\u00A2c is usually found external to the verbal complex (Wolfart 1973; Dahlstrom 1991). 233 (51) -yi can be omitted in adjoined clauses and still be disjoint a. Jeff k\u00C3\u00A2-m\u00C3\u00B4s\u00C3\u00A2hkin\u00C3\u00A2t atima, \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAyit J k\u00C3\u00A2-m\u00C3\u00B4s\u00C3\u00A2hkin -\u00C3\u00A2 -t atimw-a \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA -yi-t J C2-pick.up.VTA-DIR-3 dog -OBV C1-hungry.VAI-DS-3 \u00E2\u0080\u0098When Jeff picked up the dog, he(\u00E2\u0089\u00A0Jeff, =dog) was hungry.\u00E2\u0080\u0099 b. Jeff k\u00C3\u00A2-m\u00C3\u00B4s\u00C3\u00A2hkin\u00C3\u00A2t atima, \u00C3\u00AA-n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AAt J k\u00C3\u00A2-m\u00C3\u00B4s\u00C3\u00A2hkin -\u00C3\u00A2 -t atimw-a \u00C3\u00AA- n\u00C3\u00B4ht\u00C3\u00AAhkat\u00C3\u00AA-t J C2-pick.up.VTA-DIR-3 dog -OBV C1-hungry.VAI-3 \u00E2\u0080\u0098When Jeff picked up the dog, he(=Jeff, =dog) was hungry.\u00E2\u0080\u0099 Mediated-subject clauses actually disprefer -yi, a pattern for which I have no good explanation at the moment6. (52) -yi should be omitted in mediated subject clauses a. \u00C3\u00AA-miyw\u00C3\u00A2sik \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt Jeff \u00C3\u00AA- miyw\u00C3\u00A2si -k \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J C1-good.VII-0 DEM.INAN C1-COME-go.VAI-3 J \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that Jeff came.\u00E2\u0080\u0099 b. #? \u00C3\u00AA-miyw\u00C3\u00A2sik \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAyit Jeff-a \u00C3\u00AA- miyw\u00C3\u00A2si -k \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -yi -t J -a C1-good.VII -0 DEM.INAN C1-COME-go.VAI-DS-3 J-OBV \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that Jeff came.\u00E2\u0080\u0099 However, this does show a subject-object asymmetry for mediated argument clauses, because mediated object clauses require the different subject marker -yi in the dependent clause. For example, in (53), we have the psych predicate nitaw\u00C3\u00AAyim- \u00E2\u0080\u0098want.VTA\u00E2\u0080\u0099 introducing the dependent clause anihi iskw\u00C3\u00AAwa ka-nimihitoyit \u00E2\u0080\u0098(that) the woman dances\u00E2\u0080\u0099. (The higher predicate is inflected for an animate object (here iskw\u00C3\u00AAwa woman), meaning that we are dealing with a copy-to-object construction (Dahlstrom 1991)). (53) a. ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim\u00C3\u00AAw anihi iskw\u00C3\u00AAwa ka-nimihitoyit ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim -\u00C3\u00AA -w anihi iskw\u00C3\u00AAw -a ka- nimihito -yi -t DEM.AN man want.VTA -DIR-3 DEM.OBV woman-OBV IRR-dance.VAI-DS-3 \u00E2\u0080\u0098That man wants that woman to dance.\u00E2\u0080\u0099 6 I suspect that it has to do with the fact that the matrix clause given \u00E2\u0080\u0093 as well as all other clauses that introduce mediated subject clauses that I can think of - is an inanimate intransitive verb, which lacks the referential properties associated with switch reference; cf. M\u00C3\u00BChlbauer 2008. 234 b. * ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim\u00C3\u00AAw anihi iskw\u00C3\u00AAwa ka-nimihitot ana n\u00C3\u00A2p\u00C3\u00AAw nitaw\u00C3\u00AAyim -\u00C3\u00AA -w anihi iskw\u00C3\u00AAw -a ka- nimihito -t DEM.AN man want.VTA -DIR-3 DEM.OBV woman-OBV IRR-dance.VAI-3 --- Further confirmation of this pattern is shown in (54-55) for both VTA matrix clauses (which agree with the subject of the lower clause) and VTI matrix clauses (which show agreement for the proposition). (54) a. (ana) n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtaw\u00C3\u00AAw iskw\u00C3\u00AAwa \u00C3\u00AA-nikamoyit MEDIATED OBJECT ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtaw -\u00C3\u00AA -w iskw\u00C3\u00AAw-a \u00C3\u00AA- nikamo -yi -t DEM man hear.VTA-DIR-3 woman-OBV C1-sing.VAI-DS-3 \u00E2\u0080\u0098The man heard the woman singing.\u00E2\u0080\u0099 b. ! ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtaw\u00C3\u00AAw iskw\u00C3\u00AAw(a) \u00C3\u00AA-nikamot7 ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtaw -\u00C3\u00AA -w iskw\u00C3\u00AAw-a \u00C3\u00AA- nikamo -t DEM man hear.VTA-DIR-3 woman-OBV C1-sing.VAI-3 --- (intended: \u00E2\u0080\u0098The man heard the woman singing.\u00E2\u0080\u0099) (55) a. ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtam iskw\u00C3\u00AAwa \u00C3\u00AA-nikamoyit MEDIATED OBJECT ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtam-w iskw\u00C3\u00AAw-a \u00C3\u00AA- nikamo -yi -t DEM man hear.VTI-3 woman-OBV C1-sing.VAI-DEP-3 \u00E2\u0080\u0098The man heard that the woman was singing.\u00E2\u0080\u0099 b. ! ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtam iskw\u00C3\u00AAw \u00C3\u00AA-nikamot ana n\u00C3\u00A2p\u00C3\u00AAw pehtam-w iskw\u00C3\u00AAw \u00C3\u00AA- nikamo -t DEM man hear.VTI-3 woman C2-sing.VAI-3 --- (intended: \u00E2\u0080\u0098The man heard the woman singing.\u00E2\u0080\u0099) One possible exception to this generalization is an example like (56), in which the nominal iskw\u00C3\u00AAw \u00E2\u0080\u0098woman\u00E2\u0080\u0099 and the verbal complex \u00C3\u00AA-nikamot \u00E2\u0080\u0098\u00E2\u0080\u00A6s/he is singing\u00E2\u0080\u0099 are both unmarked: the nominal is not obviative, and there is no dependent reference marker -yi.8 7 It should be possible for \u00C3\u00AA-nikamot to be grammatical, meaning that the man (not the woman) was singing when he heard the woman (Wolfart 1973, 1996). When I asked about this possibility, the consultant said it was possible, but \u00E2\u0080\u009Cyou wouldn\u00E2\u0080\u0099t use \u00C3\u00AA-nikamot [in this position] with this sentence\u00E2\u0080\u009D. She preferred to use a k\u00C3\u00A2-clause with an overt aspectual marker indicating the simultaneity of the hearing and the singing; she also indicated that she preferred to have the k\u00C3\u00A2-clause placed immediately after n\u00C3\u00A2p\u00C3\u00AAw. (i) ana n\u00C3\u00A2p\u00C3\u00AAw k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-nikamot p\u00C3\u00AAhtaw\u00C3\u00AAw iskw\u00C3\u00AAwa ana n\u00C3\u00A2p\u00C3\u00AAw k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2-nikamo-t p\u00C3\u00AAhtaw-\u00C3\u00AA-w iskw\u00C3\u00AAw-a DEM man C2-NOW-sing.VAI-3 hear.VTA-DIR-3 woman-OBV \u00E2\u0080\u0098The man heard the woman while he was singing.\u00E2\u0080\u0099 / \u00E2\u0080\u0098The man who was singing heard the woman.\u00E2\u0080\u0099 8 Such examples are, to my knowledge, unattested in running speech. 235 (56) ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtam, iskw\u00C3\u00AAw \u00C3\u00AA-nikamot ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtam-w iskw\u00C3\u00AAw \u00C3\u00AA- nikamo -t DEM man hear.VTI-3 woman C2-sing.VAI-3 \u00E2\u0080\u0098The man heard it \u00E2\u0080\u0093 the woman was singing.\u00E2\u0080\u0099 comment: because he doesn\u00E2\u0080\u0099t know her, who that woman is, he just knows there\u00E2\u0080\u0099s a voice, a woman\u00E2\u0080\u0099s voice However, there are at least four independent reasons why I do not think such a clause has the same embedded status as the examples in which iskw\u00C3\u00AAw \u00E2\u0080\u0098woman\u00E2\u0080\u0099 is obviated and the verbal complex is obligatorily marked with -yi. First, there is a difference in interpretation, most clearly brought out in the evidential distinction (cf. Dahlstrom 1991): the clause with the switch reference marker (in 55a) must be interpreted as indirectly perceived by the subject of the higher clause; in (56), however the clause lacking the switch reference marker may be interpreted as directly perceived, and correlates with Kuno\u00E2\u0080\u0099s (1987) \u00E2\u0080\u0098direct discourse\u00E2\u0080\u0099. Second, there is a difference in prosody: the consultant noted that the clause without the switch-reference marker must be preceded by a prosodic break. As we have seen earlier, this is a property of chained clauses, but is otherwise unattested in mediated object clauses. Third, the string in (57) is ungrammatical, where the object agreement of the higher clause agrees with the subject of the lower clause. In other words, when you have agreement giving you an obligatory embedding relation, then this structure is unavailable. (57) * ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtaw\u00C3\u00AAw, iskw\u00C3\u00AAw \u00C3\u00AA-nikamot ana n\u00C3\u00A2p\u00C3\u00AAw p\u00C3\u00AAhtaw -\u00C3\u00AA -w iskw\u00C3\u00AAw \u00C3\u00AA- nikamo -t DEM man hear.VTA-DIR-3 woman C1-sing.VAI -3 --- Finally, the -yi marked vs. unmarked examples differ with respect to the \u00E2\u0080\u0098fronting\u00E2\u0080\u0099 diagnostic: if the clause is not marked with the switch-reference marker, the subject is clause- bound. Placing the plural subject iskw\u00C3\u00AAwak \u00E2\u0080\u0098women\u00E2\u0080\u0099 at the beginning of the utterance induces an interpretation that the women are doing the perceiving (even though the obligatory plural agreement on the matrix clause is missing). 236 (58) a. kahkiyaw iskw\u00C3\u00AAwa p\u00C3\u00AAhtam \u00C3\u00AA-nikamoyit kahkiyaw iskw\u00C3\u00AAw -a p\u00C3\u00AAhtam -w \u00C3\u00AA- nikamo -yi -t all woman-OBV heard.VTI-3 C1-sing.VAI-DS-3 \u00E2\u0080\u0098S/he heard all the women singing.\u00E2\u0080\u0099 b. * iskw\u00C3\u00AAwak p\u00C3\u00AAhtam \u00C3\u00AA-nikamocik iskw\u00C3\u00AAw -ak p\u00C3\u00AAhtam -w \u00C3\u00AA- nikamo -t -k woman-PL hear.VTI-3 C1-sing.VAI-3-PL --- (intended: \u00E2\u0080\u0098S/he heard all the women singing.\u00E2\u0080\u0099) Thus, (56) patterns with adjoined clauses rather than mediated argument clauses with respect to the fronting diagnostic. The obligatory presence of -yi as a diagnostic splits object-mediated clauses from chained and adjoined clauses in exactly the same way as quantifier-fronting and discontinuous arguments: while adjoined and chained clauses do not require -yi marking, mediated object clauses do. In other words, object-mediated clauses can never form an independent domain for different-subject marking. I take this data to stem from the position of object-mediated clauses: they are introduced and adjoined to VP; the entire clause is thus always in the scope of the subject of the matrix clause. 5.5 Consequences In this section I consider some of the consequences of the proposed syntactic classification of anaphoric CONJUNCT clauses. In particular, I single out the mediated-argument clauses, whose structure I repeat in (59). An object-mediated clause is associated with an object DP position and adjoined to VP (59a); a subject-mediated clause is associated with a subject DP position and adjoined to IP (59b). 237 (59) a. Object-mediated CP 5 VP 3 VP CPi 2 5 V DPi DEPENDENT b. Subject-mediated CP 5 IP 3 IP CPi 2 5 DPi 2DEPENDENT VP MEDIATED ARGUMENT CLAUSES If this is the correct analysis for this class of clauses, and the other types of clauses are adjoined and chained clauses, then there are no true argument clauses. In \u00C2\u00A75.5.1, I provide some evidence that these clauses are not sitting directly in an argument position. In \u00C2\u00A75.5.2, I show that there is VP-complementation, as evidenced by \u00E2\u0080\u009Cpre-verbs\u00E2\u0080\u009D. Finally, the analysis of mediated argument clauses forces an analysis of copy-to-object constructions (discussed for Plains Cree by Dahlstrom 1986, 1991) as local-agreement (contra the proposal made by Branigan & MacKenzie 2002 for Innu-Aimun, another Algonquian language), as discussed in \u00C2\u00A75.5.3. 5.5.1 The non-existence of argument clauses In this section I turn to the syntax of mediated argument clauses: mediated subjects and mediated objects. I take seriously the older linguistic claims such as Wolfart (1973:46), who claims that in an example like (60) the \u00C3\u00AA-clause \u00E2\u0080\u009Cfunctions as the adjunct of a verb\u00E2\u0080\u009D. (60) k\u00C3\u00AEtahtaw\u00C3\u00AA p\u00C3\u00AAyak k\u00C3\u00AEh-paw\u00C3\u00A2tam k\u00C3\u00AAtaht\u00C3\u00A2w\u00C3\u00AA p\u00C3\u00AAyak k\u00C3\u00AE- paw\u00C3\u00A2tam -w once certain PREV-dream.VTI-3 \u00E2\u0080\u0098Then at one time a certain man dreamt [\u00C3\u00AA-w\u00C3\u00AEh-kap\u00C3\u00A2yit m\u00C3\u00B4niy\u00C3\u00A2w-iyiniwa w\u00C3\u00A2piskiwiy\u00C3\u00A2sah] \u00C3\u00AA- w\u00C3\u00AE- kap\u00C3\u00A2 -yi -t m\u00C3\u00B4niyawiyiniw -a w\u00C3\u00A2piskiwiy\u00C3\u00A2s -a C1-INT-land.VAI-DS-3 White.man -OBV Canadian -OBV that the Canadian, the White Man would land here.\u00E2\u0080\u0099 Under the current proposal, the dependent clause of (61) would adjoined to VP, and associated with an object DP introduced by the verb. 238 (61) Mediated object clause CP 5 VP 3 VP CPi 2 6 V DPi \u00C3\u00AA-w\u00C3\u00AEh-kap\u00C3\u00A2yit m\u00C3\u00B4niy\u00C3\u00A2w-iyiniwa w\u00C3\u00A2piskiwiy\u00C3\u00A2sah In (62), the CONJUNCT clause \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt Jeff \u00E2\u0080\u0098Jeff came.\u00E2\u0080\u0099 is associated with the subject position of miyw\u00C3\u00A2sin \u00E2\u0080\u0098it\u00E2\u0080\u0099s good.\u00E2\u0080\u0099; it is analyzed as being adjoined to IP and associated with the subject DP (here instantiated by the demonstrative \u00C3\u00B4ma \u00E2\u0080\u0098this\u00E2\u0080\u0099). (62) Mediated subject clause a. miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt Jeff miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J good.VII DEM.INAN C1-COME-go.VAI-3 J \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that Jeff came.\u00E2\u0080\u0099 b. CP 5 IP 3 IP CPi 2 5 DPi 2 \u00C3\u00AA-p\u00C3\u00AA-toht\u00C3\u00AAt Jeff 5 VP \u00C3\u00B4ma 5 These clauses are thus analyzed as an instance of base-generated extraposition (cf. Culicover & Rochemont 1990, Wiltschko 1995); i.e., the dependent clause is generated in its adjoined position, rather than moving to it from an argument position. It should be pointed out that although the arguments for these structures are based on the properties of clauses in Plains Cree, this is not a claim specific to Plains Cree. English \u00E2\u0080\u0098argument\u00E2\u0080\u0099 clauses exhibit several behaviours that have led many syntacticians to believe they are not generated in argument position. 239 First, only some \u00E2\u0080\u0098argument-clauses\u00E2\u0080\u0099 can undergo passivization (63; Rosenbaum 1967); in particular, the infinitival \u00E2\u0080\u0098argument-clauses\u00E2\u0080\u0099 cannot. If passivization targets an object position, then the data in (64) implies that not all clauses are in an object position. (63) a. Columbus demonstrated [that the world is not flat]. b. [That the world is not flat] was demonstrated by Columbus. (64) a. She began [to cry]. b. * [To cry] was begun by her. (Rosenbaum 1967:10-11) Second, object nominals most neutrally occur before manner adverbials, but \u00E2\u0080\u0098complement-clauses\u00E2\u0080\u0099 most neutrally occur after them as in (65-66), taken from Stowell 1981. This implies that the \u00E2\u0080\u0098complement-clauses\u00E2\u0080\u0099 are attaching not as complements of V, but higher up (e.g., adjoined to VP or IP). (65) DPs are pre-adverbial a. He explained the situation very carefully. b. ? He explained that he was not going to leave very carefully. (66) Clauses are post-adverbial a. ? He explained very carefully the situation. b. He explained very carefully that he was not going to leave. Third, we see that for at least some \u00E2\u0080\u0098object\u00E2\u0080\u0099 clauses, and all \u00E2\u0080\u0098subject\u00E2\u0080\u0099 clauses, there is the possibility of having an overt pronominal form in the argument position, with the clause being in such cases obligatorily extraposed to the right (Jesperson 1937; McCawley 1988, among others). (67) Pronominal form in subject clause a. John regrets [that he quit his job]. b. John regrets it [that he quits his job]. 240 (68) Pronominal form in object clause a. [That you got sick] is unfortunate. b. It is unfortunate [that you got sick]. In fact, some \u00E2\u0080\u0098subject-clauses\u00E2\u0080\u0099 can only occur in the extraposed position, as in (69) and (70). (69) a. it happened [that John came early] b. * [that John came early] happened (Rosenbaum 1967:71-2) (70) a. I wonder whether the robbery surprised [them] b. ?/* I wonder whether that the pig was stolen surprised [them] c. I wonder whether it surprised [them] that the pig was stolen (Haegeman 1994:57) Taking the most widely-available construction to be the basic, and the restricted form to be derived, this has led many to the conclusion that the extraposed position for subject clauses is basic, and the non-extraposed position derived (Rosenbaum 1967, Ross 1967b, Williams 1974, Emonds 1976, McCawley 1981, 1988:98). Finally, even for \u00E2\u0080\u0098subject-clauses\u00E2\u0080\u0099 that are permitted in the non-extraposed position, there is a requirement that it be introduced either by that, which is also used in the nominal domain as a demonstrative, or by a nominal phrase like the fact, as in (71) (Koster 1978; Safir 1985). In other words, they require an element that is independently attested in DP-syntax. (71) a. * [John failed the final] means he failed the course. b. [That John failed the final] means he failed the course. c. [The fact that John failed the final] means he failed the course. (R. Waldie, pc) Returning to the question of Plains Cree clauses, we will see in the following section that similar sorts of problems rear their head if we try to treat clauses as arguments. First, \u00E2\u0080\u0098subject\u00E2\u0080\u0099 clauses require an overt nominal antecedent (\u00C2\u00A75.5.1.1). Second, clauses do not have the same freedom of ordering that nominals have (\u00C2\u00A75.5.1.2). Third, no predicates select for a clausal 241 argument (\u00C2\u00A75.5.1.3). Fourth, some predicates select for nominals (\u00C2\u00A75.5.1.4). Finally, incorporation targets nominal arguments, but not clauses. 5.5.1.1 \u00E2\u0080\u0098Subject\u00E2\u0080\u0099 clauses require overt nominal antecedent If \u00E2\u0080\u0098argument\u00E2\u0080\u0099 clauses are always mediated by a DP, the DP position should always be available. Thus we predict that an overt nominal antecedent to a mediated argument clause should always be possible. This is correct. In fact, object-oriented clauses always can, and very often do co-occur with an overt inanimate demonstrative (cf. Ahenakew 1987). The result is that we get the following paradigm: the matrix predicate inflected for an inanimate object argument (72a), and there is a corresponding inanimate demonstrative \u00C3\u00B4ma (72b). (72) a. nikisk\u00C3\u00AAyiht\u00C3\u00AAn ni- kisk\u00C3\u00AAyiht\u00C3\u00AA -n 1- know.VTI -SAP \u00E2\u0080\u0098I know it.\u00E2\u0080\u0099 b. nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00B4ma ni- kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00B4ma 1- know.VTI -SAP DEM .INAN \u00E2\u0080\u0098I know it/this.\u00E2\u0080\u0099 If the object argument is associated with a clause (73a), then the inanimate demostrative can precede it (73b). (73) a. nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-w\u00C3\u00AE-k\u00C3\u00AEw\u00C3\u00AAyan ni-kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- w\u00C3\u00AE- k\u00C3\u00AEw\u00C3\u00AA -yan 1- know.VTI -SAP C1-INT-go.home.VAI-2 \u00E2\u0080\u0098I know that you\u00E2\u0080\u0099re going to go home.\u00E2\u0080\u0099 b. nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00B4ma \u00C3\u00AA-w\u00C3\u00AE-k\u00C3\u00AEw\u00C3\u00AAyan ni-kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00B4ma \u00C3\u00AA- w\u00C3\u00AE- k\u00C3\u00AEw\u00C3\u00AA -yan 1-know.VTI -SAP DEM.INAN C1-INT-go.home.VAI-2 \u00E2\u0080\u0098I know that you\u00E2\u0080\u0099re going to go home.\u00E2\u0080\u0099 (Ahenakew 1987:159) 242 The demonstrative must precede the clause with which it is associated, as shown in (74).9 (74) a. nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00B4ma \u00C3\u00AA-w\u00C3\u00AE-k\u00C3\u00AEw\u00C3\u00AAyan ni-kisk\u00C3\u00AAyiht\u00C3\u00AA-n \u00C3\u00B4ma \u00C3\u00AA- w\u00C3\u00AE- k\u00C3\u00AEw\u00C3\u00AA -yan 1-know.VTI -SAP DEM.INAN C1-INT-go.home.VAI-2 \u00E2\u0080\u0098I know that you\u00E2\u0080\u0099re going to go home.\u00E2\u0080\u0099 b. ! nikisk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-w\u00C3\u00AE-k\u00C3\u00AEw\u00C3\u00AAyan \u00C3\u00B4ma ni- kisk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- w\u00C3\u00AE- k\u00C3\u00AEw\u00C3\u00AA -yan \u00C3\u00B4ma 1- know.VTI -SAP C1-INT-go.home.VAI-2 DEM.INAN --- (intended: \u00E2\u0080\u0098I know that you\u00E2\u0080\u0099re going to go home.\u00E2\u0080\u0099) In subject position, the generalization is much stronger: subject clauses must be preceded by an overt nominal, even when the agreement is such to license an inanimate nominal. Consultants can interpret but reject subject-oriented clauses that occur without \u00C3\u00B4ma (or some other inanimate demonstrative). (75) a. ?? miyw\u00C3\u00A2sin \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt Jeff miyw\u00C3\u00A2sin \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J good.VII C1-COME-go.VAI-3 J \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that Jeff came.\u00E2\u0080\u0099 b. miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt J miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J good.VII DEM.INAN C1-COME-go.VAI-3 J \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that Jeff came.\u00E2\u0080\u0099 9 In addition to the fact that the demonstrative must precede the clause with which it is associated, there is the question of whether it is possible to extract out of an extraposed clause if there is an overt antecedent (such as the demonstrative). In German, for example, such extraction is impossible and is attributed to the presence of a Novelty condition (cf. Wiltschko 1995; see also Heim 1982 on indefinites). I do not at present have the relevant data to compare Plains Cree to German on this particular point; however, if the Novelty condition were active, we would expect to see its effects across the grammar. For example, we would expect relative clauses to always follow their head, but this does not work for Plains Cree, where a relative clause may either precede or follow the head noun; this is exemplified in (i). (i) a. niw\u00C3\u00A2pam\u00C3\u00A2w k\u00C3\u00A2-m\u00C3\u00A2tot iskw\u00C3\u00AAw ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w k\u00C3\u00A2- m\u00C3\u00A2to -t iskw\u00C3\u00AAw 1- see.VTA -DIR -3 C2- cry.VAI-3 woman \u00E2\u0080\u0098I saw the woman that\u00E2\u0080\u0099s crying / crying woman.\u00E2\u0080\u0099 b. niw\u00C3\u00A2pam\u00C3\u00A2w iskw\u00C3\u00AAw k\u00C3\u00A2-m\u00C3\u00A2tot ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w iskw\u00C3\u00AAw k\u00C3\u00A2- m\u00C3\u00A2to -t 1- see.VTA -DIR -3 woman C2- cry.VAI-3 \u00E2\u0080\u0098I saw the woman that\u00E2\u0080\u0099s crying.\u00E2\u0080\u0099 I leave for further research the question of why Plains Cree exhibits some, but not all, of the precedence constraints on extraposed clauses; minimally, Plains Cree would seem to provide evidence that not all precedence can be explained in terms of the Novelty condition. 243 In corpora, clauses are associated with a subject position only if they are preceded by a nominal element (including \u00C3\u00B4ma \u00E2\u0080\u0098this.INAN\u00E2\u0080\u0099; k\u00C3\u00AEkw\u00C3\u00A2y \u00E2\u0080\u0098what\u00E2\u0080\u0099). For example, in (76) we have a clause with an inverse marker and the inanimate subject k\u00C3\u00AEkway \u00E2\u0080\u0098something\u00E2\u0080\u0099; the clause which identifies that something (enclosed in brackets) occurs in final position, discontinuous from the nominal element with which it is associated. (76) a. ..., \u00C3\u00AAwak \u00C3\u00B4ma k\u00C3\u00AEkway k-\u00C3\u00A2st\u00C3\u00A2hikoyahk, \u00C3\u00AAwakw \u00C3\u00B4ma k\u00C3\u00AEkway k\u00C3\u00A2-ast\u00C3\u00A2h -iko -yan-k TOPIC DEM something C2-worry.VTA-INV-2 -PL \u00E2\u0080\u0098... there is something which is worrying us, tahtw\u00C3\u00A2w \u00C3\u00AA-k\u00C3\u00AEsik\u00C3\u00A2k \u00C3\u00AAk\u00C3\u00A2 k\u00C3\u00AEkway k\u00C3\u00A2-miyw\u00C3\u00A2sik. tahtw\u00C3\u00A2w \u00C3\u00AA- k\u00C3\u00AEsik\u00C3\u00A2 -k \u00C3\u00AAk\u00C3\u00A2 k\u00C3\u00AEkway k\u00C3\u00A2-miyw\u00C3\u00A2si-k so.many.times C1-day.VII-0 NEG something C2-good.VII-0 the fact that day after day there are things which are not good.\u00E2\u0080\u0099 (JKN 1.1) Likewise, in (77) the clause \u00C3\u00AA-k\u00C3\u00AE-minihkw\u00C3\u00AAskit \u00E2\u0080\u0098he used to drink\u00E2\u0080\u0099 is associated with the subject of \u00C3\u00AA-k\u00C3\u00AE-kitimahikot \u00E2\u0080\u0098it gave my husband trouble\u00E2\u0080\u0099; there is an overt nominal element k\u00C3\u00AEkw\u00C3\u00A2y \u00E2\u0080\u0098something\u00E2\u0080\u0099. The utterance is ungrammatical if k\u00C3\u00AEkw\u00C3\u00A2y \u00E2\u0080\u0098something\u00E2\u0080\u0099 is removed, as in (77b). (77) a. k\u00C3\u00AEkw\u00C3\u00A2y \u00C3\u00AA-k\u00C3\u00AE-kitimahikot niw\u00C3\u00AEkim\u00C3\u00A2kan, \u00C3\u00AA-k\u00C3\u00AE-minihkw\u00C3\u00AAskit; k\u00C3\u00AEkw\u00C3\u00A2y \u00C3\u00AA- k\u00C3\u00AE- kitimah -iko -t ni- w\u00C3\u00AEkim\u00C3\u00A2kan \u00C3\u00AA- k\u00C3\u00AE- minihkw\u00C3\u00AA -ski -t thing C1-PREV-trouble.VTA-INV-3 1- spouse C1-PREV-drink.VAI -HAB-3 \u00E2\u0080\u0098What used to give my husband trouble was that he used to drink;\u00E2\u0080\u0099 (EM 28) b. * \u00C3\u00AA-k\u00C3\u00AE-kitimahikot niw\u00C3\u00AEkim\u00C3\u00A2kan, \u00C3\u00AA-k\u00C3\u00AE-minihkw\u00C3\u00AAskit \u00C3\u00AA- k\u00C3\u00AE- kitimah -iko -t ni- w\u00C3\u00AEkim\u00C3\u00A2kan \u00C3\u00AA- k\u00C3\u00AE- minihkw\u00C3\u00AA -ski -t C1-PREV-trouble.VTA-INV-3 1- spouse C1-PREV-drink.VAI -HAB-3 --- (intended: \u00E2\u0080\u0098It troubled my husband that he used to drink.\u00E2\u0080\u0099) Further, k\u00C3\u00AEkw\u00C3\u00A2y \u00E2\u0080\u0098something\u00E2\u0080\u0099 can be replaced by a nominalization, but not by the clause, as the contrast in (78) shows. 244 (78) a. ominihkw\u00C3\u00AAwin \u00C3\u00AA-kitimahikot niw\u00C3\u00AEkim\u00C3\u00A2kan o-minihkw\u00C3\u00AA-win \u00C3\u00AA- kitimah -iko-t ni-w\u00C3\u00AEkim\u00C3\u00A2kan 3-drink.VAI-NOM C1-trouble.VTA-INV-3 1-spouse \u00E2\u0080\u0098\u00E2\u0080\u00A6his drinking troubled my husband.\u00E2\u0080\u0099 b. * \u00C3\u00AA-k\u00C3\u00AE-minihkw\u00C3\u00AAskit \u00C3\u00AA-kitimahikot niw\u00C3\u00AEkim\u00C3\u00A2kan \u00C3\u00AA- k\u00C3\u00AE- minihkw\u00C3\u00AA-ski -t \u00C3\u00AA- kitimah -iko -t ni-w\u00C3\u00AEkim\u00C3\u00A2kan C1-PREV-drink.VAI-HAB-3 C1-trouble.VTA-INV-3 1-spouse --- (intended: \u00E2\u0080\u0098That he used to drink troubled my husband.\u00E2\u0080\u0099) Here we see a subject-object asymmetry where object-mediated clauses allow an overt nominal antecedent, and subject-mediated clauses require an overt nominal antecedent. I take the subject- object asymmetry in the clauses to be related to more general subject-object asymmetries where we see that, cross-linguistically, subjects are not licensed by the verb and thus require some independent mechanism for licensing. If we say that all \u00E2\u0080\u0098argument\u00E2\u0080\u0099 clauses are associated with a mediating argument expression, this accurately predicts that overt nominal elements like \u00C3\u00B4ma \u00E2\u0080\u0098this.INAN\u00E2\u0080\u0099 are always in principle available. The most widely available option (in this case, the presence of a nominal element like \u00C3\u00B4ma \u00E2\u0080\u0098this.INAN\u00E2\u0080\u0099) is basic, and the restricted option (in this case, the absence of a nominal element) is derived. However, if we claimed that clauses are in argument position, we would have to add an additional rule to derive obligatory extraposition and nominal insertion. I thus take the above data to support the current analysis.` 5.5.1.2 Clauses have different ordering properties than arguments A second argument for the mediated argument analysis is the restricted ordering properties that clauses have as opposed to nominal arguments. In Plains Cree, nominal arguments are regularly found in both preverbal and postverbal position (cf. D\u00C3\u00A9chaine 1997, 2007, M\u00C3\u00BChlbauer 2003, Reinholtz 1999, Wolvengrey 2003). 245 (79) a. niw\u00C3\u00A2pam\u00C3\u00A2w min\u00C3\u00B4s ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w min\u00C3\u00B4s 1- see.VTA-DIR -3 cat \u00E2\u0080\u0098I see a cat.\u00E2\u0080\u0099 b. min\u00C3\u00B4s niw\u00C3\u00A2pam\u00C3\u00A2w min\u00C3\u00B4s ni-w\u00C3\u00A2pam -\u00C3\u00A2 -w cat 1- see.VTA-DIR-3 \u00E2\u0080\u0098I see a cat.\u00E2\u0080\u0099 Argument-like clauses are different from nominal arguments in this respect. Recall from \u00C2\u00A75.3 that such clauses are restricted to postverbal position. The relevant data is repeated below for both subjects (80) and objects (81-82). (80) Mediated subject clause a. miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt John miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J good.VII DEM C1-COME-go.VAI-3 J \u00E2\u0080\u0098It\u00E2\u0080\u0099s good that John came.\u00E2\u0080\u0099 b. * \u00C3\u00B4ma \u00C3\u00AA-p\u00C3\u00AA-itoht\u00C3\u00AAt John, miyw\u00C3\u00A2sin \u00C3\u00B4ma \u00C3\u00AA- p\u00C3\u00AA- itoht\u00C3\u00AA -t J miyw\u00C3\u00A2sin DEM C1-COME-go.VAI-3 J good.VII --- (81) Mediated object clause a. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn ka-m\u00C3\u00AEcisoy\u00C3\u00A2n ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n ka- m\u00C3\u00AEciso -y\u00C3\u00A2n 1- want.VTI -SAP IRR- eat.VAI -1 \u00E2\u0080\u0098I would like to eat.\u00E2\u0080\u0099 b. * ka-m\u00C3\u00AEcisoy\u00C3\u00A2n ninitaw\u00C3\u00AAyiht\u00C3\u00AAn ka- m\u00C3\u00AEciso -y\u00C3\u00A2n ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n IRR-eat.VAI -1 1- want.VTI -SAP --- (intended: \u00E2\u0080\u0098I would like to eat.\u00E2\u0080\u0099) 246 (82) Mediated object clause a. niwanikiskisin \u00C3\u00AA-nip\u00C3\u00A2t aw\u00C3\u00A2sis ni-w\u00C3\u00A2nkiskisi -n \u00C3\u00AA-nip\u00C3\u00A2 -t aw\u00C3\u00A2sis 1- forget.VAI -SAP C1-sleep.VAI-3 child \u00E2\u0080\u0098I forgot that the child is sleeping.\u00E2\u0080\u0099 b. ?* \u00C3\u00AA-nip\u00C3\u00A2t aw\u00C3\u00A2sis niwanikiskisin ni- w\u00C3\u00A2nkiskisi -n \u00C3\u00AA- nip\u00C3\u00A2 -t aw\u00C3\u00A2sis 1- forget.VAI -SAP C1-sleep.VAI-3 child --- (intended: \u00E2\u0080\u0098I forgot that the child is sleeping.\u00E2\u0080\u0099) If clauses can sit in argument position, these ordering restrictions are unexpected. However, ordering restrictions are one of the cross-linguistic hallmarks of extraposed clauses, (cf. Rosenbaum 1967; Culicover & Rochemont 1990; Koster 1978; Wiltschko 1995, etc). In (83-84), for example, we see that extraposed clauses do not like to be in a preverbal position in English. (83) Ordering restrictions on extraposed subject-oriented clauses a. It happened [that John came early]. b. * It [that John came early] happened. (Rosenbaum 1967) (84) Ordering restrictions on extraposed object-oriented clauses a. John liked iti [CP that we ate all his food ]i. b. * [CP That we ate all his food, ]i John liked iti. For German, we also see that the extraposed clause has ordering restrictions; in particular, it cannot precede the pronominal expression that sits in the argument position. (85) a. Peter hat es geglaubt, dass Maria Bier trinkt Peter has it believed [that Maria beer drinks] (Wiltschko 1995:55) \u00E2\u0080\u0098Peter believed that Mary drinks beer.\u00E2\u0080\u0099 b. * Dass Maria Bier trinkt hat Peter es geglaubt [that Mary beer drinks] has Peter it believed --- (intended: \u00E2\u0080\u0098Peter believed that Mary drinks beer.\u00E2\u0080\u0099) (Wiltschko 1995:147) 247 The ordering restrictions on Plains Cree argument-like clauses thus look like a general property of extraposed clauses. If we do not analyze these clauses as extraposed, we then have ann unexplained restriction on ordering that is specific to Plains Cree. 5.5.1.3 No predicates subcategorize for clauses So far the predicates that introduce object-oriented clauses are of the morphological class VTI \u00E2\u0080\u0093 transitive verbs coded for an inanimate argument. In these cases the predicate has identical agreement to cases where an inanimate nominal argument is being introduced, and we in fact saw that it was always possible to have an additional overt nominal antecedent. One might wish to save the \u00E2\u0080\u0098clause-as-argument\u00E2\u0080\u0099 analysis by looking for clauses that subcategorize for a clausal argument. For example, when trying to understand the relation of English argument-like clauses to nominal arguments, Grimshaw (1979, 1981) argued that some verbs could syntactically select (c-select) for a clause rather than a nominal. Such verbs are restricted to interrogative verbs (Lahiri 2002) like wonder and inquire. (86) a. John wondered [CP what the time was]. b. * John wondered [NP the time]. (Grimshaw 1979, Lahiri 2002) In this section, I show that Plains Cree lacks such a class of verbs. First, like English, non-interrogative predicates can always select for nominal expressions.10 For example, the speech predicate wihtamaw- \u00E2\u0080\u0098tell.x.to.y\u00E2\u0080\u0099 can introduce either a 10 A set of data that at first looks like a wrinkle are predicates like it\u00C3\u00AAyihtam \u00E2\u0080\u0098s/he thinks thus\u00E2\u0080\u0099 and itw\u00C3\u00AAw \u00E2\u0080\u0098s/he says thus\u00E2\u0080\u0099, which may exclusively introduce propositions as opposed to nominals (although note that they can introduce deictic topics). (i) \u00C3\u00AA-n\u00C3\u00AAstosiy\u00C3\u00A2n nik\u00C3\u00AE-itw\u00C3\u00A2n \u00C3\u00AA- n\u00C3\u00AAstosi -y\u00C3\u00A2n ni-k\u00C3\u00AE- itw\u00C3\u00AA -n C1-tired.VAI-1 1-PREV- thus.say.VAI -SAP \u00E2\u0080\u0098I said I\u00E2\u0080\u0099m tired.\u00E2\u0080\u0099 (ii) ? k\u00C3\u00AEkway itw\u00C3\u00AAw k\u00C3\u00AEkway itw\u00C3\u00AA -w thing thus.say.VAI -3 \u00E2\u0080\u0098S/he said something.\u00E2\u0080\u0099 (iii) \u00C3\u00AAkosi itw\u00C3\u00AAw \u00C3\u00AAkosi itw\u00C3\u00AA -w that thus.say.VAI -3 \u00E2\u0080\u0098S/he said thattop.\u00E2\u0080\u0099 248 dependent clause \u00C3\u00AA-n\u00C3\u00AAstosiy\u00C3\u00A2n \u00E2\u0080\u0098I am tired\u00E2\u0080\u0099 as in (87), or a nominal such as k\u00C3\u00AEkway \u00E2\u0080\u0098thing\u00E2\u0080\u0099 as in (88). (87) niw\u00C3\u00AEhtamaw\u00C3\u00A2w nis\u00C3\u00AEmis \u00C3\u00AA-n\u00C3\u00AAstosiy\u00C3\u00A2n ni- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w ni- s\u00C3\u00AEmis \u00C3\u00AA- n\u00C3\u00AAstosi -y\u00C3\u00A2n 1- tell.VTA -DIR-3 1- SIBLING C1-tired.VAI -1 \u00E2\u0080\u0098I told my little brother/sister I was tired.\u00E2\u0080\u0099 (88) \u00C3\u00A2tiht ayisk ayisiyiniwak, nam\u00C3\u00B4y w\u00C3\u00AEhk\u00C3\u00A2c k\u00C3\u00AEkway aya w\u00C3\u00AEhtamaw\u00C3\u00A2wak aya,\u00E2\u0080\u00A6 \u00C3\u00A2tiht ayisk ayisiyiniw -ak nam\u00C3\u00B4y w\u00C3\u00AEhk\u00C3\u00A2c k\u00C3\u00AEkway aya w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w -ak aya some for person -PL NEG ever thing CONN tell.VTA -DIR-3 -PL CONN \u00E2\u0080\u0098for some people are never told anything, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 42) Likewise, the VTI predicate nitaw\u00C3\u00AAyiht- \u00E2\u0080\u0098want\u00E2\u0080\u0099 can introduce a dependent clause (\u00C3\u00AA-nikamot John \u00E2\u0080\u0098John is singing\u00E2\u0080\u0099 is what I want), or an inanimate nominal (maskisin \u00E2\u0080\u0098shoe\u00E2\u0080\u0099 is what I want). (89) a. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-nikamot John ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- nikamo -t J 1- want.VTI -SAP C1-sing.VAI-3 J \u00E2\u0080\u0098I want John to sing.\u00E2\u0080\u0099 b. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn maskisin ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n maskisin 1- want.VTI -SAP shoe \u00E2\u0080\u0098I want a/the shoe.\u00E2\u0080\u0099 Second, interrogative predicates like wonder are not predicates as such in Plains Cree; rather the adverbial particle sequence matw\u00C3\u00A2n c\u00C3\u00AE \u00E2\u0080\u0098I wonder if\u00E2\u0080\u0099 is used. This particle sequence can introduce nominals like one month in (90b), and the particle t\u00C3\u00A2pw\u00C3\u00AA \u00E2\u0080\u0098true\u00E2\u0080\u0099 (90a). These are \u00E2\u0080\u0098bridge verbs\u00E2\u0080\u0099, which, cross-linguistically, have anomalous syntactic behaviour compared to other propositional predicates (Erteschik-Shir 1973, Fodor 1992, Holmberg & Platzack 1995, among others). In Plains Cree, the clauses introduced by these two predicates have very different behaviour from other propositions. First, unlike complement-like clauses they usually (in fact, always in running speech) precede the proposition that introduces them. Second, also unlike complement-like clauses, they may be in the INDEPENDENT order (but only if the proposition being expressed is direct speech or direct thought). Finally, the predicates it\u00C3\u00AAyihtam \u00E2\u0080\u0098s/he thinks thus\u00E2\u0080\u0099 and itw\u00C3\u00AAw \u00E2\u0080\u0098s/he says thus\u00E2\u0080\u0099 are ungrammatical without an overt argument. This behaviour is not entirely unpredictable: notice that both of these predicates have as their root the morpheme it- \u00E2\u0080\u0098thus\u00E2\u0080\u0099. This is one of a class of relative roots (cf. Howse 1865, Wolfart 1973): pronominal forms that require some antecedent to be well formed (cf. chapter 3 for more discussion). As such, the mechanism by which these propositions are introduced is distinct from the way arguments are introduced (notice, for example, that roots are not involved in transitivity classes; cf. Hirose 2000, D\u00C3\u00A9chaine 2003). At minimum, it seems necessary to treat the clauses that precede these verbs as something different from the extrapositional clauses we are looking at here. 249 (90) a. ..., \u00E2\u0080\u009Cmatw\u00C3\u00A2n c\u00C3\u00AE t\u00C3\u00A2pw\u00C3\u00AA,\u00E2\u0080\u009D nik\u00C3\u00AE-it\u00C3\u00AAyiht\u00C3\u00AAn m\u00C3\u00A2na, \u00E2\u0080\u00A6 matw\u00C3\u00A2n c\u00C3\u00AE t\u00C3\u00A2pw\u00C3\u00AA ni- k\u00C3\u00AE- it\u00C3\u00AAyiht\u00C3\u00AA -n m\u00C3\u00A2na wonder Q true 1- PREV-thus.think.VTI-SAP usually \u00E2\u0080\u0098..., \u00E2\u0080\u009CThat will be the day,\u00E2\u0080\u009D I used to think, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 5.5) (Lit.: \u00E2\u0080\u009CI wonder if it\u00E2\u0080\u0099s true?\u00E2\u0080\u009D I used to think) b. two m-~ two months, matw\u00C3\u00A2n c\u00C3\u00AE one month, \u00E2\u0080\u00A6 two months matw\u00C3\u00A2n c\u00C3\u00AE one month two months wonder Q one month \u00E2\u0080\u0098after two months, I wonder if it was one month, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 12.14) The sequence matw\u00C3\u00A2n c\u00C3\u00AE can also introduce INDEPENDENT clauses, which as we have already seen cannot be embedded. (91) ..., matw\u00C3\u00A2n c\u00C3\u00AE ka-kaskiht\u00C3\u00A2n\u00C3\u00A2naw s\u00C3\u00B4niy\u00C3\u00A2w ka-mow\u00C3\u00A2yahk?\u00E2\u0080\u009D matw\u00C3\u00A2n c\u00C3\u00AE ka- kaskiht\u00C3\u00A2 -n\u00C3\u00A2naw s\u00C3\u00B4niy\u00C3\u00A2w ka- mow -\u00C3\u00A2 -yan -k wonder Q IRR-able.VAI -21PL money IRR-eat.VTA-DIR-2 -PL \u00E2\u0080\u0098 \u00E2\u0080\u009C..., I wonder if we will be able to eat money?\u00E2\u0080\u009D \u00E2\u0080\u0099 (EM 63) Finally, while there are no predicates that select for clauses (i.e., CPs), there are predicates that select for another verbal predicate (i.e., VPs). These constructions have a significantly different structure than the constructions we have seen so far: the higher predicate is introduced in the preverb domain, and lower predicate is inflected as the main predicate, as in (92a-c). (92) a. nikw\u00C3\u00AA-nip\u00C3\u00A2n ni- kw\u00C3\u00AA- nip\u00C3\u00A2 -n 1- try- sleep.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m trying to sleep\u00E2\u0080\u0099 b. nim\u00C3\u00A2ci-p\u00C3\u00A2hpin ni- m\u00C3\u00A2ci- p\u00C3\u00A2hpi -n 1- start- laugh.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099m starting to laugh.\u00E2\u0080\u0099 c. nik\u00C3\u00AEs-nimihiton ni- k\u00C3\u00AEs- nimihito -n 1- finish-dance.VAI-SAP \u00E2\u0080\u0098I\u00E2\u0080\u0099ve finished dancing.\u00E2\u0080\u0099 As I show below, VP-complementation is mono-clausal and is a \u00E2\u0080\u0098restructuring\u00E2\u0080\u0099 phenomena (Napoli 1981; Cinque 2006; Wurmbrand 2003, Williams 2004). 250 5.5.1.4 Predicates that subcategorize for nominals: AIt verbs As Grimshaw (1979) points out, some predicates can be characterized as picking out a semantic object (such as a proposition), regardless of its syntactic category. For example, the predicate ask in English is said to semantically select for a question, but that question can be syntactically realized as a CP, an NP, or even a null pronominal form (Grimshaw 1979, Lahiri 2002). (93) a. John asked me [CP what the time is]. b. John asked me [NP the time]. c. John wanted to know what the time was, so he asked [\u00EF\u0081\u00B8]. (Lahiri 2002:245) Restrictions on what a predicate can introduce can thus in principle be framed in terms of categorial-selection or semantic-selection. In Plains Cree, there is a class of predicates which are s-selectionally neutral, but which syntactically select only for a nominal argument, never a clausal argument. An example of this is given in (94) with the predicate m\u00C3\u00A2ciht\u00C3\u00A2- \u00E2\u0080\u0098begin\u00E2\u0080\u0099: the nominal is fine, but the clause cannot be construed with the predicate. (94) a. m\u00C3\u00A2ciht\u00C3\u00A2w w\u00C3\u00AEy\u00C3\u00A2kana m\u00C3\u00A2ciht\u00C3\u00A2 -w w\u00C3\u00AEy\u00C3\u00A2kan-a begin.VAI-3 dish -PL \u00E2\u0080\u0098He started the dishes.\u00E2\u0080\u0099 b. * m\u00C3\u00A2ciht\u00C3\u00A2w \u00C3\u00AA-p\u00C3\u00A2hpit m\u00C3\u00A2ciht\u00C3\u00A2 -w \u00C3\u00AA- p\u00C3\u00A2hpi -t begin.VAI-3 C1-laugh.VAI-3 ---(intended: \u00E2\u0080\u0098He started laughing.\u00E2\u0080\u0099) All of these predicates which have this behaviour share the intransitive transitivizer -iht\u00C3\u00A2 \u00E2\u0080\u0098do/make\u00E2\u0080\u0099: that is, they are one of the classes of predicates in which there is a mismatch between the morphology (which indicates that the predicate is intransitive) and the syntax (which allows for an object) (cf. Wolfart 1973, Dahlstrom 1991 on AIt \u00E2\u0080\u0098Animate Intransitive transitive\u00E2\u0080\u0099 predicates). Thus the predicates such as kociht\u00C3\u00A2- \u00E2\u0080\u0098try.VAI\u00E2\u0080\u0099 (95) and k\u00C3\u00AEsiht\u00C3\u00A2- \u00E2\u0080\u0098finish.VAI\u00E2\u0080\u0099 (96) below may take a nominal argument, including English loanwords like cookies, and demonstratives like \u00C3\u00B4ma \u00E2\u0080\u0098that\u00E2\u0080\u0099, but no dependent clause. 251 (95) a. kociht\u00C3\u00A2w cookies kociht\u00C3\u00A2-w cookies try.VAI -3 cookies \u00E2\u0080\u0098S/he tried the cookies.\u00E2\u0080\u0099 b. * kociht\u00C3\u00A2w ka-nip\u00C3\u00A2t kociht\u00C3\u00A2-w ka- nip\u00C3\u00A2 -t try.VAI -3 IRR- sleep.VAI-3 --- (intended: \u00E2\u0080\u0098S/he is trying to sleep.\u00E2\u0080\u0099) (96) a. k\u00C3\u00AEsiht\u00C3\u00A2w \u00C3\u00B4ma k\u00C3\u00AEsiht\u00C3\u00A2 -w \u00C3\u00B4ma finish.VAI-3 DEM \u00E2\u0080\u0098S/he finished this.\u00E2\u0080\u0099 b. * k\u00C3\u00AEsiht\u00C3\u00A2w \u00C3\u00AA-nimihitot k\u00C3\u00AEsiht\u00C3\u00A2 -w \u00C3\u00AA- nimihito -t finish.VAI-3 C1-dance.VAI-3 --- (intended: \u00E2\u0080\u0098S/he finished dancing.\u00E2\u0080\u0099) c. * k\u00C3\u00AEsiht\u00C3\u00A2w \u00C3\u00B4ma \u00C3\u00AA-nimihitot k\u00C3\u00AEsiht\u00C3\u00A2 -w \u00C3\u00B4ma \u00C3\u00AA- nimihito -t finish.VAI-3 DEM.INAN C1-dance.VAI-3 --- (intended: \u00E2\u0080\u0098S/he finished dancing.\u00E2\u0080\u0099) The last example, which shows that the demonstrative \u00C3\u00B4ma can function as an argument of the predicate, is particularly important, because we recall that \u00C3\u00B4ma was used to introduce propositions of transitive predicates like tapw\u00C3\u00AAwak\u00C3\u00AAyihtam \u00E2\u0080\u0098s/he believes it\u00E2\u0080\u0099. If the dependent CP were a complement of D (here instantiated by the demonstrative \u00C3\u00B4ma), then the grammaticality of \u00C3\u00B4ma would predict the grammaticality of the complement CP. Here, however, while \u00C3\u00B4ma is grammatical, the CP is not; the result is that we do not want to analyze the dependent CPs as complements of D. Nominal arguments have a privileged relationship to the verb that propositions do not share. 252 5.5.1.5 Incorporation is nominal A final diagnostic for separating nominal and clausal elements is specific to Plains Cree: noun- incorporation, a process whereby some \u00E2\u0080\u009Cnoun\u00E2\u0080\u009D may occur internal to the verbal complex, excludes clauses but not nominal arguments. Most widely known are examples like (97), where the object aw\u00C3\u00A2sis \u00E2\u0080\u0098child\u00E2\u0080\u0099 may occur external to the verbal complex as in (97a), but may be \u00E2\u0080\u0098incorporated\u00E2\u0080\u009911 into the verbal complex as in (97b). (97) a. nikanaw\u00C3\u00AAyim\u00C3\u00A2w aw\u00C3\u00A2sis ni-kanaw\u00C3\u00AAyim -\u00C3\u00A2 -w aw\u00C3\u00A2sis 1-watch.over.VTA-DIR-3 child \u00E2\u0080\u0098I watched over a/the kid.\u00E2\u0080\u0099 b. nikanaw\u00C3\u00AAyim\u00C3\u00A2wason ni-kanaw\u00C3\u00AAyim -aw\u00C3\u00A2s-o -n 1- watch.over.VTA-child-INTR-SAP \u00E2\u0080\u0098I babysat.\u00E2\u0080\u0099 (from Hirose 2000:132) This is a highly productive process in Plains Cree (Wolfart 2008), and can target morphosyntactically-complex nouns like nominalizations. However, inflected stems like maskisina \u00E2\u0080\u0098shoes\u00E2\u0080\u0099 in (98a) cannot be incorporated unless the inflection is removed, as in (98b). (98) a. postinam masksina postinam -w maskisin -a put.on.VTI-3 shoe -PL \u00E2\u0080\u0098He put shoes on.\u00E2\u0080\u0099 b. postaskisin\u00C3\u00AAw [post-askisin]\u00C3\u00AA -w put.on-shoe.VAI-3 \u00E2\u0080\u0098He put his shoes on.\u00E2\u0080\u0099 11 Notice that the \u00E2\u0080\u0098incorporation\u00E2\u0080\u0099 terminology assumes a transformational relation between these two forms. While I continue to use this term since it is the most recognizable term, it is not at all clear that Plains Cree incorporation should be analyzed as a transformational process; the incorporated for could also be analyzed as a base-generated form (cf. Hirose 2001). This is a topic for further research. 253 This type of incorporation is completely unavailable to clauses. (99) * nikisk-\u00C3\u00AA-itoht\u00C3\u00AAt-\u00C3\u00AA-w ni- kisk- \u00C3\u00AA- itoht\u00C3\u00AA -t -\u00C3\u00AA -w 1- know-C1-go.VAI-3-DIR-3 --- (intended: \u00E2\u0080\u0098I know that s/he went.\u00E2\u0080\u0099) The impossibility of incorporation here may be attributed to independent factors, since the incorporated form is often a truncated or even suppletive form of the unincorporated form. Of more particular interest to the present discussion is a second type of incorporation which does not have the categorial or prosodic restrictions that stem-incorporation has. In this type of incorporation, the incorporated element can be subject or object, and can include quantifiers, demonstratives, and possessed forms. For example, in (100a) we have an indefinite object k\u00C3\u00AEkway \u00E2\u0080\u0098something\u00E2\u0080\u0099, and in (100b) we have the inferential evidential \u00C3\u00AAtokw\u00C3\u00AA, and 1st- person possessed subject nisis \u00E2\u0080\u0098my father-in-law/uncle\u00E2\u0080\u0099: (100) a. \u00C3\u00AAkoni \u00C3\u00AA-masinihtat\u00C3\u00A2y\u00C3\u00A2hk k\u00C3\u00A2-w\u00C3\u00AE-aya-k\u00C3\u00AEkway-os\u00C3\u00AEht\u00C3\u00A2y\u00C3\u00A2hk, OBJECT \u00C3\u00AAkoni \u00C3\u00AA- masinihtat\u00C3\u00A2 -y\u00C3\u00A2n-k k\u00C3\u00A2 -w\u00C3\u00AE- aya- k\u00C3\u00AEkway- os\u00C3\u00AEht\u00C3\u00A2 -y\u00C3\u00A2n-k TOPIC.ref C1-pattern.VAI -1 -PL C2-INT-CONN-something- make.VAI-1 -PL \u00E2\u0080\u0098...we would use these as patterns when we were going to make something, ...\u00E2\u0080\u0099 (EM 68) b. \u00C3\u00AA-k\u00C3\u00AE-\u00C3\u00AAtokw\u00C3\u00AA-nisis-kakw\u00C3\u00AA-miskamaw\u00C3\u00A2t iskw\u00C3\u00AAwa aya, SUBJECT \u00C3\u00AA- k\u00C3\u00AE- \u00C3\u00AAtokw\u00C3\u00AA-ni-sis- kakw\u00C3\u00AA-miskamaw-\u00C3\u00A2 -t iskw\u00C3\u00AAw-a aya C1-PREV-EVID- 1- uncle-try- find.VTA -DIR-3 woman-OBV CONN \u00E2\u0080\u0098My father-in-law must have tried to find a wife for him, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 40) If clauses behaved as arguments, this is a place where we could expect clauses to show up, but they are ungrammatical. 254 (101) a. nik\u00C3\u00AE-t\u00C3\u00A2pw\u00C3\u00AAwak\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-miyosit ni-k\u00C3\u00AE- t\u00C3\u00A2pw\u00C3\u00AAwak\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- miyosi -t 1-PREV-believe.VTI -SAP C1- good.VAI-3 \u00E2\u0080\u0098I believed she was pretty.\u00E2\u0080\u0099 b. * nik\u00C3\u00AE-\u00C3\u00AA-miyosit-tapw\u00C3\u00AAwak\u00C3\u00AAyiht\u00C3\u00AAn ni-k\u00C3\u00AE- \u00C3\u00AA- miyosi -t- tapw\u00C3\u00AAwak\u00C3\u00AAyiht\u00C3\u00AA-n 1-PREV-C1-good.VAI-3-believe.VTI -SAP --- From this I conclude that mediated argument clauses have a special status syntactically as opposed to true arguments. 5.5.1.6 Summary: Clauses do not sit in argument positions Together, the noun-incorporation facts, the subcategorization facts about this class of \u00E2\u0080\u009Cintransitive-transitive\u00E2\u0080\u009D verbs, and the facts about subjects provide language-internal evidence that propositions do not function as arguments: they are not complements, but rather they must be adjoined. The analysis of mediated argument clauses makes them formally similar to right- dislocated nominals. In fact, when we compare the two constructions in Plains Cree, we see that they have similar characteristics. For example, in (102a), we have the pronominal form k\u00C3\u00AEkw\u00C3\u00A2y \u00E2\u0080\u0098what\u00E2\u0080\u0099 preceding the verbal complex and the clause associated with and modifying it following the main clause; a prosodic break precedes the dependent clause. Similarly, in (102b), k\u00C3\u00AEkway \u00E2\u0080\u0098thing\u00E2\u0080\u0099 is associated with right-dislocated arguments maskisina \u00E2\u0080\u0098moccasins\u00E2\u0080\u0099 and astisa \u00E2\u0080\u0098mittens\u00E2\u0080\u0099 (cf. M\u00C3\u00BChlbauer 2003, Wolvengrey 2007). (102) a. k\u00C3\u00AEkw\u00C3\u00A2y \u00C3\u00AA-k\u00C3\u00AE-kitimahikot niw\u00C3\u00AEkim\u00C3\u00A2kan, [\u00C3\u00AA-k\u00C3\u00AE-minihkw\u00C3\u00AAskit]; k\u00C3\u00AEkw\u00C3\u00A2y \u00C3\u00AA- k\u00C3\u00AE- kitimah -iko -t ni- w\u00C3\u00AEkim\u00C3\u00A2kan \u00C3\u00AA- k\u00C3\u00AE- minihkw\u00C3\u00AA -ski -t thing C1-PREV-trouble.VTA-INV-3 1- spouse C1-PREV-drink.VAI -HAB-3 \u00E2\u0080\u0098What used to give my husband trouble was that he used to drink;\u00E2\u0080\u0099 (EM 28) 255 b. \u00E2\u0080\u00A6 k\u00C3\u00AEkway k-\u00C3\u00B4s\u00C3\u00AEht\u00C3\u00A2cik, maskisina \u00C3\u00AAkwa aya astisa. k\u00C3\u00AEkway k\u00C3\u00A2- os\u00C3\u00AEht\u00C3\u00A2 -t -k maskisin -a \u00C3\u00AAkwa aya astis -a thing C2- make.VAI-3 -PL moccasin-PL and CONN mitten-PL \u00E2\u0080\u0098\u00E2\u0080\u00A6 the things they made, moccasins and mittens.\u00E2\u0080\u0099 (EM48) This right-dislocation position is associated with restriction: here the things being made are more narrowly identified as moccasins and mittens. 5.5.2 VP-complementation involves restructuring In the last section I argued that CPs were never in argument positions; this means they are never in a complement position. In this section I show that there is a verbal complementation structure available. I argue this is VP-complementation as in (103). (103) CP 3 IP 3 VP 3 V VP This kind of complementation involves restructuring, where two predicates are integrated into a single clause. There are approximately a dozen preverbs which have restructuring properties.12 Table 5.11 summarizes two properties that subclassify them: (i) whether they can occur in a non- restructuring environment (i.e., a position other than the preverbal one); and (ii) whether they can occur with an inanimate subject. 12 Preverbs are a heterogenous class consisting also of clause-typing tense/aspectual, adverbial, and resumptive proforms (cf. Wolfart 1973, Cook 2003a, b; 2004), 256 Restructured (mono-clausal) Full predicate Doublet? Inanim subject p\u00C3\u00B4n- \u00E2\u0080\u0098stop\u00E2\u0080\u0099 p\u00C3\u00B4niht\u00C3\u00A2- (w/ nominal) \u00E2\u009C\u0094 k\u00C3\u00AEs- \u00E2\u0080\u0098finish / complete\u00E2\u0080\u0099 k\u00C3\u00AEs\u00C3\u00AEht\u00C3\u00A2- (w/ nominal) \u00E2\u009C\u0094 m\u00C3\u00A2ci- \u00E2\u0080\u0098start\u00E2\u0080\u0099 m\u00C3\u00A2ciht\u00C3\u00A2- (w/ nominal) \u00E2\u009C\u0094 nitawi- \u00E2\u0080\u0098go and do X\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0094 w\u00C3\u00AE- \u00E2\u0080\u0098intend / going to\u00E2\u0080\u0099 \u00E2\u009C\u0096 \u00E2\u009C\u0094 n\u00C3\u00AEht\u00C3\u00A2- \u00E2\u0080\u0098do habitually well\u00E2\u0080\u0099 \u00E2\u009C\u0096 restricted n\u00C3\u00B4ht\u00C3\u00AA- \u00E2\u0080\u0098want\u00E2\u0080\u0099 \u00E2\u009C\u0096 restricted kw\u00C3\u00AA- \u00E2\u0080\u0098try\u00E2\u0080\u0099 koc\u00C3\u00AEht\u00C3\u00A2- (w/ nominal) koci- (w/ clause) restricted Table 5.11. Restructuring preverbs in Plains Cree In this section I present evidence that restructuring in Plains Cree involves a single CP (i.e., is mono-clausal), and then I present evidence that restructuring involves a VP-complement. 5.5.2.1 Restructuring involves a single set of agreement Restructured clauses have only a single set of agreement. In (104a) we have an object-mediated clause which carries an independent set of agreement (e.g., the third-person -w and third-person -t); (10b) shows that the absence of -w results in ungrammaticality. (104) a. Jeff wanikiskisiw ka-asam\u00C3\u00A2t atimwa J wanikiskisi-w ka- asam -\u00C3\u00A2 -t atimw-a J forget.VAI -3 IRR-feed.VTA-DIR-3 dog -OBV \u00E2\u0080\u0098Jeff forgot to feed the dog.\u00E2\u0080\u0099 b. * Jeff wanikiskisi_ ka-asam\u00C3\u00A2t atimwa J wanikiskisi_ ka-asam -\u00C3\u00A2 -t atimw-a J forget.VAI_ IRR-feed.VTA-DIR-3 dog -OBV --- In restructured clauses, there is only one set of agreement (105a); adding another agreement marker to the preverb results in ungrammaticality. 257 (105) a. Jeff \u00C3\u00AA-kw\u00C3\u00AA-asam\u00C3\u00A2t atimwa J \u00C3\u00AA- kw\u00C3\u00AA- asam -\u00C3\u00A2 -t atimw -a J C1-TRY- feed.VTA-DIR-3 dog -OBV \u00E2\u0080\u0098Jeff is trying to feed the dog.\u00E2\u0080\u0099 b. * Jeff \u00C3\u00AA-kw\u00C3\u00AAt-asam\u00C3\u00A2t atimwa J \u00C3\u00AA- kw\u00C3\u00AA -t -asam -\u00C3\u00A2 -t atimw-a J C1-TRY -3-feed.VTA-DIR-3 dog -OBV --- Given that agreement is always and only associated with a CP (see the discussion in chapter 2), the single set of agreement provides evidence that the the verbal complex forms a single CP. 5.5.2.2 Independent-order agreement is possible Full dependent clauses require CONJUNCT-order agreement as shown in (106). (106) a. nikiskisin \u00C3\u00AA-p\u00C3\u00A2hpiy\u00C3\u00A2n CONJUNCT ni-kiskisi -n \u00C3\u00AA- p\u00C3\u00A2hpi -y\u00C3\u00A2n 1-remember.VAI-SAP C1-laugh.VAI-1 \u00E2\u0080\u0098I remember that I laughed.\u00E2\u0080\u0099 b. * nikiskisin nip\u00C3\u00A2hpin INDEPENDENT ni- kiskisi -n ni- p\u00C3\u00A2hpi -n 1- remember.VAI-SAP 1- laugh.VAI-SAP --- If the agreement in a restructured clause were agreement for a dependent clause, we would therefore expect that it would necessarily be in the CONJUNCT order (e.g., 107a). However, the agreement of a restructured clause can be INDEPENDENT order agreement (e.g., 107b). (107) a. \u00C3\u00AA-n\u00C3\u00AEht\u00C3\u00A2-p\u00C3\u00A2hpit Lisa CONJUNCT \u00C3\u00AA- n\u00C3\u00AEht\u00C3\u00A2- p\u00C3\u00A2hpi -t L C1-good.at-laugh.VAI-3 L \u00E2\u0080\u0098Lisa is good at laughing/laughs a lot.\u00E2\u0080\u0099 b. n\u00C3\u00AEht\u00C3\u00A2-p\u00C3\u00A2hpiw Lisa INDEPENDENT n\u00C3\u00AEht\u00C3\u00A2- p\u00C3\u00A2hpi -w L good.at-laugh.VAI-3 L \u00E2\u0080\u0098Lisa is good at laughing/laughs a lot\u00E2\u0080\u0099 258 The fact that conjunct-mode agreement is unnecessary provides evidence that restructuring yields a single clause, and thus that the complement phrase is smaller than a CP. 5.5.2.3 Restructuring allows only one set of temporal marking Restructured clauses also have only one set of preverbal temporal/realis marking (k\u00C3\u00AE-; ka-; w\u00C3\u00AE-). This is shown for the irrealis ka- in (108), and shifting preverb k\u00C3\u00AE- in (109). (108) a. nika-p\u00C3\u00B4n-m\u00C3\u00A2ton w\u00C3\u00AEpac ni- ka- p\u00C3\u00B4n- m\u00C3\u00A2to -n w\u00C3\u00AEpac 1- IRR-stop- cry.VAI-SAP soon \u00E2\u0080\u0098I will stop crying soon.\u00E2\u0080\u0099 b. * nika-p\u00C3\u00B4n-ka-m\u00C3\u00A2ton w\u00C3\u00AEpac ni- ka- p\u00C3\u00B4n- m\u00C3\u00A2to -n w\u00C3\u00AEpac 1- IRR-stop-cry.VAI-SAP soon --- (109) a. nik\u00C3\u00AE-p\u00C3\u00B4n-m\u00C3\u00A2ton ni- k\u00C3\u00AE- p\u00C3\u00B4n- m\u00C3\u00A2to -n 1- PREV-stop- cry.VAI-SAP \u00E2\u0080\u0098I stopped crying.\u00E2\u0080\u0099 b. * nik\u00C3\u00AE-p\u00C3\u00B4n-k\u00C3\u00AE-m\u00C3\u00A2ton ni- k\u00C3\u00AE- p\u00C3\u00B4n- k\u00C3\u00AE- m\u00C3\u00A2to -n 1- PREV-stop- PREV- cry.VAI-SAP --- Further, the temporal marking must precede the matrix predicate (110a); it cannot occur between the higher predicate and the stem predicate (110b). 259 (110) a. \u00C3\u00AA-k\u00C3\u00AE-kw\u00C3\u00AA-w\u00C3\u00AEcihak awa aw\u00C3\u00A2sis \u00C3\u00AA- k\u00C3\u00AE- kw\u00C3\u00AA- w\u00C3\u00AEcih -ak awa aw\u00C3\u00A2sis C2-PREV-TRY- help.VTA-1>3 DEM.AN child \u00E2\u0080\u0098I had tried helping this child\u00E2\u0080\u0099 b. * \u00C3\u00AA-kw\u00C3\u00AA-k\u00C3\u00AE-w\u00C3\u00AEcihak \u00C3\u00AA- kw\u00C3\u00AA-k\u00C3\u00AE- w\u00C3\u00AEcih -ak C1-TRY- PREV-help.VTA-1>3 \u00E2\u0080\u0098I had tried helping him\u00E2\u0080\u0099 (111) a. Jeff k\u00C3\u00AE-p\u00C3\u00B4n-m\u00C3\u00AEcisow J k\u00C3\u00AE- p\u00C3\u00B4n- m\u00C3\u00AEciso -w J PREV-stop-eat.VAI -3 \u00E2\u0080\u0098Jeff had stopped eating\u00E2\u0080\u0099 b. * Jeff p\u00C3\u00B4n-k\u00C3\u00AE-m\u00C3\u00AEcisow J p\u00C3\u00B4n- k\u00C3\u00AE- m\u00C3\u00AEciso -w J stop-PREV-eat.VAI -3 --- The inability for temporal elements to modify the complement provides evidence that the complement is smaller than IP. 5.5.2.4 Restructured clauses introduce a single set of arguments Plains Cree verbal predicates are syntactically decomposable into lexical information (the root), temporal structure, and argument structure (Hirose 1999, D\u00C3\u00A9chaine 2002, 2003). The morpho- phonological unit \u00E2\u0080\u0098stem\u00E2\u0080\u0099 corresponds to \u00E2\u0080\u0098predicate\u00E2\u0080\u0099: it consists of a root, plus one or more manner suffixes and one or more valency markers that correspond with argument structure (cf. Wolfart 1973, Hirose 1999, D\u00C3\u00A9chaine 2003). A root without a manner suffix cannot be a stem, even with the appropriate agreement. 260 (112) a. nikinosin INTRANSITIVE ni- kinosi -n 1- tall.VAI-SAP \u00E2\u0080\u0098I am tall.\u00E2\u0080\u0099 b. * nikinon ni- kino -n 1- tall -SAP --- (intended: \u00E2\u0080\u0098I am tall.\u00E2\u0080\u0099) (113) a. nip\u00C3\u00AEkon\u00C3\u00AAn TRANSITIVE ni- p\u00C3\u00AEkwn\u00C3\u00AA -n 1- break.by.hand.VTI-SAP \u00E2\u0080\u0098I broke it.\u00E2\u0080\u0099 b. * nip\u00C3\u00AEkw\u00C3\u00AAn ni- p\u00C3\u00AEkw\u00C3\u00AA -n 1- break.VTI-SAP --- (intended: \u00E2\u0080\u0098I broke it.\u00E2\u0080\u0099) (Adapted from Hirose 1999; 17a & 18a) In restructuring contexts, however, the matrix \u00E2\u0080\u0098predicate\u00E2\u0080\u0099 consists of a bare root.13 In fact, a bare root is obligatory \u00E2\u0080\u0093 no manner suffixes (temporal/argument structure) or theme signs (argument structure) are allowed. (114) a. Toni \u00C3\u00AA-k\u00C3\u00AEsi-m\u00C3\u00AEcisot pahkw\u00C3\u00AAsikana T \u00C3\u00AA- k\u00C3\u00AEsi- m\u00C3\u00AEciso -t pahkw\u00C3\u00AAskan -a T C1-finish-eat.VAI-3 bread -OBV \u00E2\u0080\u0098Toni finished eating the bread.\u00E2\u0080\u0099 b. * Toni \u00C3\u00AA-k\u00C3\u00AEsiso-m\u00C3\u00AEcisot pahkw\u00C3\u00AAsikana T \u00C3\u00AA- k\u00C3\u00AEsiso -m\u00C3\u00AEciso -t pahkw\u00C3\u00AAsikan -a T C1-finish.VAI-eat.VAI-3 bread -OBV --- When these roots are used to form simple (non-restructured) clauses, they are like every other predicate in Plains Cree \u00E2\u0080\u0093 they must take a manner suffix (valency marker). All of the aspectual roots take the neutral manner suffix -\u00C3\u00AEht\u00C3\u00A2- \u00E2\u0080\u0098do\u00E2\u0080\u0099, while non-aspectual roots take other 13 Sometimes an -i appears suffixed to these roots. Historically, this was a morphological marker of an element in preverb position (see Pentland 1979); synchronically, it is completely absent from some preverbs, and appears, contingent on syllabic and foot structure, with other preverbs. It does not appear to have any impact on the predicate status of these roots. 261 transitivizers, including but not limited to -\u00C3\u00AAyim \u00E2\u0080\u0098by mind\u00E2\u0080\u0099 and -im \u00E2\u0080\u0098by mouth\u00E2\u0080\u0099, and the neutral transitivizer -ih. (115) a. * k\u00C3\u00AEsw k\u00C3\u00AEs -w finish-3 --- b. k\u00C3\u00AEsiht\u00C3\u00A2w k\u00C3\u00AEsiht\u00C3\u00A2 -w finish.VAI-3 \u00E2\u0080\u0098S/he finishes it.\u00E2\u0080\u0099 (116) a. * niht\u00C3\u00A2w niht\u00C3\u00A2 -w want -3 --- b. niht\u00C3\u00A2w\u00C3\u00AAyihtam niht\u00C3\u00A2w\u00C3\u00AAyihtam -w good.by.mind.VTI-3 \u00E2\u0080\u0098S/he is clever/resourceful (at that).\u00E2\u0080\u0099 (Wolvengrey 2001) 5.5.2.5 Restructured clauses only permit a single subject Restructured clauses in Plains Cree have obligatory subject control. This is true even when the non-restructured clausal complement is allowed to have a disjoint subject, as in the case of nitaw\u00C3\u00AAyiht- \u00E2\u0080\u0098want\u00E2\u0080\u0099 (i.e., disjoint subjects should be allowed on semantic grounds): (117) Bi-clausal constructions allow distinct subjects a. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn niya ka-nikamoy\u00C3\u00A2n SAME SUBJECT ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n niya ka- nikamo -y\u00C3\u00A2n 1- want.VTI -SAP 1.EMPH IRR-sing.VAI-1 \u00E2\u0080\u0098I want for myself to sing.\u00E2\u0080\u0099 / \u00E2\u0080\u0098I want for myself to be able to sing.\u00E2\u0080\u0099 b. ninitaw\u00C3\u00AAyiht\u00C3\u00AAn Rose-Marie ka-nikamot DISTINCT SUBJECT ni- nitaw\u00C3\u00AAyiht\u00C3\u00AA -n RM ka- nikamo -t 1- want.VTI -SAP RM IRR-sing.VAI-3 \u00E2\u0080\u0098I want Rose-Marie to sing.\u00E2\u0080\u0099 262 (118) Restructured constructions do not allow distinct subjects a. nin\u00C3\u00B4ht\u00C3\u00AA-cihk\u00C3\u00AAyiht\u00C3\u00AAn SAME SUBJECT ni- n\u00C3\u00B4ht\u00C3\u00AA- cihk\u00C3\u00AAyiht\u00C3\u00AA -n 1- want- happy.VTI -SAP \u00E2\u0080\u0098I want to be happy.\u00E2\u0080\u0099 b. * nin\u00C3\u00B4ht\u00C3\u00AA-cihk\u00C3\u00AAyiht\u00C3\u00AAn Laura DISTINCT SUBJECT ni- n\u00C3\u00B4ht\u00C3\u00AA- cihk\u00C3\u00AAyiht\u00C3\u00AA -n L 1- want- happy.VTI -SAP L --- (intended: \u00E2\u0080\u0098I want Laura to be happy.\u00E2\u0080\u0099) This piece of evidence supports my claim that the complement phrase is a VP (rather than an IP or CP, both of which should allow disjoint subjects). 5.5.2.6 Restructuring preverbs are category-sensitive Restructuring preverbs are category-sensitive: they can only select for verbal complements, not nominal complements. (119) a. * p\u00C3\u00B4n-atim p\u00C3\u00B4n-atim stop-dog --- b. p\u00C3\u00B4n-p\u00C3\u00A2hpiw p\u00C3\u00B4n-p\u00C3\u00A2hpi -w stop-laugh.VAI-3 \u00E2\u0080\u0098S/he stopped laughing.\u00E2\u0080\u0099 Notice that restructuring preverbs are different from adverbial preverbs, which are category neutral and can modify both nominal and verbal predicates. 263 (120) a. nimisi-m\u00C3\u00AEcison ni- misi- m\u00C3\u00AEciso -n 1- lot- eat.VAI-SAP \u00E2\u0080\u0098I ate a lot.\u00E2\u0080\u0099 b. nimisi-min\u00C3\u00B4sim ni- misi- min\u00C3\u00B4s -im 1- big- cat -DSJ \u00E2\u0080\u0098my big cat\u00E2\u0080\u0099 Thus, it is in restructuring environments that we find the categorial selection we were looking for with \u00E2\u0080\u0098complement-clauses.\u00E2\u0080\u0099 This means that argument-structure morphology in Plains Cree stems selects for DPs, stem-external verbal morphology selects for VPs. CPs by contrast, are never selected; they are always adjoined. 5.5.3 Copy-to-object constructions must be local agreement In the copy-to-object phenomenon, first discussed for Plains Cree by Dahlstrom (1991), we have exactly a minimal pair: in one case the higher clause is inflected for an inanimate object, in the other case, it is inflected for an animate object, and this object must be coreferential with the subject of the lower clause.14 14 \u00E2\u0080\u009CCopy-to-object\u00E2\u0080\u009D is also known as \u00E2\u0080\u009Craising-to-object\u00E2\u0080\u009D (Frantz 1978, 1980) and \u00E2\u0080\u009Ccross-clausal-agreement\u00E2\u0080\u009D (Branigan & McKenzie 2002). Note that copy-to-object is, at least in Plains Cree, explicitly restricted to subjects (cf. Dahlstrom\u00E2\u0080\u0099s 1991 discussion on Tests for subjecthood), and cannot be applied to objects (contra the claims made in Long 1999 (Plains & Swampy Cree), Bruening 2001 (Passamaquoddy), Branigan & MacKenzie 2002 (Innu-Aimun), Ritter & Rosen 2005 (Algonquian family), Bliss 2007 (Blackfoot)). Thus for example, in a direct verb form the higher verb may agree only with George (ia-b), while in an inverse verb form the higher verb form may agree only with okosisa \u00E2\u0080\u0098his sons\u00E2\u0080\u0099(iia-b) (data from Dahlstrom 1986). (i) a. nikisk\u00C3\u00AAyim\u00C3\u00A2w George \u00C3\u00AA-s\u00C3\u00A2kih\u00C3\u00A2t okosisa agree with George ni- kisk\u00C3\u00AAyim -\u00C3\u00A2 -w G \u00C3\u00AA- s\u00C3\u00A2kih -\u00C3\u00A2 -t o- kosis -a 1- know.VTA-DIR-3 G C1-love.VTA-DIR-3 3- son -OBV \u00E2\u0080\u0098I know George loves his sons.\u00E2\u0080\u0099 b. * nikisk\u00C3\u00AAyimim\u00C3\u00A2wa George \u00C3\u00AA-s\u00C3\u00A2kih\u00C3\u00A2t okosisa agree with okosisa ni- kisk\u00C3\u00AAyim -im -\u00C3\u00A2 -w -a G \u00C3\u00AA- s\u00C3\u00A2kih -\u00C3\u00A2 -t o- kosis -a 1- know.VTA-DISJ-DIR-3 -OBV G C1-love.VTA-DIR-3 3- son -OBV --- 264 (121) a. Mary kisk\u00C3\u00AAyihtam George-a \u00C3\u00AA-\u00C3\u00A2hkosiyit NON-COPY-TO-OBJECT M kisk\u00C3\u00AAyihtam -w G -a \u00C3\u00AA- \u00C3\u00A2hkosi -yi -t M know.VTI -3 G -OBV C1-sick.VAI-DS-3 \u00E2\u0080\u0098Mary knows George is sick.\u00E2\u0080\u0099 b. Mary kisk\u00C3\u00AAyim\u00C3\u00AAw George-a \u00C3\u00AA-\u00C3\u00A2hkosiyit COPY-TO-OBJECT M kisk\u00C3\u00AAyim -\u00C3\u00AA -w G -a \u00C3\u00AA- ahkosi -yi -t M know.VTA-DIR-3 G -OBV C1-sick.VAI-DS-3 \u00E2\u0080\u0098Mary knows George is sick.\u00E2\u0080\u0099 (Dahlstrom 1986:85) There are two positions in the literature on the relation between these two forms. The first position is that in non-copy-to-object constructions, the main clause agrees with a proposition, while in copy-to-object constructions it agrees with an animate object, which then must be co- referent with the subject of the lower clause (Dahlstrom 1991, 1995). I will call this the object- agreement hypothesis. (122) CP 6 AGR OBJECT CP 6 SUBJ\u00E2\u0080\u00A6 This thesis takes the position that the object-agreement in these cases is always a case of local object agreeent. The second position, argued for by Branigan & MacKenzie (2002), is that copy-to-object constructions have long-distance agreement with the topic of the embedded clause. On this view, object-agreement agrees with two types of elements: arguments, and non-arguments that are \u00E2\u0080\u009Caltruistically check[ed] \u00E2\u0080\u00A6 to allow a topicalization structure to be established in the embedded clause that could not otherwise occur\u00E2\u0080\u009D (Branigan & MacKenzie 2002:386). This means that the argument of the lower clause escapes into some higher position where it can license agreement. I call this the long-distance agreement hypothesis; the relevant structure is (ii) a. * nikisk\u00C3\u00AAyim\u00C3\u00A2w George \u00C3\u00AA-s\u00C3\u00A2kihikot okosisa agree with George ni- kisk\u00C3\u00AAyim -\u00C3\u00A2 -w G \u00C3\u00AA- s\u00C3\u00A2kih -iko -t o- kosis -a 1- know.VTA-DIR-3 G C1-love.VTA-INV -3 3- son -OBV --- b. nikisk\u00C3\u00AAyimim\u00C3\u00A2wa George \u00C3\u00AA-s\u00C3\u00A2kihikot okosisa agree with okosisa ni-kisk\u00C3\u00AAyim -im -\u00C3\u00A2 -w -a G \u00C3\u00AA- s\u00C3\u00A2kih -iko -t o- kosis -a 1- know.VTA-DISJ-DIR-3 -OBV G C1-love.VTA-INV -3 3- son -OBV \u00E2\u0080\u0098I know that his sons love George.\u00E2\u0080\u0099 265 given in (123). Here the agreement in the upper clause (Xprobe) targets the subject of the lower clause (Xgoal), even though this crosses a CP-boundary. (123) Structure for cross-clausal agreement (Branigan & MacKenzie 2002) CP 5 Xprobe CP 2 Xgoali 2 IP 5 ti Branigan & MacKenzie posit this structure because they believe there are a number of problems with the local agreement hypothesis that cannot be overcome. However, as they themselves point out, this analysis means that object-agreement in Innu-Aimun must be able to code two different kinds of relations \u00E2\u0080\u0093 one regular local object agreement, and one long-distance agreement, although there is no independent evidence that it is doing both. Thus, if the problems with the local object agreement turn out not to be problems, the local object agreement is more economic. I believe the local object agreement analysis can be maintained, at least in Plains Cree. First, many of arguments used to argue against the local object agreement hypothesis show that there may be more divergence between the so-called \u00E2\u0080\u0098dialects\u00E2\u0080\u0099 of Cree than is usually assumed. The syntax of copy-to-object constructions (and cross-clausal syntax more generally) in Innu-Aimun is quite different than Plains Cree on a number of grounds. For example, Branigan & MacKenzie show that in Innu-Aimun, discontinuous quantifiers are only allowed in copy-to-object constructions, but in Plains Cree, as we have already seen, quantifiers may be discontinuous in any mediated argument clause (\u00C2\u00A75.4.2). Thus, in Plains Cree there is nothing special about copy-to-object constructions. Similarly, the evidence that the prothetic object analysis cannot work is based on cases where plural referents are used in conjunction with singular agreement on the verb to get a distributive interpretation; in copy-to-object constructions, the agreement must match in both clauses, a fact which Branigan & MacKenzie claim cannot be accounted for by a prothetic object analysis. 266 (124) P\u00C3\u00BBn m\u00C3\u00A2k M\u00C3\u00A2n\u00C3\u00AE nikam\u00C3\u00BBpan Paul and Mary sing.3SG = Paul sang and Mary sang. (Branigan & MacKenzie 2002 (27b)) (125) a. N-u\u00C3\u00AE -tshissenim-\u00C3\u00A2u tsheku\u00C3\u00A2nnu kuet m\u00C3\u00BBpisht\u00C3\u00BBt P\u00C3\u00BBn m\u00C3\u00A2k M\u00C3\u00A2n\u00C3\u00AE. 1-want-know-3SG why visited-2SG/3SG Paul and Marie \u00E2\u0080\u0098I want to know why you visited Paul and Marie.\u00E2\u0080\u0099 (B&M 2002 (28b)) b. * N-u\u00C3\u00AE-tshissenim-\u00C3\u00A2u tsheku\u00C3\u00A2nnu kuet m\u00C3\u00BBpisht\u00C3\u00BBt-\u00C3\u00A2u P\u00C3\u00BBn m\u00C3\u00A2k M\u00C3\u00A2n\u00C3\u00AE. 1-want-know-3SG why visited-2SG/3PL Paul and Marie --- (B&M 2002 (29)) In Plains Cree, however, singular agreement is impossible in this context, and thus this test is inapplicable. (126) a. Paul \u00C3\u00AAkwa Mary nikamowak P \u00C3\u00AAkwa M nikamo-w-ak P and M sing.VAI -3-PL \u00E2\u0080\u0098Paul and Mary sang.\u00E2\u0080\u0099 b. * Paul \u00C3\u00AAkwa Mary nikamow P \u00C3\u00AAkwa M nikamo-w P and M sing.VAI -3 --- (intended: \u00E2\u0080\u0098Paul and Mary sang.\u00E2\u0080\u0099) A final place where Plains Cree seems to diverge from Innu-Aimun is with respect to the linear position of the overt nominal associated with both clauses. Branigan & MacKenzie point out that the overt nominal is sometimes linearized within the lower clause, and conclude that the prothetic object analysis (or any local object agreement analysis) would predict this to be a violation of condition C, since the pronominal in the matrix clause would presumably bind the R- expression in the dependent clause. However, as we have seen earlier, there are reasons to think that at least some overt nominals are introduced at the text-level in the sense of McCawley (1970). In fact, in general, Plains Cree nominals are much free-er in their linearization than English (or Innu-Aimun) appear to be. For example, in (128) the nominal associated with the initial (embedded) clause is discontinuous across a superordinate intransitive clause. 267 (127) \u00E2\u0080\u009C \u00E2\u0080\u0098miskahkw\u00C3\u00A2wi, nika-misihon, maskihk\u00C3\u00AEwiyiniwak,\u00E2\u0080\u0099 nititik \u00E2\u0080\u00A6\u00E2\u0080\u009D miskam -k-w\u00C3\u00A2w-i ni-ka-misiho -n maskihk\u00C3\u00AEwiyiniw-ak ni(t)-it -ik -w find.VTI-0-3PL -SUBJ 1- IRR-trouble.VAI-SAP doctor -PL 1- say.VTA-INV-3 \u00E2\u0080\u0098 \u00E2\u0080\u009C \u00E2\u0080\u0098If the doctors find it, I\u00E2\u0080\u0099ll be in trouble,\u00E2\u0080\u0099 he said to me ...\u00E2\u0080\u009D \u00E2\u0080\u0099 (AA 4.8) (Lit: if they find it, I\u00E2\u0080\u0099ll be in trouble, (the) doctors,\u00E2\u0080\u0099 he said to me.) Conversely, the fact that the overt nominal can appear in the higher clause in Innu- Aimun\u00E2\u0080\u0099s copy-to-object construction is not limited to copy-to-object constructions in Plains Cree. As we saw earlier, overt nominals can be productively be situated to the left of an intervening clause. In Plains Cree, the position of the overt nominal is simply not sensitive to copy-to-object, and does not tell us much about local- vs. long-distance agreement. Some of the other arguments that Branigan & MacKenzie use to argue against the local agreement anlaysis are in fact arguments against the prothetic object hypothesis, based on a construction that occurs in English. There are, in fact, several constructions in English that have what also looks like object agreement. These include the prothetic object construction (128a), the exceptional case-marking (ECM) construction (128b), and the direct percept constructions (128c). (128) a. I saw of Mary that she was sleeping prothetic object b. I saw Mary sleeping. ECM c. I saw Mary \u00E2\u0080\u0093 she was sleeping. direct percept Branigan & McKenzie (2002) argue that copy-to-object cannot be local object agreement paralleling either of the structures in (128a-b). For example, prothetic object constructions do not allow conjoined embedded clauses, but copy-to-object constructions do. (129) a. * Peter believed of her that Marie fixed the car and Paula washed it b. Peter believed of them that Marie fixed the car and Paula washed it. 268 (130) a. Pien u\u00C3\u00AEtshi-eu M\u00C3\u00A2n\u00C3\u00AEua ni\u00C3\u00A2tiniminitsh\u00C3\u00AE ut\u00C3\u00A2p\u00C3\u00A2nnu \u00C3\u00A2ku \u00C3\u0082n\u00C3\u00AEua u\u00C3\u00AEtshinit k\u00C3\u00AEe P\u00C3\u00BBna Peter help-3 Marie push truck rear Annie house and Paul \u00E2\u0080\u0098Peter helped Marie push the truck behind Annie\u00E2\u0080\u0099s house and Paul uieuesht\u00C3\u00A2nitsh\u00C3\u00AE ishk\u00C3\u00AEt\u00C3\u00BBnnu. fix Ski-Doo (to) fix the Ski-Doo.\u00E2\u0080\u0099 b. * N-u\u00C3\u00AEtshi\u00C3\u00A2uat Pien uieuesht\u00C3\u00A2t ishk\u00C3\u00AEt\u00C3\u00BBnnu m\u00C3\u00A2k \u00C3\u0082n\u00C3\u00AE 1-help-3PL Peter fix Ski-Doo and Annie ni\u00C3\u00A2tin\u00C3\u00A2k ut\u00C3\u00A2p\u00C3\u00A2nnu \u00C3\u00A2ku \u00C3\u0082n\u00C3\u00AE u\u00C3\u00AEtsh\u00C3\u00AEt. push truck rear Annie house --- (no intended translation given) Conversely, prothetic object clauses allow the prothetic object to be coreferential with embedded DPs (e.g., a possessor), but copy-to-object constructions do not. The problem above, however, seems to be a problem with associating the copy-to-object construction in Innu-Aimun (or other Algonquian languages) with the prothetic object construction in English. It is clear that there is more than one way to have object agreement \u00E2\u0080\u0093 even within English there are multiple object-agreement constructions (cf. 128a-c). Plains Cree\u00E2\u0080\u0099s copy-to-object constructions share some properties with each of the English constructions, but the fact that Plains Cree does not share all of the properties of any of English\u00E2\u0080\u0099s object-agreement constructions does not rule out the possibility that Plains Cree\u00E2\u0080\u0099s object- agreement is local. Summing up, then, the analysis of mediated argument clauses adopted here, where there is a uniform object-agreement syntax for Plains Cree copy-to-object and non-copy-to-object constructions, can be maintained. 5.6 Summary In this chapter I argued that anaphoric clauses in Plains Cree subclassify into three types based on their syntactic properties: (i) CHAINS, which are sensitive to precedence but not c-command; (ii) ADJUNCTS, which are sensitive to c-command but not precedence; 269 (iii) MEDIATED ARGUMENTS, which are sensitive to both c-command and precedence. I presented 2 kinds of diagnostics to motivate this classification. First, exclusion tests consistently fail with chained clauses, picking out sensitivity to c-command. Second, fronting tests consistently pick out mediated argument clauses. Finally, I reported on a variety of consequences for this typology of clausal relations, including the lack of argument clauses, the lack of complementation, and uniform object agreement. If there is no clausal complementation in Plains Cree, then we have to conclude that the term \u00E2\u0080\u0098complementizer\u00E2\u0080\u0099 (i.e., element that creates a complement) in Plains Cree is misleading: the CP layer of clauses in Plains Cree does not create complements. This result is in line with Reinholtz (2007) who claims that the function of C is parameterized and that the Cree complementizer system does not have functions that other, more studied systems have: for example, Plains Cree\u00E2\u0080\u0099s complementizer system lacks the illocutionary distinctions between declarative, interrogative and imperative, made in a language like English. On a broader view, Plains Cree\u00E2\u0080\u0099s complementizer system is doing exactly what we might expect: \u00E2\u0080\u009C[it is] the interface between a propositional content (expressed by the IP) and the superordinate structure (a higher clause, or, possibly, the articulation of discourse, if we consider a root clause)\u00E2\u0080\u009D (Rizzi 1997:283). Rather than coding a syntactic relation of complementation, Plains Cree\u00E2\u0080\u0099s clause-typing system codes a relation of anaphora, which is subject to the general properties of anaphora. 270 CHAPTER 6 THE SEMANTICS OF ANAPHORIC CLAUSES 6.1 Proposal: Presuppositional, a-veridical, and unspecified clauses Up until now, I have developed an analysis that captures the distinction between indexical clauses (instantiated by Plains Cree\u00E2\u0080\u0099s INDEPENDENT order) and anaphoric clauses (instantiated by Plains Cree\u00E2\u0080\u0099s CONJUNCT order). In chapter 5, we saw that anaphoric clauses come in three different syntactic \u00E2\u0080\u0098flavours\u00E2\u0080\u0099: (i) they may be in an anaphoric chain; (ii) they may be adjoined to a CP; or (iii) they may be introduced via a mediated argument and adjoined to VP or IP. We further saw that this syntactic classification cross-cut the morpho-syntactic classification (i.e., the form of the clause-typing proclitic) of the CONJUNCT order. In this chapter I turn to the classification of semantic functions of anaphoric clauses. I argue that there are three semantic \u00E2\u0080\u0098flavours\u00E2\u0080\u0099 of Plains Cree CONJUNCT clauses, corresponding to the three forms of the clause-typing proclitic: k\u00C3\u00A2- introduces a presuppositional clause; the covert \u00EF\u0081\u00B8 introduces a a-veridical clause, and \u00C3\u00AA- is an elsewhere case, i.e., introduces a semantically- unspecified clause. FUNCTION FORM (i) Presuppositionality To presuppose a proposition in the pragmatic sense is to take its truth for granted, and to assume that others involved in the context do the same. (Stalnaker 1999:38) k\u00C3\u00A2-CONJUNCT (ii) A-veridicality A proposition is a-veridical if and only if there is no possible entailment of p or \u00C2\u00ACp in any individual\u00E2\u0080\u0099s epistemic model ME(x). \u00EF\u0081\u00B8-CONJUNCT (simple CONJUNCT) (iii) Elsewhere To be elsewhere is to be unspecified, occurring in the absence of a specified element. \u00C3\u00AA-CONJUNCT 271 There is therefore a one-to-one mapping between the semantic function of the clause and the form of the clause-typing proclitic. Almost nothing has been said in the Plains Cree literature about the function and distribution of CONJUNCT clauses. Wolfart (1973, 1996) provides a cursory summary of the function of each clause type. Blain (1997) provides a syntactic account of clause-typing with respect to wh-constructions but does not discuss how the analysis she proposes generalizes to other constructions with the same clause-typing. She also does not discuss the interpretational differences between different kinds of wh-questions. Long (1999) provides a syntactic analysis of complement-like clauses, but limits herself to discussing only CONJUNCT clauses with the complementizer \u00C3\u00AA-, and does not discuss the semantics of the complementizer. This chapter, therefore, marks the first time these generalizations have been presented, and the first attempt to develop an analysis. I will start by introducing the clause-type I am claiming is semantically unspecified (\u00C2\u00A76.2). I then examine each of two specified types of CONJUNCT clauses: the presuppositionality of k\u00C3\u00A2- (\u00C2\u00A76.3) and the a-veridicality of the simple CONJUNCT (\u00C2\u00A76.4). 6.2 \u00C3\u00AA- as the unspecified complementizer Clauses introduced by the \u00C3\u00AA-complementizer are by far the most ubiquitous in Plains Cree: they occur most frequently, and are in the widest distribution. When we consider all of the possible contexts \u00C3\u00AA-clauses can appear in, there seems little hope of providing a single property that unifies these contexts. Here I take the position that \u00C3\u00AA- is a semantically unspecified complementizer: its interpretation is a function of the contrast it provides to some other clause- type (cf. Goddard 2004 on animacy). In terms of its distribution, it occurs where the more- specified clause-type is infelicitous, and its distribution acquires the meaning that the specified clause-type doesn\u00E2\u0080\u0099t have. We then end up with three different contrasts. First, there is the contrast between \u00C3\u00AA- and k\u00C3\u00A2- clauses; this is a contrast between two overt complementizers. The k\u00C3\u00A2- complementizer is specified for presuppositionality (\u00C2\u00A76.3). 272 (1) a. CP 3 3 k\u00C3\u00A2- XP [presuppositionality] b. CP 3 3 \u00C3\u00AA- XP Second, there is a contrast between \u00C3\u00AA- and \u00EF\u0081\u00B8-clauses. This contrast is between an overt complementizer and a null complementizer. (2) a. CP 3 3 \u00EF\u0081\u00B8 XP [a-veridical] b. CP 2 2 \u00C3\u00AA- XP As with k\u00C3\u00A2-, \u00EF\u0081\u00B8 is the specified member of the pair. The \u00EF\u0081\u00B8-clause (simple CONJUNCT) codes a- veridicality (\u00C2\u00A76.4), while the \u00C3\u00AA-clause is unspecified. In the next section, I show how positing \u00C3\u00AA- as an unspecified complementizer accounts for its behaviour. 6.2.1 Distributional evidence for \u00C3\u00AA- being unspecified In terms of their distribution, unspecified elements are predicted to occur in a wider set of contexts than their specified counterparts (Battistella 1990:37; see also Fort 1919, Jakobsen 1929, Trubetzkoy 1969, Aronoff 1976, Williams 1997). Distribution can be divided into two parts: the contexts that the element occurs in, and the other elements it combines with1. If we consider the distribution of clauses in Plains Cree, it is clear that the \u00C3\u00AA- clauses have the widest distribution: they occur in unembedded contexts, in relative clause contexts, and in mediated argument contexts. It is the only clause-type in the language that can do this. Further, every time two or more clause-types are grammatically possible, the \u00C3\u00AA- clause is one of the possibilities. 1 Some linguists use frequency as another distribution measure. Since frequency alone provides only limited information, I do not include it here, but a quick glance through any Plains Cree text will suffice to show that the \u00C3\u00AA- clause-type is by far the most ubiquitous. 273 In unembedded contexts, we have indexical INDEPENDENT clauses, and anaphoric CONJUNCT clauses: the anaphoric CONJUNCT clauses have the complementizer \u00C3\u00AA-, as summarized in table 6.1. CLAUSE RELATION IND \u00C3\u008A- K\u00C3\u0082- \u00EF\u0081\u00B8 Indexical \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 unembedded Elsewhere (=anaphoric) \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Table 6.1. \u00C3\u00AA-clauses are anaphoric in unembedded contexts In relative-clause contexts, both k\u00C3\u00A2- and \u00C3\u00AA- clauses are possible. The k\u00C3\u00A2- complementizer occurs in presuppositional relative clauses, and the \u00C3\u00AA- complementizer occurs in the other, non- presuppositional relative clauses. CLAUSE RELATION K\u00C3\u0082 \u00C3\u008A- IND \u00EF\u0081\u00B8 presuppositional \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 relative clause: argument modification non-presuppositional \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 presuppositional \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 wh-clefts non-presuppositional \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 clause 1 presuppositional \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 correlatives clause 2 non-presuppositional \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Table 6.2. \u00C3\u00AA-clauses are non-presuppositional in relative-clause contexts Finally, in mediated argument clauses, both simple CONJUNCT and \u00C3\u00AA- clauses are possible. The simple CONJUNCT specifies a-veridicality, and the \u00C3\u00AA- clause is again unspecified. CLAUSE RELATION \u00EF\u0081\u00B8 \u00C3\u008A- IND K\u00C3\u0082- a-veridical \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0096 mediated argument clauses veridical \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 Table 6.3. \u00C3\u00AA-clauses are veridical in mediated argument clauses Similarly, \u00C3\u00AA- clauses have the widest ability to combine with other elements. For example the particle os\u00C3\u00A2m has two functions: it can be a intensify the quantifier mistahi \u00E2\u0080\u0098much\u00E2\u0080\u0099 (os\u00C3\u00A2m mistahi \u00E2\u0080\u0098too much\u00E2\u0080\u0099), or it can be a subordinator indicating \u00E2\u0080\u0098because\u00E2\u0080\u0099. As an intensifier, we find it in both indexical INDEPENDENT clauses (3a), and anaphoric CONJUNCT clauses (3b). 274 (3) a. \u00E2\u0080\u0093 \u00C3\u00AAkosi m\u00C3\u00A2n \u00C3\u00AA-itw\u00C3\u00AAy\u00C3\u00A2n, INDEPENDENT \u00C3\u00AAkosi m\u00C3\u00A2na \u00C3\u00AA- itw\u00C3\u00AA -y\u00C3\u00A2n TOP usually C1-say.VAI-1 \u00E2\u0080\u0098\u00E2\u0080\u0093 that is what I usually say, os\u00C3\u00A2m mistah \u00C3\u00A2tiht ayiw\u00C3\u00A2k\u00C3\u00AAyim\u00C3\u00AAwak s\u00C3\u00B4niy\u00C3\u00A2wa. os\u00C3\u00A2m mistahi atiht ayiw\u00C3\u00A2k\u00C3\u00AAyim -\u00C3\u00AA -w -ak s\u00C3\u00B4niy\u00C3\u00A2w-a too much some emphasize.VTA-DIR-3 -PL money -OBV some people put too much emphasis on money.\u00E2\u0080\u0099 (EM 63) b. ..., os\u00C3\u00A2m m\u00C3\u00A2na mistahi m\u00C3\u00AEn \u00C3\u00AA-k\u00C3\u00AE-atosk\u00C3\u00AAt aya w\u00C3\u00A2sak\u00C3\u00A2m n\u00C3\u00AEkin-~ CONJUNCT os\u00C3\u00A2m m\u00C3\u00A2na mistahi m\u00C3\u00AEna \u00C3\u00AA- k\u00C3\u00AE- atosk\u00C3\u00AA -t aya w\u00C3\u00A2sak\u00C3\u00A2m n- \u00C3\u00AEkin too usually much also C1-PREV-work.VAI-3 CONN around 1- house \u00E2\u0080\u0098..., and she also worked too hard around our house \u00E2\u0080\u0093~ \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 56) As a subordinator, however, it occurs with only one clause-type. That clause-type is the unspecified one: the \u00C3\u00AA- CONJUNCT, as exemplified in (4). (4) ..., \u00C3\u00AAwako \u00C3\u00B4ma, os\u00C3\u00A2m \u00C3\u00AA-n\u00C3\u00AAhiyaw\u00C3\u00AAy\u00C3\u00A2n mitoni, ... \u00C3\u00AA-CONJUNCT \u00C3\u00AAwakw \u00C3\u00B4ma os\u00C3\u00A2m \u00C3\u00AA- n\u00C3\u00AAhiyaw\u00C3\u00AA -y\u00C3\u00A2n mitoni TOP DEM.INAN because C1-speak.Cree.VAI -1 really \u00E2\u0080\u0098..., that is the reason, because I truly speak Cree, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (SW 1.2) Assuming that this is not a case of homophony, anaphoric \u00C3\u00AA-clauses thus allow both interpretations (i.e., they do not restrict the interpretation), while other clause-types restrict the meaning to one part of the term. In the following discussion, we will see in greater detail the interaction of the function of particles with clause-typing. 6.2.2 Interpretational evidence that \u00C3\u00AA- is unspecified The interpretational evidence for determining specification has to do with the fact that a semantically unspecified form does not have any dedicated interpretation. Where the grammar prohibits one member of the contrast from occurring, the contrast is neutralized, and we expect that the unspecified member will lose its contextual value2. 2 Marked elements, on the other hand, should retain their specification is contexts where the contrast is neutralized. This is accurate for k\u00C3\u00A2- and simple-CONJUNCT clauses; see \u00C2\u00A76.3 and \u00C2\u00A76.4 for details. 275 Applying this criteria to the clause-typing system in Plains Cree, the prediction is that there will be contexts where \u00C3\u00AA- clauses lack non-presuppositional force (which it has in opposition to k\u00C3\u00A2-) and contexts where they lack veridical force (which it has in opposition to \u00EF\u0081\u00B8). Specifically, the \u00C3\u00AA-clauses\u00E2\u0080\u0099 contextually-given force will disappear when the other clause-type is grammatically impossible. This prediction is borne out for both the presuppositional contrast and the veridicality contrast. For example, in (5), the simple CONJUNCT introduces an a-veridical proposition, and the \u00C3\u00AA- CONJUNCT, which is occurring in the same syntactic context, introduces a veridical proposition; here the \u00C3\u00AA- clause is in contrast with the simple CONJUNCT, and has the complementary interpretation. (5) a. piko ka-w\u00C3\u00A2pamak n\u00C3\u00A2p\u00C3\u00AAw SIMPLE CONJUNCT piko ka- w\u00C3\u00A2pam -ak n\u00C3\u00A2p\u00C3\u00AAw be.necessary IRR-see.VTA-1>3 man \u00E2\u0080\u0098I have to see that man.\u00E2\u0080\u0099 = (i) I have not necessarily seen that man, but it is necessary that I do at some point. \u00E2\u0089\u00A0 (ii) I have seen that man and it was necessary. b. piko \u00C3\u00AA-w\u00C3\u00A2pamak n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u008A-CONJUNCT piko \u00C3\u00AA- w\u00C3\u00A2pam -ak n\u00C3\u00A2p\u00C3\u00AAw be.necessary C1-see.VTA -1>3 man \u00E2\u0080\u0098I have to see that man.\u00E2\u0080\u0099 \u00E2\u0089\u00A0 (i) I have not seen that man, but it is necessary that I do at some point = (ii) I have seen that man and it was necessary. However, when a simple CONJUNCT clause is impossible, \u00C3\u00AA- can be used in an a-veridical context; for example, an irrealis concessive clause introduced by kiy\u00C3\u00A2m \u00E2\u0080\u0098although\u00E2\u0080\u0099 (6a). This means crucially that we cannot assign a particular veridicality value to the \u00C3\u00AA-clause: rather, it is unspecified. (6) a. * kiy\u00C3\u00A2m ka-mam\u00C3\u00A2y\u00C3\u00AEy\u00C3\u00AAk \u00C3\u00AA-p\u00C3\u00AEkiskw\u00C3\u00AAy\u00C3\u00AAk SIMPLE CONJUNCT kiy\u00C3\u00A2m \u00C3\u00AA- mam\u00C3\u00A2y\u00C3\u00AE -y\u00C3\u00AAk \u00C3\u00AA- p\u00C3\u00AEkiskw\u00C3\u00AA -y\u00C3\u00AAk even C1-make.mistake.VAI-2PL C1-speak.VAI -2PL --- (intended: \u00E2\u0080\u0098\u00E2\u0080\u00A6even if you make mistakes when you speak.\u00E2\u0080\u0099) b. ..., kiy\u00C3\u00A2m \u00C3\u00AA-mam\u00C3\u00A2y\u00C3\u00AEy\u00C3\u00AAk \u00C3\u00AA-p\u00C3\u00AEkiskw\u00C3\u00AAy\u00C3\u00AAk, ... \u00C3\u008A-CONJUNCT kiy\u00C3\u00A2m \u00C3\u00AA-mam\u00C3\u00A2y\u00C3\u00AE -y\u00C3\u00AAk \u00C3\u00AA- p\u00C3\u00AEkiskw\u00C3\u00AA -y\u00C3\u00AAk even C1-make.mistake.VAI-2PL C1-speak.VAI-2PL \u00E2\u0080\u0098\u00E2\u0080\u00A6, even if you make mistakes when you speak, ...\u00E2\u0080\u0099 (SW 1.2) 276 Likewise, the presuppositionality contrast shows that \u00C3\u00AA- does not have a fixed value. In particular, when a k\u00C3\u00A2- clause is syntactically impossible, an \u00C3\u00AA- clause is used to introduce clauses that look presuppositional, e.g., factive mediated argument clauses. (7) a. * Laura kisk\u00C3\u00AAyihtam Sam k\u00C3\u00A2-m\u00C3\u00AEcisoyit cookies L kisk\u00C3\u00AAyihtam -w S k\u00C3\u00A2- m\u00C3\u00AEciso -yi -t cookies L know.VTI -3 S C1- eat.VAI-DS-3 cookies --- (intended: \u00E2\u0080\u0098Laura knows that Sam ate the cookies.\u00E2\u0080\u0099) b. Laura kisk\u00C3\u00AAyihtam Sam \u00C3\u00AA-m\u00C3\u00AEcisoyit cookies L kisk\u00C3\u00AAyihtam -w S \u00C3\u00AA- m\u00C3\u00AEciso -yi -t cookies L know.VTI -3 S C1-eat.VAI-DS -3 cookies \u00E2\u0080\u0098Laura knows that Sam ate the cookies.\u00E2\u0080\u0099 This data argues against analyzing \u00C3\u00AA- as having a dedicated non-presuppositional semantic value. The most we can say is that \u00C3\u00AA- is non-presuppositional when it occurs in a context where a marked presuppositional k\u00C3\u00A2- clause is also possible. The criteria of \u00E2\u0080\u0098indeterminateness\u00E2\u0080\u0099 also gets at the semantic value of the unmarked member of the value (i.e., that it doesn\u00E2\u0080\u0099t have one). This means that there are places where the unmarked member of the opposition can be substituted for the marked one in some contexts. Taking this criteria seriously may help think about the cases where the \u00C3\u00AA- clause and some other clause occur in contexts that are difficult to tease apart. For example, some of the temporal sequencers, such as isp\u00C3\u00AE \u00E2\u0080\u0098at.that.time\u00E2\u0080\u0099 or mayaw \u00E2\u0080\u0098as.soon.as\u00E2\u0080\u0099 can introduce both \u00C3\u00AA- and k\u00C3\u00A2- clauses (exemplified in (8)). I am not implying that there is no difference between these two forms, but rather saying that the presupposition / non-presupposition distinction seems to have disappeared. 277 (8) a. \u00E2\u0080\u00A6, s\u00C3\u00AAm\u00C3\u00A2k aya \u00C3\u00AA-k\u00C3\u00AE-kiskinohamawit, mayaw k\u00C3\u00A2-p\u00C3\u00AA-w\u00C3\u00AEc\u00C3\u00AAwakik, \u00E2\u0080\u00A6 s\u00C3\u00AAm\u00C3\u00A2k aya \u00C3\u00AA- k\u00C3\u00AE- kiskinhoamaw-it right.away CONN C1-PREV-teach.VTA -3>1 mayaw k\u00C3\u00A2- p\u00C3\u00AA- w\u00C3\u00AEc\u00C3\u00AAw -ak -ik as.soon.as C2-come-dwell.VTA-1>3-PL \u00E2\u0080\u0098\u00E2\u0080\u00A6 she taught me right away, as soon as I came to live with them, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 48) b. ..., \u00C3\u00AA-k\u00C3\u00AE-kak\u00C3\u00AAskimit ana s\u00C3\u00AAm\u00C3\u00A2k aya, mayaw \u00C3\u00AA-k\u00C3\u00AEsi-kihci-w\u00C3\u00AEkihtoy\u00C3\u00A2hk aya, \u00E2\u0080\u00A6 \u00C3\u00AA- k\u00C3\u00AE- kak\u00C3\u00AAskim -it ana s\u00C3\u00AAm\u00C3\u00A2k aya C1-PREV-counsel.VTA-3>1 DEM.AN right.away CONN \u00E2\u0080\u0098\u00E2\u0080\u00A6, [she] immediately began to counsel me, mayaw \u00C3\u00AA- k\u00C3\u00AEsi- kihciw\u00C3\u00AEkih-to -y\u00C3\u00A2n-k aya as.soon.as C1-finish-marry.VTA -REFL-1 -PL CONN as soon as we had gotten married, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 42) 6.3 k\u00C3\u00A2- as a presuppositional complementizer In this section I argue that the complementizer k\u00C3\u00A2- introduces a presuppositional clause: the proposition is presupposed. (9) To presuppose a proposition in the pragmatic sense is to take its truth for granted, and to assume that others involved in the context do the same. (Stalnaker 1999:38) According to this definition, presupposed propositions differ from non-presupposed ones in terms of how they relate to the speaker and hearer. In a non-presupposed proposition, part of the information being conveyed is an explicit claim about the truth of the proposition3. In a presupposed proposition, no such explicit claim is being made; the truth is assumed. One test for the presuppositionality of a proposition is whether the assumed truth of the proposition survives under negation. If negation can deny the truth of the proposition, the proposition is taken to be non-presupposed; if negation cannot deny the truth of the proposition, the proposition is taken to be presupposed. For example, in (10), the factive verb remember introduces a factive presupposed complement (Kiparsky & Kiparksy 1971). In (10b), the 3 Regardless of whether this truth is relative to an individual (Lasersohn 2005, Stephenson 2007, K\u00C3\u00B6lbel 2003; see also D\u00C3\u00A9chaine 2007, M\u00C3\u00BChlbauer 2008 for Plains Cree). 278 negation in the main clause denies the main clause proposition, but not the embedded clause proposition. (10) a. John remembered that Bill left for Chicago yesterday. Presupposition: Bill left for Chicago yesterday. b. John didn\u00E2\u0080\u0099t remember that Bill left for Chicago yesterday. Presupposition: Bill left for Chicago yesterday. In the following sections I show that k\u00C3\u00A2- clauses are found in a range of presuppositional contexts, including relative clauses (\u00C2\u00A76.3.1), wh-clauses (\u00C2\u00A76.3.2), temporal modification clauses (\u00C2\u00A76.3.3), concessive clauses (\u00C2\u00A76.3.4) and correlative constructions (\u00C2\u00A76.3.5). 6.3.1 Relative clauses: k\u00C3\u00A2- and \u00C3\u00AA- Relative clauses \u00E2\u0080\u0093 clauses which modify and restrict a nominal \u00E2\u0080\u0093 are abundant in Plains Cree, particularly since adjectival modification is usually expressed via a full clause. I take a relative clause to be a CP with an operator in spec, CP that is adjoined to the NP serving as its antecedent, as in (11) (Heim & Kratzer 1998, Alexiadou et al. 2000, Bhatt 2002, Bianchi 1999, 2000). (11) NP 3 NPi CPi 4 2 Opi 2 k\u00C3\u00A2-/\u00C3\u00AA- 5 \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 As the examples in (12a-b) illustrate, a relative clause can modify both subjects and objects of the predicate. 279 (12) a. ana apisis-iskw\u00C3\u00AAsis k\u00C3\u00A2-miyosit \u00C3\u00AA-p\u00C3\u00A2hpit SUBJECT RC ana apisis-iskw\u00C3\u00AAs -is k\u00C3\u00A2-miyosi -t \u00C3\u00AA- p\u00C3\u00A2hpi -t DEM.AN little- girl -DIM C2-pretty.VAI-3 C1-laugh.VAI-3 \u00E2\u0080\u0098The little girl who is pretty smiled.\u00E2\u0080\u0099 b. ni-nistaw\u00C3\u00AAyim\u00C3\u00A2w ana iskw\u00C3\u00AAw k\u00C3\u00A2-m\u00C3\u00A2tot OBJECT RC ninistaw\u00C3\u00AAyim -\u00C3\u00A2 -w ana iskw\u00C3\u00AAw k\u00C3\u00A2-m\u00C3\u00A2to -t 1-know.VTA -DIR-3 DEM.AN woman C2-cry.VAI-3 \u00E2\u0080\u0098I know that woman who is crying.\u00E2\u0080\u0099 While the relative clause follows the noun in both examples above, this is not by any means necessary; as with other adjoined clauses, both orderings are possible. (13) shows the preceding/following alternation for object relative clauses. (13) a. Misti w\u00C3\u00A2pam\u00C3\u00AAw k\u00C3\u00A2-w\u00C3\u00A2piskisiyit min\u00C3\u00B4s4 PRENOMINAL RC M w\u00C3\u00A2pam -\u00C3\u00AA -w k\u00C3\u00A2- w\u00C3\u00A2piskisi -yi -t min\u00C3\u00B4s M see.VTA-DIR-3 C2- white.VAI -DS-3 cat \u00E2\u0080\u0098Misti saw that cat that is white.\u00E2\u0080\u0099 b. ninistaw\u00C3\u00AAyim\u00C3\u00A2w ana n\u00C3\u00A2p\u00C3\u00AAw k\u00C3\u00A2-maskisit POSTNOMINAL RC ni-nistaw\u00C3\u00AAyim -\u00C3\u00A2 -w ana n\u00C3\u00A2p\u00C3\u00AAw k\u00C3\u00A2-maskisi -t 1-know.VTA -DIR-3 DEM.AN man C2-lame.VAI-3 \u00E2\u0080\u0098I know the man who\u00E2\u0080\u0099s lame.\u00E2\u0080\u0099 I claim that when a relative k\u00C3\u00A2- clause is used, the proposition restricting the referent is presupposed. This analysis is used to capture the contrast between relative clauses introduced by the k\u00C3\u00A2- complementizer, and those introduced by the \u00C3\u00AA- conjunct clauses, which are also used in modificational structures. For example, in (14a-b) below, which differ minimally in the choice of complementizer of the relative clause, k\u00C3\u00A2- clauses cannot be used to introduce referents into a discourse: thus speakers reject k\u00C3\u00A2- relative clauses in contexts where the referent is previously unknown to the hearer. 4 Note that the object of the transitive verb, min\u00C3\u00B4s \u00E2\u0080\u0098cat\u00E2\u0080\u0099, in (37a and 38a-b) is not marked for obviation, although the corresponding verb is marked for dependent reference. This is a common occurrence in elicitation contexts; cf. Cook & M\u00C3\u00BChlbauer (2006). 280 (14) context: want to tell someone about a cat that Misti saw a. ?# w\u00C3\u00A2pam\u00C3\u00AAw Misti k\u00C3\u00A2-w\u00C3\u00A2piskisiyit p\u00C3\u00B4s\u00C3\u00AEsa w\u00C3\u00A2pam -\u00C3\u00AA -w M k\u00C3\u00A2-w\u00C3\u00A2piskisi -yi -t p\u00C3\u00B4s\u00C3\u00AEs-a see.VTA -DIR-3 M C2-white.VAI -DS-3 cat -OBV \u00E2\u0080\u0098Misti saw the cat that was white.\u00E2\u0080\u0099 b. w\u00C3\u00A2pam\u00C3\u00AAw Misti \u00C3\u00AA-w\u00C3\u00A2piskisiyit p\u00C3\u00B4s\u00C3\u00AEsa w\u00C3\u00A2pam -\u00C3\u00AA -w M \u00C3\u00AA- w\u00C3\u00A2piskisi -yi -t p\u00C3\u00B4s\u00C3\u00AEs -a see.VTA -DIR-3 M C1-white.VAI -DEP-3 cat -OBV \u00E2\u0080\u0098Misti sees a cat that\u00E2\u0080\u0099s white.\u00E2\u0080\u0099 (15) context: Hearer comes up and asks Speaker what happened to an injured girl a. # iskw\u00C3\u00AAsis \u00C3\u00AA-w\u00C3\u00AEhtamawit k\u00C3\u00A2-kaskit\u00C3\u00AAsiyit atim \u00C3\u00AA-tahkamikot iskw\u00C3\u00AAsis \u00C3\u00AA- w\u00C3\u00AEhtamaw -it k\u00C3\u00A2- kaskit\u00C3\u00AAsi -yi -t atim \u00C3\u00AA- tahkam -iko -t girl C1-tell.VTA -3>1 C2- black.VAI-DS-3 dog C1- bite.VTA-INV -3 --- (intended: \u00E2\u0080\u0098The girl told me a black dog bit her.\u00E2\u0080\u0099) b. iskw\u00C3\u00AAsis \u00C3\u00AA-w\u00C3\u00AEhtamawit atim \u00C3\u00AA-kaskitesiyit \u00C3\u00AA-tahkamikot iskw\u00C3\u00AAsis \u00C3\u00AA- w\u00C3\u00AEhtamaw -it \u00C3\u00AA- kaskit\u00C3\u00AAsi -yi -t atim \u00C3\u00AA- tahkam -iko -t girl C1-tell.VTA -3>1 C1- black.VAI-DS-3 dog C1- bite.VTA-INV -3 \u00E2\u0080\u0098The girl told me that a dog that\u00E2\u0080\u0099s black bit her.\u00E2\u0080\u0099 This is consistent with there being a presupposition on the k\u00C3\u00A2- relative clause: in order for the information about the referent to be presupposed, the referent itself must already exist in the discourse. By contrast, if the referent and the relevant proposition already exist either in the immediate spatio-temporal context, or in the previous discourse, then the k\u00C3\u00A2- relative clause is felicitous and the \u00C3\u00AA- relative clause is not. (16) context: specifically pointing at the white cat Misti saw a. Misti w\u00C3\u00A2pam\u00C3\u00AAw k\u00C3\u00A2-w\u00C3\u00A2piskisiyit min\u00C3\u00B4s M w\u00C3\u00A2pam -\u00C3\u00AA -w k\u00C3\u00A2-w\u00C3\u00A2piskisi -yi -t min\u00C3\u00B4s M see.VTA-DIR-3 C2-white.VAI -DS-3 cat \u00E2\u0080\u0098Misti saw that cat that is white.\u00E2\u0080\u0099 b. # Misti w\u00C3\u00A2pam\u00C3\u00AAw \u00C3\u00AA-w\u00C3\u00A2piskisiyit min\u00C3\u00B4s M w\u00C3\u00A2pam -\u00C3\u00AA -w \u00C3\u00AA-w\u00C3\u00A2piskisi -yi -t min\u00C3\u00B4s M see.VTA-DIR-3 C1-white.VAI -DS-3 cat --- (intended: \u00E2\u0080\u0098Misti saw that cat that is white.\u00E2\u0080\u0099) 281 (17) context: picking the black cat out of a set of two cats that have been established in discourse a. John w\u00C3\u00A2pam\u00C3\u00AAw anhi k\u00C3\u00A2-kast\u00C3\u00AAsiyit min\u00C3\u00B4s J w\u00C3\u00A2pam -\u00C3\u00AA -w anihi k\u00C3\u00A2- kast\u00C3\u00AAsi -yi -t min\u00C3\u00B4s J see.VTA -DIR-3 DEM.OBV C2-black.VAI-DS-3 cat \u00E2\u0080\u0098John saw the black cat.\u00E2\u0080\u0099 b. # John w\u00C3\u00A2pam\u00C3\u00AAw anhi \u00C3\u00AA-kast\u00C3\u00AAsiyit min\u00C3\u00B4s J w\u00C3\u00A2pam -\u00C3\u00AA -w anihi \u00C3\u00AA- kast\u00C3\u00AAsi -yi -t min\u00C3\u00B4s J see.VTA -DIR-3 DEM.OBV C1-black.VAI-DS-3 cat ---(intended: \u00E2\u0080\u0098John saw the black cat.\u00E2\u0080\u0099) (18) context: Rosie\u00E2\u0080\u0099s been telling me about this big black dog she\u00E2\u0080\u0099s been seeing; after some time, I finally see this dog, and want to tell Rosie about it5 a. niw\u00C3\u00A2pam\u00C3\u00A2w ana atim k\u00C3\u00A2-kast\u00C3\u00AAsit ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w ana atim k\u00C3\u00A2-kast\u00C3\u00AAsi -t 1- see.VTA-DIR-3 DEM.AN dog C2-black.VAI-3 \u00E2\u0080\u0098I saw that black dog (that you\u00E2\u0080\u0099ve been talking about in the neighborhood).\u00E2\u0080\u0099 b. # niw\u00C3\u00A2pam\u00C3\u00A2w atim \u00C3\u00AA-kast\u00C3\u00AAsit ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w atim \u00C3\u00AA-kast\u00C3\u00AAsi -t 1- see.VTA-DIR-3 dog C1-black.VAI-3 \u00E2\u0080\u0098I saw that black dog.\u00E2\u0080\u0099 As the difference in interpretation and necessity for context in the above example implies, the choice of complementizer correlates a contrast in definiteness (cf. Blain 1999). More specifically, because k\u00C3\u00A2- CONJUNCT clauses are presupposed the referent must also exist in the discourse (i.e., with respect to both speaker and hearer).6 Thus, when introducing a referent that is previously unknown to the hearer, modificational k\u00C3\u00A2- clauses are infelicitous. Another example is given in (19), where the main verb nakiskaw- \u00E2\u0080\u0098meet someone\u00E2\u0080\u0099 facilitates a context for introducing a new referent; only the modificational \u00C3\u00AA- clause is felicitous. 5 As a side note, when I asked about using an \u00C3\u00AA- clause with the deictic demonstrative, this was also ruled infelicitous by the consultant but given a different translation (one that suggests the \u00C3\u00AA- clause is being interpreted as mediated complement clause). (i) # niw\u00C3\u00A2pam\u00C3\u00A2w ana atim \u00C3\u00AA-kast\u00C3\u00AAsit ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w ana atim \u00C3\u00AA- kast\u00C3\u00AAsi -t 1- see.VTA-DIR-3 DEM.AN dog C1-black.VAI-3 \u00E2\u0080\u0098I saw the dog is black.\u00E2\u0080\u0099 6 The implicational relationship between presupposition of the proposition and existence of the referent goes only one direction, i.e., it is possible for a referent to exist without the proposition with which it is associated to be presupposed. 282 (19) a. ninakiskaw\u00C3\u00A2w anohc toni \u00C3\u00AA-\u00C3\u00A2hkosit n\u00C3\u00A2p\u00C3\u00AAw ni-nakiskaw -\u00C3\u00A2 -w anohc mitoni \u00C3\u00AA- ahkosi -t n\u00C3\u00A2p\u00C3\u00AAw 1-meet.VTA -DIR-3 today very C1-sick.VAI-3 man \u00E2\u0080\u0098Today I met a very sick man.\u00E2\u0080\u0099 b. # ninakiskaw\u00C3\u00A2w anohc toni k\u00C3\u00A2-\u00C3\u00A2hkosit n\u00C3\u00A2p\u00C3\u00AAw ni- nakiskaw -\u00C3\u00A2 -w anohc mitoni k\u00C3\u00A2- ahkosi -t n\u00C3\u00A2p\u00C3\u00AAw 1- meet.VTA -DIR-3 today very C2-sick.VAI-3 man \u00E2\u0080\u0098Today I met a very sick man.\u00E2\u0080\u0099 comment: bad because we haven\u00E2\u0080\u0099t been talking about this guy, and [the listener] doesn\u00E2\u0080\u0099t know him On the other hand, if the speaker is talking to the hearer about a referent who is known to the hearer, and referring to properties known by both the speaker and hearer, then the modificational \u00C3\u00AA- clause is no longer felicitous; a k\u00C3\u00A2- CONJUNCT clause must be used. (20) a. nikiyokaw\u00C3\u00A2w kist\u00C3\u00AAs k\u00C3\u00A2-m\u00C3\u00A2skisit ni-kiyokaw -\u00C3\u00A2 -w ki-st\u00C3\u00AAs k\u00C3\u00A2- m\u00C3\u00A2skisi -t 1-visit.VTA -DIR-3 2- brother C2-lame.VAI-3 \u00E2\u0080\u0098I visited your lame brother.\u00E2\u0080\u0099 b. # nikiyokaw\u00C3\u00A2w kist\u00C3\u00AAs \u00C3\u00AA-m\u00C3\u00A2skisit ni-kiyokaw -\u00C3\u00A2 -w ki-st\u00C3\u00AAs \u00C3\u00AA- m\u00C3\u00A2skisi -t 1-visit.VTA -DIR-3 2- brother C1-lame.VAI-3 \u00E2\u0080\u0098I visited your lame brother.\u00E2\u0080\u0099 comment: funny, because [if you say it that way] you\u00E2\u0080\u0099re announcing that he\u00E2\u0080\u0099s lame, but they would already know that In this context, the noun phrase containing the \u00C3\u00AA- clause has an indefinite interpretation, but notice that we don\u00E2\u0080\u0099t have to say that the \u00C3\u00AA- clause is inherently specified. Under the principles of blocking, the use of an \u00C3\u00AA- form in a context where k\u00C3\u00A2- is felicitous induces a complementary interpretation for \u00C3\u00AA-; the indefinite meaning is derived from its context (cf. Heim 1982, Diesing 1992, Matthewson 1999 on the non-specification of indefinites). These two analyses (i.e., specification of indefinite vs. derived indefiniteness) make different predictions: the former predicts that \u00C3\u00AA- will always be indefinite (or more generally, non-presupposed); the latter predicts that \u00C3\u00AA- will be interpreted differently when it is part of a different contrast set (cf. \u00C2\u00A76.4). 283 6.3.2 Wh-questions In Plains Cree, wh-questions systematically use \u00C3\u00AA- or k\u00C3\u00A2- CONJUNCT forms, rather than INDEPENDENT forms (Blain 1997). Thus, for example, a reason wh-question has the form in (21): an initial wh-word, an \u00E2\u0080\u0098optional\u00E2\u0080\u0099 demonstrative7, and the k\u00C3\u00A2- CONJUNCT clause-type. (21) t\u00C3\u00A2n\u00C3\u00AEhk\u00C3\u00AA (\u00C3\u00B4ma) k\u00C3\u00A2-t\u00C3\u00AApayan t\u00C3\u00A2n-ihk\u00C3\u00AA \u00C3\u00B4ma k\u00C3\u00A2 -t\u00C3\u00AApa -yan Q -why DEM.INAN C2-yell.VAI -2 \u00E2\u0080\u0098Why did you yell?\u00E2\u0080\u0099 In Blain\u00E2\u0080\u0099s (1997) work on argument wh-questions, she found that all wh-questions tested alternated between the clause-typing proclitics \u00C3\u00AA- and k\u00C3\u00A2-. (22) a. aw\u00C3\u00AEna k\u00C3\u00A2-oc\u00C3\u00AAm\u00C3\u00A2t John-a aw\u00C3\u00AEna k\u00C3\u00A2-oc\u00C3\u00AAm -\u00C3\u00A2 -t J -a who C2-kiss.VTA-DIR-3 J -OBV \u00E2\u0080\u0098Who is it that kissed John / Who kissed John?\u00E2\u0080\u0099 b. aw\u00C3\u00AEna \u00C3\u00AA-oc\u00C3\u00AAm\u00C3\u00A2t John-a aw\u00C3\u00AEna \u00C3\u00AA- oc\u00C3\u00AAm -\u00C3\u00A2 -t J -a who C1-kiss.VTA-DIR-3 J -OBV \u00E2\u0080\u0098Who kissed John?\u00E2\u0080\u0099 (from Blain 1997:66) Based on the arguments made in Blain (1997) and adopting her proposal, I take the structure of wh-questions to be as in (23): the wh-word is generated in a higher nominal predicate, and there is null operator movement to spec, CP within the dependent clause. (23) [ WHi ]PRED [SUBJ pro [CP Opi [C k\u00C3\u00A2- [IP \u00E2\u0080\u00A6 ti \u00E2\u0080\u00A6 ] There are several pieces of evidence for this structure; here I review two of the arguments (see Blain 1997 for fuller discussion). First, the wh-word cannot be \u00E2\u0080\u0098in-situ\u00E2\u0080\u0099, which would be surprising if the wh-word were moved (cf. English in-situ). 7 The presence/absence of the demonstrative is determined by discourse-context; its presence seems to correlate with discourse-linking in the sense of Pesetsky (1987). When no overt demonstrative appears, I assume that there is a null pro argument (cf. Blain 1997). 284 (24) a. What did you see? b. You saw WHAT? (25) a. aw\u00C3\u00AEniwa John k\u00C3\u00A2-oc\u00C3\u00AAm\u00C3\u00A2t aw\u00C3\u00AEni-wa J k\u00C3\u00A2-oc\u00C3\u00AAm -\u00C3\u00A2 -t who -OBV J C2-kiss.VTA-DIR-3 \u00E2\u0080\u0098Who did John kiss?\u00E2\u0080\u0099 b. * John k\u00C3\u00A2-oc\u00C3\u00AAm\u00C3\u00A2t aw\u00C3\u00AEniwa (Blain 1997: 60) J k\u00C3\u00A2-oc\u00C3\u00AAm -\u00C3\u00A2 -t aw\u00C3\u00AEni-wa J C2-kiss.VTA-DIR-3 who -OBV --- This is especially surprising given that Plains Cree word order is generally much free-er than that of English. However, if the wh-word is in a separate clause, then the ordering restrictions make sense. Second, in languages where the wh-words undergo movement, multiple wh-questions are possible (with the second and third wh-words either in-situ or moved; cf. Richards 1997), as illustrated for English in (26). (26) a. Who said what? b. Who spoke to who? However, this is impossible in Plains Cree. The examples in (27) show sentences constructed analogously to the English examples immediately above. (27) a. * aw\u00C3\u00AEna \u00C3\u00AA-itw\u00C3\u00AAt k\u00C3\u00AEkw\u00C3\u00A2y aw\u00C3\u00AEna \u00C3\u00AA- itw\u00C3\u00AA -t k\u00C3\u00AEkw\u00C3\u00A2y who C1-thus.say.VAI-3 what --- (intended: \u00E2\u0080\u0098Who said what?\u00E2\u0080\u0099) b. * aw\u00C3\u00AEna k\u00C3\u00A2-p\u00C3\u00AEkiskw\u00C3\u00A2t\u00C3\u00A2t aw\u00C3\u00AEna aw\u00C3\u00AEna k\u00C3\u00A2-p\u00C3\u00AEkiskw\u00C3\u00A2t -\u00C3\u00A2 -t aw\u00C3\u00AEna who C2-speak.VTA-DIR-3 who --- (intended: \u00E2\u0080\u0098Who spoke to who?\u00E2\u0080\u0099) (from Blain 1997:90, (52a-b)) Neither can multiple wh-words be in initial position, as shown in (28). 285 (28) * aw\u00C3\u00AEna k\u00C3\u00AEkw\u00C3\u00A2y \u00C3\u00AA-itw\u00C3\u00AAt aw\u00C3\u00AEna k\u00C3\u00AEkw\u00C3\u00A2y \u00C3\u00AA- itw\u00C3\u00AA -t who what C1-say.VAI-3 --- (intended: \u00E2\u0080\u0098Who said what?\u00E2\u0080\u0099) In the cleft structure, however, the wh-word is a predicate, rather than an argument, and given that there can only be one predicate per predication (Calabrese 1984, 1987), the ban on multiple wh-questions is explained. Now, although Blain (1997) makes a claim about the syntax of Plains Cree wh-questions, she does not address their semantics. Here I argue that, although clefts in English are generally treated as inherently presuppositional, and questions are also sometimes treated as inherently presuppositional (Katz & Postal 1964, Karttunen & Peters 1976, among others) only the k\u00C3\u00A2- wh- questions are presuppositional in Plains Cree (cf. Rooth 1996, D\u00C3\u00A9chaine 2002b; Davis, Matthewson, and Shank 2004 on non-presuppositional clefts). Wh-questions using \u00C3\u00AA- are non- presupposed even though they have the same syntactic structure as their k\u00C3\u00A2- counterparts. 6.3.2.1 k\u00C3\u00A2- wh-questions as presuppositional The distributional differences between \u00C3\u00AA- clauses and k\u00C3\u00A2- clauses have not received much discussion in the literature for Plains Cree (see e.g., Blain 1997). On the account that the k\u00C3\u00A2- proclitic is marked for a presupposition, it may be surprising at first that there is any variation at all, given that many linguists consider questions to have an existential presupposition in them (cf. Katz & Postal 1964, Postal 1971, Karttunen & Peters 1976, Comorovski 1996). For example, Katz (1972) claims that \u00E2\u0080\u009Ca presupposition of a question is a necessary condition for a successful interrogative speech act.\u00E2\u0080\u009D On the other hand, Fitzpatrick 2005 argues that there is no inherent presupposition for any wh-question except how come wh-questions. He shows that how come wh-questions have systematic presuppositional asymmetries with other wh-questions, such as the inability to be used in rhetorical questions where a negative answer is expected, and the inability of how come, but not other wh-questions to license negative polarity items (see also Chang 1997 on presuppositional and non-presuppositional wh-questions in French). On this account, a 286 syntactically marked structure (in English, a cleft with a definite determiner it, cf. Rooth 1996) is necessary to code a presupposition in other wh-questions. If we take seriously that there are presuppositional and non-presuppositional questions, my claim that k\u00C3\u00A2- is presuppositional leads us to expect that presuppositional questions in Plains Cree will have the k\u00C3\u00A2- clause-typing, and by extension, that non-presuppositional questions will have the \u00C3\u00AA- clause-typing. This analysis helps to explain several otherwise puzzling asymmetries in the distribution of \u00C3\u00AA- vs. k\u00C3\u00A2- wh-questions. Property \u00C3\u00AA- k\u00C3\u00A2- out-of-the-blue context \u00E2\u009C\u0094 \u00E2\u009C\u0096 presuppositional context \u00E2\u009C\u0096 \u00E2\u009C\u0094 allows overt demonstrative \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 6.4. Asymmetries between two wh-clause types First, in wh-questions that have an alternation between an \u00C3\u00AA- proclitic and a k\u00C3\u00A2- proclitic, the k\u00C3\u00A2- version is rejected in out-of-the-blue contexts. For example, when consultants are asked to form a wh-question using aw\u00C3\u00AEna \u00E2\u0080\u0098who\u00E2\u0080\u0099, they give the \u00C3\u00AA- form (29a); when presented with the k\u00C3\u00A2- form, they react to it as \u00E2\u0080\u0098funny\u00E2\u0080\u0099. (29) context: out-of-the-blue translation of \u00E2\u0080\u0098Who\u00E2\u0080\u0099s tired?\u00E2\u0080\u0099 a. ? aw\u00C3\u00AEna k\u00C3\u00A2-n\u00C3\u00AAstosit aw\u00C3\u00AEna k\u00C3\u00A2- n\u00C3\u00AAstosi -t who C2- tired.VAI-3 --- b. aw\u00C3\u00AEna \u00C3\u00AA-n\u00C3\u00AAstosit aw\u00C3\u00AEna \u00C3\u00AA- n\u00C3\u00AAstosi -t who C1-tired.VAI-3 \u00E2\u0080\u0098Who is tired?\u00E2\u0080\u0099 287 (30) context: out-of-the-blue translation of \u00E2\u0080\u0098Who\u00E2\u0080\u0099s sleeping?\u00E2\u0080\u0099 a. *? aw\u00C3\u00AEna k\u00C3\u00A2-nip\u00C3\u00A2t aw\u00C3\u00AEna k\u00C3\u00A2- nip\u00C3\u00A2 -t who C2-sleep.VAI-3 --- b. aw\u00C3\u00AEna \u00C3\u00AA-nip\u00C3\u00A2t aw\u00C3\u00AEna \u00C3\u00AA- nip\u00C3\u00A2 -t who C1-sleep.VAI-3 \u00E2\u0080\u0098Who is sleeping?\u00E2\u0080\u0099 However, the \u00E2\u0080\u0098funniness\u00E2\u0080\u0099 can be fixed by introducing a presuppositional context. In (31), the speaker has in mind that someone is tired, and is trying to identify that person out of a group. Here, in a converse context, the clause-typing flips: the k\u00C3\u00A2- form is entirely natural. (31) context: you know someone is tired, but you don\u00E2\u0080\u0099t know who it is a. aw\u00C3\u00AEn \u00C3\u00A2wa k\u00C3\u00A2-n\u00C3\u00AAstosit aw\u00C3\u00AEna awa k\u00C3\u00A2-n\u00C3\u00AAstosi -t who DEM.AN C2-tired.VAI-3 \u00E2\u0080\u0098Who is this person that is tired?\u00E2\u0080\u0099 (Does anybody know this person that is tired?) b. # aw\u00C3\u00AEna \u00C3\u00AA-n\u00C3\u00AAstosit aw\u00C3\u00AEna \u00C3\u00AA- n\u00C3\u00AAstosi -t who C1-tired.VAI-3 \u00E2\u0080\u0098Who is tired?\u00E2\u0080\u0099@ Likewise, in (32), speaker A reports that he found speaker B\u00E2\u0080\u0099s hat, whereupon speaker B asks about the time that this hat was found. In such cases, consultants volunteer the k\u00C3\u00A2- clause. (32) A. kitastotin nik\u00C3\u00AE-misk\u00C3\u00AAn ki(t)- astotin ni- k\u00C3\u00AE- misk\u00C3\u00AA -n 2- hat 1- PREV-find.VTI-SAP \u00E2\u0080\u0098I found your hat.\u00E2\u0080\u0099 B. t\u00C3\u00A2nisp\u00C3\u00AE k\u00C3\u00A2-niskaman8 t\u00C3\u00A2n-isp\u00C3\u00AE k\u00C3\u00A2-miskam -an Q- TEMP C2-find.VTI -2 \u00E2\u0080\u0098When did you find it?\u00E2\u0080\u0099 8 The form in elicitation was k\u00C3\u00A2-niskaman with a stem-initial [n], rather than the expected [m]. 288 In one of the above examples, a demonstrative intervenes between the question word and the verbal complex (awa \u00E2\u0080\u0098this.AN\u00E2\u0080\u0099 in (31a)). A third difference between wh-questions with the k\u00C3\u00A2- clause-type and those with the \u00C3\u00AA- clause-type is that k\u00C3\u00A2- clauses readily permit the overt demonstrative, but \u00C3\u00AA- clauses do not. Blain reports that they are impossible for the consultants she worked with (33). (33) a. aw\u00C3\u00AEna ana k\u00C3\u00A2-oc\u00C3\u00AAm\u00C3\u00A2t John-a aw\u00C3\u00AEna ana k\u00C3\u00A2- oc\u00C3\u00AAm -\u00C3\u00A2 -t J-a who DEM.AN C2- kiss.VTA-DIR-3 J-OBV \u00E2\u0080\u0098Who is it that kissed John?\u00E2\u0080\u0099 b. * aw\u00C3\u00AEna ana \u00C3\u00AA-oc\u00C3\u00AAm\u00C3\u00A2t John-a aw\u00C3\u00AEna ana \u00C3\u00AA- oc\u00C3\u00AAm -\u00C3\u00A2 -t J-a who DEM.AN C1-kiss.VTA-DIR-3 J-OBV ---(intended: \u00E2\u0080\u0098Who is it that kissed John?\u00E2\u0080\u0099) (Blain 1997:68) While there are examples like (33b) in running speech, they are quite rare, and it is not clear what conditions the choice of \u00C3\u00AA- as opposed to k\u00C3\u00A2-. In the following example, the speaker used an \u00C3\u00AA- clause, hesitated, and then immediately repeated the question with a k\u00C3\u00A2- clause. (34) ... (-- t\u00C3\u00A2nis \u00C3\u00B4m \u00C3\u00AA-isiy\u00C3\u00AEhk\u00C3\u00A2t\u00C3\u00AAk anima kotak, aya nik\u00C3\u00AE-~, -- \u00C3\u00A2, t\u00C3\u00A2n -isi \u00C3\u00B4ma \u00C3\u00AA- isiy\u00C3\u00AEhk\u00C3\u00A2t\u00C3\u00AA -k anima kotak aya ni- k\u00C3\u00AE \u00C3\u00A2 Q -THUS DEM.INAN C1-be.called.VII-0 DEM.INAN other CONN 1- PREV HES \u00E2\u0080\u0098... (what is that other place called \u00E2\u0080\u0093 well, t\u00C3\u00A2nis \u00C3\u00B4ma k-\u00C3\u00AAsiy\u00C3\u00AEhk\u00C3\u00A2t\u00C3\u00AAk, niwanikiskisin --); t\u00C3\u00A2n -isi \u00C3\u00B4ma k\u00C3\u00A2- isiy\u00C3\u00AEhk\u00C3\u00A2t\u00C3\u00AA -k ni-wanikiski -n Q -THUS DEM.INAN C2- be.called.VII-0 1- forget.VAI-SAP what is it called, I have forgotten);\u00E2\u0080\u0099 (AA 3.4) I do not have an analysis that captures exactly why \u00C3\u00AA- clauses cannot be used with a demonstrative; however, I believe using an demonstrative as an overt subject of the nominal predicate also has a discourse-linking effect that works in tandem with the k\u00C3\u00A2- clause-typing, and I suspect that because of the non-presuppositional interpretation that \u00C3\u00AA- clauses have in this context (i.e., as a result of their contrast with the k\u00C3\u00A2- clauses), it is difficult to use the overt demonstrative with them. Fourth, reason wh-questions seem to only allow the k\u00C3\u00A2- clause-type, as in (35). 289 (35) a. t\u00C3\u00A2n\u00C3\u00AAhki k\u00C3\u00A2-m\u00C3\u00A2tot Tomio REASON t\u00C3\u00A2n -\u00C3\u00AAhki k\u00C3\u00A2- m\u00C3\u00A2to -t T Q -RAT C2-cry.VAI-3 T \u00E2\u0080\u0098Why was Tomio crying?\u00E2\u0080\u0099 b. * t\u00C3\u00A2n\u00C3\u00AAhki \u00C3\u00AA-m\u00C3\u00A2tot Tomio t\u00C3\u00A2n -\u00C3\u00AAhki \u00C3\u00AA- m\u00C3\u00A2to -t T Q -RAT C1-cry.VAI-3 T --- Reason wh-questions are restricted in particular ways across many languages (Huang 1982, Cheng 1991, among others). Lawler (1971) and Collins (1991) argue that why questions in English are somehow more presuppositional than other wh-questions, and Fitzpatrick (2005) argues that how come wh-questions are the only ones that have a presupposition at all. If this is correct, then t\u00C3\u00A2n\u00C3\u00AAhki wh-questions in Plains Cree appear to behave exactly like English how come wh-questions in being inherently presuppositional. Finally, for other adjunct wh-questions there is a systematic difference in translation: the \u00C3\u00AA- wh-questions are translated as simple English wh-questions, but the k\u00C3\u00A2- wh-questions are translated as English definite clefts. (36) a. t\u00C3\u00A2nisi k\u00C3\u00A2-isi-sipw\u00C3\u00AAht\u00C3\u00AAt W\u00C3\u00A2pastim MANNER t\u00C3\u00A2n -isi k\u00C3\u00A2 -isi -sipw\u00C3\u00AAht\u00C3\u00AA -t W Q -THUS C2 -THUS-leave.VAI-3 W \u00E2\u0080\u0098How was it that W\u00C3\u00A2pastim left?\u00E2\u0080\u0099 b. t\u00C3\u00A2nisi \u00C3\u00AA-isi-sipw\u00C3\u00AAht\u00C3\u00AAt W\u00C3\u00A2pastim t\u00C3\u00A2n -isi \u00C3\u00AA- isi- sipw\u00C3\u00AAht\u00C3\u00AA -t W Q -THUS C1-THUS-leave.VAI-3 W \u00E2\u0080\u0098How did W\u00C3\u00A2pastim leave?\u00E2\u0080\u0099 (37) a. t\u00C3\u00A2nisp\u00C3\u00AE k\u00C3\u00A2-sipw\u00C3\u00AAht\u00C3\u00AAt W\u00C3\u00A2pastim TEMPORAL t\u00C3\u00A2n -isp\u00C3\u00AE k\u00C3\u00A2- sipw\u00C3\u00AAht\u00C3\u00AA -t W Q -TEMP C2-leave.VAI -3 W \u00E2\u0080\u0098When was it that W\u00C3\u00A2pastim left?\u00E2\u0080\u0099 b. t\u00C3\u00A2nisp\u00C3\u00AE \u00C3\u00AA-sipw\u00C3\u00AAht\u00C3\u00AAt W\u00C3\u00A2pastim t\u00C3\u00A2n -isp\u00C3\u00AE \u00C3\u00AA- sipw\u00C3\u00AAht\u00C3\u00AA -t W Q -TEMP C1-leave.VAI -3 W \u00E2\u0080\u0098When did W\u00C3\u00A2pastim leave?\u00E2\u0080\u0099 290 (38) a. t\u00C3\u00A2nit\u00C3\u00AA k\u00C3\u00A2-p\u00C3\u00A2hpit W\u00C3\u00A2pastim LOCATIVE t\u00C3\u00A2n -it\u00C3\u00AA k\u00C3\u00A2- p\u00C3\u00A2hpi -t W Q -LOC C2-laugh.VAI-3 W \u00E2\u0080\u0098Where is W\u00C3\u00A2pastim laughing from?\u00E2\u0080\u00999 b. t\u00C3\u00A2nit\u00C3\u00AA \u00C3\u00AA-p\u00C3\u00A2hpit W\u00C3\u00A2pastim t\u00C3\u00A2n -it\u00C3\u00AA \u00C3\u00AA- p\u00C3\u00A2hpi -t W Q -LOC C1-laugh.VAI-3 W \u00E2\u0080\u0098Where is W\u00C3\u00A2pastim laughing?\u00E2\u0080\u0099 6.3.3 Temporal modification A third construction in which k\u00C3\u00A2- clauses are used is temporal modification, which gives a temporal relation between one event and another. In English, temporal modification is usually coded by the subordinators while (for overlap relations) and when (more general); both introduce a presupposed proposition (Declerck 1991). The temporal relations expressed by English when and while are expressed in Plains Cree by means of the clause-typing k\u00C3\u00A2-. A k\u00C3\u00A2- clause, without any additional subordination elements, is interpreted as giving an unspecified temporal relation between the two clauses10. (39) a. iskw\u00C3\u00AAw \u00C3\u00AAsa k\u00C3\u00A2-pimoht\u00C3\u00AAt, piy\u00C3\u00AAs\u00C3\u00AEs w\u00C3\u00A2pam\u00C3\u00AAw k\u00C3\u00A2-clause, IND iskw\u00C3\u00AAw \u00C3\u00AAsa k\u00C3\u00A2- pimoht\u00C3\u00AA -t piy\u00C3\u00AAs\u00C3\u00AEs w\u00C3\u00A2pam -\u00C3\u00AA -w woman EVID C2- walk.VAI -3 bird see.VTA -DIR-3 \u00E2\u0080\u0098As the woman was walking, she saw a bird.\u00E2\u0080\u0099 b. w\u00C3\u00A2pahtam cik\u00C3\u00A2st\u00C3\u00AApayin \u00E2\u0080\u0098Survivor\u00E2\u0080\u0099 k\u00C3\u00A2-n\u00C3\u00AAstosit. IND, k\u00C3\u00A2-clause w\u00C3\u00A2pahtam -w cik\u00C3\u00A2st\u00C3\u00AApayin S k\u00C3\u00A2- n\u00C3\u00AAstosi -t see.VTI -3 show S C2-tired.VAI-3 \u00E2\u0080\u0098He watches the show \u00E2\u0080\u0098Survivor\u00E2\u0080\u0099 when he\u00E2\u0080\u0099s tired.\u00E2\u0080\u0099 9 I do not know exactly where the locative preposition in the English translation is coming from, but I suspect the consultant is adding some sort of overt locative element to correspond to the question word; i.e., something like \u00E2\u0080\u0098W\u00C3\u00A2pastim cried somewhere; where is that place?\u00E2\u0080\u0099 which would be an alternative way (as opposed to English clefting) to mark the presupposition in English. 10 Usually the aspectual value of the predicate yields a \u00E2\u0080\u0098default\u00E2\u0080\u0099 relation between the two clauses (cf. Hinrichs 1986, Kamp & Rohrer 1983, Declerck 1991), but these can be reversed by setting up a context. The relation can also be specified in a variety of ways: for example, the preverb m\u00C3\u00AAkw\u00C3\u00A2- provides an explicit overlap relation; the preverb k\u00C3\u00AE- provides a sequencing relation. 291 (40) a. k\u00C3\u00A2-p\u00C3\u00AA-p\u00C3\u00AEhtikoy\u00C3\u00A2n, nikosis \u00C3\u00AA-m\u00C3\u00AEcit cookies k\u00C3\u00A2-clause, CONJUNCT k\u00C3\u00A2-p\u00C3\u00AA- p\u00C3\u00AEhtiko-y\u00C3\u00A2n ni-kosis \u00C3\u00AA- m\u00C3\u00AEci -t cookies C2-COME-inside -1 1- son C1-eat.VTI-3 cookies \u00E2\u0080\u0098When I came in, he was eating the cookies.\u00E2\u0080\u0099 b. n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA-pimp\u00C3\u00A2ht\u00C3\u00A2t k\u00C3\u00A2-itoht\u00C3\u00AAt ataw\u00C3\u00AAwikamikohk CONJUNCT, k\u00C3\u00A2-clause n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA- pimp\u00C3\u00A2ht\u00C3\u00A2-t k\u00C3\u00A2-itoht\u00C3\u00AA -t ataw\u00C3\u00AAw-kamikw -hk man C1-run.VAI -3 C2-go.VAI-3 buy- building-LOC \u00E2\u0080\u0098The man ran when he went to the store.\u00E2\u0080\u0099 Notice that \u00C3\u00AA- clauses are rejected when translating when or while clauses into Plains Cree. (41) context: English-to-Cree translation task: \u00E2\u0080\u0098I was sleeping when he arrived.\u00E2\u0080\u0099 a. ninip\u00C3\u00A2n k\u00C3\u00A2-takosinit ni- nip\u00C3\u00A2 -n k\u00C3\u00A2- takosin -t 1- sleep.VAI-SAP C2- arrive.VAI-3 \u00E2\u0080\u0098I was sleeping when s/he arrived.\u00E2\u0080\u0099 b. * ninip\u00C3\u00A2n \u00C3\u00AA-takosinit ni- nip\u00C3\u00A2 -n \u00C3\u00AA- takosin -t 1- sleep.VAI-SAP C1-arrive.VAI-3 --- This doesn\u00E2\u0080\u0099t have to do with temporal sequencing; \u00C3\u00AA- clauses can provide temporal sequencing just fine. Rather, it has to do with the presupposition: when translating a presuppositional clause, \u00C3\u00AA- is an infelicitous choice. The difference between the two is illustrated in the minimal pair of (42) and (43). If the k\u00C3\u00A2- clause is used, only denial of the \u00C3\u00AA- clause is possible. (42) A. n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA-pimp\u00C3\u00A2ht\u00C3\u00A2t k\u00C3\u00A2-itoht\u00C3\u00AAt ataw\u00C3\u00AAwikamikohk n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA- pimp\u00C3\u00A2ht\u00C3\u00A2 -t k\u00C3\u00A2- itoht\u00C3\u00AA -t ataw\u00C3\u00AAwi- kamikw -hk man C1-run.VAI -3 C2-go.VAI-3 buy- building -LOC \u00E2\u0080\u0098The man ran when he went to the store.\u00E2\u0080\u0099 B. m\u00C3\u00B4ya NEG \u00E2\u0080\u0098No.\u00E2\u0080\u0099 = I disagree that he ran \u00E2\u0089\u00A0 I disagree that he went to the store If the k\u00C3\u00A2- clause is replaced by an \u00C3\u00AA- clause, m\u00C3\u00B4ya \u00E2\u0080\u0098no\u00E2\u0080\u0099 can negate either clause. 292 (43) A. n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA-pimp\u00C3\u00A2ht\u00C3\u00A2t \u00C3\u00AA-it\u00C3\u00B4ht\u00C3\u00AAt ataw\u00C3\u00AAwikamikohk n\u00C3\u00A2p\u00C3\u00AAw \u00C3\u00AA-pimp\u00C3\u00A2ht\u00C3\u00A2-t \u00C3\u00AA- it\u00C3\u00B4ht\u00C3\u00AA -t \u00C3\u00A2taw\u00C3\u00AAw-i -kamikw-hk man C1-run.VAI -3 C1-go.VAI-3 buy.VAI-PV-building-LOC \u00E2\u0080\u0098The man is running to the store.\u00E2\u0080\u0099 B. m\u00C3\u00B4ya NEG \u00E2\u0080\u0098No.\u00E2\u0080\u0099 = I disagree that he ran = I disagree that he went to the store comment: could be no to both; could be either one. 6.3.4 Concessive clauses Concessive clauses \u00E2\u0080\u009Cindicate that the situation in the matrix clause is contrary to expectation in the light of what is said in the concessive clause\u00E2\u0080\u009D (Quirk et al. 1985:1098). Relevant for our purposes is the fact that concessive clause is also presupposed: in a sentence like (44) the clause introduced by although concedes the allergy. (44) Although I\u00E2\u0080\u0099m allergic to them, I love peanuts. Concessive clauses in Plains Cree are often introduced by the particle sequence kiy\u00C3\u00A2m (\u00C3\u00A2ta). Like other particles we have seen in Plains Cree (see for example the discussion of \u00C3\u00AAsa \u00E2\u0080\u0098evidential\u00E2\u0080\u0099 in chapter 3), the meaning of these particles is not fixed; rather, it is determined by a combination of its position and the clause-type that it introduces. For example, when kiy\u00C3\u00A2m combines with a simple CONJUNCT or INDEPENDENT clause, the (presuppositional) concessive meaning is gone. With a simple CONJUNCT clause, the result is a type of weak imperative (45a), and with an INDEPENDENT clause it has a modifier role (45b)11. 11 Out of four texts, this was the only example of kiy\u00C3\u00A2m occurring with the INDEPENDENT order that I could find, and my consultants were unable to give me an interpretation that demonstrated the role of kiy\u00C3\u00A2m in such a clause. At any rate, it is not acting as a concessive. 293 (45) a. ..., \u00E2\u0080\u009Ckiy\u00C3\u00A2m \u00C3\u00AAkota ka-nipah\u00C3\u00A2hkatosocik,\u00E2\u0080\u009D \u00C3\u00AA-itihcik. SIMPLE CONJ. kiy\u00C3\u00A2m \u00C3\u00AAkota ka-nipah\u00C3\u00A2hkatoso -t -k \u00C3\u00AA- it -ih -t -k HORT there IRR-starve.VAI -3 -PL C1-say.VTA-USC-3 -PL \u00E2\u0080\u0098..., \u00E2\u0080\u009CLet them starve to death there,\u00E2\u0080\u009D it was said of them.\u00E2\u0080\u0099 (AA 9.7) b. ..., kiy\u00C3\u00A2m kik\u00C3\u00A2h-nan\u00C3\u00A2skomitin\u00C3\u00A2w\u00C3\u00A2w, INDEPENDENT kiy\u00C3\u00A2m ki- k\u00C3\u00A2h- nan\u00C3\u00A2skom -iti -n\u00C3\u00A2w\u00C3\u00A2w MOD 2- would-thank.VTA -1>2 -PL \u00E2\u0080\u0098... and I would thank you ...\u00E2\u0080\u0099 (SW 1.2) The data in (45) above shows that the particle kiy\u00C3\u00A2m does not inherently have presuppositional force. When kiy\u00C3\u00A2m (\u00C3\u00A2ta) is combined with a k\u00C3\u00A2- clause, however, the clause has a concessive force and a representative gloss would be \u00E2\u0080\u0098although\u00E2\u0080\u0099. In (46), the kiy\u00C3\u00A2m clause describes a situation that the speaker presents as given information: she is not stating that the father sometimes spoke to the son in anger, but rather taking it for granted, and making a statement about what the son did in those circumstances. (46) ..., kiy\u00C3\u00A2m \u00C3\u00A2ta k\u00C3\u00A2-kis\u00C3\u00AEkitotikot, k\u00C3\u00AE-man\u00C3\u00A2cih\u00C3\u00AAw \u00C3\u00B4ht\u00C3\u00A2wiya, ... kiy\u00C3\u00A2m \u00C3\u00A2ta k\u00C3\u00A2- kis\u00C3\u00AEkitot -iko -t k\u00C3\u00AE- man\u00C3\u00A2cih -\u00C3\u00AA -w o- oht\u00C3\u00A2wiy -a although even C2-speak.angrily.VTA-INV-3 PREV-respect.VTA-DIR-3 3- father -OBV \u00E2\u0080\u0098..., even when [his father] spoke to him in anger, he used to respect his father ...\u00E2\u0080\u0099 (EM 30) The distribution of kiy\u00C3\u00A2m (\u00C3\u00A2ta) \u00E2\u0080\u0098although\u00E2\u0080\u0099 clauses in corpus sources is given in table 6.5. For all speakers where this form is attested (EM, AA, and SW), the clause type introduced by kiy\u00C3\u00A2m (\u00C3\u00A2ta) can be k\u00C3\u00A2-; for two of the speakers, k\u00C3\u00A2- is the only choice. This is expected by an analysis of k\u00C3\u00A2- where it introduces a presupposition: there is a one-to-one mapping between the presuppositional interpretation of the clause and the clause-type. kiy\u00C3\u00A2m (\u00C3\u00A2ta) \u00E2\u0080\u0098although\u00E2\u0080\u0099 IND \u00C3\u00AA- k\u00C3\u00A2- \u00EF\u0081\u00B8 IC EM \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 AA \u00E2\u009C\u0096 \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 SW \u00E2\u009C\u0096 \u00E2\u009C\u0094 \u00E2\u009C\u0094 \u00E2\u009C\u0096 \u00E2\u009C\u0096 JKN -- -- -- -- -- Table 6.5. Clause-typing and kiy\u00C3\u00A2m Notice that one text illustrates kiy\u00C3\u00A2m introducing concessive clauses with the clause-typing proclitic \u00C3\u00AA-. However, there is a difference in interpretation: the \u00C3\u00AA- clause is irrealis (even if), and 294 the k\u00C3\u00A2- clause is realis. In an irrealis clause, the event has not been realized, and so its truth cannot be evaluated (cf. \u00C2\u00A76.4); thus the proposition (47a) is not classified as presupposed. (47) a. ..., kiy\u00C3\u00A2m \u00C3\u00AA-mam\u00C3\u00A2y\u00C3\u00AEy\u00C3\u00AAk \u00C3\u00AA-p\u00C3\u00AEkiskw\u00C3\u00AAy\u00C3\u00AAk, ... kiy\u00C3\u00A2m \u00C3\u00AA- mam\u00C3\u00A2y\u00C3\u00AE -y\u00C3\u00AAk \u00C3\u00AA- p\u00C3\u00AEkiskw\u00C3\u00AA -y\u00C3\u00AAk even C1-make.mistake.VAI-2PL C1-speak.VAI-2PL \u00E2\u0080\u0098..., even if you make mistakes when you speak, ...\u00E2\u0080\u0099 (SW 1.2) b. \u00C3\u00AAkwa m\u00C3\u00AEna kiy\u00C3\u00A2m awiyak k\u00C3\u00A2-kakw\u00C3\u00AA-kak\u00C3\u00AAskimiht, ... \u00C3\u00AAkwa m\u00C3\u00AEna kiy\u00C3\u00A2m awiyak k\u00C3\u00A2- kakw\u00C3\u00AA- kak\u00C3\u00AAskim -ih -t and also even someone C2- TRY- counsel.VTA-USC-3 \u00E2\u0080\u0098And even when one tries to counsel them, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (\u00E2\u0080\u0098Jean\u00E2\u0080\u0099 in SW 31) Finally, notice that although kiy\u00C3\u00A2m often occurs with \u00C3\u00A2ta, and although \u00C3\u00A2ta is sometimes defined as a concessive (W&A 2000:200), we cannot say that \u00C3\u00A2ta is responsible for the concessive interpretation. First, we see examples where \u00C3\u00A2ta is not present (e.g., 47b above) and the concessive meaning is still there. Second, when we compare the clauses it occurs in, we see that the concessive meaning shows up in k\u00C3\u00A2-clauses (47b), but not in other clause types, such as the INDEPENDENT clause in (48b) where it is part of an emphatic phrase. (48) a. ..., ak\u00C3\u00A2mask\u00C3\u00AEhk \u00C3\u00A2ta k\u00C3\u00AEkway k-\u00C3\u00B4hpikik aya, K\u00C3\u0082-CONJUNCT ak\u00C3\u00A2mask\u00C3\u00AE -hk \u00C3\u00A2ta k\u00C3\u00AEkway k\u00C3\u00A2- ohpiki -k aya overseas -LOC although something C2-grow.VII-0 CONN ..., even when something grows in these places overseas, manic\u00C3\u00B4sak \u00C3\u00AAsa m\u00C3\u00A2n \u00C3\u00AA-kit\u00C3\u00A2cik kistik\u00C3\u00A2na, ... manic\u00C3\u00B4s -ak \u00C3\u00AAsa m\u00C3\u00A2na \u00C3\u00AA- kit\u00C3\u00A2 -t -k kistik\u00C3\u00A2n -a insect -PL EVID usually C1-eat.VAI-3 -PL crop -PL insects eat the entire crops, ... (EM 63) b. \u00C3\u00A2ta wiya n\u00C3\u00AEsta piyisk nik\u00C3\u00AE-t\u00C3\u00B4t\u00C3\u00AAn \u00C3\u00AAwakw \u00C3\u00A2nima, ... INDEPENDENT \u00C3\u00A2ta wiya n\u00C3\u00AEsta piyisk ni- k\u00C3\u00AE- t\u00C3\u00B4t\u00C3\u00AA -n \u00C3\u00AAwakw anima indeed EMPH 1.EMPH finally 1- PREV-do.VTI-SAP TOP DEM.INAN \u00E2\u0080\u0098I, too, finally used to do that, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 50) 295 6.3.5 Correlatives The identifying feature of a correlative structure is that it is a relative clause that is dissociated from the nominal it modifies in the main clause. These occur in many languages, though they are perhaps most widely discussed for Hindi (Srivastav 1991). For example, in (49) the relative clause jo khaRii hai \u00E2\u0080\u0098who is standing\u00E2\u0080\u0099 which occurs with its head laRkii \u00E2\u0080\u0098girl\u00E2\u0080\u0099 in (49a), can be separated from it as in (49b). The two clauses are linked morphologically, by a demonstrative (here vo) on the main clause and a relative clause marker (here jo) on the relative clause. (49) a. [jo laRkii khaRii hai] [ vo lambii hai] Hindi REL girl standing is DEM tall is \u00E2\u0080\u0098The girl who is standing is tall.\u00E2\u0080\u0099 b. [vo laRkii lambii hai] [ jo khaRii hai] DEM girl tall is REL standing is \u00E2\u0080\u0098The girl who is standing is tall.\u00E2\u0080\u0099 (Srivastav 1991:639-40) Srivastav argues that in a structure like (49b), the relative clause is adjoined to the main clause. She argues that this an IP for Hindi, but since nothing crucial hinges on that (p. 674) and the data she gives justifying it in Hindi does not extend to Plains Cree, we can say that it is adjoined to CP (i.e., like any other clausal adjunct). (50) CP1 3 CP2i CP1 6 6 correlative clause main clause In Plains Cree we get parallel structures to the Hindi examples. (51a-b) provides an example: the predicate that the relative clause modifies has a manner component, and both clauses are marked with isi- \u00E2\u0080\u0098thus\u00E2\u0080\u0099. The initial clause has a head (e.g., p\u00C3\u00AAyakwan \u00E2\u0080\u0098same\u00E2\u0080\u0099 and/or isi- \u00E2\u0080\u0098thus\u00E2\u0080\u0099), and is clause-typed with the proclitic k\u00C3\u00A2-; the second clause has a topic marker \u00C3\u00AAkosi \u00E2\u0080\u0098that way\u00E2\u0080\u0099 and is clause-typed with the proclitic \u00C3\u00AA-. 296 (51) a. p\u00C3\u00AAyakwan k\u00C3\u00A2-k\u00C3\u00AE-isi-w\u00C3\u00AEhtamawit, \u00C3\u00AAkos \u00C3\u00AA-isi-w\u00C3\u00AEhtamawit awa kis\u00C3\u00AAyiniw, ... p\u00C3\u00AAyakwan k\u00C3\u00A2- k\u00C3\u00AE- isi- w\u00C3\u00AEhtamaw-it same C2-PREV-THUS-tell.VTA -3>1 \u00C3\u00AAkosi \u00C3\u00AA- isi- w\u00C3\u00AEhtamaw-it awa kis\u00C3\u00AAyiniw TOPIC C1-THUS-tell.VTA -3>1 DEM.AN old.man \u00E2\u0080\u0098this old man ... told me the same thing as my father had told me:\u00E2\u0080\u0099 (JKN 1.5) (lit.: \u00E2\u0080\u0098the same as he thus told me, so did this old man tell me\u00E2\u0080\u0099) b. kis\u00C3\u00AA-manitow k-\u00C3\u00AAsi-kit\u00C3\u00A2pamikoyahk, \u00C3\u00AAkos \u00C3\u00A2nim \u00C3\u00AA-isi-kit\u00C3\u00A2pam\u00C3\u00A2t; kis\u00C3\u00AA-manitow k\u00C3\u00A2- isi- kit\u00C3\u00A2pam -iko -yahk God C2-THUS-look.VTA-INV -1/2.PL \u00C3\u00AAkosi anima \u00C3\u00AA- isi- kit\u00C3\u00A2pam -\u00C3\u00A2 -t TOPIC DEM.INAN C1-THUS-look.VTA-DIR-3 \u00E2\u0080\u0098the way God looks upon us, that is the way she looks upon them;\u00E2\u0080\u0099 (EM 38) Adopting Srivastav\u00E2\u0080\u0099s analysis, we represent these clauses as in (52). (52) CP1 3 CP2i CP1 6 6 p\u00C3\u00AAyakwan \u00C3\u00AAkos \u00C3\u00AA-isi-w\u00C3\u00AEhtamawit k\u00C3\u00A2-k\u00C3\u00AE-isi-w\u00C3\u00AEhtamawit awa kis\u00C3\u00AAyiniw Of particular interest here is the clause typing of each clause. While both clauses introduce an open variable that must be bound (hence, both are anaphoric CONJUNCT clauses), in particular the relative clause is represented by the k\u00C3\u00A2- clause \u00E2\u0080\u0093 this is the clause which is presupposed. This property fits both with the independent facts we find about k\u00C3\u00A2- clauses as relative clauses, and with the more general semantic properties of k\u00C3\u00A2- as introducing presuppositional content. 6.4 \u00EF\u0081\u00B8 as a-veridical I now turn to the semantics of the simple CONJUNCT, which I am claiming has a null complementizer (\u00EF\u0081\u00B8). Whereas k\u00C3\u00A2- codes a preupposition of the truth of the proposition, I claim that the simple CONJUNCT codes that the truth of the proposition cannot be established at all. 297 Consider the following four sentences in English. In (53a), the proposition \u00E2\u0080\u0098I like apples\u00E2\u0080\u0099 is entailed to be true relative to me (the speaker). In (53b) and (53c), there is no entailment that the proposition \u00E2\u0080\u0098you like apples\u00E2\u0080\u0099 is true. And in (53d), there is an entailment that the proposition \u00E2\u0080\u0098I like apples\u00E2\u0080\u0099 is false. (53) a. I like apples b. Do you like apples? c. Eat an apple! d. I don\u00E2\u0080\u0099t like apples. There are two ways we could subclassify these sentences. On the basis of negative polarity items in a number of languages, Giannakidou (1998, 2008) proposes that the relevant feature is whether there is an entailment of truth or not. This means that the basic division is between (53a), which she terms veridical and (53b-d), which she terms nonveridical; the nonveridical class is an \u00E2\u0080\u0098elsewhere\u00E2\u0080\u0099 class. The relevant definition is given in (54). (54) Veridicaldef: a propositional operator F is veridical if and only if F(p) entails or presupposes that p is true in some individual\u00E2\u0080\u0099s epistemic model ME(x); otherwise F is nonveridical (Giannakidou 2008:13) The utterance in (53d), which entails that the proposition is false, is a subclass of nonveridical cases, and is defined as in (55). (55) Anti-veridicaldef: a nonveridical operator F is antiveridical if and only if F(p) entails that not p in some individual\u00E2\u0080\u0099s epistemic model: Fp \u00EF\u0083\u00A0 \u00C2\u00ACp in some ME(x) (Giannakidou 2008:13) Since anti-veridicality is a subclass of nonveridicality, this analysis predicts that nonveridical elements will be triggered in antiveridical contexts (for example, in the context of negation). However, in Plains Cree\u00E2\u0080\u0099s clause-typing system, there is evidence that the distinction is between propositions which have an entailment (whether that is veridical or antiveridical) and those which do not. Simple CONJUNCT clauses are found in contexts where there is no entailment of truth, as in the dependent clause in (56), where there is no entailment that Jeff either did or did not smile. 298 (56) niw\u00C3\u00AEhtamaw\u00C3\u00A2w Jeff ka-p\u00C3\u00A2hpit ni- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w J ka- p\u00C3\u00A2hpi -t 1- tell.VTA -DIR-3 J IRR-laugh.VAI-3 \u00E2\u0080\u0098I told Jeff to smile.\u00E2\u0080\u0099 Crucially, the simple CONJUNCT cannot be used in antiveridical contexts (i.e., when there is an entailment of falsehood). For example, in (57) we have the predicate w\u00C3\u00AEhtamaw- \u00E2\u0080\u0098tell.x.to.y.VTA\u00E2\u0080\u0099 introducing a negated embedded clause. The simple CONJUNCT cannot be used here (57b). (57) a. niw\u00C3\u00AEhtam\u00C3\u00A2k Laura \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA-m\u00C3\u00A2toyit Martha ni-w\u00C3\u00AEhtam-aw-ikw L \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- m\u00C3\u00A2to -yi -t M 1-tell.VTA-BEN-INV L NEG C1-cry.VAI-DS-3 M \u00E2\u0080\u0098Laura told me that Martha isn\u00E2\u0080\u0099t crying.\u00E2\u0080\u0099 b. ! niw\u00C3\u00AEhtam\u00C3\u00A2k Laura \u00C3\u00AAk\u00C3\u00A2 ka-m\u00C3\u00A2toyit Martha ni-w\u00C3\u00AEhtam-aw-ikw L \u00C3\u00AAk\u00C3\u00A2 ka- m\u00C3\u00A2to -yi -t M 1-tell.VTA-BEN-INV L NEG IRR-cry.VAI-DS-3 M --- (intended: \u00E2\u0080\u0098Laura told me that Martha isn\u00E2\u0080\u0099t crying.\u00E2\u0080\u0099) Giannakidou (1998:117) also comments that there are polarity items that have this distribution, and suggests that they are licensed by nonveridicality but anti-licensed by anti-veridicality. As an alternative to Giannakidou\u00E2\u0080\u0099s (1998, 2008) classification, I suggest on the basis of Plains Cree\u00E2\u0080\u0099s data that veridicality could be organized in some languages around whether there is an entailment or not. I propose the term a-veridicality to talk about those things which do not have an entailment of any kind. (58) A-veridicaldef: a proposition is a-veridical if and only if there is no possible entailment of p or \u00C2\u00ACp in any individual\u00E2\u0080\u0099s epistemic model ME(x) In terms of the English sentences above, the proposed organization would class (53a) and (53d) together in one subclass, and (53b-c) in a second subclass. In the following subsections I show that this definition accounts for the distribution of Plains Cree simple CONJUNCT clauses in a variety of contexts. These include mediated argument clauses, corresponding in much detail to the subjunctive of Romance (specifically, Romanian) (\u00C2\u00A76.4.1), purpose clauses (\u00C2\u00A76.4.2), and the antecedents of conditionals (\u00C2\u00A76.4.3). 299 6.4.1 Mediated argument clauses split along a-veridicality As we saw in chapter five, mediated argument clauses (associated with an argument position) sometimes were realized by a clause with the complementizer \u00C3\u00AA-, and sometimes with just the irrealis marker ka-. Returning to the distributional criteria that distinguish these two clauses, we see that they parallel very closely the distributional difference between indicative and subjunctive clauses in Indo-European (and specifically, in Romanian, based on Farkas 1985, 1992). Like the Romanian subjunctive, Plains Cree simple CONJUNCT clauses occur (i) under weak intensional predicates; (ii) with the weak intensional meaning of an intensionally ambiguous predicate; (iii) under certain negated strong intensional predicates. Also like the Romanian subjunctive (but unlike other Romance languages), simple CONJUNCT clauses cannot occur (iv) under factive-emotive predicates; or (v) to indicate lack of speaker knowledge. The veridicality analysis of the Romanian subjunctive, here extended to Plains Cree, thus captures both the idea that the subjunctive mood is marked for a semantic property, and the identity of that property (i.e., a-veridicality). 6.4.1.1 Sensitivity to weak intensional predicates In Romance, verbal predicates can be systematically categorized according to whether they introduce an indicative or subjunctive clause (see, e.g., Farkas 1985, 1992). In chapter 5, we saw that verbal predicates in Plains Cree can also be classified according to whether the embedded clause has the complementizer \u00C3\u00AA- or is simple CONJUNCT. If we compare the two classification systems, we see that the Romanian subjunctive is triggered by the same class of predicates that triggers the ka- prefixed form of the simple CONJUNCT clause12. 12 I have excluded some Romanian contexts, such as predicates of uncertainty, that do not apply to Plains Cree (i.e., Plains Cree does not have a comparable predicate of uncertainty). 300 Class of superordinate predicates Romanian Subjunctive Plains Cree ka- simple CONJUNCT declaratives (say, tell, announce) \u00E2\u009C\u0096 \u00E2\u009C\u0096 epistemic (know, believe, be sure) \u00E2\u009C\u0096 \u00E2\u009C\u0096 perceptual (see, hear, smell, taste) \u00E2\u009C\u0096 \u00E2\u009C\u0096 fiction (dream, imagine) \u00E2\u009C\u0096 \u00E2\u009C\u0096 factive-emotives (regret, be sad) \u00E2\u009C\u0096 \u00E2\u009C\u0096 desideratives (try, want, wish, desire) \u00E2\u009C\u0094 \u00E2\u009C\u0094 directives (order, ask, request) \u00E2\u009C\u0094 \u00E2\u009C\u0094 fear (be afraid of) \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 6.6. Comparison of Plains Cree simple CONJUNCT and Romanian subjunctive The individual predicates falling into these classes includes the following members (this is not an exhaustive list, but includes all predicates I had data for). Class of predicates Cree class members Clause-type declaratives itw\u00C3\u00AA- \u00E2\u0080\u0098say.VAI\u00E2\u0080\u0099; it- \u00E2\u0080\u0098say.to.VTA\u00E2\u0080\u0099 w\u00C3\u00AEhtamaw- \u00E2\u0080\u0098tell.VTA\u00E2\u0080\u0099 \u00C3\u00AA- epistemic t\u00C3\u00A2pw\u00C3\u00AAwak\u00C3\u00AAyiht- \u00E2\u0080\u0098believe.VTI\u00E2\u0080\u0099 kisk\u00C3\u00AAyiht- \u00E2\u0080\u0098know.VTI\u00E2\u0080\u0099 it\u00C3\u00AAyiht- \u00E2\u0080\u0098think.VTI\u00E2\u0080\u0099 kiskisi- \u00E2\u0080\u0098remember.VAI\u00E2\u0080\u0099 w\u00C3\u00A2nkiskisi- \u00E2\u0080\u0098forget.VAI\u00E2\u0080\u0099 \u00C3\u00AA- perceptual w\u00C3\u00A2paht- \u00E2\u0080\u0098see.VTI\u00E2\u0080\u0099 p\u00C3\u00AAht- \u00E2\u0080\u0098hear.VTI\u00E2\u0080\u0099 \u00C3\u00AA- fiction pakwat- \u00E2\u0080\u0098dream.VTI\u00E2\u0080\u0099 \u00C3\u00AA- factive- emotives miyw\u00C3\u00AAyiht- \u00E2\u0080\u0098be.glad.VTI\u00E2\u0080\u0099 kost- \u00E2\u0080\u0098be.afraid.VTI\u00E2\u0080\u0099 takahk\u00C3\u00AAyiht- \u00E2\u0080\u0098approve.VTI\u00E2\u0080\u0099 miyw\u00C3\u00A2sin \u00E2\u0080\u0098be.good.VII\u00E2\u0080\u0099 kis\u00C3\u00AE- \u00E2\u0080\u0098angry\u00E2\u0080\u0099 \u00C3\u00AA- desideratives nitaw\u00C3\u00AAyiht- \u00E2\u0080\u0098want.VTI\u00E2\u0080\u0099 nitaw\u00C3\u00AAyim- \u00E2\u0080\u0098want.VTA\u00E2\u0080\u0099 koci- \u00E2\u0080\u0098try.VAI\u00E2\u0080\u0099 SIMPLE w/ ka- directives kaw\u00C3\u00AAcim- \u00E2\u0080\u0098ask.VTA\u00E2\u0080\u0099 nitotam\u00C3\u00A2- \u00E2\u0080\u0098request.VAI\u00E2\u0080\u0099 kak\u00C3\u00AAskim- \u00E2\u0080\u0098counsel.VTA\u00E2\u0080\u0099 SIMPLE w/ ka- fear kost- \u00E2\u0080\u0098fear.VTI\u00E2\u0080\u0099 s\u00C3\u00AAkisi- \u00E2\u0080\u0098fear.VAI\u00E2\u0080\u0099?? SIMPLE w/ ka- Table 6.7. Classification of predicates by clause-type of subordinate clause 301 6.4.1.2 Sensitivity to weak intensional meanings A second property of the indicative/subjunctive split in Romance is that there are a number of verbs which can take either kind of embedded clause. In these cases, the form of the clause correlates with the weak vs. strong intensional meaning of the higher verb. For example, Farkas (1992:70) reports \u00E2\u0080\u009Cin [59a] the verb is a declarative: it reports an assertion made by Ion; in (59b) the verb is a directive: it reports a directive of Ion\u00E2\u0080\u0099s.\u00E2\u0080\u009D (59) a. Ion a spus [c\u00C4\u0083 Maria a plecat]. I. has said that M. has left. b. Ion a spus [c\u00C4\u0083 Maria s\u00C4\u0083 plece imediat]. I. has said that M. SUBJ leave immediately. (Farkas 1992:70) Plains Cree likewise has predicates which can introduce either kind of embedded clause, including verbs of speaking and generic verbs. Table 6.8 presents the predicates I have found with such an alternation. Verb \u00C3\u00AA- clause ka- CONJUNCT clause w\u00C3\u00AEhtamaw- (tell.VTA) \u00E2\u0080\u0098tell Y that X\u00E2\u0080\u0099 directive \u00E2\u0080\u0098tell Y to X\u00E2\u0080\u0099 it- (say.to.VTA) \u00E2\u0080\u0098say to Y that X\u00E2\u0080\u0099 directive \u00E2\u0080\u0098say to Y to X\u00E2\u0080\u0099 miyw\u00C3\u00AAyiht- (be.happy.VTI) factive-emotive \u00E2\u0080\u0098happy that X\u00E2\u0080\u0099 generic-emotive \u00E2\u0080\u0098happy Xing\u00E2\u0080\u0099 w\u00C3\u00A2nkiskisi- (forget.VAI) factive a-veridical kiskisi- (remember.VAI) factive a-veridical kost- (afraid.VTI); s\u00C3\u00AAkisi- (afraid.VAI) factive-emotive fear of unknown Table 6.8. Unselective predicates The behaviour of speech predicates is exemplified in (60): when the embedded clause is marked by \u00C3\u00AA-, the predicate wihtamaw- \u00E2\u0080\u0098tell\u00E2\u0080\u0099 presents information (predicting what Jeff will do); when the embedded clause has no overt marking, wihtamaw- \u00E2\u0080\u0098tell\u00E2\u0080\u0099 directs someone to do something (telling Jeff what he must do). 302 (60) a. niw\u00C3\u00AEhtamaw\u00C3\u00A2w Jeff \u00C3\u00AA-w\u00C3\u00AE-p\u00C3\u00A2hpit ni-w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w J \u00C3\u00AA- w\u00C3\u00AE- p\u00C3\u00A2hpi -t 1- tell.VTA -DIR-3 J C1-INT-laugh.VAI-3 \u00E2\u0080\u0098I\u00E2\u0080\u0099m telling Jeff he\u00E2\u0080\u0099s going to laugh.\u00E2\u0080\u0099 comment: you\u00E2\u0080\u0099re going to tell him a joke that will make him laugh b. niw\u00C3\u00AEhtamaw\u00C3\u00A2w Jeff ka-p\u00C3\u00A2hpit ni- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w J ka- p\u00C3\u00A2hpi -t 1- tell.VTA -DIR-3 J IRR-laugh.VAI-3 \u00E2\u0080\u0098I told Jeff to smile.\u00E2\u0080\u0099 With psychological predicates, the embedded simple CONJUNCT clause gets an a-veridical interpretation, while the embedded \u00C3\u00AA- clause consistently gets a factive (veridical) interpretation. For example, in (61), the simple CONJUNCT form (61a) does not entail anything about Jeff\u00E2\u0080\u0099s eating, while \u00C3\u00AA- clause entails (61b) that he did eat. (61) a. Jeff \u00C3\u00AA-wanikiskisit ka-m\u00C3\u00AEcisot J \u00C3\u00AA- wanikiskisi -t ka- m\u00C3\u00AEciso -t J C1-forget.VAI -3 IRR-eat.VAI-3 \u00E2\u0080\u0098Jeff forgot to eat.\u00E2\u0080\u0099 b. Jeff \u00C3\u00AA-wanikiskisit \u00C3\u00AA-m\u00C3\u00AEcisot J \u00C3\u00AA- wanikiskisi -t \u00C3\u00AA- m\u00C3\u00AEciso -t J C1-forget.VAI -3 C1-eat.VAI -3 \u00E2\u0080\u0098Jeff forgot that he had eaten.\u00E2\u0080\u0099 Likewise, in (62), the simple CONJUNCT does not entail that the speaker has ever been left home alone. The \u00C3\u00AA- CONJUNCT does entail that the speaker is alone. (62) a. nis\u00C3\u00AAkisin ka-p\u00C3\u00AAyakwapiy\u00C3\u00A2n ni-s\u00C3\u00AAkisi -n ka- p\u00C3\u00AAyokw\u00C3\u00A2pi -y\u00C3\u00A2n 1- afraid.VAI-SAP IRR-be.alone.VAI-1 \u00E2\u0080\u0098I\u00E2\u0080\u0099m afraid to be home alone.\u00E2\u0080\u0099 comment: you would say this even if you\u00E2\u0080\u0099ve never been left alone before b. nis\u00C3\u00AAkisin \u00C3\u00AA-p\u00C3\u00AAyakwapiy\u00C3\u00A2n ni-s\u00C3\u00AAkisi -n \u00C3\u00AA- p\u00C3\u00AAyokw\u00C3\u00A2pi -y\u00C3\u00A2n 1- afraid.VAI-SAP C1-be.alone.VAI-1 \u00E2\u0080\u0098I was afraid being home alone.\u00E2\u0080\u0099 comment: I\u00E2\u0080\u0099m home alone right now 303 6.4.1.3 Sensitivity to negation A third piece of evidence for the semantic properties of the Romance clause-typing is its sensitivity to negation. An epistemic predicate that introduces the indicative (63a), can introduce the subjunctive if it is negated (64a-b). (63) Epistemic predicates in Romanian introduce indicative Ion crede [c\u00C4\u0083 a venit Ana]. I. believes that has come A \u00E2\u0080\u0098Ion believes Ana came.\u00E2\u0080\u0099 (64) Negative epistemic predicates in Romanian introduce indicative or subjunctive a. Nu cred [s\u00C4\u0083 fi venit Ana ]. no believe.1 SUBJ past come A. \u00E2\u0080\u0098I don\u00E2\u0080\u0099t believe Ana came.\u00E2\u0080\u0099 b. Nu cred [c\u00C4\u0083 a veniit Ana ]. no believe.1 that has come A \u00E2\u0080\u0098I don\u00E2\u0080\u0099t believe Ana came.\u00E2\u0080\u0099 (from Farkas 1992:70) Likewise, negation introduces the possibility for either clause-type in Plains Cree. As we saw in chapter 5, the two clause-types correspond to two distinct contexts that are not distinguished in English clause-typing. In (65) the simple CONJUNCT expresses a proposition whose truth is not entailed according to any individual, while in (66) the \u00C3\u00AA- CONJUNCT expresses a proposition whose truth is entailed, but not relative to the speaker\u00E2\u0080\u0099s visual perception (i.e., the higher predicate). (65) context: when no-one knows where little brother is, someone asks if you saw him m\u00C3\u00B4ya \u00C3\u00AA-w\u00C3\u00A2pamak nis\u00C3\u00AEmis wayaw\u00C3\u00AEhtamihk ka-m\u00C3\u00AAtaw\u00C3\u00AAt m\u00C3\u00B4ya \u00C3\u00AA- w\u00C3\u00A2pam -ak ni- s\u00C3\u00AEmis wayaw\u00C3\u00AEhtamihk ka- m\u00C3\u00AAtaw\u00C3\u00AA -t NEG C1-see.VTA-1>3 1- sibling outside IRR-play.VAI-3 \u00E2\u0080\u0098I didn\u00E2\u0080\u0099t see my little brother playing outside.\u00E2\u0080\u0099 comment: nobody knows if he\u00E2\u0080\u0099s playing outside or not 304 (66) context: if some accident happened outside where your brother got hurt through you not seeing him. You say this afterwards m\u00C3\u00B4ya \u00C3\u00AA-w\u00C3\u00A2pamak nis\u00C3\u00AEmis wayaw\u00C3\u00AEhtamihk \u00C3\u00AA-m\u00C3\u00AAtaw\u00C3\u00AAt m\u00C3\u00B4ya \u00C3\u00AA- w\u00C3\u00A2pam -ak ni- s\u00C3\u00AEmis wayaw\u00C3\u00AEhtamihk \u00C3\u00AA- m\u00C3\u00AAtaw\u00C3\u00AA -t NEG C1-see.VTA -1>3 1- sibling outside C1-play.VAI-3 \u00E2\u0080\u0098I didn\u00E2\u0080\u0099t see my little brother playing outside.\u00E2\u0080\u0099 comment: he was playing outside, but you didn\u00E2\u0080\u0099t see him 6.4.1.4 Insensitivity to factive-emotive predicates So far, I have compared Plains Cree only to Romanian. If we broaden our perspective to include other Romance languages (e.g., French, Spanish, & Italian), and German, we see that in crucial places where Romance languages show variation, Plains Cree patterns with Romanian. One such context is clauses embedded under a factive-emotive predicate like be sad (that), or be glad (that). Farkas (2003) reports that in Romanian, these clauses are always indicative, while in French (and Spanish), they vary with the subjunctive clause being preferred. (67) a. Ion e trist c\u00C4\u0083 Maria e bolnav\u00C4\u0083. Romanian Ion is sad that Maria is.INDIC sick \u00E2\u0080\u0098Ion is sad that Maria is sick.\u00E2\u0080\u0099 (Farkas 2003:2) (68) a. Jean regrette que Marie est mal Jean regrets that Marie is.INDIC badly \u00E2\u0080\u0098Jean regrets that Marie is sick.\u00E2\u0080\u0099 b. Jean regrette que Marie soit mal. French Jean regrets that Marie is.SUBJ badly \u00E2\u0080\u0098Jean regrets that Marie is sick.\u00E2\u0080\u0099 (Farkas 2003:2) Plains Cree patterns with Romanian here. For example, when a predicate like miyw\u00C3\u00AAyiht- \u00E2\u0080\u0098like.VTI\u00E2\u0080\u0099 introduces an embedded factive proposition, the dependent clause is introduced by \u00C3\u00AA-. (69) \u00C3\u00AAwak \u00C3\u00B4hci mistahi nimiyw\u00C3\u00AAyiht\u00C3\u00AAn ayisiyiniw \u00C3\u00AAkotowahk \u00C3\u00AA-nit\u00C3\u00B4skahk. \u00C3\u00AAwakw ohci mistahi ni- miyw\u00C3\u00AAyiht\u00C3\u00AA -n ayisiyiniw \u00C3\u00AAkotowahk \u00C3\u00AA- nit\u00C3\u00B4skam -k TOP ORIG much 1- glad.VTI -SAP person that.kind C1-search.VTI-0 \u00E2\u0080\u0098For this reason I am very glad that people are searching for this kind.\u00E2\u0080\u0099 (JKN 3.16) 305 A list of the relevant predicates whose dependent-clause-typing I have documented is in Table 6.9. Predicate Clause-type factive-emotives miyw\u00C3\u00AAyiht- \u00E2\u0080\u0098be.glad.VTI\u00E2\u0080\u0099 kost- \u00E2\u0080\u0098be.afraid.VTI\u00E2\u0080\u0099 takahk\u00C3\u00AAyiht- \u00E2\u0080\u0098approve.VTI\u00E2\u0080\u0099 miyw\u00C3\u00A2sin \u00E2\u0080\u0098be.good.VII\u00E2\u0080\u0099 \u00C3\u00AA- Table 6.9. Factive-emotive predicates in Plains Cree These are a subset of the psychological predicates that we saw above; the important point here is that when the embedded proposition is factive, simple CONJUNCT clause-typing cannot be used. (70) context: my brother came to visit me; I want to express my approval a. nitakahk\u00C3\u00AAyiht\u00C3\u00AAn \u00C3\u00AA-p\u00C3\u00AA-kiyok\u00C3\u00AAt nis\u00C3\u00AEmis ni- takahk\u00C3\u00AAyiht\u00C3\u00AA -n \u00C3\u00AA- p\u00C3\u00AA- kiyok\u00C3\u00AA -t ni- s\u00C3\u00AEmis 1- approve.VTI -SAP C1-COME-visit.VAI-3 1- sibling \u00E2\u0080\u0098I\u00E2\u0080\u0099m glad my brother came to visit.\u00E2\u0080\u0099 b. ! nitakahk\u00C3\u00AAyiht\u00C3\u00AAn ka-p\u00C3\u00AA-kiyok\u00C3\u00AAt nis\u00C3\u00AEms ni- takahk\u00C3\u00AAyiht\u00C3\u00AA -n ka- p\u00C3\u00AA- kiyok\u00C3\u00AA -t ni- s\u00C3\u00AEmis 1- approve.VTI -SAP IRR-COME-visit.VAI-3 1- sibling --- (intended: \u00E2\u0080\u0098I\u00E2\u0080\u0099m glad my brother came to visit.\u00E2\u0080\u0099) To account for the difference between the subjunctive in Romanian versus the subjunctive in French and Spanish, Farkas proposes that the subjunctive is triggered by different semantic contexts. For French and Spanish, the subjunctive occurs in contexts where the proposition must be evaluated (i.e., judged as good or bad; cf. Heim 1992) by some individual (either the subject or the speaker). Since factive-emotives, desideratives, and directives all have this evaluation component, they all introduce the subjunctive. Analyzing Plains Cree\u00E2\u0080\u0099s simple CONJUNCT clauses as a-veridical correctly predicts that they cannot occur in this context. 6.4.1.5 Insensitivity to lack-of-speaker-knowledge The other context where variation in mood selection has been widely discussed is with clauses introduced by epistemic (perceptual & belief) and communcative predicates (cf. Farkas 2003, 306 Schlenker 2003, Giorgi & Pianesi 1997, among others). In German and Italian, these predicates introduce a subjunctive clause \u00E2\u0080\u009Cto mark the the absence of speaker commitment to the complement\u00E2\u0080\u009D (Farkas 2003:17). In (70a), the indicative is used and the proposition is represented as true according to the speaker; in (70b) the subjunctive indicates that the truth of the proposition cannot be determined relative to the speaker (even if its truth can be determined relative to another individual). (71) a. Ich habe geh\u00C3\u00B6rt, da\u00C3\u009F er krank ist. German I have heard that he sick is.INDIC \u00E2\u0080\u0098I heard that he\u00E2\u0080\u0099s sick.\u00E2\u0080\u0099 (I do not doubt it.) b. Ich habe geh\u00C3\u00B6rt, da\u00C3\u009F er krank sei/w\u00C3\u00A4re. I have heard that he sick is.SUBJ1/is.SUBJ2 \u00E2\u0080\u0098I heard that he\u00E2\u0080\u0099s sick. (I don\u00E2\u0080\u0099t know whether this is true.) (Lederer 1969:118) Here again, Plains Cree does not make a distinction. Epistemic predicates introduce a proposition whose truth can be judged according to someone: in (71) the truth of \u00E2\u0080\u0098Jeff is sick\u00E2\u0080\u0099 is judged according to whoever the speaker heard this from. The a-veridicality analysis correctly predicts that the simple CONJUNCT is unavailable. (72) a. \u00C3\u00AA-p\u00C3\u00AAhtam\u00C3\u00A2n Jeff \u00C3\u00AA-nikamot ot\u00C3\u00A2kosihk \u00C3\u00AA- p\u00C3\u00AAhtam -\u00C3\u00A2n J \u00C3\u00AA- nikamo -t ot\u00C3\u00A2kosin -k C1-hear.VTI -1 J C1-sing.VAI-3 be.evening.VII -0 \u00E2\u0080\u0098I heard that Jeff sang yesterday\u00E2\u0080\u0099 b. * \u00C3\u00AA-p\u00C3\u00AAhtam\u00C3\u00A2n Jeff, ka-nikamot ot\u00C3\u00A2kosihk \u00C3\u00AA- p\u00C3\u00AAhtam -\u00C3\u00A2n J \u00C3\u00AA- nikamo -t ot\u00C3\u00A2kosin -k C1-hear.VTI -1 J C1-sing.VAI-3 be.evening.VII -0 --- (intended: \u00E2\u0080\u0098I heard that Jeff sang yesterday (I don\u00E2\u0080\u0099t know whether it\u00E2\u0080\u0099s true).\u00E2\u0080\u0099) Farkas points out that in a language like German, the subjunctive is sensitive to the speaker\u00E2\u0080\u0099s knowledge or belief state; thus veridicality must either be (re-)defined in terms of the speaker or veridicality is not the right way to talk about the subjunctive in German. For Romanian and Plains Cree, however, a-veridicality correctly excludes the subjunctive in contexts where there is some individual (e.g., the subject of a higher verb) that the truth of the proposition can be judged according to. 307 6.4.2 Adverbial clauses split along a-veridicality Adverbial clauses (i.e., clauses adjoined to CP; cf. Chapter 5) in the simple CONJUNCT are also used for clauses that express an a-veridical event: i.e., an event that cannot be judged as true or false with respect to any individual. These include irrealis temporal modification clauses (if/when), purpose/future result clauses, the antecedents of conditionals, and before-clauses. 6.4.2.1 Irrealis temporal modification Just as there is a realis/irrealis split in Plains Cree\u00E2\u0080\u0099s matrix clauses (with irreality needing to be overtly marked, e.g., via ka-), there is also a realis/irrealis split in dependent clauses. For example, clauses that provide a temporal relation to an irrealis event must use the simple CONJUNCT with subjunctive -i. In (72a) the temporal relation is overtly marked with the temporal sequencer isp\u00C3\u00AE \u00E2\u0080\u0098then\u00E2\u0080\u0099 and is related to a clause marked with w\u00C3\u00AE \u00E2\u0080\u0098going to\u00E2\u0080\u0099; in (72b) we get the same sequencer isp\u00C3\u00AE, but the main clause is in the past (note the preverb k\u00C3\u00AE-), so the clause-type is \u00C3\u00AA-. (73) a. \u00E2\u0080\u009C \u00E2\u0080\u0098\u00E2\u0080\u00A6, \u00C3\u00AAkonik aniki piko \u00E2\u0080\u00A6 \u00C3\u00AA-w\u00C3\u00AE-pim\u00C3\u00A2tisicik \u00C3\u00B4t\u00C3\u00AA ati-n\u00C3\u00AEk\u00C3\u00A2n, \u00C3\u00AAkonik aniki piko \u00C3\u00AA- w\u00C3\u00AE- pim\u00C3\u00A2tisi -t -k \u00C3\u00B4t\u00C3\u00AA ati- n\u00C3\u00AEk\u00C3\u00A2n TOPIC.AN DEM.AN QUANT C1-INT-live.VAI -3-PL LOC DIR- future \u00E2\u0080\u0098 \u00E2\u0080\u009C \u00E2\u0080\u0098\u00E2\u0080\u00A6, only they are going to be alive then in the future, isp\u00C3\u00AE \u00C3\u00A2yimahki mistahi,\u00E2\u0080\u0099 k\u00C3\u00A2-isit,\u00E2\u0080\u009D itw\u00C3\u00AAw. isp\u00C3\u00AE ayiman -k -i mistahi k\u00C3\u00A2- it -it itw\u00C3\u00AA -w TEMP difficult.VII-0 -SUBJ very C2-say.to.VTA-3>1 say.VAI-3 when life will become very hard,\u00E2\u0080\u0099 he said to me,\u00E2\u0080\u009D she said.\u00E2\u0080\u0099 (AA 10.7) b. k\u00C3\u00AE-\u00C3\u00A2yiman m\u00C3\u00A2k \u00C3\u00A2ya, isp\u00C3\u00AE \u00C3\u00AA-kakw\u00C3\u00AAcimit aya n\u00C3\u00B4ht\u00C3\u00A2wiy aya, \u00E2\u0080\u00A6 k\u00C3\u00AE- \u00C3\u00A2yiman m\u00C3\u00A2ka aya isp\u00C3\u00AE \u00C3\u00AA- kakw\u00C3\u00AAcim -it aya n- \u00C3\u00B4ht\u00C3\u00A2wiy aya PREV-difficult.VII but CONN TEMP C1-ask.VTA -3>1 CONN 1- father CONN \u00E2\u0080\u0098But it was difficult then, when my father asked me, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 40) The same pattern is seen for mayaw \u00E2\u0080\u0098as soon as\u00E2\u0080\u0099. If the mayaw clause is subordinated to an irrealis clause (marked with ka-; 73a), it is in the simple CONJUNCT. If the mayaw clause is subordinated to a realis clause (e.g., marked with k\u00C3\u00AE-; 73b), it has the complementizer \u00C3\u00AA-. 308 (74) a. [ka-IND [mayaw CONJ] ] ..., \u00E2\u0080\u009Cka-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAhtahitin s\u00C3\u00AAm\u00C3\u00A2k, mayaw p\u00C3\u00B4ni-n\u00C3\u00AEmihitohk[i],\u00E2\u0080\u009D ... ki-ka- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AAhtah -iti -n s\u00C3\u00AAm\u00C3\u00A2k mayaw p\u00C3\u00B4ni-n\u00C3\u00AEmihit -oh -k -i 2-IRR-DIR-home.VTA-1>2-SAP immediately as.soon.as stop-dance.VAI-USC-3 -SUBJ \u00E2\u0080\u0098..., \u00E2\u0080\u009CI will bring you home right away, as soon as the dance is over,\u00E2\u0080\u009D \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 2.2) b. [ [mayaw \u00C3\u00AA-CONJ ] k\u00C3\u00AE-IND ] \u00C3\u00AAwak \u00C3\u00B4hci, mayaw \u00C3\u00AA-w\u00C3\u00A2pahtahkik \u00C3\u00AAkotowahk, k\u00C3\u00AE-otinamwak kis\u00C3\u00AAyiniwak, \u00E2\u0080\u00A6 \u00C3\u00AAwakw ohci mayaw \u00C3\u00AA- w\u00C3\u00A2pahtam -k -k \u00C3\u00AAkotawahk TOP ORIG as.soon.as C1-see.VTI -0 -PL that.kind k\u00C3\u00AE- otinam -w-ak kis\u00C3\u00AAyiniw-ak PREV-take.VTI-3-PL old.man -PL \u00E2\u0080\u0098For this reason the old men used to take it, as soon as they saw that kind, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (JKN 3.9) 6.4.2.2 Unrealized alternatives The particle iyikohk is also sensitive to clause-typing. When it introduces an \u00C3\u00AA- or k\u00C3\u00A2- clause, it is a degree marker akin to English so/such. An example is given in (74). (75) ... iyikohk \u00C3\u00AA-k\u00C3\u00AE-miyokiht\u00C3\u00A2y\u00C3\u00A2hk askipw\u00C3\u00A2wa, \u00C3\u00AAkosi m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-isi-tipaham\u00C3\u00A2hk, \u00E2\u0080\u00A6 iyikohk \u00C3\u00AA- k\u00C3\u00AE- miyokiht\u00C3\u00A2 -y\u00C3\u00A2n-k askipw\u00C3\u00A2w-a so C1-PREV-good.grow.VAI-1 -PL potato -PL \u00C3\u00AAkosi m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- isi- tipaham -\u00C3\u00A2n-k TOPIC usually C1-PREV-THUS-measure.VTI-1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6 we grew such a good crop of potatoes, that is how we measured them, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (EM 54) However, when iyikohk introduces a simple CONJUNCT clause, it presents an unrealized alternative to the realized event. For example, in (75), the alternatives are letting the meat spoil, and sharing it with other people (mentioned earlier in the discourse). (76) iyikohk ka-misiwan\u00C3\u00A2taniyik anima wiy\u00C3\u00A2s, \u00C3\u00AAkos \u00C3\u00A2nima m\u00C3\u00A2n \u00C3\u00AA-k\u00C3\u00AE-t\u00C3\u00B4tahkik, \u00E2\u0080\u00A6 iyikohk ka- misiwan\u00C3\u00A2tan-yi -k anima wiy\u00C3\u00A2s instead.of IRR-spoil.VII -DS-0 DEM.INAN meat \u00C3\u00AAkosi anima m\u00C3\u00A2na \u00C3\u00AA- k\u00C3\u00AE- t\u00C3\u00B4tam -k \u00E2\u0080\u0093k TOPIC DEM.INAN usually C1-PREV-do.VTI-0-PL \u00E2\u0080\u0098Instead of letting the meat spoil, that is what they used to do, \u00E2\u0080\u00A6 \u00E2\u0080\u0099 (EM 57) 309 In this case, the function of the preverb is determined on the basis of the clause-type it combines with: with \u00C3\u00AA- clauses, the event is realis, and the iyikohk functions as a degree-marker, while with ka- clauses, the event is irrealis. This is consistent with analyzing the simple CONJUNCT as having an a-veridical meaning \u00E2\u0080\u0093 the subordinating particles it combines with must have an a-veridical function. 6.4.2.3 \u00E2\u0080\u0098before\u00E2\u0080\u0099 clauses \u00E2\u0080\u0098Before\u00E2\u0080\u0099 clauses are a slightly more complicated case. \u00E2\u0080\u0098Before\u00E2\u0080\u0099 clauses in English are ambiguous between a reading where the event in the \u00E2\u0080\u0098before\u00E2\u0080\u0099 clause temporally follows the event in the main clause (i.e., this is just a case of adverbial temporal sequencing, akin to after, when, and while clauses), and a case where there is no statement that the event in the \u00E2\u0080\u0098before\u00E2\u0080\u0099 clause happened at all. These two readings are illustrated by the utterance in (76). (77) I left before I started crying. = I left, then I started crying. = I left in order to stop from crying (implies I didn\u00E2\u0080\u0099t cry). As many linguists have pointed out, \u00E2\u0080\u0098before\u00E2\u0080\u0099 clauses cannot be simple inverses of \u00E2\u0080\u0098after\u00E2\u0080\u0099 clauses. For example, regardless of whether the event in the \u00E2\u0080\u0098before\u00E2\u0080\u0099 clause is realized or not, before licenses negative polarity items (NPIs), while after does not (Landman 1991, Giannakidou 1998, Condorovdi & Beaver 2003): (78) a. I wrote John before he told anyone the news. b. * I wrote John after he told anyone the news. In (77a), anyone is fine in a context where I wrote John, and then he started telling people the news (e.g., I gave him instructions on what to say); it is also fine in a context where my writing prevented him from telling people the news (e.g., I wrote as a warning). Thus a system could be sensitive to before in one of two different ways. It could be sensitive to the distinction between whether the event is realized or not, in which case we would expect the two contexts in (77) to be grammatically coded in distinct ways: (a-)veridicality would depend on context. Alternatively, before could always trigger the a-veridical operator, 310 regardless of context, as happens with NPIs in English and many other languages (Giannakidou 1998). In Plains Cree, both possibilities are attested, depending on the speaker (i.e., there are two subgrammars). Some speakers treat clauses introduced by pamw\u00C3\u00A2y\u00C3\u00AAs \u00E2\u0080\u0098before\u00E2\u0080\u0099 as typed for a- veridicality: they are always simple conjunct clauses, regardless of whether the event is realized or not. This is illustrated in (78a-b); in (78a) the narrator did convince her husband of her point. In (78b), on the other hand, a command is given about something in the future; the event is not realized. Both clause are simple conjunct: they lack an overt complementizer. (79) a. realized event \u00EF\u0083\u00A0 a-veridical clause-typing kinw\u00C3\u00AAs nik\u00C3\u00AE-tasim\u00C3\u00A2w, p\u00C3\u00A2mway\u00C3\u00AAs ta-kaskimak anima; kinw\u00C3\u00AAs ni- k\u00C3\u00AE- tasim -\u00C3\u00A2 -w p\u00C3\u00A2mway\u00C3\u00AAs ta- kaskim -ak anima long.time 1- PREV-persevere.VTA-DIR-3 before IRR-convince.VTA-1>3 DEM.INAN \u00E2\u0080\u0098I kept at it a long time before I convinced him of it;\u00E2\u0080\u0099 (AA 4.6) b. unrealized event \u00EF\u0083\u00A0 a-veridical clause-typing \u00E2\u0080\u00A6 kis\u00C3\u00AEp\u00C3\u00AAkinihkok kipih\u00C3\u00AAmiw\u00C3\u00A2wak, p\u00C3\u00A2mway\u00C3\u00AAs p\u00C3\u00AA-nitaw-\u00C3\u00A2yamih\u00C3\u00A2y\u00C3\u00AAk!\u00E2\u0080\u009D kis\u00C3\u00AEp\u00C3\u00AAkin -ihkok ki- pih\u00C3\u00AAm -w\u00C3\u00A2w -ak p\u00C3\u00A2mway\u00C3\u00AAs p\u00C3\u00AA- nitaw- \u00C3\u00A2yamih\u00C3\u00A2 -y\u00C3\u00AAk wash.VTA-2PL>3 2- bird -2PL -PL before COME-GO- pray.VAI -2PL \u00E2\u0080\u00A6 wash your prairie-chickens before you come here to church!\u00E2\u0080\u009D (AA 7.2) Here the clause is being typed as a-veridical by the operator and thus it lacks the overt complementizer. In this system, we expect that p\u00C3\u00A2mw\u00C3\u00A2y\u00C3\u00AAs \u00E2\u0080\u0098before\u00E2\u0080\u0099 will not introduce other clause-types; this is borne out in both narratives I checked where the narrator had this grammar (Minde 1997, Ahenakew 2000). Other speakers, by contrast, differentiate the two clauses according to whether the event in the dependent clause is simply temporally sequenced (and therefore veridical), or a-veridical. For such a speaker, the veridical clause is marked as presuppositional and is clause-typed with k\u00C3\u00A2-, analogously to the other temporally sequenced clauses (cf. \u00C2\u00A76.3). 311 (80) realized event \u00EF\u0083\u00A0 presuppositional clause-typing a. \u00C3\u00AAkwa p\u00C3\u00B4n-mispon p\u00C3\u00A2mway\u00C3\u00AAs k\u00C3\u00A2-kisk\u00C3\u00B4payit \u00C3\u00AAkwa p\u00C3\u00B4n-mispon p\u00C3\u00A2mway\u00C3\u00AAs k\u00C3\u00A2-kisk\u00C3\u00B4payi -t and stop-snow.VII before C2-awake.VAI-3 \u00E2\u0080\u0098It stopped snowing before he woke up.\u00E2\u0080\u0099 b. n\u00C3\u00A2p\u00C3\u00AAw ana \u00C3\u00AA-\u00C3\u00B4c\u00C3\u00AAmit, p\u00C3\u00A2mway\u00C3\u00AAs k\u00C3\u00A2-sipw\u00C3\u00AAht\u00C3\u00AAt n\u00C3\u00A2p\u00C3\u00AAw ana \u00C3\u00AA-oc\u00C3\u00AAm -it p\u00C3\u00A2mway\u00C3\u00AAs k\u00C3\u00A2-sipw\u00C3\u00AAht\u00C3\u00AA -t man DEM C1-kiss.VTA-3>1 before C2-leave.VAI-3 \u00E2\u0080\u0098The man kissed me before he left.\u00E2\u0080\u0099 The other clause gets the default clause-typing of \u00C3\u00AA-, as for example in (80), where I take the knife away and thereby prevent a young girl from hurting herself. Here \u00C3\u00AA- is functioning in contrast to k\u00C3\u00A2-: k\u00C3\u00A2- presupposes that the event did occur, and \u00C3\u00AA-, by blocking, does not. (81) unrealized event \u00EF\u0083\u00A0 elsewhere clause-typing nimaskam\u00C3\u00A2w m\u00C3\u00B4hkom\u00C3\u00A2n aps\u00C3\u00AEs iskw\u00C3\u00AAsis pamway\u00C3\u00AAs \u00C3\u00AA-m\u00C3\u00A2nsisot. ni- maskam -\u00C3\u00A2 -w m\u00C3\u00B4hkom\u00C3\u00A2n aps\u00C3\u00AEs iskw\u00C3\u00AAsis p\u00C3\u00A2mway\u00C3\u00AAs \u00C3\u00AA- m\u00C3\u00A2niso -t 1- take.away.VTA-DIR-3 knife little girl before C1-cut.self.VAI-3 \u00E2\u0080\u0098I forcefully took away the knife from the little girl before she cut herself.\u00E2\u0080\u0099 In this second type of system, we correctly predict that a context where it is not possible that the second event happened will exclude the k\u00C3\u00A2- form. (82) context: out on the town for fun, but knowing I should save some of my money, return home early and still have some left a. nik\u00C3\u00AEw\u00C3\u00A2n p\u00C3\u00A2mway\u00C3\u00AAs nis\u00C3\u00B4niy\u00C3\u00A2s kahkiyaw \u00C3\u00AA-m\u00C3\u00AAstopayiy\u00C3\u00A2n13 ni- k\u00C3\u00AEw\u00C3\u00AA -n p\u00C3\u00A2mway\u00C3\u00AAs ni-s\u00C3\u00B4niy\u00C3\u00A2s kahkiyaw \u00C3\u00AA- m\u00C3\u00AAstopayi -y\u00C3\u00A2n 1- go.home.VAI-SAP before 1- money all C1-spend.VAI -1 \u00E2\u0080\u0098I went home before I spent all my money.\u00E2\u0080\u0099 b. # nik\u00C3\u00AEw\u00C3\u00A2n p\u00C3\u00A2mway\u00C3\u00AAs nis\u00C3\u00B4niy\u00C3\u00A2s kahkiyaw k\u00C3\u00A2-m\u00C3\u00AAstopayiy\u00C3\u00A2n ni- k\u00C3\u00AEw\u00C3\u00AA -n p\u00C3\u00A2mway\u00C3\u00AAs ni-s\u00C3\u00B4niy\u00C3\u00A2s kahkiyaw k\u00C3\u00A2- m\u00C3\u00AAstopayi-y\u00C3\u00A2n 1- go.home.VAI-SAP before 1- money all C2- spend.VAI -1 intended: I went home before I spent all my money. comment: that k\u00C3\u00A2- has to do with past, \u00E2\u0080\u00A6 so you wouldn\u00E2\u0080\u0099t use it here\u00E2\u0080\u00A6 K\u00C3\u00A2- mestopayiy\u00C3\u00A2n is \u00E2\u0080\u0098When I spent it\u00E2\u0080\u0099 \u00E2\u0080\u0098\u00E2\u0080\u00A6it\u00E2\u0080\u0099s all gone\u00E2\u0080\u0099 13 I do not know why the quantifier follows the noun it quantifies over in these examples, although it was relatively common for this speaker. 312 Thus, Plains Cree provides examples of both kinds of sensitivity to before in its clause- typing system. 6.4.2.4 Antecedents of conditionals The antecedent of a conditional must be typed as such in one of two ways in Plains Cree: either it must be introduced by the conditional particle k\u00C3\u00AEspin \u00E2\u0080\u0098if\u00E2\u0080\u0099, which can introduce any clause-type14, or the clause must be in the simple CONJUNCT (cf. D\u00C3\u00A9chaine & Wolfart 2005, Barczak et. al 2006). Put another way, k\u00C3\u00AEspin can be omitted from the antecedent only if the clause-typing is simple CONJUNCT. Notice that for antecedents of conditionals, the clause may either be in the subjunctive15 (marked with the suffix -i) or not16. (83) k\u00C3\u00AEspin is optional in simple CONJUNCT clauses a. mummy, (k\u00C3\u00AEspin) ka-p\u00C3\u00AA-kiyok\u00C3\u00A2wiyan, t\u00C3\u00B4nit\u00C3\u00B4ni nika-cihk\u00C3\u00AAyiht\u00C3\u00AAn M k\u00C3\u00AEspin ka- p\u00C3\u00AA- kiyok\u00C3\u00A2wi -yan mitoni ni-ka- c\u00C3\u00AEhk\u00C3\u00AAyiht\u00C3\u00AA -n M if IRR-COME-visit.VAI -2 very 1- IRR-happy.VTI -SAP \u00E2\u0080\u0098Mom, if you came to visit (me), I would be very happy.\u00E2\u0080\u0099 b. (k\u00C3\u00AEspin) w\u00C3\u00A2pamaki John, nika-w\u00C3\u00AEhtamaw\u00C3\u00A2w k\u00C3\u00A2-itw\u00C3\u00AAyan k\u00C3\u00AEspin w\u00C3\u00A2pam -ak -i J ni- ka- w\u00C3\u00AEhtamaw -\u00C3\u00A2 -w k\u00C3\u00A2- itw\u00C3\u00AA -yan if see.VTA -1>3-SUBJ J 1- IRR- tell.VTA -DIR-3 C2- say.VAI-2 \u00E2\u0080\u0098Should I see John, I\u00E2\u0080\u0099ll tell him what you said.\u00E2\u0080\u0099 If the antecedent is, for example, an indexical INDEPENDENT clause, k\u00C3\u00AEspin \u00E2\u0080\u0098if\u00E2\u0080\u0099 is obligatory. 14 Syntactically, conditionals have been analyzed as topic/comment structures in Plains Cree (D\u00C3\u00A9chaine & Wolfart 2005). The choice of clause-type with k\u00C3\u00AEspin depends on the kind of conditional being used. INDEPENDENT clauses are used in indicative conditionals, changed (\u00C3\u00AA- or k\u00C3\u00A2-) CONJ for realis non-subjunctive conditionals, and simple CONJ for realis subjunctive and irrealis conditionals. See D\u00C3\u00A9chaine & Wolfart 2005, Barczak et. al 2006 for details. 15 In Plains Cree, the term subjunctive is used for simple CONJUNCT forms that are suffixed with -i (Wolfart 1973). Unlike the Indo-European subjunctive, it is used (almost?) exclusively in if/when clauses. 16 One of the consultants I worked with noted that the subjunctive form carries more certainty about the antecedent being fulfilled than the ka-prefixed form. 313 (84) a. k\u00C3\u00AEspin niw\u00C3\u00A2pam\u00C3\u00A2w Jeff, nika-cihk\u00C3\u00AAyiht\u00C3\u00AAn k\u00C3\u00AEspin ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w J ni- ka- c\u00C3\u00AEhk\u00C3\u00AAyiht\u00C3\u00AA -n if 1- see.VTA-DIR-3 J 1- IRR-happy.VTI -SAP \u00E2\u0080\u0098If/when I see Jeff, I\u00E2\u0080\u0099ll be happy.\u00E2\u0080\u0099 b. ! niw\u00C3\u00A2pam\u00C3\u00A2w Jeff, nika-cihk\u00C3\u00AAyiht\u00C3\u00AAn k\u00C3\u00AEspin ni- w\u00C3\u00A2pam -\u00C3\u00A2 -w J ni- ka- c\u00C3\u00AEhk\u00C3\u00AAyiht\u00C3\u00AA -n if 1- see.VTA-DIR-3 J 1- IRR-happy.VTI -SAP --- (intended: \u00E2\u0080\u0098If/when I see Jeff, I\u00E2\u0080\u0099ll be happy.\u00E2\u0080\u0099) Since the antecedent of the conditional is only stating a condition on some other event, and says nothing about whether that condition has been realized or not, the clause-typing pattern is fully consistent with claiming that simple CONJUNCT clauses are a-veridical. 6.4.2.5 Purpose clauses Purpose clauses indicate the (perhaps unfilled) purpose that some event happens in order to fulfill. In (85) the purpose of being in graduate school is getting a PhD; purpose clauses in English are introduced by either just the infinitival to or by the more extended in order to. (85) I was in graduate school (in order) to get a PhD. Like with the mediated argument clauses we saw above, there is no way to evaluate the truth of propositions introduced by the purpose clause. There is no way to judge from (85) the truth or the falsity of the proposition \u00E2\u0080\u0098I got a PhD.\u00E2\u0080\u0099 These clauses, like other a-veridical clauses, lack the overt complementizer. In (86a-b), we see two examples of purpose clauses: the purpose of the man in waiting [to take the narrator home], and the purpose of the twin in training [to be a pilot]. (86) a. ..., \u00C3\u00AA-p\u00C3\u00AAhit ta-naskomak ka-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AEhtahit. \u00C3\u00AA- p\u00C3\u00AAh -it ta- naskom -ak ka- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AEhtah -it C1-wait.VTA-3>1 IRR-respond.VTA-1>3 IRR-COME-take.home.VTA-3>1 \u00E2\u0080\u0098..., waiting for my response so that he could take me home.\u00E2\u0080\u0099 (AA 2.1) 314 b. \u00C3\u00AAkwa ana p\u00C3\u00AAyak wiy \u00C3\u00A2na n\u00C3\u00AEs\u00C3\u00B4t\u00C3\u00AAw ana, \u00C3\u00AA-k\u00C3\u00AE-~ \u00C3\u00AA-k\u00C3\u00AE-kiskinoham\u00C3\u00A2ht ta-pimih\u00C3\u00A2t, ... \u00C3\u00AAkwa ana p\u00C3\u00AAyak wiya ana n\u00C3\u00AEs\u00C3\u00B4t\u00C3\u00AA -w ana and DEM.AN one EMPH DEM.AN twin.VAI-3 DEM.AN \u00C3\u00AA- k\u00C3\u00AE- kiskinoham -\u00C3\u00A2h -t ta- pimih\u00C3\u00A2 -t C1-PREV-train.VTA -USC-3 IRR-pilot.VAI -3 \u00E2\u0080\u0098And one of the twins trained to be a pilot, \u00E2\u0080\u00A6\u00E2\u0080\u0099 (AA 5.2) When speakers are asked how to express a purpose clause, they volunteer the simple CONJUNCT and reject other clause-types like \u00C3\u00AA-clauses. (87) context: translation task: \u00E2\u0080\u0098I did it to make her happy.\u00E2\u0080\u0099 a. \u00C3\u00AA-it\u00C3\u00B4tam\u00C3\u00A2n \u00C3\u00AAwakw \u00C3\u00A2nima, ka-cihk\u00C3\u00AAyiht\u00C3\u00A2hk \u00C3\u00AA- it\u00C3\u00B4tam -\u00C3\u00A2n \u00C3\u00AAwakw anima ka- cik\u00C3\u00AAyihtam -k C1-do.VTI -1 TOPIC DEM.INAN IRR- happy.VTI -0 \u00E2\u0080\u0098I did this so in order for her to be happy.\u00E2\u0080\u0099 b. # \u00C3\u00AA-it\u00C3\u00B4tam\u00C3\u00A2n \u00C3\u00AAwakw \u00C3\u00A2nima, \u00C3\u00AA-cihk\u00C3\u00AAyiht\u00C3\u00A2hk \u00C3\u00AA- it\u00C3\u00B4tam -\u00C3\u00A2n \u00C3\u00AAwakw anima \u00C3\u00AA- cik\u00C3\u00AAyihtam -k C1-do.VTI -1 TOPIC DEM.INAN C1-happy.VTI -0 \u00E2\u0080\u0098I did it, she was happy.\u00E2\u0080\u0099 comment: these are two separate sentences, they\u00E2\u0080\u0099re not connected 6.5 Summary There are three robust clause-typing alternations in Plains Cree embedded clauses. I have shown that k\u00C3\u00A2- clauses introduce a presupposed proposition, simple CONJUNCT clauses introduce an a- veridical proposition, and that \u00C3\u00AA- clauses are an elsewhere case, where the semantics of the proposition depends on the context in which it is introduced. I showed that these characterizations help us understand more about the distribution and function of Plains Cree clause-types. A final question to be addressed is: why this division? Why do the clause-types in Plains Cree have the functions they do? Is there a higher-order principle that can derive this split? I do not have a full answer to this question. However, we do see that the system can be thought of as organized around the issue of the truth of the proposition relative to the discourse. 315 According to the proposal in this chapter, (a-)veridicality distinguishes between propositions that have an entailment (of either truth or falsity), and those who don\u00E2\u0080\u0099t. (88) Proposition 3 Veridical A-veridical (CHANGED (SIMPLE CONJUNCT) CONJUNCT) The distinction between a-veridicality and the more familiar non-veridicality, was made on the basis of data like (57), repeated here as (89), where negation, entailing the falsity of p does not license the clause-type under discussion. (89) a. niw\u00C3\u00AEhtam\u00C3\u00A2k Laura \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA-m\u00C3\u00A2toyit Martha ni-w\u00C3\u00AEhtam-aw-ikw L \u00C3\u00AAk\u00C3\u00A2 \u00C3\u00AA- m\u00C3\u00A2to -yi -t M 1-tell.VTA-BEN-INV L NEG C1-cry.VAI-DS-3 M \u00E2\u0080\u0098Laura told me that Martha isn\u00E2\u0080\u0099t crying.\u00E2\u0080\u0099 b. ! niw\u00C3\u00AEhtam\u00C3\u00A2k Laura \u00C3\u00AAk\u00C3\u00A2 ka-m\u00C3\u00A2toyit Martha ni-w\u00C3\u00AEhtam-aw-ikw L \u00C3\u00AAk\u00C3\u00A2 ka- m\u00C3\u00A2to -yi -t M 1-tell.VTA-BEN-INV L NEG IRR-cry.VAI-DS-3 M --- (intended: \u00E2\u0080\u0098Laura told me that Martha isn\u00E2\u0080\u0099t crying.\u00E2\u0080\u0099) Among veridical propositions, the clause-typing system in Plains Cree distinguishes among those whose truth is presupposed, and those whose truth is introduced. (90) Proposition 3 Veridical A-veridical (OTHER) (SIMPLE CONJUNCT) 3 Introduced Presupposed (OTHER) (K\u00C3\u0082-CONJUNCT) Finally, returning to the contrast between CONJUNCT and INDEPENDENT order clauses that we examined in chapters 3 and 4, those propositions whose truth is introduced into the discourse divide between those whose truth conditions are evaluated with respect to the speech situation, and those whose truth conditions are evaluated relative to an anaphorically given situation. 316 (91) Proposition 3 Veridical A-veridical (OTHER) (SIMPLE CONJUNCT) 3 Introduced Presupposed (OTHER) (K\u00C3\u0082-CONJUNCT) 3 Anaphoric Deictic (OTHER) (INDEPENDENT) On this view, the clause-typing system in Plains Cree, then, is fundamentally concerned with the evaluation of the truth of a proposition by the participants in a discourse. 317 CHAPTER 7 CONCLUSIONS 7.1 The syntax and semantics of clause-typing in Plains Cree In this thesis, I have argued that clause-typing in Plains Cree codes a fundamental distinction between indexical CPs, which are anchored to the speech act, and anaphoric CPs, which are not anchored, and thus must be licensed by general principles of anaphora. Chapters 2, 3, and 4 were concerned with arguing for the CP-status of both INDEPENDENT and CONJUNCT clauses, the indexical status of INDEPENDENT clauses, and the anaphoric status of CONJUNCT clauses, respectively. This yields the split in (1). (1) CLAUSE-TYPING 3 Deictic Anaphoric (INDEPENDENT) (CONJUNCT) In the last two chapters, I developed the syntax and semantics of anaphoric clauses in more detail. Syntactically we end up with the typology in (2), which is based on their distribution and island-like tests. (2) CLAUSE-TYPING qp Deictic Anaphoric (=Matrix) (=Elsewhere) (INDEPENDENT) (CONJUNCT) Precedence C-command Precedence & c-command (=CHAINS) (=ADJUNCTS) (=MEDIATED ARGUMENTS) 318 The semantics of anaphoric clauses cross-cuts their syntactic classification and corresponds to the choice of complementizer. I claimed there were two specified values (presupposed and a- veridical), as well as an elsewhere complementizer whose meaning is determined by the context it appears in. We can organize the classification of anaphoric clauses based on how the proposition is introduced into the discourse. (3) CLAUSE-TYPING 3 Deictic Anaphoric \u00EF\u0083\u009F Proposition is presented (INDEPENDENT) (CONJUNCT = ELSE) 3 Introduced Presupposed \u00EF\u0083\u009F Proposition is presupposed (\u00C3\u008A- = ELSE) (K\u00C3\u0082-CONJ) 3 (IC-CONJ = iterative) Veridical Non-veridical \u00EF\u0083\u009F Proposition is unevaluated (\u00C3\u008A- = ELSE) (SIMPLE CONJ) There are many ways in which this work could be expanded. One important line of research concerns how the clause-typing system presented here maps onto systems that use the other clause-typing divisions discussed in the literature. In Plains Cree, these other divisions (matrix vs. embedded, declarative vs. interrogative, etc.) overlay the indexical/anaphoric split, but some of the diagnostics developed here could be applied to other languages which lack the overt morpho-syntactic coding Plains Cree exhibits. The notions of \u00E2\u0080\u0098indexicality\u00E2\u0080\u0099 and \u00E2\u0080\u0098anaphora\u00E2\u0080\u0099 have traditionally been applied to argument expressions (DPs). In \u00C2\u00A77.2, I briefly discuss some further lines of research that could be pursued with respect to the parallel between CPs and DPs; I also discuss the link between the syntax and semantics of indexicals. Within Plains Cree, the current analysis both offers a new analytic possibility for k\u00C3\u00AE- (\u00C2\u00A77.3), and opens up questions with respect to the construction of modality (\u00C2\u00A77.4) A third line of research is to examine the range of variation in clause-typing found in languages closely-related to Plains Cree. This thesis concentrated closely on Plains Cree; in \u00C2\u00A77.5 I lay out how the forms that play a role in Plains Cree\u00E2\u0080\u0099s clause-typing are the same or different in Ojibwe and Blackfoot. 319 7.2 The parallels between CPs and DPs In this thesis, I used the property of indexicality to drive the syntax and semantics of the clause- type that is morpho-syntactically represented by Plains Cree\u00E2\u0080\u0099s INDEPENDENT order. In particular, an indexical expression is obligatorily free \u00E2\u0080\u0093 it cannot be bound. With respect to Plains Cree\u00E2\u0080\u0099s INDEPENDENT order, I showed that indexical clauses could not be either c-commanded or preceded by some other clause. In this respect, indexical clauses behave like R-expressions in the nominal domain; compare the conditions on indexical clauses with condition C of binding theory (Chomsky 1981), given in (4). (4) Condition C: R-expressions must be free Likewise, the property of anaphoricity can account for the distribution and interpretation of Plains Cree\u00E2\u0080\u0099s CONJUNCT order. Thus, parallel to anaphoric pronominals, we have evidence for anaphoric clauses. Just as an anaphoric pronominal is infelicitous without having a referent supplied, so an anaphoric clause is infelicitous without having a context supplied. Notice that in both domains (CP and DP) we have a syntactic requirement (i.e., \u00E2\u0080\u009Cmust be free\u00E2\u0080\u009D vs. \u00E2\u0080\u009Ccannot be free\u00E2\u0080\u009D) proceeding in hand with a semantic function (i.e., deixis vs. non- deixis). The parallel between these two separate domains of the grammar provides support for the idea that the syntax and semantics of notions like indexicality and anaphoricity are inextricable. Thus, although a systematic contrast and comparison of CPs and DPs is beyond the scope of this thesis, the overarching similarities warrant further research. 7.3 (Im)possible analyses of k\u00C3\u00AE- In the discussion of temporal relations in chapter 3, I made frequent use of the contrast between a bare clause and one with the preverbal element k\u00C3\u00AE- added. I would like to say a few words about what I think k\u00C3\u00AE- is, and what I am quite sure it is not. There are generally two \u00E2\u0080\u0093 or three, depending on how you count \u00E2\u0080\u0093 hypotheses of k\u00C3\u00AE-. One is that it is a past tense marker; the other is that it is an aspectual marker (i.e., perfect or 320 perfective). Previous literature for these analyses include Edwards 1954; Wolfart 1973; Dahlstrom 1986, 1991; and Hunter & Karpinski 1991. The hypothesis taken here is closer to the tense analysis then the aspect analysis, but crucially does not involve specification of speech time. (5) Hypotheses for k\u00C3\u00AE- H1: disjunction and precedence PRECEDE (Tref, Teval) H2: deictic past tense PRECEDE (Tref, T0) H3: aspect (perfect) INCLUDE (Tref, T0) and PRECEDE (Tsit, Tref) H4: aspect (perfective) 7.3.1 k\u00C3\u00AE- marks disjunction and precedence I model k\u00C3\u00AE- as marking a temporal non-coincidence relation between the reference time and the evaluation time. In the INDEPENDENT order, this is non-coincidence between the reference time and the speech time; in the CONJUNCT order, the non-coincidence is between the reference time and whatever time is given by the antecedent of the anaphoric time. In addition, as I briefly noted, but abstracted away from in chapter 3, we need to specify an ordering relation between the two times, since k\u00C3\u00AE- always marks a relation of precedence. (6) - COIN (Tref, Teval) Tref < T0 INDEPENDENT Tref < T CONJUNCT This captures the fixed \u00E2\u0080\u0098simple past\u00E2\u0080\u0099 interpretation of k\u00C3\u00AE- marked INDEPENDENT clauses (\u00C2\u00A73.3.1), and the shifting \u00E2\u0080\u0098past/pluperfect\u00E2\u0080\u0099 interpretation of k\u00C3\u00AE- marked CONJUNCT clauses (\u00C2\u00A74.4.1). 321 Syntactically, k\u00C3\u00AE- is modelled as a functional head very high in the clause (e.g., as a \u00E2\u0080\u0098low complementizer\u00E2\u0080\u0099 in the sense of Rizzi 1997)1. This reflects not only its function, but also captures its linear order and (in)sensitivity to other elements in the clause. Linearly, k\u00C3\u00AE- can only be preceded by clause-typing and the irrealis marker ka-. All other temporal elements and adverbials must follow it (cf. Edwards 1954, Wolfart 1973, Dahlstrom 1991, Cook 2004). A schematized template is given in (7); elements with a star \u00E2\u0080\u0098*\u00E2\u0080\u0099 can be iterated. (7) [ CLAUSE-TYPE] [IRREALIS] [k\u00C3\u00AE-] [restructure*] [p\u00C3\u00AA-] [restructure*] [adverbial*] [STEM] The interpretation of k\u00C3\u00AE- is not only sensitive to clause-typing (e.g., the contrast between INDEPENDENT and CONJUNCT), but also to irrealis operators. If negation appears in the clause, the temporal operator has a suppletive form oh(ci) (cf. chapter 3). (8) a. k\u00C3\u00AE-w\u00C3\u00A2pam\u00C3\u00AAwak k\u00C3\u00AE- w\u00C3\u00A2pam -\u00C3\u00AA -w -ak PREV-see.VTA-DIR-3 -PL \u00E2\u0080\u0098They saw him/her.\u00E2\u0080\u0099 b. nam\u00C3\u00B4ya ohci-w\u00C3\u00A2pam\u00C3\u00AAwak nam\u00C3\u00B4ya ohci- w\u00C3\u00A2pam -\u00C3\u00AA -w -ak NEG ORIG-see.VTA -DIR-3 -PL \u00E2\u0080\u0098They never saw him.\u00E2\u0080\u0099 If k\u00C3\u00AE- is left under negation (or under the irrealis ka-) it gives rise to a modal reading. (9) a. nam\u00C3\u00B4ya k\u00C3\u00AE-w\u00C3\u00A2pam\u00C3\u00AAwak nam\u00C3\u00B4ya k\u00C3\u00AE- w\u00C3\u00A2pam -\u00C3\u00AA -w -ak NEG PREV-see.VTA -DIR-3 -PL \u00E2\u0080\u0098They can\u00E2\u0080\u0099t see him/her.\u00E2\u0080\u0099 b. ka-k\u00C3\u00AE-w\u00C3\u00A2pam\u00C3\u00AAwak ka- k\u00C3\u00AE- w\u00C3\u00A2pam -\u00C3\u00AA -w -ak IRR-PREV-see.VTA-DIR-3 -PL \u00E2\u0080\u0098They can see him/her.\u00E2\u0080\u0099 1 I do not discuss the relation between Plains Cree\u00E2\u0080\u0099s clause-typing system proper (i.e., those complementizers that relate the proposition to the superstructure of the discourse, and other elements which seem to be low complementizers having to do with mood/modality/finiteness. This is a question for further research. 322 Finally, as I show down below, k\u00C3\u00AE- is completely insensitive to aspectual information such as predicate class or other aspectual operators. This is consistent with k\u00C3\u00AE- being associated with tense & modality, but surprising if k\u00C3\u00AE- being an aspectual operator. 7.3.2 k\u00C3\u00AE- is not a deictic past tense When one talks about whether an element is an instantiation of \u00E2\u0080\u0098tense\u00E2\u0080\u0099, there are often quite different criteria used to make the decision; although these criteria are related, at this point there is not much agreement in the literature as to which are necessary and/or sufficient. With respect to the semantic treatments of tense, it is important to distinguish between existential theories of tense which claim that tense involves existential quantification over times (e.g., Prior 1957, 1967, Montague 1973, Dowty 1979, Ogihara 1995), and deictic theories of tense that claim tense is referential (akin to pronouns) and is the relation of some time relative to speech time (Partee 1973, Enc 1987, Klein 1994, Kratzer 1998). Plains Cree\u00E2\u0080\u0099s k\u00C3\u00AE- cannot be analyzed as a deictic past tense, since it crucially has only an ordering relation, with no inherent reference to speech time. Recall from the discussion in chapter 3 that in CONJUNCT clauses, k\u00C3\u00AE- can sequence one event with respect to an event in a preceding or superordinating clause. The relevant data is repeated in (10): when there is no dedicated temporal marking, the linear order of events reflects their temporal sequencing (cf. Kamp & Rohrer 1983, Hinrichs 1986 for English); but the presence of k\u00C3\u00AE- on the second clause reverses the temporal ordering (similar to a past perfect in English). 323 (10) a. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff, \u00C3\u00AA-m\u00C3\u00AEcisoy\u00C3\u00A2hk \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J \u00C3\u00AA- m\u00C3\u00AEciso -y\u00C3\u00A2n -k C1-come-go.home.VAI-3 J C1-eat.vai -1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6Jeff came home, we had eaten.\u00E2\u0080\u0099 (we = speaker & someone else) = come home < eat b. \u00C3\u00AA-p\u00C3\u00AA-k\u00C3\u00AEw\u00C3\u00AAt Jeff, \u00C3\u00AA-k\u00C3\u00AE-m\u00C3\u00AEcisoy\u00C3\u00A2hk \u00C3\u00AA- p\u00C3\u00AA- k\u00C3\u00AEw\u00C3\u00AA -t J \u00C3\u00AA- k\u00C3\u00AE- m\u00C3\u00AEciso -y\u00C3\u00A2n -k C1-come-go.home.VAI-3 J C1-PREV-eat.vai -1 -PL \u00E2\u0080\u0098\u00E2\u0080\u00A6Jeff came home, we had eaten.\u00E2\u0080\u0099 (we = speaker & someone else) = eat < come home If k\u00C3\u00AE- were a deictic past tense marker, it should always shift with respect to T0, but here it behaves like a past perfect and shifts with respect to the time established in the previous clause. If we compare Plains Cree\u00E2\u0080\u0099s k\u00C3\u00AE- with, for example, English\u00E2\u0080\u0099s past tense marker (the analysis of which itself is the subject of much debate), we also see a difference between the two in modal constructions. In English, past tense (-ed) is used in antecedents of conditionals to indicate hypothetical situations (If I talked that way...) (cf. James 1982, Iatridou 2000, von Fintel 2005, among others). Plains Cree k\u00C3\u00AE- is also sensitive to modal constructions (cf. \u00C2\u00A77.3 for discussion), but it corresponds with existential quantificational force. For example, in (11a), we have an irrealis marker in an INDEPENDENT order clause, and the interpretation is a future tense, which corresponds to a universally quantified circumstantial modal. (11b) provides a minimal contrast: k\u00C3\u00AE- has been added, and the resulting interpretation is an ability interpretation, corresponding to an existentially quantified circumstantial modal. (11) a. ka-t\u00C3\u00A2hkon\u00C3\u00AAw acim\u00C3\u00B4sisa UNIVERSAL ka- t\u00C3\u00A2hkon -\u00C3\u00AA -w atimw -sis -a IRR-carry.VTA-DIR-3 dog -DIM-OBV \u00E2\u0080\u0098S/he will carry the puppies.\u00E2\u0080\u0099 b. ka-k\u00C3\u00AE-t\u00C3\u00A2hkon\u00C3\u00AAw acim\u00C3\u00B4sisa I EXISTENTIAL ka- k\u00C3\u00AE- t\u00C3\u00A2hkon -\u00C3\u00AA -w atimw -sis -a IRR-PREV-carry.VTA-DIR-3 dog -DIM-OBV \u00E2\u0080\u0098S/he is able to carry the puppies.\u00E2\u0080\u0099 While this interpretation appears to be compatible with an existential quantificational analysis of tense (here the effect of k\u00C3\u00AE- unambiguously contributes existential force), it is hard to see how such data would straightforwardly be accounted for with a deictic theory of tense. 324 There is also a great deal of variation in the literature on the syntax of tense. Based on English, tense is often taken to be (cross-linguistically, and (semi-)independent of its interpretation) a functional head located in the IP-domain and the assigner of nominative case. If we take these diagnostics to be criterial, then I do not believe that k\u00C3\u00AE- can be treated as tense in the syntactic sense. First, it does not assign nominative case (case-assignment, if it is a property of Algonquian languages, seems most closely to relate to the theme-sign system; see D\u00C3\u00A9chaine & Reinholtz 1997, 2008 for discussion). Second, the pervasive syntactic conditioning of k\u00C3\u00AE- by CP- level elements including the INDEPENDENT/CONJUNCT clause-typing distinction, modality, and CP-level negation, paired with the complete lack of syntactic conditioning by IP-level elements like the direct-inverse system and switch-reference, strongly suggest that k\u00C3\u00AE- is external to the IP- domain. If this is accurate, and if k\u00C3\u00AE- nevertheless has a semantic tense function, then Plains Cree\u00E2\u0080\u0099s grammar opens the door to a rethinking of the relation between the syntax and semantics of tense. 7.3.3 k\u00C3\u00AE- is not a perfect Recall that a perfect tense is a combination of tense and aspect (Comrie 1976; Fenn 1987; Klein 1994): it sequences the reference time both with respect to the speech time and with respect to the situation time. (12) + COIN (Tref, T0) - COIN (Tsit, Tref) If we compare the distribution of k\u00C3\u00AE- marked clauses to the distribution of the present perfect in English, we see that they are, in the contexts discussed for English, in complete complementary distribution. 325 Property k\u00C3\u00AE- ENGLISH PERFECT co-occur w/ \u00E2\u0080\u0098yesterday\u00E2\u0080\u0099 \u00E2\u009C\u0094 \u00E2\u009C\u0096 perfect of result \u00E2\u009C\u0096 \u00E2\u009C\u0094 experiential perfect \u00E2\u009C\u0096 \u00E2\u009C\u0094 recent perfect (news) \u00E2\u009C\u0096 \u00E2\u009C\u0094 persistent perfect \u00E2\u009C\u0096 \u00E2\u009C\u0094 Table 7.1. Patterning of k\u00C3\u00AE- relative to the English perfect First, k\u00C3\u00AE- is compatible with both ot\u00C3\u00A2kosihk \u00E2\u0080\u0098yesterday\u00E2\u0080\u0099 and anohc \u00E2\u0080\u0098today.\u00E2\u0080\u0099 (13) a. ot\u00C3\u00A2kosihk nik\u00C3\u00AE-w\u00C3\u00A2pam\u00C3\u00A2w atim min\u00C3\u00B4sa ot\u00C3\u00A2kosin -k ni-k\u00C3\u00AE- w\u00C3\u00A2pam -\u00C3\u00A2 -w atim min\u00C3\u00B4s-a be.evening.VII-0 1-PREV-see.VTA-DIR-3 dog cat -OBV \u00E2\u0080\u0098Yesterday I saw a dog kill a cat.\u00E2\u0080\u0099 (cf. * \u00E2\u0080\u0098Yesterday I have seen a dog kill a cat.\u00E2\u0080\u0099) b. anohc nik\u00C3\u00AE-itoht\u00C3\u00A2n ataw\u00C3\u00AAwikamikohk anohc ni-k\u00C3\u00AE- itoht\u00C3\u00A2-n ataw\u00C3\u00AAw-kamikw -hk today 1- PREV-go.VAI-SAP buy- building-LOC \u00E2\u0080\u0098Today I went to the store.\u00E2\u0080\u0099 Second, k\u00C3\u00AE- is not appropriate in \u00E2\u0080\u0098perfect of result\u00E2\u0080\u0099 contexts. (14) A: Tom c\u00C3\u00AE p\u00C3\u00AA-takosin T c\u00C3\u00AE p\u00C3\u00AA- takosin -3 T Q COME-arrive.VAI-3 \u00E2\u0080\u0098Has Tom arrived yet?\u00E2\u0080\u0099 B1: \u00C3\u00AAh\u00C3\u00A2, \u00C3\u00A2say p\u00C3\u00AA-takosin \u00C3\u00AAh\u00C3\u00A2 \u00C3\u00A2say p\u00C3\u00AA- takosin -w yes already COME-arrive.VAI-3 \u00E2\u0080\u0098Yes, he has arrived\u00E2\u0080\u0099 comment: this one is implying that he\u00E2\u0080\u0099s still here, that we\u00E2\u0080\u0099re ready to dance B2: # \u00C3\u00AAh\u00C3\u00A2, \u00C3\u00A2say k\u00C3\u00AE-p\u00C3\u00AA-takosin \u00C3\u00AAh\u00C3\u00A2 \u00C3\u00A2say k\u00C3\u00AE- p\u00C3\u00AA- takosin -w yes already PREV-COME-arrive.VAI-3 \u00E2\u0080\u0098Yes, he had already arrived.\u00E2\u0080\u0099 (before, previously, some time ago) comment: it could imply that he had already come and gone. I would probably be tempted to say \u00E2\u0080\u009Cwell, he is still here?\u00E2\u0080\u009D Third, k\u00C3\u00AE- is at best marginal with the experiential perfect. Experiential perfects describe an event that happened at some time previous to now, so depending on how the context is set, even in English there is variation between the past and perfect. For example, the context for (15) is 326 vague enough to allow both in English (cf. 15b). In this context, consultants find k\u00C3\u00AE-marked clauses appropriate. (15) context: talking about big things we\u00E2\u0080\u0099ve done in our lives a. Jeff (k\u00C3\u00AE-)itoht\u00C3\u00AAw ak\u00C3\u00A2maskiy J k\u00C3\u00AE- itoht\u00C3\u00AA -w ak\u00C3\u00A2maskiy J PREV-go.VAI-3 across.land \u00E2\u0080\u0098Jeff went overseas (e.g., to Europe).\u00E2\u0080\u0099 b. Jeff went overseas vs. Jeff has gone overseas But if the context is more specified, the bare clause is used instead. For example, in asking an open question as in (16), with w\u00C3\u00AEhkac \u00E2\u0080\u0098ever\u00E2\u0080\u0099, the bare form is used for both the question and the answer. (16) A: wihk\u00C3\u00A2c kitayim\u00C3\u00AEht\u00C3\u00A2n \u00C3\u00B4ma masinahikan2? wihk\u00C3\u00A2c ki(t)- ayim\u00C3\u00AEht\u00C3\u00A2 -n \u00C3\u00B4ma masinahikan ever 2- read.VAI -SAP DEM.INAN book \u00E2\u0080\u0098Have you ever read this book?\u00E2\u0080\u0099 B: \u00C3\u00AAh\u00C3\u00A2, nitayim\u00C3\u00AEht\u00C3\u00A2n m\u00C3\u00AEhc\u00C3\u00AAtw\u00C3\u00A2w \u00C3\u00AAh\u00C3\u00A2 ni(t)- ayim\u00C3\u00AEht\u00C3\u00A2 -n m\u00C3\u00AEhcetw\u00C3\u00A2w yes 1- read.VAI -SAP many.times \u00E2\u0080\u0098Yes, I have read it many times.\u00E2\u0080\u0099 Similarly, when the extra-linguistic context is more carefully specified, k\u00C3\u00AE- is completely bad for conveying a present perfect meaning. In (17), a young child John is gravely ill, and other people are outside talking about experiences John has not yet had (here holding puppies). The bare form leaves open the possibility that he still could have this experience; the k\u00C3\u00AE-marked form does not. 2 This was volunteered as a Y/N question. It lacks the Y/N interrogative c\u00C3\u00AE, which was quite standard for this particular consultant. The question form can be independently identified by a combination of the negative polarity element w\u00C3\u00AEhk\u00C3\u00A2c \u00E2\u0080\u0098ever\u00E2\u0080\u0099, and a rising pitch on the final syllable of masinahikan \u00E2\u0080\u0098book\u00E2\u0080\u0099, which can be used to implicate a phonologically null syllable (cf. M\u00C3\u00BChlbauer 2006). 327 (17) context: John is very gravely ill in the hospital a. m\u00C3\u00B4hk\u00C3\u00A2c John \u00C3\u00AA-t\u00C3\u00A2hkon\u00C3\u00A2t acimosisa BARE CLAUSE m\u00C3\u00B4ya wihkac J \u00C3\u00AA- t\u00C3\u00A2hkon -\u00C3\u00A2 -t atim -sis -a NEG ever J C1-carry.VTA-DIR-3 dog -DIM-OBV \u00E2\u0080\u0098John never carries puppies/small dogs.\u00E2\u0080\u0099 \u00E2\u0080\u0098John has never held a small dog.\u00E2\u0080\u0099 b. m\u00C3\u00B4hk\u00C3\u00A2c John \u00C3\u00AA-k\u00C3\u00AE-t\u00C3\u00A2hkon\u00C3\u00A2t acimosisa CLAUSE W/ k\u00C3\u00AE- m\u00C3\u00B4ya wihkac J \u00C3\u00AA- k\u00C3\u00AE- t\u00C3\u00A2hkon -\u00C3\u00A2 -t atim -sis -a NEG ever J C1-PREV-carry.VTA-DIR-3 dog -DIM -OBV \u00E2\u0080\u0098John had never carried a puppy.\u00E2\u0080\u0099 comment: the person thinks he\u00E2\u0080\u0099s gonna die Turning next to the \u00E2\u0080\u0098perfect of recent events\u00E2\u0080\u0099, we see that k\u00C3\u00AE- is again not appropriate, as illustrated by (18). (18) context: a child\u00E2\u0080\u0099s father has just broken his leg; child runs in to tell the news/get help a. nip\u00C3\u00A2pa w\u00C3\u00AEsakisin, p\u00C3\u00AEkonam osk\u00C3\u00A2t, p\u00C3\u00AA-w\u00C3\u00AEcihin\u00C3\u00A2n BARE CLAUSE ni- p\u00C3\u00A2pa w\u00C3\u00AEsakisin -w p\u00C3\u00AEkonam -w o- sk\u00C3\u00A2t p\u00C3\u00AA- w\u00C3\u00AEcihi -n\u00C3\u00A2n 1- papa fall.VAI -3 break.VTI-3 3- leg come-help.VTA-1.PL \u00E2\u0080\u0098My dad got hurt, he broke his leg, come and help us!\u00E2\u0080\u0099 b. # np\u00C3\u00A2pa k\u00C3\u00AE-wisaksin, k\u00C3\u00AE-p\u00C3\u00AEkonam osk\u00C3\u00A2t, p\u00C3\u00AA-w\u00C3\u00AEcihin\u00C3\u00A2n CLAUSE W/ k\u00C3\u00AE- ni- p\u00C3\u00A2pa k\u00C3\u00AE- w\u00C3\u00AEsakisin -w k\u00C3\u00AE- p\u00C3\u00AEkonam -w o- sk\u00C3\u00A2t p\u00C3\u00AA- w\u00C3\u00AEcihi -n\u00C3\u00A2n 1- papa PREV-fall.VAI -3 PREV-break.VTI-3 3- leg come-help.VTA-1.PL \u00E2\u0080\u0098My dad got hurt, he broke his leg, come and help us!\u00E2\u0080\u0099 Finally, k\u00C3\u00AE- cannot be used for events that still hold, often termed \u00E2\u0080\u0098persistent perfect\u00E2\u0080\u0099 contexts. This is shown in by the example in (19). 328 (19) context: talking about house I\u00E2\u0080\u0099m living in right now a. niw\u00C3\u00AEkin \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwaskiy BARE CLAUSE ni-w\u00C3\u00AEki -n \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwo askiy 1-live.AI-SAP here DEM house four year \u00E2\u0080\u0098I\u00E2\u0080\u0099ve lived in this house for four year.\u00E2\u0080\u0099 b. # nik\u00C3\u00AE-w\u00C3\u00AEkin \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwaskiy CLAUSE W/ k\u00C3\u00AE- ni-k\u00C3\u00AE- w\u00C3\u00AEki -n \u00C3\u00B4ta \u00C3\u00B4ma w\u00C3\u00A2skahikan n\u00C3\u00AAwo askiy 1-prev-live.AI-SAP here DEM house four year \u00E2\u0080\u0098I lived at this house for four years.\u00E2\u0080\u0099 comment: can\u00E2\u0080\u0099t use this to talk about the house you\u00E2\u0080\u0099re still living in; you\u00E2\u0080\u0099d use this if you were showing someone the house you used to live in Since k\u00C3\u00AE- fails to be felicitous in any of the \u00E2\u0080\u0098perfect\u00E2\u0080\u0099 contexts, I conclude that k\u00C3\u00AE- is not a perfect. 7.3.4 k\u00C3\u00AE- is not a perfective Perfectivity is an aspectual distinction that focuses on the endpoint of an event or treats an event as a whole (Comrie 1976; Bybee 1985; etc.), and is in opposition to imperfective aspect, which look at the internal structure of the event. Since they are sensitive to endpoints, they characteristically differentiate between aspectual classes of predicates, which are defined in large part dependent on the presence or absence of such endpoints. Thus, if k\u00C3\u00AE- were a perfective, we would expect in some form the following properties. Again, k\u00C3\u00AE- behaves exactly contrary to expectations. Property k\u00C3\u00AE- PERFECTIVE Sensitivity to verb classes \u00E2\u009C\u0096 \u00E2\u009C\u0094 co-occur with imperfective? \u00E2\u009C\u0094 \u00E2\u009C\u0096 co-occur with inceptive \u00E2\u009C\u0094 \u00E2\u009C\u0096 Table 7.2. Patterning of k\u00C3\u00AE- relative to perfective aspect First, k\u00C3\u00AE- attaches to verbs of all aspectual classes. More importantly, perhaps, it does not act sensitive to the aspectual class at all \u00E2\u0080\u0093 e.g., it is not interpreted differently on different predicates. This is shown for permanent statives (20a), temporary statives (20b), activities (20c), and accomplishments (20d) below. 329 (20) a. Jeff \u00C3\u00AA-k\u00C3\u00AE-kinosit, \u00C3\u00AAkwa \u00C3\u00AA-k\u00C3\u00AE-tak\u00C3\u00A2hk\u00C3\u00A2p\u00C3\u00AAwit. J \u00C3\u00AA- k\u00C3\u00AE- kinosi -t \u00C3\u00AAkwa \u00C3\u00AA- k\u00C3\u00AE- tak\u00C3\u00A2hk\u00C3\u00A2p\u00C3\u00AAwi -t J C1-PREV-tall.VAI-3 and C1-PREV-beautiful.man.VAI-3 \u00E2\u0080\u0098Jeff was/had been tall, and he was/had been a good-looking man.\u00E2\u0080\u0099 b. k\u00C3\u00AE-cihk\u00C3\u00AAyihtam Betilo k\u00C3\u00AE- cihk\u00C3\u00AAyihtam -3 B PREV-happy.VTI -3 B \u00E2\u0080\u0098Betilo was happy/Betilo had been happy.\u00E2\u0080\u0099 c. k\u00C3\u00AE-p\u00C3\u00A2hpiw nip\u00C3\u00A2pa k\u00C3\u00AE- p\u00C3\u00A2hpi -w ni- p\u00C3\u00A2pa PREV-laugh.VAI-3 1- dad \u00E2\u0080\u0098My dad laughed/My dad had laughed.\u00E2\u0080\u0099 d. nik\u00C3\u00AE-kisip\u00C3\u00AAkin\u00C3\u00AAn \u00C3\u00B4ma w\u00C3\u00AEy\u00C3\u00A2kan ni- k\u00C3\u00AE- kisip\u00C3\u00AAkinam -n \u00C3\u00B4ma w\u00C3\u00AEy\u00C3\u00A2kan 1- PREV-wash.VTI -SAP DEM.INAN dish \u00E2\u0080\u0098I washed this dish/I had washed this dish.\u00E2\u0080\u0099 Second, k\u00C3\u00AE- is compatible with the preverb m\u00C3\u00AAkw\u00C3\u00A2- \u00E2\u0080\u0098midst\u00E2\u0080\u0099. In (21a) we see m\u00C3\u00AAkw\u00C3\u00A2-3 being used to denote imperfectivity in the temporal modifying clause. In (21b) we see k\u00C3\u00AE- co- occurring with m\u00C3\u00AAkw\u00C3\u00A2-; notice that k\u00C3\u00AE- does not change the imperfectivity of the clause it occurs in. (21) a. Jen k\u00C3\u00A2-m\u00C3\u00AAkw\u00C3\u00A2c-atosk\u00C3\u00AAt \u00C3\u00A2hkosiwipayiw. J k\u00C3\u00A2- m\u00C3\u00AAkw\u00C3\u00A2c- atosk\u00C3\u00AA -t \u00C3\u00A2hkosiwipayi -w J C2-MIDST- work.VAI-3 get.sick.VAI -3 \u00E2\u0080\u0098While Jen was working, she got ill / sick\u00E2\u0080\u0099 b. Jen k\u00C3\u00A2-k\u00C3\u00AE-m\u00C3\u00AAkw\u00C3\u00A2c-m\u00C3\u00A2yi-t\u00C3\u00B4tahk, J k\u00C3\u00A2- k\u00C3\u00AE- m\u00C3\u00AAkw\u00C3\u00A2c- m\u00C3\u00A2yi- t\u00C3\u00B4tam -k J C2-PREV-MIDST- bad- do.VTI-3 moy \u00C3\u00AA-kisk\u00C3\u00AAyihtahk ayisiyiniwak \u00C3\u00AA-miyo-m\u00C3\u00A2miton\u00C3\u00AAyihtahk m\u00C3\u00B4ya \u00C3\u00AA- kisk\u00C3\u00AAyihtam -k ayisiyiniw-ak \u00C3\u00AA- miyo- m\u00C3\u00A2mitoni\u00C3\u00AAyihtam -k NEG C1-know.VTI -3 person -PL C1-good-think.of.VTI -3 \u00E2\u0080\u0098While Jen was doing bad things she didn\u00E2\u0080\u0099t know people were wishing her well.\u00E2\u0080\u0099 Given that a predicate cannot simultaneously be perfective and continuous, this is evidence that k\u00C3\u00AE- is not perfective. Another aspectual preverb that k\u00C3\u00AE- co-occurs with is ati- \u00E2\u0080\u0098start/become/in.process\u00E2\u0080\u0099. For 3 Or, as in the case of this speaker, the particle form m\u00C3\u00AAkw\u00C3\u00A2c. 330 example, in (22) the unmarked stative is interpreted as holding at speech time; the ati- marked stative is interpreted as coming into being at speech time. (22) a. Jane kinosiw J kinosi -w J tall.VAI-3 \u00E2\u0080\u0098Jane is tall.\u00E2\u0080\u0099 b. Jane \u00C3\u00A2t-kinosiw J ati- kinosi -w J INCEP-tall.VAI-3 \u00E2\u0080\u0098Jane is getting taller\u00E2\u0080\u0099 In (23), we see that k\u00C3\u00AE- can be combined with ati- without chainging the inchoative interpretation of the predicate. Rather the change seems to be in the temporal anchoring of the clause. This again provides evidence that k\u00C3\u00AE- is not a perfective marker. (23) a. Clare \u00C3\u00AAkwa Jeff ati-m\u00C3\u00A2towak C \u00C3\u00AAkwa J ati- m\u00C3\u00A2to -w -ak C and J INCEP-cry.VAI-3 -PL \u00E2\u0080\u0098Clare and Jeff are teary-eyed/just starting to cry/almost crying.\u00E2\u0080\u0099 b. Clare \u00C3\u00AAkwa Jeff \u00C3\u00AA-k\u00C3\u00AE-ati-m\u00C3\u00A2tocik C \u00C3\u00AAkwa J \u00C3\u00AA- k\u00C3\u00AE- ati- m\u00C3\u00A2to -t -k C and J C1-PREV-INCEP-cry.VAI-3-PL \u00E2\u0080\u0098Clare and Jeff were going to start crying.\u00E2\u0080\u0099 All other aspectual morphemes are also compatible with k\u00C3\u00AE-; there is no known morpheme that is incompatible with k\u00C3\u00AE-. 7.4 Deconstructing modality: Clause-typing, irreality, and k\u00C3\u00AE- More evidence for the a-veridicality analysis of simple CONJ clauses \u00E2\u0080\u0093 for which, however, I do not have a full analysis \u00E2\u0080\u0093 comes from their role in the construction of modalized propositions. I take modality here to describe what is possible or necessary (or somewhere in between) given some state-of-affairs (or situation, or context) (Kratzer 1981, 1991, von Fintel 2006). The purpose of this section is three-fold. First, it provides an initial descriptive characterization of the expression of modality in Plains Cree. Second, it makes the point that 331 modality is constructed via the interaction of several components of the grammar (i.e., there are no dedicated modal morphemes in Plains Cree), and that one of these components is clause- typing. Third, it shows that, among the anaphoric CONJUNCT clause-types, the a-veridical simple CONJUNCT is used for modality, Since modality is about what is possible or necessary, rather than about what exists, the fact that simple CONJUNCT clauses are used in modal contexts supports the generalization that the simple CONJUNCT introduces a-veridical propositions. A particular modal interpretation in Cree depends on (i) clause-typing; (ii) the preverbal markers ka- and k\u00C3\u00AE-; and (iii) negation. Putting these together, we get the modal interpretations in table 7.3. Modal Base Quantificational Force Circumstantial Deontic \u00E2\u0088\u0083 (existential) ABILITY PERMISSION \u00E2\u0088\u0080 (universal) FUTURE UNSTOPPABLE INTERNAL OBLIGATION EXTERNAL NECESSITY Table 7.3. Modal interpretations These interpretations map onto specific sets of forms in Plains Cree, as shown in table 7.4. Modal Base Quantificational Force Circumstantial Deontic \u00E2\u0088\u0083 (existential) ABILITY ka- + k\u00C3\u00AE- + IND PERMISSION ka- + k\u00C3\u00AE- + A-VERIDICAL \u00E2\u0088\u0080 (universal) FUTURE ka- + IND UNSTOPPABLE piko + ka- + A-VERIDICAL INTERNAL OBLIGATION ka- + A-VERIDICAL EXTERNAL NECESSITY piko + ka + A-VERIDICAL Table 7.4. Mapping of Plains Cree forms to modal interpretations In the following sections I consider the role of each formal contrast in turn. 7.4.1 The role of clause-typing: Circumstantial vs. deontic modality4 Recall that the simple CONJUNCT either has the prefix ka- or the suffix -i (Plains Cree\u00E2\u0080\u0099s subjunctive). The ka- prefix also occurs in the INDEPENDENT order, but is ungrammatical with 4 Thanks to Lisa Matthewson for discussion of this point. 332 other CONJUNCT clause-types.5 The presence of irrealis ka- has different effects on INDEPENDENT vs. CONJUNCT clauses. Modal Base Clause-type circumstantial INDEPENDENT deontic simple CONJUNCT Table 7.5. Clause-typing affects modal base In the INDEPENDENT order, the result is a future expression, as in (24); this is usually considered a type of universal circumstantial modality in that it conveys what must necessarily happen given a particular set of circumstances (Kratzer 1991, Copley 2002, Davis et. al to appear). (24) ka- + INDEPENDENT = \u00E2\u0080\u0098future\u00E2\u0080\u0099 (universal circumstantial) a. ..., \u00E2\u0080\u009Cm\u00C3\u00A2ka pikw \u00C3\u00AEsp\u00C3\u00AE ka-takosin,\u00E2\u0080\u009D ... m\u00C3\u00A2ka piko isp\u00C3\u00AE ka- takosin -w but all TEMP IRR-arrive.VAI-3 \u00E2\u0080\u0098..., \u00E2\u0080\u009Cbut he will be back any time now,\u00E2\u0080\u009D ... \u00E2\u0080\u0099 (AA 4.11) b. \u00E2\u0080\u009Cnika-m\u00C3\u00A2kohikwak kisk\u00C3\u00AAyihtahkw\u00C3\u00A2wi,\u00E2\u0080\u009D ... ni-ka- m\u00C3\u00A2koh -ikw-ak kisk\u00C3\u00AAyihtam-k-w\u00C3\u00A2w-i 1- IRR-trouble.VTA-INV-PL know.VTI -0-PL -SUBJ \u00E2\u0080\u0098 \u00E2\u0080\u009Cthey will give me trouble if they find out,\u00E2\u0080\u009D ...\u00E2\u0080\u0099 (AA 4.8) By contrast, in simple CONJUNCT clauses, the result is an expression of necessary obligation. I take this to be a kind of universal deontic modal construction (where deontic modality expresses what is possible or necessary given some set of laws, rules, or other context). In elicitation, a simple CONJUNCT clause is dispreferred without an overt higher predicate (cf. Ahenakew 1987); this is consistent with what we have seen about CONJUNCT clauses being embedded. In running speech, where more long-distance dependencies are found, ka- prefixed CONJUNCT forms are found, not frequently, but regularly. 5 For some speakers, the prefix ka- alternates with (ki)ta-. The latter form seems to be restricted to third-persons and the ka-/kita- alternation is in fact often analyzed as a person-split (Wolfart 1973, Wolvengrey 2006). However, the distribution seems to be more complicated than that: (i) both ka- and (ki)ta- occur with third-persons (cf. Ahenakew 2000); (ii) speakers report that the ta- form has a stronger sense of obligation. I leave this for further research. 333 (25) ka-p-~ ka-p\u00C3\u00A2hpiyahk ka-p\u00C3\u00AEkiskw\u00C3\u00A2t\u00C3\u00A2yahkik, ka- p\u00C3\u00A2hpi -yahk ka- p\u00C3\u00AEkiskw\u00C3\u00A2t\u00C3\u00A2 -yahk -k IRR-laugh.VAI-21PL IRR-speak.VTA -21PL -3PL \u00E2\u0080\u0098We should laugh and speak to these young people, ka-w\u00C3\u00A2pahtih\u00C3\u00A2yahkik \u00C3\u00AA-kitim\u00C3\u00A2k\u00C3\u00AAyim\u00C3\u00A2yahkik. ka- w\u00C3\u00A2pahtih\u00C3\u00A2-yahk -k \u00C3\u00AA- kitim\u00C3\u00A2k\u00C3\u00AAyim\u00C3\u00A2 -yahk -k IRR-show.vta -21PL-3PL c1-care.VTA -21PL -3PL we should show them that we care for them.\u00E2\u0080\u0099 (EM 35) Thus, both INDEPENDENT clauses and simple CONJUNCT clauses express universal quantification, but the modal force is different. The preverb k\u00C3\u00AE- \u00E2\u0080\u0098PREVIOUS\u00E2\u0080\u0099, which otherwise has a temporal shifting function, can be added to a modal clause to change the quantificational force. Table 7.6 summarizes; k\u00C3\u00AE- is used in existential quantification contexts; universal quantification contexts have no morphological exponent. Quantificational Force Morphology \u00E2\u0088\u0083 (existential) k\u00C3\u00AE- \u00E2\u0088\u0080 (universal) \u00EF\u0081\u00B8 Table 7.6. k\u00C3\u00AE- codes existential force in modal contexts In (26) there is still a contrast between INDEPENDENT and SIMPLE CONJUNCT, and it can still be characterized as a difference between circumstantial and deontic modality. With the addition of k\u00C3\u00AE-, however, the circumstantial modality is existential (ability) rather than universal (future), and the deontic modality is more akin to should than have to (cf. below for relevant contexts). Notice that other forms of the CONJUNCT are ungrammatical. 334 (26) a. ka- + k\u00C3\u00AE- + INDEPENDENT = existential circumstantial ka-k\u00C3\u00AE-t\u00C3\u00A2hkon\u00C3\u00AAw acim\u00C3\u00B4sisa INDEPENDENT ka- k\u00C3\u00AE- t\u00C3\u00A2hkon -\u00C3\u00AA -w atimw -isis -a IRR-PREV-carry.VTA-DIR-3 dog -DIM-OBV \u00E2\u0080\u0098S/he is able to carry the puppies.\u00E2\u0080\u0099 b. ka- + k\u00C3\u00AE- + CONJUNCT = existential deontic ka-k\u00C3\u00AE-t\u00C3\u00A2hkon\u00C3\u00A2t acim\u00C3\u00B4sisa SIMPLE CONJUNCT ka- k\u00C3\u00AE- t\u00C3\u00A2hkon -\u00C3\u00A2 -t atimw -isis -a IRR-PREV-carry.VTA-DIR-3 dog -DIM-OBV \u00E2\u0080\u0098S/he is supposed to carry the puppies.\u00E2\u0080\u0099 c. * \u00C3\u00AA-ka-k\u00C3\u00AE-t\u00C3\u00A2hkon\u00C3\u00A2t acim\u00C3\u00B4sisa CHANGED CONJUNCT \u00C3\u00AA- ka- k\u00C3\u00AE- t\u00C3\u00A2hkon -\u00C3\u00A2 -t atimw -isis -a C1-IRR-PREV-carry.VTA-DIR-3 dog -DIM-OBV --- (27) context: asking permission to go to party (existential deontic) a. # nika-k\u00C3\u00AE-itoht\u00C3\u00A2n c\u00C3\u00AE ni-ka- k\u00C3\u00AE- itoht\u00C3\u00AA -n c\u00C3\u00AE 1- IRR-PREV-go.VAI-SAP Q \u00E2\u0080\u0098Can I go?\u00E2\u0080\u0099 b. nika-k\u00C3\u00AE-itoht\u00C3\u00AAy\u00C3\u00A2n c\u00C3\u00AE ni- ka- k\u00C3\u00AE- itoht\u00C3\u00AA -y\u00C3\u00A2n c\u00C3\u00AE 1- IRR- PREV-go.VAI-1 Q \u00E2\u0080\u0098Can I go?\u00E2\u0080\u0099 7.4.2 k\u00C3\u00AE- has existential modal force under negation The interpretation of k\u00C3\u00AE- when it co-occurs with irrealis ka- is not the only place where k\u00C3\u00AE- interacts with modality. It also has modal force under negation, in both INDEPENDENT and CONJUNCT clauses.6 These negated modal clauses always have a circumstantial interpretation, not a deontic interpretation. 6 The scope of negation over k\u00C3\u00AE- is clause-bound. Consider the minimal pair in (i): in (ia) negation is in the higher clause, but the k\u00C3\u00AE- in the lower mediated argument clause still acts as an anaphoric temporal shifting device. In (ib), the negation is in the same clause and men are unable to leave. 335 (28) a. NEG + k\u00C3\u00AE- + INDEPENDENT = negative existential circumstantial m\u00C3\u00B4y nik\u00C3\u00AE-kiskisin m\u00C3\u00B4y ni- k\u00C3\u00AE- kiskisi -n NEG 1- PREV-remember.VAI-SAP \u00E2\u0080\u0098I can\u00E2\u0080\u0099t remember.\u00E2\u0080\u0099 b. NEG + k\u00C3\u00AE- + \u00C3\u008A- CONJUNCT = negative existential circumstantial m\u00C3\u00B4y \u00C3\u00AA-k\u00C3\u00AE-t\u00C3\u00A2hkon\u00C3\u00A2t acim\u00C3\u00B4sisa m\u00C3\u00B4y \u00C3\u00AA- k\u00C3\u00AE- t\u00C3\u00A2hkon -\u00C3\u00A2 -t atimw -isis -a NEG C1-PREV-carry.VTA-DIR-3 dog -DIM -OBV \u00E2\u0080\u0098S/he can\u00E2\u0080\u0099t carry the puppies.\u00E2\u0080\u0099 (too small, not strong enough) 7.4.3 Negation widens possible interpretations The addition of negation to a simple CONJUNCT clause, which already has a deontic modal force, introduces the possibility of a circumstantial modal force. This construction is ambiguous, as illustrated by the pair of data points in (29a-b). (29) NEG + ka- + k\u00C3\u00AE- + CONJUNCT = negative existential circumstantial OR deontic modality a. ..., \u00E2\u0080\u009Ct\u00C3\u00A2pisk\u00C3\u00B4t, t\u00C3\u00A2pisk\u00C3\u00B4c m\u00C3\u00B4y ka-k\u00C3\u00AE-miyw-\u00C3\u00A2y\u00C3\u00A2cik,\u00E2\u0080\u009D... CIRCUMST. t\u00C3\u00A2pisk\u00C3\u00B4t t\u00C3\u00A2pisk\u00C3\u00B4t-i m\u00C3\u00B4y ka- k\u00C3\u00AE- miyw- \u00C3\u00A2y\u00C3\u00A2 -t -k seem seem NEG IRR-PREV-good- be.VAI-3-PL \u00E2\u0080\u0098 ..., \u00E2\u0080\u009Cit seems, it seems as though they cannot recover,\u00E2\u0080\u009D ... \u00E2\u0080\u0099(AA 4.10) b. ..., \u00C3\u00AAwakw \u00C3\u00A2nim \u00C3\u00A2yisiyiniwak, nam\u00C3\u00B4y ka-k\u00C3\u00AE-wanikiskisicik; DEONTIC \u00C3\u00AAwakw anima ayisiyiniw -ak nam\u00C3\u00B4y ka- k\u00C3\u00AE- wanikiskisi -t -k TOPIC DEM.INAN person -PL NEG IRR-PREV-forget.VAI -3-PL \u00E2\u0080\u0098..., that is something people should not forget;\u00E2\u0080\u0099 (EM 42) (i) a. [CP neg \u00E2\u0080\u00A6 [CP \u00E2\u0080\u00A6 k\u00C3\u00AE- \u00E2\u0080\u00A6 ] ] n\u00C3\u00A2p\u00C3\u00AAwak m\u00C3\u00B4y kisk\u00C3\u00AAyihtamwak \u00C3\u00AA-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAcik n\u00C3\u00A2p\u00C3\u00AAw -ak m\u00C3\u00B4y kisk\u00C3\u00AAyihtam -w -ak \u00C3\u00AA- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -t -k man -PL NEG know.VTI -3 -PL C1-PREV-leave.VAI -3 -PL \u00E2\u0080\u0098the men didn\u00E2\u0080\u0099t know they had left.\u00E2\u0080\u0099 comment: laughing \u00E2\u0080\u0093 they didn\u00E2\u0080\u0099t know that they left \u00E2\u0080\u0093 like they\u00E2\u0080\u0099re drunk b. [CP \u00E2\u0080\u00A6 [CP neg \u00E2\u0080\u00A6 k\u00C3\u00AE- \u00E2\u0080\u00A6 ] ] n\u00C3\u00A2p\u00C3\u00AAwak kisk\u00C3\u00AAyihtamwak m\u00C3\u00B4y \u00C3\u00AA-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAcik n\u00C3\u00A2p\u00C3\u00AAw -ak kisk\u00C3\u00AAyihtam -w -ak m\u00C3\u00B4y \u00C3\u00AA- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -t -k man -PL know.VTI -3 -PL NEG C1-PREV-leave.VAI -3 -PL \u00E2\u0080\u0098The men knew they couldn\u00E2\u0080\u0099t go.\u00E2\u0080\u0099 comment: they were in jail, where it is impossible to get out 336 Conversely, the addition of negation to an INDEPENDENT clause, which already has circumstantial modal force, introduces the possibility of a deontic force. (30) NEG + ka- + k\u00C3\u00AE- + CONJUNCT = negative existential circumstantial/deontic modality a. ..., m\u00C3\u00B4y pikw \u00C3\u00AEsp\u00C3\u00AE ka-k\u00C3\u00AE-k\u00C3\u00A2hcitin\u00C3\u00A2wak, ... CIRCUMST. m\u00C3\u00B4y pikw isp\u00C3\u00AE ka- k\u00C3\u00AE- k\u00C3\u00A2hcitin -\u00C3\u00A2 -w -ak NEG Q TEMP IRR-PREV-take.hold.VTA-USC-3 -PL \u00E2\u0080\u0098..., you cannot get ahold of that kind just any time, ...\u00E2\u0080\u0099 (AA 4.6) b. ... (m\u00C3\u00B4y nika-k\u00C3\u00AE-\u00C3\u00A2kay\u00C3\u00A2s\u00C3\u00AEmon aya), ... DEONTIC m\u00C3\u00B4y ni-ka- k\u00C3\u00AE- \u00C3\u00A2kay\u00C3\u00A2simo -n aya NEG 1- IRR-PREV-English.VAI-SAP CONN \u00E2\u0080\u0098... (I must not say it in English), ...\u00E2\u0080\u0099 (EM 32) (31) moy ka-ki-m\u00C3\u00A2tow Sarah m\u00C3\u00B4y ka- k\u00C3\u00AE- m\u00C3\u00A2to -w S NEG IRR-PREV-cry.VAI -3 S \u00E2\u0080\u0098Sarah can\u00E2\u0080\u0099t cry.\u00E2\u0080\u0099 context: Sarah is unable to cry. context: Sarah is not allowed to cry. In both of these cases, the quantificational force is not changing, but the modal base is. 7.4.4 Embedding neutralizes modal distinctions Recall that INDEPENDENT clauses, which we have seen express circumstantial modality, cannot occur in embedded contexts. In embedded contexts, the circumstantial/deontic distinction in CONJUNCT clauses is neutralized. For example, (32) provides examples of the CONJUNCT being interpreted as an ability modal. In both of these cases, we have both the irrealis ka- and the preverb k\u00C3\u00AE-; consistent with what we saw earlier, these are modal constructions involving existential quantification. 337 (32) a. [CP \u00C3\u00AA-it\u00C3\u00AAyihtahkik [CP ka-k\u00C3\u00AE-w\u00C3\u00AEcihakik ] ] EXISTENTIAL CIRCUM. \u00C3\u00AAkwa, \u00C3\u00AA-it\u00C3\u00AAyihtahkik \u00C3\u00AAtokw\u00C3\u00AA, ka-k\u00C3\u00AE-w\u00C3\u00AEcihakik ohc \u00C3\u00A2ya, ... \u00C3\u00AAkwa \u00C3\u00AA- it\u00C3\u00AAyihtam -k -k \u00C3\u00AAtokw\u00C3\u00AA ka- k\u00C3\u00AE- w\u00C3\u00AEcih -ak -k ohci aya and C1-think.VTI -3 -PL DUB IRR-PREV-help.VTA-1>3-pl ORIG CONN \u00E2\u0080\u0098And perhaps they think I can help them ...\u00E2\u0080\u0099 (EM 7) b. [CP kisk\u00C3\u00AAyihtamwak [CP ka-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAcik ] ] EXISTENTIAL CIRCUM. n\u00C3\u00A2p\u00C3\u00AAwak m\u00C3\u00B4y kisk\u00C3\u00AAyihtamwak ka-k\u00C3\u00AE-sipw\u00C3\u00AAht\u00C3\u00AAcik n\u00C3\u00A2p\u00C3\u00AAw -ak m\u00C3\u00B4y kisk\u00C3\u00AAyihtam -w -ak ka- k\u00C3\u00AE- sipw\u00C3\u00AAht\u00C3\u00AA -t -k man -PL NEG know.VTI -3 -PL IRR-PREV-leave.VAI -3-PL \u00E2\u0080\u0098The men didn\u00E2\u0080\u0099t know they could leave.\u00E2\u0080\u0099 context: men think they are prisoners, but the door is open The particle piko as a predicative element also introduces a modal clause whose type of modality is neutralized. I take piko to be a universal quantifier7: in non-predicative positions it has a meaning of \u00E2\u0080\u0098all\u00E2\u0080\u0099 (preceding the element it quantifies over) or \u00E2\u0080\u0098only\u00E2\u0080\u0099 (in second position). All of its uses are exemplified in the following example taken from Minde (1998). (33) a. ..., pik \u00C3\u00B4ma ka-m\u00C3\u00A2maw\u00C3\u00B4hkam\u00C3\u00A2toyahk, ..., we must work together kwayask ka-kakw\u00C3\u00AA-isi-pim\u00C3\u00A2tisiyahk, to try to lead a good life now\u00C3\u00A2hc ka-kakw\u00C3\u00AA-isi-pim\u00C3\u00A2tisiyahk; to try to lead a better life b. \u00C3\u00AAkosi piko k-\u00C3\u00AAs-\u00C3\u00A2ya-miyaw\u00C3\u00A2t\u00C3\u00AAn\u00C3\u00A2naw, that is the only way we will be happy c. pikw \u00C3\u00A2wiyak nawasw\u00C3\u00A2tam miyaw\u00C3\u00A2tamowin; and everyone chases after happiness; (EM37) On independent grounds, (the modal form of) piko seems to be a predicator (i.e., something that introduces a dependent clause) (cf. Wolfart 1973, 1998). Evidence for this analysis comes from the strict requirement that it be in initial position, and the fact that it can never introduce an IND clause. When we look at the modal interpretations, they pattern with other locally-embedded clauses: piko combined with a simple CONJUNCT clause appears to be compatible with either circumstantial or deontic force. 7 Thanks to Lisa Matthewson for pointing out the connection between these different uses of piko and universal quantification. 338 (34) Universal deontic modal a. piko ka-kan\u00C3\u00A2cihcik\u00C3\u00AAy\u00C3\u00A2n, piko ka- kan\u00C3\u00A2cihtcik\u00C3\u00AA-y\u00C3\u00A2n necessary IRR-clean.VAI -1 ayis nis\u00C3\u00AEmsak, oskin\u00C3\u00AEkiwak, \u00C3\u00AA-w\u00C3\u00AE-p\u00C3\u00AA-k\u00C3\u00AEyok\u00C3\u00AAcik ayis ni-s\u00C3\u00AEmis -ak oskin\u00C3\u00AEkiw-ak \u00C3\u00AA- w\u00C3\u00AE- p\u00C3\u00AA- k\u00C3\u00AEyok\u00C3\u00AA -t -k for 1- sibling-PL youth -PL C1-int-DIR-visit.VAI-3-PL \u00E2\u0080\u0098I have to clean the house, because my siblings, young men, are coming to visit.\u00E2\u0080\u0099 b. context: I am scheduled to work today; if I don\u00E2\u0080\u0099t go in, they fire me piko ka-nitawi-atosk\u00C3\u00AAy\u00C3\u00A2n anohc piko ka-nitawi- atosk\u00C3\u00AA -y\u00C3\u00A2n anohc be.necessary IRR-go- work.VAI-1 today \u00E2\u0080\u0098I have to go to work.\u00E2\u0080\u0099 (35) Universal circumstantial a. context: on the phone, feel cough coming on, say this to excuse yourself piko ka-atohoy\u00C3\u00A2n piko ka- atoho -y\u00C3\u00A2n necessary IRR-cough.VAI-1 \u00E2\u0080\u0098I have to cough.\u00E2\u0080\u0099 b. context: falling asleep while talking to someone, keep falling forward, can\u00E2\u0080\u0099t stay awake any longer, about to pass out piko ka-nip\u00C3\u00A2y\u00C3\u00A2n piko ka- nip\u00C3\u00A2 -y\u00C3\u00A2n necessary IRR-sleep.VAI-1 \u00E2\u0080\u0098I have to sleep.\u00E2\u0080\u0099 This is unexpected if the ka- clause is not locally-embedded, but looks familiar when considered with other locally-embedded clauses. Thus, this data provides additional support for analyzing piko as a predicative element, and provides another instance of neutralization of modal force. 339 7.4.5 Summary One of the key aspects to modal interpretation in Plains Cree is the clause-typing system: in matrix clauses, the indexical INDEPENDENT order expresses circumstantial modality, and a- veridical CONJUNCT clauses express deontic modality. In the CONJUNCT order, the simple CONJ is used to express modality, which is consistent with my claim that the simple CONJUNCT is used to introduce a-veridical propositions. In order to have a full semantic account of modality in Plains Cree, there are several puzzles that remain unsolved. I here raise two such questions, leaving the answers for further research. First, what are the contributions of each piece of Plains Cree\u00E2\u0080\u0099s modal constructions to the overall meaning? For example, clause-typing distinguishes modal force, but in negated and embedded clauses, this distinction is neutralized. In terms of quantificational force, k\u00C3\u00AE- seems to specify existential quantification, but there is no morphological marking of universal quantification. In addition, the irrealis marker ka- is a key component of modal constructions, but crosses both the circumstantial/deontic modal distinction and the existential/universal quantificational distinction. Second, putting the distinction between modal base and quantificational force together, we expect the following four-way typology, all of which are attested in Plains Cree: Quantification Circumstantial Deontic Existential \u00E2\u009C\u0094 \u00E2\u009C\u0094 Universal \u00E2\u009C\u0094 \u00E2\u009C\u0094 Table 7.7. Modal base vs. quantification in Plains Cree In fact, Plains Cree exhibits more contrasts than this model would lead us to expect; in particular, we saw that Cree had two ways of constructing a universal circumstantial, and two ways of constructing an existential deontic. 340 Quantification Circumstantial Deontic Existential ABILITY PERMISSION OBLIGATION Universal FUTURE UNSTOPPABLE NECESSITY Table 7.8. Modal interpretations in Plains Cree Thus, in order to fully capture Cree\u00E2\u0080\u0099s system, we need a more fine-grained analysis of modality. 7.5 Variation in clause-typing across Algonquian Documentation of the microvariation between Algonquian languages at the syntactic and semantic level remains very coarse-grained at this point, but the relatively detailed analysis developed here for Plains Cree offers a good starting point for understanding much more about the clausal domain of this family. In this section, I want to particularly address in more detail some of the variation that was briefly mentioned in chapter two about the mapping of forms, such as the pronominal proclitics and initial change, to syntax and semantics across Algonquian. 7.5.1 Variation of the pronominal proclitics: Plains Cree vs. Blackfoot With respect to forms in the INDEPENDENT order, I have claimed that the pronominal proclitics ni- and ki- are in spec, CP. Working on Blackfoot, a somewhat distantly related but geographically adjacent Algonquian language, Ritter & Wiltschko (2005, 2007) have argued that the pronominal proclitics sit head the IP layer of the clause. Ignoring for the moment the difference between the spec and head position, there are several crucial ways that Plains Cree and Blackfoot forms differ in their distribution. Table 7.9 shows five diagnostics that could theoretically distinguish between an IP-level element and a CP-level element; we see that Plains Cree and Blackfoot differ on three of the five diagnostics. 341 Properties of INDEPENDENT proclitics Plains Cree Blackfoot Peripheral position? \u00E2\u009C\u0094 \u00E2\u009C\u0094 Sensitive to temporal value? \u00E2\u009C\u0096 \u00E2\u009C\u0096 Interacts with inverse? \u00E2\u009C\u0096 \u00E2\u009C\u0094 Sensitive to embedding? \u00E2\u009C\u0094 \u00E2\u009C\u0096 Complementary with clause-typing? \u00E2\u009C\u0094 \u00E2\u009C\u0096 Table 7.9. Diagnostics for determining the position of pronominal proclitics Both Plains Cree and Blackfoot pronominal proclitics occur on the far left-peripheral position of the verbal complex, and neither set of proclitics are sensitive to temporal value. However, on at least three counts they differ. First, Blackfoot pronominal proclitics have a particular interaction with the inverse marker of the theme system that Plains Cree (and, to my knowledge, all other Algonquian languages) lack. In (35), we see that the Blackfoot inverse marker is used when a third person acts on a first person (35a) or when a first person acts on a second person (35b). (36) a. nitsik\u00C3\u00A1komimmokinnaani nits- ik\u00C3\u00A1komimm-ok -innaan -i 1- love -INV-1PL -3PL \u00E2\u0080\u0098They love us(excl).\u00E2\u0080\u0099 (D\u00C3\u00A9chaine 1999, (70a)) b. kitsik\u00C3\u00A1komimmoki kits- ik\u00C3\u00A1komimm -oki 2- love -INV \u00E2\u0080\u0098I love you.\u00E2\u0080\u0099 (D\u00C3\u00A9chaine 1999 (72c), translation corrected) In (36) we see that the Plains Cree inverse marker is used when a third person acts on a first person, but that a different form, -iti, is used when a first person acts on a second person (36b). (37) a. nis\u00C3\u00A2kihik ni- s\u00C3\u00A2kih -ikw 1- love.VTA-INV \u00E2\u0080\u0098S/he loves me.\u00E2\u0080\u0099 b. kis\u00C3\u00A2kihitin ki- s\u00C3\u00A2kih -iti -n 2- love.VTA-1>2-SAP \u00E2\u0080\u0098I love you.\u00E2\u0080\u0099 342 Given that the theme-sign system is sensitive to case (D\u00C3\u00A9chaine & Reinholtz 1997, 2007), and more generally that the inverse-marker involves a raising operation (Bruening 2001), the data in (35-36) suggests that pronominal proclitics are lower down in Blackfoot than in Plains Cree. Second, as we saw in chapter 2, verbal stems carrying pronominal proclitics cannot be embedded clauses in Plains Cree. In Blackfoot, however, there is no such prohibition. Third, while the pronominal proclitics in Plains Cree are in complementary distribution with clause-typing elements (in Plains Cree, these are the left-edge elements \u00C3\u00AA- and k\u00C3\u00A2-), the pronominal proclitics in Blackfoot are not in complementary distribution with clause-typing elements. For example, the pronominal proclitics occur in matrix clauses, which do not have any other clause-typing (37a); in factive embedded clauses with the -hsi clause-typing suffix (37b), and in non-affirmative clauses with the -hpa clause-typing suffix (37c). (38) a. nit\u00C3\u00A1akahkayi nit- \u00C3\u00A1ak- ahkayi 1- will- go^home \u00E2\u0080\u0098I\u00E2\u0080\u0099m going home.\u00E2\u0080\u0099 (Frantz 1991:21) b. nits\u00C3\u00ADkohtaahs\u00C3\u00AD\u00E2\u0080\u0099taki kik\u00C3\u00A1\u00C3\u00B3\u00E2\u0080\u0099toohsi nit- ik- oht- yaahs -i\u00E2\u0080\u0099taki k- ik\u00C3\u00A1\u00C3\u00A1- o\u00E2\u0080\u0099too -hs -yi 1- very-source-good -feel.VAI 2- PERF- arrive.VAI-conj-conj \u00E2\u0080\u0098I\u00E2\u0080\u0099m glad that you have arrived.\u00E2\u0080\u0099 (Frantz 1991:112) c. kik\u00C3\u00A1ta\u00E2\u0080\u0099y\u00C3\u00A1aka\u00E2\u0080\u0099po\u00E2\u0080\u0099takihpa k- Ik\u00C3\u00A1ta\u00E2\u0080\u0099- y\u00C3\u00A1ak- a\u00E2\u0080\u0099p- o\u00E2\u0080\u0099taki -hpa 2-interrog-FUT- PREF-work -nonaffirm \u00E2\u0080\u0098Will you work?\u00E2\u0080\u0099 (Frantz 1991:133) Taken together, the facts above indicate that although Plains Cree and Blackfoot share the same forms, the way these forms are mapped into the syntax can be quite different from language to language. Thus, when working on the clause-typing system of one of these languages, it is important to understand not just what the forms are, but the larger distributional patterns. 343 7.5.2 Variation in initial change: Plains Cree vs. Blackfoot vs. Ojibwe Within anaphoric clauses, I claimed that there was both syntactic and semantic subclassification; in particular, we saw that there was a direct mapping between the semantics of the proposition and the form of a Plains Cree CONJUNCT order clause. If we look at this set of forms paradigmatically, we see that the relevant distinctions have to do with the presence/absence of a morpho-phonological process termed initial change in the Algonquianist literature, and whether this process targets the initial syllable of the word (very rare), is realized over some underlyingly contentless preverb (form: \u00C3\u00AA-), or realized on the preverb k\u00C3\u00AE- (form: k\u00C3\u00A2-). The process of initial change is attested across the Algonquian language. It is posited for Proto-Algonquian, cf. Costa 1996; for discussion of individual languages see the following (incomplete) list: Blackfoot (Taylor 1967, Proulx 2005); Cree/Montaignais/Naskapi (MacKenzie 1980); Menominee (Bloomfield 1962:98); Ojibwe (Bloomfield 1958, Valentine 2004); Passamaquoddy (Mitchell 1921); Plains Cree (Wolfart 1973; Rogers 1978), among many others. Despite the significant literature on the form of initial change, its syntax and function in different Algonquian languages remains poorly understood. The below table summarizes five properties of initial change which differ from language to language. Properties of initial change Plains Cree Blackfoot Ojibwe Peripheral position of verbal complex? \u00E2\u009C\u0094 \u00E2\u009C\u0096 (Frantz 1991:36) \u00E2\u009C\u0094 (Valentine 2004:156) Co-occurs with pronominal proclitics \u00E2\u009C\u0096 \u00E2\u009C\u0094 (Proulx 2005:17) \u00E2\u009C\u0096 (Valentine 2004:156) Cued to realis/irrealis \u00E2\u009C\u0094 \u00E2\u009C\u0094 (Proulx 2005:17) \u00E2\u009C\u0096 (Valentine 2004:166) Used in forms with nominal morphology? \u00E2\u009C\u0096 \u00E2\u009C\u0094 (Frantz 1991:36) \u00E2\u009C\u0094 (Valentine 2004: 510) Able to iterate? \u00E2\u009C\u0096 \u00E2\u009C\u0094 (Taylor 1967:125) \u00E2\u009C\u0096 (Valentine 2004) Table 7.10. Properties of initial change in Plains Cree, Blackfoot, and Ojibwe Significantly, I could find no literature that systematically describes the relation of initial change to clausal relations, although it is widely acknowledged that there is some relation. It is 344 hoped that the current observations and claims made about the syntax and semantics of initial change in Plains Cree will provide a broader backdrop as well as a set of more specific diagnostics and contexts for understanding initial change in other Algonquian languages. 345 REFERENCES Abusch, Dorit. 1998. Generalizing tense semantics for future contexts. Events and grammar, ed. by Susan Rothstein, pp. 13-33. Dordrecht: Kluwer Academic Publishers. Ahenakew, Alice. 2000. \u00C3\u00A2h-\u00C3\u00A2y\u00C3\u00AEtaw isi \u00C3\u00AA-k\u00C3\u00AE-kisk\u00C3\u00AAyihtahkik maskihkiy / They knew both sides of medicine: Cree tales of curing and cursing told by Alice Ahenakew, ed. by H.C. Wolfart and Freda Ahenakew, ed. and tr. by H.C. Wolfart and Freda Ahenakew. Publications of the Algonquian Text Society / Collection de la Soci\u00C3\u00A9t\u00C3\u00A9 d\u00E2\u0080\u0099\u00C3\u00A9dition des textes algonquiens, Winnipeg: University of Manitoba Press. Ahenakew, Freda. 1987. Cree Language Structures: A Cree Approach. Pemmican Publications, Inc. Ahenakew, Freda, and H.C. Wolfart. 1997. (eds.) kwayask \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-kiskinow\u00C3\u00A2pahtihicik / Their example showed me the way, told by Emma Minde. Edmonton:University of Alberta Press. Ahenakew, Freda, and H.C. Wolfart. 1998. (eds.) Ana k\u00C3\u00A2-pimw\u00C3\u00AAw\u00C3\u00AAhahk okak\u00C3\u00AAskihk\u00C3\u00AAmowina: The Counselling Speeches of Jim K\u00C3\u00A2-Nip\u00C3\u00AEt\u00C3\u00AAht\u00C3\u00AAw. Winnipeg: University of Manitoba Press. Aksu-Ko\u00C3\u00A7, Ayhan A., and Dan I. Slobin. 1986. A psychological account of the development and use of evidentials in Turkish. Evidentiality: The linguistic coding of epistemology, ed. by Wallace Chafe and Johanna Nichols, pp. 159-167. Norwood, New Jersey: Ablex. Aksu-Ko\u00C3\u00A7, Ayhan A., and Dan I. Slobin. 1988. The Acquisition of Aspect and Modality: The Case of Past Reference in Turkish. Cambridge: Cambridge University Press. Aikhenvald, Alexandra. 2004. Evidentiality. Oxford University Press. Alexiadou, Artemis, Paul Law, Andr\u00C3\u00A9 Meinunger, and Chris Wilder. 2000. Editors, The syntax of relative clauses. Berlin: John Benjamins. Aronoff, Mark. 1976. Word formation in generative grammar. Linguistic Inquiry monograph 1. Cambridge, MA: MIT Press. Austin, John L. 1950. Truth. Proc. of the Aristotelian Society. Supp. vol. xxiv. Austin, David. 1990. What's the Meaning of \"This\"? Ithaca, NY: Cornell University Press. Baker, Mark. 1985. The mirror principle and morphosyntactic explanation. Linguistic Inquiry 16, pp. 373-415. Banfield, Anne. 1982. Unspeakable Sentences: Narration and Representation in the Language of Fiction. Boston: Routledge & Kegan Paul. Bar-Hillel, Y. 1954. Indexical expressions. In Mind, Vol. 63, pp. 359-379. 346 Barczak, Leszek, Rose-Marie D\u00C3\u00A9chaine, and H.C. Wolfart. 2006. Open conditionals in Plains Cree. Paper given at WSCLA 11, Vancouver, British Columbia. Barwise, Jon. 1981. Scenes and other situations. In Journal of Philosophy, Vol. 78, pp. 369-397. Barwise, Jon and J. Etchemendy. 1987. The liar. Oxford: Oxford University Press. Barwise, Jon and Perry, John. 1983. Situations and Attitudes. Cambridge, MA: MIT Press. Battistella, Edwin L. 1990. Markedness: The evaluative superstructure of language. SUNY Press. Beaver, D. and C. Condoravdi. 2003. A uniform analysis of \u00E2\u0080\u0098before\u00E2\u0080\u0099 and \u00E2\u0080\u0098after\u00E2\u0080\u0099. Proceedings of SALT. Bennett, M. and Barbara H. Partee 1972. Toward the logic of tense and aspect in English. System Development Corporation, Santa Monica, Calif. Reprinted in Partee, B. Compositionality in Formal Semantics. 2004. Oxford: Blackwell Benveniste, \u00C3\u0089mil. 1950. A phrase nominale. Bulletin de la Soci\u00C3\u00A9te de Linguistique de Paris 46. pp. 19-36. Bhatt, Rajesh. 2002. The raising analysis of relative clauses: Evidence from adjectival modification. Natural Language Semantics 10; pp. 43-90. Bianchi, Valentina. 1999. Consequences of antisymmetry: Headed relative clauses. Berlin: Mouton de Gruyter. Bianchi, Valentina. 2000. The raising analysis of relative clauses: A reply to Borsley. Linguistic Inquiry 31; pp. 123-140. Bittner, Maria. 2008. Aspectual universals of temporal anaphora. In Theoretical and cross- linguistic approaches to the semantics of aspect, ed. by Susan Rothstein. Blain, Eleanor. 1997. WH-constructions in N\u00C3\u00AAhiyaw\u00C3\u00AAwin. Ph.D. thesis, University of British Columbia. Blain, Eleanor. 1999. Complementizer k\u00C3\u00A2- in N\u00C3\u00AAhiyaw\u00C3\u00AAwin (Plains Cree). Proceedings from the Workshop on the Structure and Constituency in Native American Languages, ed. by Leora Bar-el, Rose-Marie D\u00C3\u00A9chaine, and Charlotte Reinholtz, pp. 1-12. Cambridge MA: MIT Occasional Papers in Linguistics 17. Blain, Eleanor, Clare Cook, Rose-Marie D\u00C3\u00A9chaine, and Jeff M\u00C3\u00BChlbauer. 2006. Plains Cree evidentials: major and minor modes. Paper read at the annual meeting of the Society for the Study of the Indigenous Languages of the Americas, Albuquerque, NM. Blain, Eleanor and Rose-Marie D\u00C3\u00A9chaine. 2006. The evidential domain hypothesis. In Proceedings of WSCLA XI, ed. by Atsushi Fujimori & Maria Am\u00C3\u00A9lia Reis Silva, pp.12- 25. Vancouver: University of British Columbia Working Papers in Linguistics. 347 Blain, Eleanor and Rose-Marie. D\u00C3\u00A9chaine. 2007. Evidential types: Evidence from Cree dialects. IJAL 73: pp. 257-291. Bliss, Heather. 2007. Object Agreement in Blackfoot: Sentient and Non-Sentient Controllers. Proceedings of the 38th Algonquian Conference, ed. by H.C. Wolfart, pp. 11-28. Winnipeg: University of Manitoba Press. Bloomfield, Leonard. 1928. The Plains Cree Langugae. International Congress of Americanists, Proceedings 22, 2. pp. 427-431. Rome. Bloomfield, Leonard. 1930. Ed., Sacred stories of the Sweet Grass Cree: Bulletin / National Museum of Canada ; no. 60. Anthropological series ; no. 11. Ottawa: F. A. Acland. Bloomfield, Leonard. 1946. Alonquian. Linguistic structures of native America, ed. by Harry Hoijer et al.; Viking Fund Publications in Anthropology, 6, New York. pp. 85-129. Bloomfield, Leonard. 1958. Eastern Ojibwa: Grammatical sketch, texts, and word lists. ed. by Charles F. Hockett. Ann Arbor, Michigan. Bloomfield, Leonard. 1962. The Menomini language, ed. by Charles F. Hockett. New Haven: Yale University Press. Bohnemeyer, Jurgen and M.D. Swift. 2004. Event realization and default aspect. Linguistics and Philosophy 27; pp. 263-296. Branigan, Phil and Marguerite MacKenzie. 2002. Altruism, A\u00E2\u0080\u0099-movement, and object agreement in Innu-Aimun. Linguistic Inquiry 33(3). pp. 385-407. Branigan, Phil and Marguerite MacKenzie. 2007. The morpho-syntax of verbal orders in Innu- Aimun. Paper read at the 12th Workshop on the Structure and Constituency of the Languages of the Americas, University of Lethbridge. Brody, Michael. 1994. Phrase structure and dependence. Working papers in the theory of grammar 1(1). Budapest: Research Institute for Linguistics, Hungarian Academy of Science. Bruening, Benjamin. 2001. Syntax at the Edge: Cross-Clausal Phenomena and the Syntax of Passamaquoddy. Ph.D Dissertation. MIT. B\u00C3\u00BChler, K. 1934. Sprachtheorie. Jena: Fischer. Bybee, Joan. 1985. Morphology: A study of the relation between meaning and form. Amsterdam: Benjamins. Calabrese, Andrea. 1984. Multiple Questions and Focus in Italian. In Sentential Complementation, W. de Geest and Y. Putseys, eds., Calabrese, Andrea. 1987. Focus Structure in Berber: A comparative Analysis with Italian. In M. Guerssel K. Hale, eds., Studies in Berber Syntax, Lexicon Project Working Papers, Center for Cognitive Sciences, Cambridge: MIT Press. Pp. 103-120. 348 Carden, Guy. 1986. Blocked forwards coreference; unblocked forwards anaphora: Evidence for an abstract model of coreference. Chicago Linguistics Society 22; pp. 262-276. Carden, Guy, and Thomas Dieterich. 1981. Introspection, Observation, and Experiment: An Example where Experiment Pays Off, in Proceedings of the 1980 Biennial Meeting of the Philosophy of Science Association, Volume 2, ed. Peter Asquith and Ronald Giere, pp. 583-597. East Lansing, Michigan: Philosophy of Science Association. Chafe, Wallace. 1988. Linking intonation units in spoken English. Clause-combining in discourse and grammar, ed. by John Haiman and Sandra Thompson, pp. 1-27. Amsterdam: John Benjamins. Chang, Lisa. 1997. Wh-in situ phenomena in French. MA thesis, University of British Columbia Cheng, Lisa Lai-Shen. 1991. On the typology of wh-questions. PhD dissertation, Massachussetts Institute of Technology. Chomsky, Noam. 1977. On wh-movement. Formal syntax, ed. by P. Culicover, T. Wasow, A. Akmajian, pp. 71-132. New York: Academic Press. Chomsky, Noam. 1981. Lectures on Government and Binding. Chomsky, Noam. 1986. Barriers. Linguistic Inquiry monograph 13. Cambridge, MA: MIT Press. Chomsky, Noam. 1995. The minimalist program. Cambridge, MA: MIT Press. Chomsky, Noam and Howard Lasnik. 1993. Chung, Kyung-sook. 2005. Space in Tense: The Interaction of Tense, Aspect, Evidentiality, and Speech Act in Korean. Ph.D. dissertation, Simon Fraser University. Cinque, Guglielmo. 1999. Adverbs and functional heads: a cross-linguistic perspective. London: Oxford University Press. Cinque, Guglielmo. 2006. Restructuring and functional heads: The cartography of syntactic structures, vol. 4. Oxford University Press Collins, Chris. 1991. Why and How Come. In Lisa Cheng and Hamida Demirdache (eds.), More Papers on Wh- Movement; pp. 31\u00E2\u0080\u009345, MIT Working Papers in Linguistics 15. Comrie, Bernard. 1976. Aspect. An introduction to the study of verbal aspect and related problems. Cambridge etc.: Cambridge University Press. Comrie, Bernard. 1985. Tense. Cambridge: Cambridge University Press. Cook, Clare. 2003a. Menominee preverbs as functional heads. Proceedings of the Workshop on Structure and Constituency in Languages of the Americas VIII, UBC Working Papers in Linguistics, vol. 11; Yunhee Chung, Carrie Gillon, and Rachel Wojdak, eds., pp. 1-16; Vancouver. 349 Cook, Clare. 2003b. Relative roots in Plains Cree and Passamaquoddy. Paper given at the 35th Algonquian Conference; London, Ontario. Cook, Clare. 2005. Extraction marking vs. scope marking in N\u00C3\u00AAhiyaw\u00C3\u00AAwin and Yor\u00C3\u00B9b\u00C3\u00A1. Proceedings of the 2004 annual conference of the Canadian Linguistic Association, Yves Roberge, Martha McGinnis, and Marie-Odile Junker, eds., 11 pages. http://www.carleton.ca/~mojunker/ACL-CLA/pdf/Cook-CLA-2004.pdf Cook, Clare. 2006. Prosodic correlates of Plains Cree clause boundaries: Patterns from two speakers. Actes du trente-septi\u00C3\u00A8me congr\u00C3\u00A8s des algonquinistes, ed. by H.C. Wolfart, pp. 75-101. Winnipeg: Universit\u00C3\u00A9 du Manitoba. Cook, Clare. 2007. Distinguishing modes in Plains Cree. Proceedings of the 38th Algonquian Conference, ed. by H.C. Wolfart, pp. 47-80. Winnipeg: University of Manitoba Press. Cook, Clare, Rose-Marie D\u00C3\u00A9chaine, and Jeff M\u00C3\u00BChlbauer. 2003. Rhetorical structure of a Plains Cree counselling speech. Paper given at the 35th Algonquian Conference; London, Ontario. Cook, Clare and Jeff M\u00C3\u00BChlbauer. 2006. The behaviour of obviation in elicitation. Actes du trente- septi\u00C3\u00A8me congr\u00C3\u00A8s des algonquinistes, ed. by H.C. Wolfart, pp. 103-129. Winnipeg: Universit\u00C3\u00A9 du Manitoba. Cook, Clare and Jeff M\u00C3\u00BChlbauer. 2007. Constructing (Co-)Presence in Plains Cree. In Proceedings of WSCLA XII, ed. by Seok Koon Chin and Hudu Fusheini; pp. 12-25. Copley, Bridget. 2002. The semantics of the future. PhD dissertation, Massachussetts Institute of Technology. Costa, David. 1996. Reconstructing initial change in Algonquian. Anthropological Linguistics 38(1); pp. 39-72. Culicover, P. and Michael Rochemont. 1990. Extraposition and the complement principle. Linguistic Inquiry 21; pp. 23-47. Dahlstrom, Amy. 1986. Plains Cree morphosyntax. PhD dissertation, University of California \u00E2\u0080\u0093 Berkeley. Dahlstrom, Amy. 1991. Plains Cree morphosyntax. New York: Garland. Dahlstrom, Amy. 1995. Topic, focus, and other word order problems in Algonquian. Voices of Rupert\u00E2\u0080\u0099s Land. Winnipeg: University of Manitoba. Dahlstrom, Amy. 2006. The syntax of Algonquian Ethnopoetics. Actes du trente-septi\u00C3\u00A8me congr\u00C3\u00A8s des algonquinistes, ed. by H.C. Wolfart, pp. 131-147. Winnipeg: Universit\u00C3\u00A9 du Manitoba. Davis, Henry, Lisa Matthewson and Hotze Rullmann. to appear. A unified modal semantics for out-of-control in St\u00E2\u0080\u0099at\u00E2\u0080\u0099imcets. Proceedings of the TAMTAM conference ed. by Lotte Jogeweg, Helen de Hoop and Andrey Malchukov. 350 Davis, Henry, Lisa Matthewson and Scott Shank. 2004. Presuppositionless Clefts. Festschrift for Dale Kinkade. Dayal, Vaneeta. 1996. Locality in wh-quantification: Questions and relative clauses in Hindi. Studies in Linguistics and Philosophy 62. Dordrecht: Kluwer. D\u00C3\u00A9chaine, Rose-Marie. 1997. Nominal predication in Plains Cree. Papers of the 28th Algonquian Conference, ed. by D. Pentland, pp. 105-135. Winnipeg: University of Manitoba Press. D\u00C3\u00A9chaine, Rose-Marie. 1999. What Algonquian morphology is really like: Hockett Revisited. Proceedings from the Workshop on the Structure and Constituency in Native American Languages, ed. by Leora Bar-el, Rose-Marie D\u00C3\u00A9chaine, and Charlotte Reinholtz, pp. 25- 72. Cambridge MA: MIT Occasional Papers in Linguistics 17. D\u00C3\u00A9chaine, Rose-Marie. 2001. Asymmetric agreement. Talk given at CUNY, 8 Feb 2001. D\u00C3\u00A9chaine, Rose-Marie. 2002. Decomposing focus: evidence from Yor\u00C3\u00B9b\u00C3\u00A1. Paper given at the \u00E2\u0080\u009CTriggers for Movement\u00E2\u0080\u009D workshop, Tilburg University. D\u00C3\u00A9chaine, Rose-Marie. 2003. The structure of vP in Plains Cree. Paper given at Workshop on Structure and Constituency of Languages of the Americas VIII. Brandon, MB. D\u00C3\u00A9chaine, Rose-Marie. 2007. Order in a free word order language: the conceptual necessity of discourse configurationality. Paper given at the Linearization Workshop. Nagoya, Japan. D\u00C3\u00A9chaine, Rose-Marie & Charlotte Reinholtz. 1997. The direct/inverse as split case. Paper given at the Second Workshop on the Structure and Constituency of the Languages of the Americas, Brandon, MB. D\u00C3\u00A9chaine, Rose-Marie and Charlotte Reinholtz 2008. Case theory meets linking theory: Algonquian direct/inverse as split case. Paper given at WSCLA XIII. Kingston: ON. D\u00C3\u00A9chaine, Rose-Marie and Martina Wiltschko. 2002a. Decomposing pronouns. Linguistic Inquiry 33(3); pp. 409-442. D\u00C3\u00A9chaine, Rose-Marie and Martina Wiltschko. 2002b. The position of negation and its consequences. Proceedings of WSCLA 7, ed. by L.Bar-el, L, Watt and I. Wilson. Vancouver: UBCWPL. D\u00C3\u00A9chaine, R.-M. & M. Wiltschko. 2002c. Negation at the Left Periphery: Evidence from Algonquian and Salish. In Proceedings of WECOL 2001. D\u00C3\u00A9chaine, Rose-Marie and H.C. Wolfart. 1998. Towards a syntax of negation in Plains Cree. Paper presented at the 29th Algonquian Conference. D\u00C3\u00A9chaine, Rose-Marie and H.C. Wolfart. 2000. N\u00C3\u00A9gation comme indice de structure syntaxique dans le Cris des Plaines. Paper presented at the 31st Algonquian Conference. D\u00C3\u00A9chaine, Rose-Marie and H.C. Wolfart. 2005. On the topic of conditionals in Plains Cree. Paper presented at the 37th Algonquian Conference, Ottawa, Ontario. 351 Declerck, Renaat. 1991. Tense in English. Its structure and use in discourse. Routledge. DeLancey, S. 1997. Mirativity: the grammatical marking of unexpected information. In Linguistic Typology, Vol. 33, pp. 52. DeLancey, S. 2001. The mirative and evidentiality. In Journal of Pragmatics, Vol. 33, pp. 369- 382. Demirdache, Hamida, and Myriam Uribe-Etxebarria. 2000a. The primitives of temporal relations. In Step by Step:Essays on Minimalist Syntax in honor of Howard Lasnik, eds. Roger Martin, David Michaels and Juan Uriagereka, 157-186. Cambridge, Massachusetts: MIT Press. Demirdache, Hamida and Myriam Uribe Etxebarria. 2000b. Tenses, aspects and time adverbs as spatio-temporal predicates. Presented at International Round Table `The Syntax of Tense and Aspect', Universit\u00C3\u00A9 Paris III - Sorbonne Nouvelle, 5-18 November 2000. Demirdache, Hamida and Myriam Uribe Etxebarria. 2007. The syntax of time arguments. Lingua 117(2). pp. 330-366. Dickinson, Connie. 2000. Mirativity in Tsafiki. Studies in Language 24(2):379-421. Diesing, Molly. 1992. Indefinites. Cambridge, MA: MIT Press. Di Sciullo, Anna Maria and Edwin Williams. 1987. On the definition of word. Cambridge, MA: MIT Press. Drapeau, Lynn. 1996. Conjurors: The use of evidentials in Montagnais secondhand narratives. nikotw\u00C3\u00A2sik iskw\u00C3\u00A2ht\u00C3\u00AAm, p\u00C3\u00A2skiht\u00C3\u00AApayih! Studies in honour of H.C. Wolfart, ed. by John D. Nichols and Arden C. Ogg, pp. 171-194. Algonquian and Iroquoian Linguistics Memoir 13. Winnipeg. Dowty, David. 1979. Word meaning in Montague grammar: the semantics of verbs and times in generative semantics and in Montague\u00E2\u0080\u0099s PTQ. Dordrecht: Reidel. Edwards, Mary. 1954. Cree: An Intensive Language Course. Alberta: Northern Canada Evangelical Mission. Ehlich, Konrad. 1982. Anaphora and Deixis: Same, Similar, or Different? In Speech, Place, and Action: Studies in Deixis and Related Topics, Robert Jarvella and Wolfgang Klein, eds., pp.315-337. New York: John Wiley & Sons, Ltd. Ellis, C. Douglas. 1983. Spoken Cree. Revised edition. Edmonton, AB: Pica Pica Press. Emonds, Joseph. 1976. A transformational approach to English syntax. New York: Academic Press. Enc, M\u00C3\u00BCrvet. 1987. Anchoring Conditions for Tense. Linguistic Inquiry 18. pp. 633-57. Erteschik-Shir, N. 1973. On the nature of island constraints. PhD dissertation, Massachussetts Institute of Technology. 352 Faller, Martina. 2002. Semantics and pragmatics of evidentials in Cuzco Quechua. Ph.D. thesis, Stanford University. Faller, Martina. 2003. Spatio-temporal deixis and evidentiality in Quechua. Paper presented at SULA 2003; Vancouver, BC. Faller, Martina. 2004. The deictic core of \u00E2\u0080\u0098non-experienced past\u00E2\u0080\u0099 in Cuzco Quechua. Journal of Semantics 21. pp. 45-85. Oxford University Press. Farkas, Donka. 1985. Intensional descriptions and the Romance subjunctive mood. New York: Garland. Farkas, Donka. 1992. On the semantics of subjunctive complements. Romance Languages and modern linguistic theory, ed. by P. Hirschb\u00C3\u00BChler, pp. 69-105. John Benjamins. Farkas, Donka and Henriette de Swart. 2003. The semantics of incorporation. Stanford: CSLI. Fenn, P. 1987. A semantic and pragmatic examination of the English perfect. T\u00C3\u00BCbingen: Narr. Fillmore, Charles J. 1975. Santa Cruz Lectures on Deixis. Distributed by Indiana University Linguistics Club, Bloomington, IN. Fillmore, Charles J. 1982. Towards a descriptive framework for spatial deixis. In Speech, Place, and Action: Studies in Deixis and Related Topics, Robert Jarvella and Wolfgang Klein, eds., pp. 31-56. New York: John Wiley & Sons, Ltd. Finer, D. 1985. The formal grammar of switch-reference. Outstanding dissertations in Linguistics. New York: Garland. von Fintel, Kai. 2006. Modality and language. Encyclopedia of philosophy \u00E2\u0080\u0093 second edition ed. by Donald Borchert. MacMillan. Fitzpatrick, Justin. 2005. The Whys and How Comes of Presupposition and NPI Licensing in Questions. Proceedings of the 24th West Coast Conference on Formal Linguistics, ed. by John Alderete et al., Somerville, MA: Cascadilla Proceedings Project. Fodor, Janet. 1992. Islands, learnability and the lexicon. In Island constraints, Theory, Acquisition, and Processing, Helen Goodluck & Michael Rochemont (eds.), pp. 109-180. Dordrecht: Kluwer. Foley, William and Robert D. Van Valin. 1984. Functional syntax and universal grammar. Cambridge: Cambridge University Press. Fort, Charles. 1919. The book of the damned. New York: Boni and Liveright. [London: John Brown, (1995)] Frantz, Donald. 1991. Blackfoot grammar. Toronto: University of Toronto Press. Garrett, Edward. 2001. Evidentiality and assertion in Tibetan. Ph.D. thesis, University of California \u00E2\u0080\u0093 Los Angeles. 353 Gennari, Silvia. 2003. Tense meanings and temporal interpretation. Journal of Semantics 20. pp. 35-71. Oxford University Press. Giorgi, Alessandra and Fabio Pianesi. 1997. Tense and aspect: From semantics to morpho- syntax. New York: Oxford University Press. Giannakidou, Anastasia. 1998. Polarity sensitivity as (non)veridical dependency. Amsterdam- Philadelphia: John Benjamins. Giannakidou, Anastasia. 2008. The dependency of the subjunctive revisited: temporal semantics and polarity. Lingua, ed. by Josep Quer. (to appear) Givon, Talmy. 2001. Syntax. Amsterdam-Philadelphia: John Benjamins. Goddard, Ives. 1967. The Algonquian Independent Indicative. in Bulletin 214, Contributions to Anthropology: linguistics 1, pages 66-106. Ottawa: National Museum of Canada. Goddard, Ives. 1979. Delaware verbal morphology: A descriptive and comparative study. New York: Garland. Goddard, Ives. 1984. The obviative in Fox narrative discourse. Papers of the 15th Algonquian Conference, ed. by William Cowan, pp. 273-286. Ottawa: Carleton University. Goddard, Ives. 2002. Grammatical gender in Algonquian, in Proceedings of 33rd Algonquian Conference, H.C. Wolfart (ed.), pp. 195-231. Winnipeg: University of Manitoba Press. Green, Georgia. 1989. Pragmatics and Natural Language Understanding. Hillsdale, NJ: Lawrence Erlbaum Associates. Grimshaw, Jane. 1979. Complement selection and the lexicon. Linguistic Inquiry 10(2); pp. 279-326. Grimshaw, Jane. 1981. Form, function, and the language acquisition device. The logical problem of language acquisition, ed. by C.L. Baker and J. McCarthy; pp. 165-182. Cambridge, MA: MIT Press. Haegeman, Liliane. 1994. Introduction to government and binding theory. Blackwell Publishing. Haegeman, Liliane. 1995. The syntax of negation. Cambridge: Cambridge University Press. Hagstrom, Paul. 1998. Decomposing questions. PhD dissertation, Massachussetts Institute of Technology. Haider, Hubert. 1996. Downward to the right. On extraction and extraposition in German, ed. by Uli Lutz and J\u00C3\u00BCrgen Pafel, pp. 245-271. Amsterdam: John Benjamins. Heim, Irene. 1982. The semantics of definite and indefinite noun phrases. PhD dissertation, Massachussetts Institute of Technology. Heim, Irene. 1992. Presupposition projection and the semantics of attitude verbs. Journal of Semantics 9; pp. 183-221. 354 Heim, Irene, and Angelika Kratzer. 1998. Semantics in generative grammar. Oxford: Blackwell publishers. Higginbotham, James. 1983. Logical form, binding, and nominals. Linguistic Inquiry 14; pp. 395-420. Hinrichs, E. 1986. Temporal Anaphora in Discourses of English. Linguistics and Philosophy 9. pp. 63-82. Hirose, Tomio. 1999. Origins of Predicates: Evidence from Plains Cree. University of British Columbia Ph.D Thesis. Hockett, Charles F. 1950. The conjunct modes in Ojibwa and Potawatomi. Language 26:278- 282. Hockett, Charles F. 1958. A course in modern linguistics. New York: The Macmillan Company. Hockett, Charles F. 1966. What Algonquian is really like. International Journal of American Linguistics 32; pp. 59-73. Holmberg, Anders and Christer Platzack. 1995. The role of inflection in Scandanavian syntax. Oxford University Press. Hornstein, Norbert. 1990. As time goes by. Cambridge, Mass.: MIT Press. Howse, Joseph. 1865. A Grammar of the Cree Language: With which is combined an Analysis of the Chippeway Dialect. London: Trubner & Co. Huang, C.-T. 1982. Logical Relations in Chinese and the Theory of Grammatical Relations. Ph.D Dissertation, MIT. Huddleston, Rodney. 1969. Some observations on tense and deixis in English. Language 45(4). pp. 777-804. Linguistic Society of America. Hunter, Emily, and Betty Karpinski. 1999. Introductory Cree: Part 1. School of Native Studies, University of Alberta. Iatridou, Sabine. 2000. The grammatical ingredients of counterfactuality. Linguistic Inquiry 31(2). pp. 231-270. James, Deborah. 1982. Past Tense, Imperfective Aspect, and Irreality in Cree. In Papers of the 13th Algonquian Conference, William Cowan, ed. pp. 143-160. Ottawa: Carleton University. James, Deborah. 1982b. Past tense and the hypothetical: a cross linguistic study. Studies in Language 6, pp. 375-403. James, Deborah, Sandra Clarke, and Marguerite MacKenzie. 1996. Indirect evidentials in the Cree/Montagnais/Naskapi of Quebec and Labrador. In Papers of the 27th Algonquian Conference, H.C. Wolfart, ed., pp. 135-151. Winnipeg: University of Manitoba. 355 Jakobsen, Roman. 1929. Remarques sur l'evolution phonologique du russe compar\u00C3\u00A9e \u00C3\u00A0 celle des autres langues slaves. Prague. Janzen, Terry. 2004. Space rotation, perspective shift, and verb morphology in ASL. Cognitive Linguistics 15(2). pp.149-174. Walter de Gruyter. Jespersen, Otto. 1924. The philosophy of grammar. London: Allen and Unwin. Jespersen, Otto. 1937. Analytic syntax. London: Allen & Erwin. Junker, Marie-Odile. 2003. East Cree dependent nouns and disjoint reference, in Algonquian and Iroquoian Linguistics 28, pp 11-13. Kamp, H. 1981. A theory of truth and semantic representation. In Formal methods in the study of language, J. Groenendik, T. Janssen and M. Stokhof (eds.), pp. 277-322. Amsterdam: Mathematics Centre Tracts. Kamp, Hans, and Uwe Reyle. 1993. From discourse to logic. Dordrecht: Kluwer Academic Publishers. Kamp, Hans, and Christian Rohrer. 1983. Tense in texts. Meaning, use and interpretation of language, ed. by R. B\u00C3\u00A4uerle, pp. 250-269. Berlin: de Gruyter. K\u00C3\u00A2-Nip\u00C3\u00AEt\u00C3\u00AAht\u00C3\u00AAw, Jim. 1998. Ana k\u00C3\u00A2-pimw\u00C3\u00AAw\u00C3\u00AAhahk okak\u00C3\u00AAskihk\u00C3\u00AAmowina: The Counselling Speeches of Jim K\u00C3\u00A2-Nip\u00C3\u00AEt\u00C3\u00AAht\u00C3\u00AAw. Ed. and translated by H. C. Wolfart and Freda Ahenakew. Winnipeg: University of Manitoba Press. Kaplan, David. 1989. Demonstratives. In Themes from Kaplan, Joseph Almog, John Perry, and Howard Wettstein, eds., pp. 481-563, Oxford: Oxford University Press. Karttunen, Laurie and Peters, Stanley. 1976. What Indirect Questions Conventionally Implicate. In Salikoko S. Mufwene, Carol A. Walker, and Stanford B. Steever (eds.), Papers from the 12th Regional Meeting Chicago Linguistics Society (pp. 351\u00E2\u0080\u0093368). Katz, Jerrold J. 1972. Semantic Theory. New York: Harper & Row. Katz, Jerrold J. and Postal, Paul M. 1964. An Integrated Theory of Linguistic Descriptions. Cambridge, MA: MIT Press. Kayne, Richard. 1994. The Antisymmetry of Syntax. MIT Press. Kazenin, Konstantin and Yakov G. Testelets. 2004. Where coordination meets subordination: Converb constructions in Tsakhur (Daghestanian). Coordinating constructions ed. by Martin Haspelmath. pp. 227-239. The Netherlands: John Benjamins. Kiparsky, Paul and Carol Kiparsky. 1971. Fact. Semantics: an interdisciplinary reader in philosophy, linguistics, and psychology, ed. by Danny D. Steinberg and Leon A. Jakobvitz, pp. 345-69. Cambridge: MIT Press. Klein, Wolfgang. 1992. The present perfect puzzle. Language 68; pp. 525-52. 356 Klein, Wolfgang. 1994. Time in language. New York: Rutledge. K\u00C3\u00B6lbel, Max. 2002. Truth without objectivity. London: Routledge. Koster, Jan. 1978. Why subject sentences don\u00E2\u0080\u0099t exist. Recent transformational studies in European languages, ed. by S.J. Keyser. Cambridge, MA: MIT Press. Kratzer, Angelika. 1977. What \u00E2\u0080\u0098Must\u00E2\u0080\u0099 and \u00E2\u0080\u0098Can\u00E2\u0080\u0099 must and can mean. Linguistics and Philosophy 1; pp. 337-355. Kratzer, Angelika. 1989. An investigation of lumps of thought. Linguistics and Philosophy 12; pp. 607-653. Kratzer, Angelika. 1998. More structural analogies between pronouns and tenses. Proceedings of semantic and linguistic theory VIII, ed. by D. Strolovitch and A. Lawson. pp. 92-110. New York: CLC Publications. Kratzer, Angelika. 2007. Situations in Natural Language Semantics. Stanford encyclopedia of philosophy. Palo Alto, CA: CSLI Publications. Kuno, Susumu. 1987. Functional Syntax: anaphor, discourse, and empathy. Chicago: University of Chicago Press. Lacombe, Albert. 1874. Dictionnaire de las Langue des Cris. Montreal. Lahiri, Utpal. 2003. Questions and answers in embedded contexts. Oxford studies in theoretical linguistics. Oxford: Oxford University Press. Laka, I. 1990. Negation in syntax: On the nature of functional categories and projections. PhD thesis, Massachussetts Institute of Technology. Lakoff, George. 1968. Pronouns and reference. Distributed by IULC. Lakoff, Robin. 1970. Tense and its relation to participants. Language 46:838-849. Landman, Fred. 1991. Structures for semantics. Dordrecht: Kluwer Academic Publishers. Langacker, Ronald. 1969. Pronomilization and the chain of command, in Modern Studies in English, D. Reibel and S. Schane, eds., pp. 160-186. New Jersey: Prentice Hall. LaPolla, R.J.. 2003. Evidentiality in Qiang. Studies in evidentiality, ed. by A. Aikhenvald and R.M.W. Dixon, pp. 63-78. Amsterdam: John Benjamins. Lasersohn, Peter. 2005. Context Dependence, Disagreement, and Predicates of Personal Taste. In Linguistics and Philosophy 28: 643\u00E2\u0080\u0093686. Lawler, John M. 1971. Any Questions? Papers from the Seventh Regional Meeting of the Chicago Linguistic Society, pp. 163\u00E2\u0080\u0093173. 357 Lederer, Lederer. 1969. Reference grammar of the German language. New York: Charles Scribner\u00E2\u0080\u0099s Sons. Long, Michelle. 1999. Obviation and clausal relations in Plains Cree. MA thesis, University of Manitoba. Longacre, Robert. 1983. Switch reference systems from two distinct linguistic areas: Wojokeso (Papua New Guinea) and Guanano (Northern South America). Switch reference and universal grammar, ed. by John Haiman and Pamela Munro. pp. 185-207. Amsterdam: John Benjamins. Macaulay, Monica. 2004. Negation, dubitatives, and mirativity in Menominee. In Papers of the Thirty-fourth Algonquian Conference, H.C. Wolfart, ed., pp. 217-240, Winnipeg: Papers of the Algonquian Conference. Mandelbaum, David. 1940. The Plains Cree. Anthropological Papers of the American Museum of Natural History, Volume XXXVII, Part II, New York. Manzini, Rita. 1992. Locality. Linguistic Inquiry monograph X. Cambridge, MA: MIT Press. Matthewson, Lisa. 1999. On the interpretation of wide-scope indefinites. In Natural Language Semantics 7:79-134. Matthewson, Lisa. 2005. Temporal semantics in a superficially tenseless language. ms, University of British Columbia Matthewson, Lisa and Charlotte Reinholtz. 1996. The syntax and semantics of determiners: A comparison of Salish and Cree. Paper given at the International Conference on Salish and Neighbouring Languages, Vancouver. May, Robert. 1985. Logical form. Cambridge, MA: MIT Press. McCawley, James. 1970. Where do nouns come from? Readings in English transformational grammar, ed. by Roderick Jacobs and Peter Rosenbaum. Waltham, MA: Ginn and Company. McCawley, James. 1981. The syntax and semantics of English relative clauses. Lingua 53; pp. 99-149. McCawley, James. 1988. The syntactic phenomena of English. 2 volumes. Chicago: Chicago University Press. Minde, Emma. 1997. kwayask \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-kiskinow\u00C3\u00A2pahtihicik / Their example showed me the way, ed. by H.C. Wolfart and Freda Ahenakew, ed. and tr. by H.C. Wolfart and Freda Ahenakew. Edmonton:University of Alberta Press. Montague, Richard. 1973. The proper treatment of quantification in ordinary English. Approaches to natural language, proceedings of the 1970 Stanford workshop on grammar and semantics, ed. by J. Hintikka and J.M.E. Moravcsik, P. Suppes. pp. 221- 242. Dordrecht: Reidel. 358 Moravcsik, Edith and Jessica Wirth. 1986. Markedness: An overview. Markedness, ed. by Fred R. Eckman, Edith Moravcsik, and Jessica Wirth, pp. 1-11. New York: Plenum. Moro, Andrea. 1997. The raising of predicates. Cambridge:Cambridge University Press. M\u00C3\u00BChlbauer, Jeff. 2003. Word Order and the Interpretation of Nominals in Plains Cree. Ms, University of British Columbia. M\u00C3\u00BChlbauer, Jeff. 2007. The speaker\u00E2\u0080\u0099s knowledge and obviation in Plains Cree. Proceedings of the 38th Algonquian Conference, ed. by H.C. Wolfart, Winnipeg: University of Manitoba Press. pp. 343-390. M\u00C3\u00BChlbauer, Jeff. In prep. k\u00C3\u00A2-y\u00C3\u00B4sk\u00C3\u00A2tahk \u00C3\u00B4ma n\u00C3\u00AAhiyaw\u00C3\u00AAwin: The representation of intentionality in Plains Cree. UBC PhD thesis. Napoli, Donna. 1981. Semantic interpretation vs. lexical governance: clitic climbing in Italian. Language 57; pp. 841-887. Nunberg, Geoffrey. 1993. Indexicality and deixis. Linguistics and Philosophy 16; pp. 1-43. Ogg, Arden. 1991. Connective particles and temporal cohesion in Plains Cree Narrative. Master\u00E2\u0080\u0099s thesis, University of Manitoba. Ogihara, Toshiyuki. 1995. The semantics of tense in embedded clauses. Linguistic Inquiry 26, pp. 663-679. Okimasis, Jean & Solomon Ratt. 1999. Cree, Language of the Plains: Nehiyawewin, Paskwawi- Pikiskwewin. Regina, SK: University of Regina Canadian Plains Research Center. Partee, Barbara H. 1973. Some Structural Analogies between Tenses and Pronouns in English. The Journal of Philosophy 70. pp. 601-609. Partee, Barbara H. 1984. Nominal and Temporal Anaphora. Linguistics and Philosophy 7. pp. 243-286. Partee, Barbara H. 1989. Binding implicit variables in quantified contexts. Chicago Linguistics Society 25; pp. 342-265. Paul, Hermann. 1886. Prinzipien der Sprachgeschichte. Jena: Niemeyer. Pentland, David H. 1979. Algonquian historical phonology. PhD dissertation, University of Toronto. Pentland, David H. 1999. The morphology of the Algonquian independent order. In Papers of the 30th Algonquian Conference, ed. by David Pentland, pp. 222-266. Winnipeg: University of Manitoba. Pesetsky, David. 1982. Paths and categories. PhD dissertation, Massachussetts Institute of Technology. 359 Pesetsky, David. 1987. Wh-in-situ: movement and unselective binding. In The Representation of (In)definiteness, ed. Eric Reuland and Alice ter Meulen, 98-129. Cambridge, Mass: MIT Press. Piggott, Glyne. 1971. Some implications of Algonquian languages. Odawa Language Project, First Report, ed. by JD. Kaye, et al. pp. 11-38. Anthropological Series, 9, University of Toronto. Pollock, Jean-Yves. 1989. Verb movement, UG, and the structure of IP. Linguistic Inquiry 20. pp. 365-424. Portner, Paul. 1997. The semantics of mood, complementation, and conversational force. Natural Language Semantics 5: 167-212. Postal, Paul. 1969. Anaphoric islands. Chicago Linguistics Society 5; pp. 205-239. Postal, Paul. 1971. Crossover phenomena. New York: Academic Press. Prior, Arthur. 1957. Time and modality. Oxford: Clarendon Press. Prior, Arthur. 1967. Past, present and the future. Oxford: Clarendon Press. Quirk, R, S. Greenbaum, G. Leech, and J. Svartvik. 1985. A comprehensive grammar of the English language. Longman. Reichenbach, Hans. 1947. Elements of Symbolic Logic. New York: Macmillan. Reinhart, Tanya. 1976. The syntactic domain of anaphora. PhD dissertation, Massachussetts Institute of Technology. Reinhart, Tanya. 1983. Anaphora and Semantic Interpretation. Chicago: University of Chicago Press. Reinholtz, Charlotte. 1995. Discontinuous constituents in Swampy Cree. Papers of the 26th Algonquian Conference, ed. by David H. Pentland, pp. 394-412. Winnipeg: University of Manitoba. Reinholtz, Charlotte. 1999. On the characterization of discontinuous constituents: Evidence from Swampy Cree. International Journal of American Linguistics 65; pp. 201-227. Reinholtz, Charlotte. 2007. The minimal scope of interrogative: Variable question marking in Eastern Swampy Cree. Paper given at Workshop on Structure and Consitutency of the Languages of the Americas XII. Lethbridge, AB. Reinholtz, Charlotte and H.C. Wolfart. 1996. Sentences. In HC Wolfart, Sketch of Cree, an Algonquian language, pp. 391-398. Languages, ed. by Ives Goddard. Pp. 390-439. W.C. Sturtevant, gen. ed., Handbook of North American Indians, vol. 17. Washington, DC: Smithsonian Institution. 360 Reinholtz, Charlotte, and H.C. Wolfart. 2001. The syntax of emphatic ani in Eastern Swampy Cree and in Plains Cree. Actes du 32e Congr\u00C3\u00A8s des Algonquinistes, ed. by John D. Nichols, pp. 427- 454. Winnipeg: University of Manitoba. Rhodes, Richard A. 1976. The morphosyntax of the Central Ojibwa verb. Ph.D. thesis, University of Michigan. Rhodes, Richard. 2006. Clause-structure, core arguments, and the Algonquian relative root construction. The Belcourt Lecture, University of Manitoba, March 20, 1998. Winnipeg: Voices of Rupert\u00E2\u0080\u0099s Land. Richards, Norvin. 1997. What moves where when in which language? PhD Dissertation, Massachussetts Institute of Technology. Ritter, Elizabeth. 1995. On the syntactic category of pronouns and agreement. Natural Language and Linguistic Theory 13(3); pp. 405-443. Ritter, Elizabeth. and Sara. Rosen. 2005. Agreement without A positions: Another look at Algonquian. In Linguistic Inquiry, Vol. 36, pp. 648-660. Ritter, Elizabeth and Martina Wiltschko. 2005. Anchoring events to utterances without Tense. Proceedings of the 24th West Coast Conference on Formal Linguistics, ed by John Alderete, et al., pp. 343-351. Somerville, MA: Cascadilla Proceedings Project. Ritter, Elizabeth and Martina Wiltschko. 2007. Tenseless languages. Rizzi, Luigi. 1997. The fine structure of the left periphery. Elements of Grammar, ed. by L. Haegeman, pp. 281-337. Dordrecht: Kluwer Academic Publishers. Roberts, Craige. 1989. Modal subordination and pronominal anaphora in discourse. Linguistics and Philosophy 12(6); pp. 683-721. Rooth, Matts. 1985. Association with focus. PhD dissertation, University of Massachussetts \u00E2\u0080\u0093 Amherst. Rooth, Matts. 1992. A theory of focus interpretation. Natural Language Semantics 1; pp. 75-116. Rooth, Matts. 1996. Focus. The handbook of contemporary semantic theory. ed. by S. Lappin. pp. 271-297. Oxford: Blackwell. Rosenbaum, Peter. 1967. The grammar of English predicate-complement constructions. Cambridge, MIT Press. Ross, John Robert. 1967a. Constraints on variables in syntax. PhD dissertation, Massachussetts Institute of Technology. Ross, John Robert. 1967b. On the cyclic nature of English pronominalization. To honor Roman Jakobson. pp. 1669-82. The Hague: Mouton. Ross, John Robert. 1969. A proposed rule of tree pruning. Modern studies in English, ed. by David Reibel and Sanford Schane, pp. 286-299. Englewood Cliffs, NJ: Prentice-Hall. 361 Ross, John Robert. 1970. On declarative sentences. Readings in English transformational grammar, ed. by Roderick Jacobs and Peter Rosenbaum, pp. 222\u00E2\u0080\u0093272. Waltham, MA: Ginn and Company. Russell, Kevin and Charlotte Reinholtz. 1996. Hierarchical structure in a non-configurational language: Asymmetries in Swampy Cree. Proceedings of the Fourteenth West Coast Conference on Formal Linguistics, ed. by Jose Camacho et al. Pp. 431-445. Stanford: CSLI. Rutherford, William. 1970. Some observations concerning subordinate clauses in English. Language 46: pp. 96-115. Rullmann, Hotze. 2003. Bound variable pronouns and the semantics of number. Proceedings of the Western Conference on Linguistics 2002, ed. by Brian Agbayani et al., pp. 243-254. Fresno, CA: California State University. Rullmann, Hotze. 2004. First and second person pronouns as bound variables. Linguistic Inquiry 35(1): pp. 159-168. Safir, Kenneth. 1985. Syntactic chains. Cambridge: Cambridge University Press. Safir, Ken. 2004. The syntax of (in)dependence. Cambridge, MA: MIT Press. de Saussure, Ferdinand. 1959. Course in general linguistics. New York: The Philosophical Library. Saxon, Leslie. 1986. The syntax of pronouns in Dogrib. PhD dissertation, University of California, San Diego. Schlenker, Philippe. 2003. A plea for monsters. In Linguistics and Philosophy 26; pp. 29-120. Searle, John. 1965. What is a Speech Act? Philosophy in America, ed. by Max Black. Ithaca, N.Y.: Cornell University Press. Smith, Carlota. 2003. Modes of discourse. Cambridge: Cambridge University Press. Speas, Peggy and Carol Tenny. 2003. Configurational Properties of Point of View Roles. In Asymmetry in Grammar, DiSciullo, A. (ed.), pp. 315-343. Amsterdam: John Benjamins. Srivastav, Vaneeta. 1991. The syntax and semantics of correlatives. Natural Language and Linguistic Theory 9, pp. 637-686. Stalnaker, Robert. 1999. Context and content: Essays on intentionality in speech and thought. Oxford: Oxford University Press. Stephenson, Tamina. 2007. Towards a theory of subjective meaning. PhD dissertation, Massachusetts Institute of Technology. Stirling, Leslie. 1993. Switch reference and discourse representation theory. Cambridge: Cambridge University Press. 362 Stowell, Timothy. 1981. Origins of Phrase Structure. Ph.D. thesis, Massachusetts Institute of Technology. Stowell, Timothy. 1993. Syntax of Tense. Unpublished ms, University of California, Los Angeles. Trubetzkoy, N.S. 1969. Principles of phonology. Berkeley: University of California Press. Translation by C. Baltaxe of \u00E2\u0080\u009CGrundz\u00C3\u00BCge der Phonologie.\u00E2\u0080\u009D G\u00C3\u00A4ttingen: Vandenhoek and Ruprecht (1939). Valentine, J. Randolph. 2001. Nishnaabemwin reference grammar. Toronto: University of Toronto Press. Whitecalf, Sarah. 1993. kin\u00C3\u00AAhiy\u00C3\u00A2wiwininaw n\u00C3\u00AAhiyaw\u00C3\u00AAwin: The Cree Language is Our Identity. The La Ronge Lectures. Wolfart, H.C., and Freda Ahenakew, eds. University of Manitoba Press. Willett, Thomas. 1988. A cross-linguistic survey of the grammaticalization of evidentiality. Studies in Language 12:55-97. Williams, Edwin. 1974. Rule ordering in syntax. PhD dissertation, Massachussetts Institute of Technology. Williams, Edwin. 1997. Blocking and anaphora. Linguistic Inquiry 28:577-628. Williams, Edwin. 2004. Representation Theory. Cambridge, MA: MIT Press. Wiltschko, Martina. 1995. IDs in syntax and discourse: An analysis of extraposition in German. PhD dissertation, Universit\u00C3\u00A4t Wien. Wolfart, H.C. 1973. Plains Cree: A Grammatical Study. Transactions of the American Philosophical Society, new series, vol. 63, part 2. Philadelphia. Wolfart, H.C. 1996. Sketch of Cree, an Algonquian language. In W.C. Sturtevant, ed., \u00E2\u0080\u009CHandbook of North American Indians 17, pp. 390-439. Washington D.C.: Smithsonian Institution. Wolfart, H.C. 1997. Introduction & Notes. kwayask \u00C3\u00AA-k\u00C3\u00AE-p\u00C3\u00AA-kiskinow\u00C3\u00A2pahtihicik. Their Example Showed Me the Way, Told by Emma Minde. ed. by H.C. Wolfart and F. Ahenakew, University of Alberta Press. Wolfart, H.C. 2000. Introduction & Notes. \u00C3\u00A2h-\u00C3\u00A2y\u00C3\u00AEtaw isi \u00C3\u00AA-k\u00C3\u00AE-kisk\u00C3\u00AAyihtahkik maskihkiy / They knew both sides of medicine: Cree tales of curing and cursing told by Alice Ahenakew, ed. by H.C. Wolfart and Freda Ahenakew, ed. and tr. by H.C. Wolfart and Freda Ahenakew. Winnipeg: University of Manitoba Press. Wolfart, H.C. 2008. Syntactic relations within and beyond the noun. Talk given at Algonquian Round-table, University of British Columbia; March 2008. 363 Wolfart, H.C. and Freda Ahenakew. 1987. Notes. In w\u00C3\u00A2skahikaniwiyiniw-\u00C3\u00A2cimowina / Stories of the House People, Told by Peter Vandall and Joe Douquette. ed. by H.C. Wolfart and Freda Ahenakew, Winnipeg: University of Manitoba Press. Wolfart, H.C. and Freda Ahenakew. 1993. (eds.) kin\u00C3\u00AAhiy\u00C3\u00A2wiwininaw n\u00C3\u00AAhiyaw\u00C3\u00AAwin: The Cree Language is Our Identity. The La Ronge Lectures. Given by Sarah Whitecalf. Winnipeg: University of Manitoba Press. Wolfart, H.C. and Freda Ahenakew. 1998. The student\u00E2\u0080\u0099s dictionary of literary Plains Cree. Memoir 15. Winnipeg: Algonquian and Iroquoian Linguistics. Wolfart, H.C. and Freda Ahenakew. 2000. (eds.) \u00C3\u00A2h-\u00C3\u00A2y\u00C3\u00AEtaw isi \u00C3\u00AA-k\u00C3\u00AE-kisk\u00C3\u00AAyihtahkik maskihkiy / They knew both sides of medicine: Cree tales of curing and cursing told by Alice Ahenakew. Publications of the Algonquian Text Society / Collection de la Soci\u00C3\u00A9t\u00C3\u00A9 d\u00E2\u0080\u0099\u00C3\u00A9dition des textes algonquiens, Winnipeg: University of Manitoba Press. Wolfart, H.C. and Janet F. Carroll. 1981. Meet Cree: a guide to the Cree language. New and completely revised edition. Edmonton: University of Alberta Press. Wolvengrey, Arok. 2001. Cree: Words. Canadian Plains Research Center, University of Regina, vols. 1-2 . Wolvengrey, Arok. 2003. Demonstratives and word order. Algonquian and Iroquoian Linguistics 23; pp. Wolvengrey, Arok. 2006. Prospective aspect in the western dialects of Cree. International Journal of American Linguistics 73(3). pp. 397-407. Wolvengrey, Arok. 2007. The placement of locatives: An example of the function of Cree word order. Invited paper given at Workshop on Structure and Constituency of the Languages of the Americas XII. Lethbridge, AB. Wurmbrand, Susanne. 2003. Infinitives: Restructuring and clause-structure. Walter de Gruyter. Zagona, Karen. 2004. Tense construal in complement clauses: verbs of communication and the double access reading. Proceedings of the Paris roundtable on the syntax of tense and aspect, ed. by J. Gu\u00C3\u00A9ron and J. LeCarme, pp. 637-54. Cambridge, Massachusetts: MIT Press. Zanuttini, R. 1997. Negation and clause structure: A comparative study of Romance languages. New York/Oxford: Oxford University Press. Zwicky, Arnold. 1985. Clitics and particles. In Language, Vol. 61, pp. 283-305."@en . "Thesis/Dissertation"@en . "2008-11"@en . "10.14288/1.0066456"@en . "eng"@en . "Linguistics"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "Attribution-NonCommercial-NoDerivatives 4.0 International"@en . "http://creativecommons.org/licenses/by-nc-nd/4.0/"@en . "Graduate"@en . "The syntax and semantics of clause-typing in Plains Cree"@en . "Text"@en . "http://hdl.handle.net/2429/951"@en .