UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Quantifiers in Kwak'wala Moewaki, Ayako 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_november_moewaki_ayako.pdf [ 212.14MB ]
Metadata
JSON: 24-1.0314095.json
JSON-LD: 24-1.0314095-ld.json
RDF/XML (Pretty): 24-1.0314095-rdf.xml
RDF/JSON: 24-1.0314095-rdf.json
Turtle: 24-1.0314095-turtle.txt
N-Triples: 24-1.0314095-rdf-ntriples.txt
Original Record: 24-1.0314095-source.json
Full Text
24-1.0314095-fulltext.txt
Citation
24-1.0314095.ris

Full Text

   Quantifiers in Kwak’wala  by Ayako Moewaki  B.A., Dokkyo University, 1997  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF ARTS in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Linguistics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2016  © Ayako Moewaki, 2016   ii Abstract  This thesis describes and analyzes the syntax and semantics of quantification in Kwak’wala, a Northern Wakashan language spoken on Vancouver Island and the adjacent mainland of British Columbia. In standard accounts of quantification, quantifiers can be divided into two categories: strong quantifiers and weak quantifiers (Barwise and Cooper (1981), Keenan (1996), Milsark (1979), Partee (1988)). Taking the strong/weak distinction as a starting point, this thesis documents the syntactic and semantic features of Kwak’wala quantifiers, focusing on wi'la (‘all’), ḵ̓ina̱m (‘many’), and numerals, and examining their behavior at the syntax-semantics interface. In line with Milsark’s predictions, only the Kwak’wala equivalent of English weak quantifiers are allowed to stand as main predicates in sentences which are the equivalents of existential sentences in English. This thesis describes the distinction between strong and weak readings for the weak quantifier ḵ̓ina̱m (‘many’): there is no correlation between its positions and strong vs. weak interpretations, except in sentences which are the equivalent of existential sentences in English. The lack of a correlation between interpretation and DP-internal position is different from the prediction of standard accounts of quantification, by e.g. Partee (1988). This thesis also investigates quantifier scope. I show that Kwak’wala has collective, distributive, and cumulative readings. The presence of distributive and inverse distributive readings shows that this language possesses quantifiers which undergo real scopal interactions (unlike, for example, St'át'imcets as shown by Davis (2010)).   iii Preface  The thesis is original and unpublished work by the author, A. Moewaki. The fieldwork presented within it was collected in accordance with UBC Ethics Certificate number H16-00193.   iv Table of Contents  Abstract .......................................................................................................................................... ii	Preface ........................................................................................................................................... iii	Table of Contents ......................................................................................................................... iv	List of Tables ............................................................................................................................... viii	List of Figures ................................................................................................................................. x	Glossary ........................................................................................................................................ xi	Acknowledgements .................................................................................................................... xiii	Dedication ..................................................................................................................................... xv	1. INTRODUCTION ............................................................................................... 1	1.1	 Objectives .......................................................................................................................... 1	1.2	 Diagnostics ......................................................................................................................... 2	1.2.1	 Syntax ......................................................................................................................... 3	1.2.2	 Semantics .................................................................................................................... 5	1.2.2.1	 Semantic Correlates of the Strong vs. Weak Distinction in English .................... 6	1.2.2.2	 Quantifier Scope .................................................................................................. 9	1.2.2.3	 Collective, Distributive, and Cumulative Readings of English Quantified DPs 10	1.3	 Methodology .................................................................................................................... 14	2. KWAK’WALA ................................................................................................... 16	2.1	 Orthography ..................................................................................................................... 17	2.2	 Core Word Order .............................................................................................................. 18	2.3	 Determiner System ........................................................................................................... 25	2.4	 Reduplication ................................................................................................................... 27	3. Ḵ̓INA̱M (‘MANY’) IS A WEAK QUANTIFIER ............................................. 28	3.1	 Predicative Status of Ḵ̓ina̱m ............................................................................................. 28	3.2	 The Internal Structure of Quantified DPs: Ḵ̓ina̱m ........................................................... 29	3.2.1	 Ḵ̓ina̱m in Pre- vs. Post-determiner Position .............................................................. 29	  v 3.2.2	 Interaction of Ḵ̓ina̱m and Adjectives ........................................................................ 32	3.2.3	 Summary of Predicative Status and Distribution of DP-Internal Qs: Ḵ̓ina̱m ........... 34	3.3	 Clause-Initial Quantified DPs and Bare Quantifiers: Ḵ̓ina̱m ........................................... 36	3.3.1	 Quantified DPs .......................................................................................................... 36	3.3.2	 Bare Quantifiers ........................................................................................................ 37	3.3.2.1	 Bare Quantifiers with Determiners: Ḵ̓ina̱m ....................................................... 37	3.3.2.2	 Bare Quantifiers without Determiners: Ḵ̓ina̱m .................................................. 39	3.4	 Summary of Distribution: Ḵ̓ina̱m ..................................................................................... 42	4. WI'LA (‘ALL’) IS A STRONG QUANTIFIER ............................................... 43	4.1	 Predicative Status of Wi'la ............................................................................................... 43	4.2	 The Internal Structure of Quantified DPs: Wi'la .............................................................. 43	4.2.1	 Wi'la in Pre- vs. Post-determiner Position ................................................................ 44	4.2.2	 Summary of Predicative Status and Distribution of DP-Internal Qs: Wi'la .............. 47	4.3	 Clause-Initial Quantified DPs and Bare Quantifiers: Wi'la ............................................. 49	4.3.1	 Quantified DPs .......................................................................................................... 49	4.3.2	 Bare Quantifiers ........................................................................................................ 49	4.3.2.1	 Bare Quantifiers with Determiners: Wi'la .......................................................... 50	4.3.2.2	 Bare Quantifiers without Determiners: Wi'la ..................................................... 51	4.3.3	 Summary of Distribution: Wi'la ................................................................................ 53	5. NUMERALS ARE WEAK QUANTIFIERS ................................................... 54	5.1	 Predicative Status of Numerals ........................................................................................ 54	5.2	 The Internal Structure of Quantified DPs: Numerals ...................................................... 55	5.2.1	 Numerals in Pre- vs. Post-determiner Position ......................................................... 55	5.2.2	 Interaction of Numerals and Adjectives .................................................................... 59	5.2.3	 Summary of Predicative Status and Distribution of DP-Internal Qs: Numerals ...... 59	5.3	 Clause-Initial Quantified DPs and Bare Quantifiers: Numerals ...................................... 61	5.3.1	 Quantified DPs .......................................................................................................... 61	5.3.2	 Bare Quantifiers ........................................................................................................ 62	5.3.2.1	 Bare Quantifiers with Determiners: Numerals .................................................. 62	  vi 5.3.2.2	 Bare Quantifiers without Determiners: Numerals ............................................. 63	5.4	 Summary of Distribution: Numerals ................................................................................ 65	6. THE SYNTAX OF KWAK’WALA QUANTIFIERS ...................................... 67	6.1	 Clausal Syntactic Structure .............................................................................................. 67	6.1.1	 Core Sentence Structure (Predicate Subject Object) ................................................ 69	6.2	 Syntactic Structure of Clause-Initial Quantified DPs ...................................................... 72	6.2.1	 Ḵ̓ina̱m (‘many’) ......................................................................................................... 73	6.2.2	 Wi'la (‘all’) ................................................................................................................ 78	6.2.3	 Numerals ................................................................................................................... 81	6.3	 DP-Internal Structure with Quantifiers ............................................................................ 83	6.4	 Active and Passive Sentence Structures .......................................................................... 85	6.4.1	 Active and Passive Sentence Structure ..................................................................... 86	6.4.2	 Passive Sentence Structure ....................................................................................... 88	6.5	 Structure of Bare Quantifiers ........................................................................................... 89	6.5.1	 Bare Quantifiers without Determiners ...................................................................... 89	7. STRONG VS. WEAK QUANTIFIER DISTINCTIONS ............................... 92	7.1	 Theoretical Background ................................................................................................... 92	7.2	 Cardinal and Proportional Readings in Kwak’wala ......................................................... 97	7.2.1	 S1’s Judgments ......................................................................................................... 97	7.2.2	 S2’s Judgments ....................................................................................................... 101	7.3	 Analysis .......................................................................................................................... 104	7.3.1	 Strong/Weak Distinction and Predicative Status of Kwak’wala Quantifiers .......... 104	7.3.2	 Strong/Weak Distinction and Cardinal and Proportional Readings in Kwak’wala 105	7.3.3	 Position Mismatches ............................................................................................... 106	8. QUANTIFIER SCOPE ................................................................................... 108	8.1	 Theoretical Background ................................................................................................. 108	8.2	 Collective, Distributive, and Cumulative Readings in Kwak’wala ............................... 109	8.3	 Analysis .......................................................................................................................... 120	  vii 9. IMPLICATIONS AND CONCLUSION ........................................................ 123	9.1	 Differences in Quantifier Status between Salish and Kwak’wala ................................. 123	9.2	 Conclusion ..................................................................................................................... 123	References ...................................................................................................................................125	Appendices ..................................................................................................................................127	Appendix A Partitive Expressions ........................................................................................................ 127	Appendix B Collective, Distributive, and Cumulative Readings (Additional) ..................................... 129	Appendix C Tsitsa̱ḵala̱m sa Ruby ......................................................................................................... 137	   viii List of Tables  Table 1 Properties of Strong/Weak Quantifiers .............................................................................. 3	Table 2 Cardinal/Proportional Conditions and Felicity Judgments ................................................ 8	Table 3 Collective, Distributive, and Cumulative Readings ......................................................... 13	Table 4 Inverse Distributive Reading ........................................................................................... 13	Table 5 Consonant Inventory ........................................................................................................ 18	Table 6 Vowel Inventory ............................................................................................................... 18	Table 7 Determiner System ........................................................................................................... 26	Table 8 Summary of Distribution: Predicative Ḵ̓ina̱m .................................................................. 34	Table 9 Summary of Distribution: DP-Internal Ḵ̓ina̱m (Non-Subject) ......................................... 35	Table 10 Summary of Distribution: DP-Internal Ḵ̓ina̱m (Subject) ............................................... 35	Table 11 Summary of Distribution: DP-Internal Ḵ̓ina̱m (interaction with adjectives) ................. 36	Table 12 Summary of Distribution: Bare Quantifier Ḵ̓ina̱m with Determiners ............................ 41	Table 13 Summary of Distribution: Bare Quantifier Ḵ̓ina̱m without Determiners ....................... 42	Table 14 Summary of Distribution: Predicative Wi'la .................................................................. 48	Table 15 Summary of Distribution: DP-Internal Wi'la (Non-Subject) ......................................... 48	Table 16 Summary of Distribution: DP-Internal Wi'la (Subject) .................................................. 48	Table 17 Summary of Distribution: Bare Quantifier Wi'la with Determiners .............................. 53	Table 18 Summary of Distribution: Bare Quantifier Wi'la without Determiners ......................... 53	Table 19 Summary of Distribution: Predicative Numerals ........................................................... 60	Table 20 Summary of Distribution: DP-Internal Numerals (Non-Subject) .................................. 60	Table 21 Summary of Distribution: DP-Internal Numerals (Subject) .......................................... 60	Table 22 Summary of Distribution: DP-Internal Numerals (interaction with adjectives) ............ 61	Table 23 Summary of Distribution: Bare Quantifier Numerals with Determiners ....................... 65	Table 24 Summary of Distribution: Bare Quantifier Numerals without Determiners .................. 65	Table 25 Summary of the Distribution of Quantificational Elements .......................................... 67	Table 26 Summary of Distribution: Bare Quantifiers without Determiners ................................. 90	Table 27 Properties of Strong/Weak Quantifiers .......................................................................... 92	Table 28 Truth Conditions of Cardinal/Proportional Readings (Partee 1988) .............................. 95	  ix Table 29 S1’s Judgments on Cardinal/Proportional Readings .................................................... 101	Table 30 S2’s Judgments on Cardinal/Proportional Readings .................................................... 104	Table 31 Strong and Weak Distinction in Kwak’wala Quantifiers ............................................. 106	Table 32 Readings with Nominative Marker and Accusative Marker for S1 .............................. 111	Table 33 Readings with Nominative Marker and Accusative Marker for S2 ............................. 112	Table 34 Readings with Nominative Marker and Oblique Marker for S1 .................................. 113	Table 35 Readings with Nominative Marker and Oblique Marker for S2 .................................. 114	Table 36 Readings with Nominative Marker and Preposition Marker ....................................... 115	Table 37 Readings with Accusative Marker and Oblique Marker for S1 ................................... 116	Table 38 Readings with Accusative Marker and Oblique Marker for S2 ................................... 116	Table 39 Readings with Accusative Marker and Preposition Marker for S1 .............................. 117	Table 40 Readings with Accusative Marker and Preposition Marker for S2 .............................. 118	Table 41 Readings with Oblique Marker and Preposition Marker for S1 .................................. 119	Table 42 Readings with Oblique Marker and Preposition Marker for S2 .................................. 120	Table 43 Readings with Nominative Marker and Accusative Marker ........................................ 130	Table 44 Readings with Nominative Marker and Oblique Marker for S1 .................................. 131	Table 45 Readings with Nominative Marker and Oblique Marker for S2 .................................. 131	Table 46 Readings with Nominative Marker and Preposition Marker for S1 ............................ 132	Table 47 Readings with Nominative Marker and Preposition Marker for S2 ............................ 133	Table 48 Readings with Nominative Marker and Accusative Marker for S1 ............................. 134	Table 49 Readings with Nominative Marker and Oblique Marker for S1 .................................. 135	Table 50 Readings with Nominative Marker and Preposition Marker for S1 ............................ 136	   x List of Figures  Figure 1 DP Internal Structure with all and many .......................................................................... 4	Figure 2 Map of Kwakwaka’wakw Territories (Pasco et al., 1998) ............................................. 17	Figure 3 Basic Sentence Structure ................................................................................................ 70	Figure 4 Relationship between Pre-Predicative Subject and Auxiliaries ...................................... 72	Figure 5 Hypothesized Structure of (67a) ..................................................................................... 74	Figure 6 Hypothesized Structure of (67b) .................................................................................... 75	Figure 7 AuxSV structure (67c) .................................................................................................... 76	Figure 8 DP Internal Structure with Wi'la and Ḵ̓ina̱m/Numerals .................................................. 83	Figure 9 Possibility for Pre-determiner Ḵ̓ina̱m and Numerals ...................................................... 85	Figure 10 Subject can move up ..................................................................................................... 86	Figure 11 Object cannot move up ................................................................................................. 88	Figure 12 Object fills in the subject position ................................................................................ 89	Figure 13 Bare Ḵ̓ina̱m at Aux ....................................................................................................... 91	Figure 14 Set Theoretic Representation of “All women danced.” ................................................ 93	Figure 15 Set Theoretic Representation of “Many women danced.” ........................................... 94	Figure 16 DP Internal Structure with Wi'la and Ḵ̓ina̱m/Numerals .............................................. 107	Figure 17 Logical Form: Relations of Quantifiers in English .................................................... 109	Figure 18 Logical Form: Relations of Kwak’wala Quantifiers .................................................. 121	   xi Glossary ACC    accusative marker AUX   auxiliary DET    determiner DISC   discourse marker DIST   distal invisible FUT    future LOC    locative MED   medial visible NEG    negation NOM   nominative OBL    oblique marker OBJ    object PART   partitive PASS   passive PERF   perfective POSS   possessive PREP   preposition PRES   present PRON   pronoun PST    past RED    reduplication SUB    subject   xii TEMP   temporal VIS    visibility   xiii Acknowledgements  I am deeply grateful to my wonderful consultants Ruby Dawson-Cranmer and the late Freda Shaughnessy for their patience in sharing their great language and culture with me. Without their kindness, knowledge, and willingness to share, this work would not have been possible.  I am grateful to my thesis committee: Henry Davis, Lisa Matthewson, and Rose-Marie Déchaine.  Henry, you introduced me not only to Kwak’wala in my first year Field Methods class, but also to Linguistics in undergrad. Without you, I never would have wanted to be a linguist, and work on a First Nations language.  Lisa, the semantics classes you taught both in undergrad and grad school were the most enjoyable and inspiring times I have ever had at UBC. You also gave me a lot of academic and emotional encouragement as a co-supervisor. Rose-Marie, you have been an invaluable resource as my first audience and have shown me different, important perspectives along the way. Thank you all.  Thanks are also due to Martina Wiltschko for her illuminating course with Rose-Marie on pronouns and noun phrases, and Michael Rochemont for his encouragement and support in my first year working on Kwak’wala. I am also indebted to the Linguistics Department of the University of British Columbia for support with various aspects of my program.  A big thanks goes to the audiences at the 45th International Conference on Salish and Neighboring Languages, the 2010 Linguistic Society of Japan Summer Seminar, and the participants in the Field Methods course at Kyusyu University, where I received many comments and suggestions that were helpful in writing this thesis. I would also like to thank Jonathan Janzan for sharing his expertise in Kwak’wala and the unpublished stories collected from our consultant, and Kento Nagatsugu and Noriyuki Ikeda for discussion after the talk at Kyusyu   xiv University. I am deeply grateful to Ayumi Ueyama, for having long discussions with me on the syntax chapter when I visited Kyusyu University.  I would also like to thank the Jacobs Research Fund who supported my fieldwork.  Finally, I am very grateful to my company for giving me extra leave to complete this thesis. Thank you to all of my friends and family for your support, encouragement and patience throughout my time in this program.   xv Dedication  To Ruby and Freda, who are very cheerful, fun, kind, and patient when working with me. This thesis is a memorial to Freda.    1 1. INTRODUCTION This chapter outlines the objectives, the diagnostics, and the fieldwork methodology of this thesis.  1.1 Objectives The goal of this thesis is to describe and analyze the syntax and semantics of quantification in Kwak’wala. Although there are lists of Kwak’wala quantifiers and numerals in various places, including the Kwak'wala Community Portal of FirstVoices (http://www.firstvoices.com/en/Kwakwala), previous research dedicated to quantification in Kwak’wala is non-existent. During my fieldwork, I elicited the following set of quantifiers (1).  1 Lists of Kwak’wala Quantifiers  wi'la (‘all’)           naxwa (‘everything’) alax̱ (‘most’)  -g̱awe (‘most’)       ḵ̓ina̱m  ( ‘many’) 'walas (‘big/lots’) o'lan (‘only’) higa'a̱m (‘only’) hulatli (‘few’) hulat (‘less’) hulałgawe' (‘least’) =kas (‘lots’)         ´na̱m (‘one’) ma'ł (‘two’) yuda̱xw (‘three’) mu (‘four’) sa̱k'a (‘five’) k'a̱t'ła (‘six’) adła̱bu (‘seven’) 'ma̱'łgwa̱'nał (‘eight’) na'na̱'ma (‘nine’) la̱'stu (‘ten’)      This thesis focuses on a subset of these quantifiers, namely, wi'la (‘all’), ḵ̓ina̱m (‘many’), and the numerals ('na̱m (‘one’) - la̱'stu (‘ten’)).    2 This thesis examines the syntax-semantics interface of quantifiers in Kwak’wala, and in particular whether certain semantic properties of quantificational expressions can be predicted from their structure and/or syntactic position. The specific goals of this thesis are the following:  2 (a)  Describe the distribution of quantifiers inside DPs in argument position.  (b)  Characterize the semantic properties of three quantifiers.   (c)  Investigate the mechanism that is involved when quantified DPs and bare quantifiers occur clause-initially, preceding the main predicate, and examine the syntactic  position of clause-initial quantifier expressions.   (d)  Provide an analysis of syntax-semantics interactions and syntax-semantics mismatches.  1.2 Diagnostics This thesis assumes a standard account of quantification, according to which quantifiers can be divided into two classes: strong and weak (Barwise and Cooper (1981), Keenan (1996), Milsark (1979), Partee (1988)). Strong quantifiers include, for example, all, while weak quantifiers include, for example, many or few. Table 1 lists the properties of English strong and weak quantifiers: the first and second properties listed are related to syntax, and the third property is related to semantics. In the table, the weak quantifier many is listed into two columns according to whether it behaves semantically as weak or strong, i.e., whether it has a strong (proportional) reading or a weak (cardinal) reading. This is further discussed in section 1.2.2.1.   3 Property Strong all Strong many Weak many Can be one-place main predicate No No Yes1 DP-Internal Position Pre-Determiner Pre-Determiner Post-Determiner Cardinal or Proportional Readings Proportional Proportional Cardinal  Table 1 Properties of Strong/Weak Quantifiers  Taking the strong/weak distinction as a starting point, this thesis will document the syntactic and semantic features of Kwak’wala quantifiers.  This thesis will also investigate another characteristic of quantifiers: quantifier scope, which is described in 1.2.2.2.  1.2.1 Syntax In this section I outline the diagnostics which are applied in later chapters. I investigate the syntactic features of each quantifier and discuss the distinctions between the quantifiers. In particular, I lay out the possible positions for each quantifier either within or outside argument positions in chapters 3, 4, and 5, and I examine the reasons for this distribution and provide a syntactic account of Kwak’wala quantifiers in chapter 6. External to the DP, the most notable distinction between strong and weak quantifiers is whether or not they can be one-place main predicates, as shown in the first property in Table 1. Milsark (1979) uses the there-insertion test to examine if a quantifier can function as a one-place predicate. For example, the weak quantifier many is compatible with there-insertion (3a), but the strong quantifier all is not (3b).  3 (a)  There are many cats. (b)  *There are all the cats.                                                 1 It is quite awkward to say e.g. ‘The men are many’ in English, so instead we use there-insertion: ‘there are many men’, which is the equivalent of ‘predicative’ uses of weak quantifiers in NW Languages. See the discussion in section 1.2.1 below.   4 This test relates to the number of arguments the quantifiers can take. Since (the) cats is an argument, but there is not, (3a) is felicitous as there is only one argument which the quantifier many takes. (3b), on the other hand, is infelicitous since the quantifier all needs to take two arguments. The distinction between strong and weak quantifiers in terms of being internal to the DP, as shown in the second line of Table 1, is exemplified in (4). In (4a), the quantifier precedes the determiner, while in (4b) the quantifier follows the determiner.  4 (a)  all the men  (b)  the many men  The position of quantifiers in relation to determiners has been investigated in many languages. Weak quantifiers, such as many, typically behave like adjectives in adjoining to NP, while strong quantifiers like all typically adjoin to DP.  (a)          DP      QP                   DP       Q              D          NP    all            the                                    men (b)          DP       D                    NP   the                     AdjP         NP                         Adj          men                    many  Figure 1 DP Internal Structure with all and many    5 As the tree in Figure 1 (b) indicates, weak quantifiers like many are often treated as adjectives. For this reason, the interaction of weak quantifiers with genuine adjectives will be described in sections 3.2.2 and 5.2.2. In addition, this thesis describes the distribution of quantifiers when they appear clause-initially. Kwak’wala is usually characterized as a VSO language but quantified DP subjects prefer to appear clause-initially. My consultants also allow bare quantifiers to appear at the beginning of the clause. Therefore, this thesis aims to examine the syntactic position of clause-initial quantifier expressions, and to find the mechanism that causes these expressions to occur clause-initially.  1.2.2 Semantics This thesis provides a semantic description of quantifiers in Kwak’wala, as shown in (5).  5 (a)  Semantic Correlates of the Strong vs. Weak Distinction  (b)  Quantifier Scope  To investigate these characteristics, it is necessary to examine the kinds of readings that quantified expressions have, including those given in (6).  6 (a)  Cardinal and Proportional readings (b)  Collective, Distributive, and Cumulative Readings of quantified DPs    6 1.2.2.1 Semantic Correlates of the Strong vs. Weak Distinction in English The third property in Table 1 has to do with a well-known semantic difference between strong and weak quantifiers. As the names indicate, a cardinal reading is a reading related to the cardinality of elements which are quantified over (in other words, how many elements there are in total: for example, 5), while a proportional reading is related to the proportionality of elements (in other words, how many elements there are out of a total number of elements: for example, 5 out of 9). Semantically, strong quantifiers, such as all or most, have only a proportional reading; example (7) is true when the number of dogs who barked is proportionally big, say 10 out of 12.  7 Most of the dogs barked.  Weak quantifiers such as many, on the other hand, are ambiguous between a cardinal reading and a proportional reading. For example, consider the following sentences. The sentence (8a) has a cardinal reading while the sentence in (8b) has a proportional reading.  8 (a)  There are many dogs barking. (b)  Many of the dogs barked.  Example (8a) is true when the number of dogs who barked is cardinally big but proportionally small, say 100 out of 500, but it is false when the number of dogs who barked is cardinally small, say 5. Example (8b) is false in both these contents. In order to understand why this happens, I examine cardinal and proportional readings, as illustrated in the English example   7 below. (Note that when I elicited these judgments, discourse contexts were always accompanied by visual representations.)  9 Sentence: John gave many bunches of grapes to Pamela. Question to the Speaker: Do you accept or reject the sentence in the following situation? (i) where John has 5 bunches of grapes and gave Pamela 2 of them (i.e., small cardinality, small proportion).  (1)  (2)   (ii) where John has 5 bunches of grapes and gave Pamela 4 of them (i.e., small cardinality, large proportion).  (1)  (2)      8 (iii) where John has 40 bunches of grapes and gave Pamela 10 of them (i.e., large cardinality, small proportion).  (1) (2)  (iv) where John has 40 bunches of grapes and gave Pamela 35 of them (i.e., large cardinality, large proportion).     (1)  (2)   The predictions and results for English are shown in Table 2.  Situation Cardinal Prediction Proportional Prediction Result for English (i) small cardinality, small proportion # # # (ii) small cardinality, large proportion # √ √ (iii) large cardinality, small proportion √ # √ (iv) large cardinality, large proportion √ √ √ Table 2 Cardinal/Proportional Conditions and Felicity Judgments  The symbol “#” in the table indicates that the example is infelicitous in the given context, and not that it is ungrammatical.  The prediction for any language is that if the sentence has only   9 a proportional reading, then that sentence is felicitous in the second and fourth situations, but not in the third situation. On the other hand, if a sentence has only a cardinal reading, then that sentence is felicitous in the third and fourth situations, but not in the second situation. If a sentence is ambiguous between both proportional and cardinal readings, then that sentence is felicitous in the second, the third, and the fourth situations. Therefore, the judgment of English speakers is that sentences like (9) are felicitous in situations with small cardinality and large proportion, and also in situations with large cardinality and small proportion, because many has both cardinal and proportional readings. Whenever either reading fits the context, the sentence is felicitous, as predicted. This thesis explores the same type of distinction between strong and weak readings for the Kwak’wala weak quantifier ḵ̓ina̱m ‘many’.  1.2.2.2 Quantifier Scope May (1985) argues that scope is to be defined syntactically. If so, then ambiguous scope must involve two different structural relations. This means that there are two syntactic structures for a sentence when the sentence is ambiguous, since a quantifier has scope over everything it c-commands. Consider the following English sentence:  10 Someone loves everyone.  This sentence has two readings. In one reading, someone has wide scope over everyone [someone > everyone], which means that there is one person who loves everyone. In the other reading, everyone has wide scope over someone [everyone > someone], which means that   10 everyone is loved by a potentially different person. One goal of this thesis is to determine whether or not Kwak’wala has similar patterns of quantifier scope.  1.2.2.3 Collective, Distributive, and Cumulative Readings of English Quantified DPs To investigate quantifier scope, I examine collective, distributive, and cumulative readings, as shown by the English examples in (11-12). The discourse contexts are always presented with visual representations as in these examples. I tested this with English speakers.  11 Situation 1: Ann, Jessica, Kate, and Beth helped their teacher move.  Sentence: Four women carried five boxes to her house.  Question to the Speaker: Do you accept or reject the sentence in the following situation?  Context (i): four women carried five boxes together (i.e., 4 women, 5 boxes).  (1)  (2)    11 (3)  (4)  (5)  (6)    Context (ii): four women each carried five boxes  (i.e., 4 women, 20 boxes).   (1)  (2)    12 (3)  (4)  (5)  (6)   Context (iii): there are four women and they carried a total of five boxes to another woman’s house, but you do not know who carried how many boxes (i.e., 4 women, 5 boxes).  (1)  (2)        13 The Number of Boxes Possible Reading Scopal Configuration Ann Jessica Kate Beth 5 (Together) Collective √ [four, five] (scopeless) 5 5 5 5 Distributive √ [four > five] 5 (Separately) Cumulative √ [four, five] (scopeless) Table 3 Collective, Distributive, and Cumulative Readings  12 Situation 2: The teacher asked more women to help her because the boxes are heavy.  Sentence: Four women carried five boxes to her (another woman’s) house.  Question to the Speaker: Do you accept/reject the sentence in the following situation?  Context (iv): four different women carried five boxes to another woman’s house  (i.e., 20 women, 5 boxes).  (1) (2)    The Number of Women Possible Reading Scopal  Configuration BoxA BoxB BoxC BoxD BoxE 4 4 4 4 4 Inverse Distributive √ [five > four] Table 4 Inverse Distributive Reading  The notation ‘>’ in Table 3 and Table 4 means the quantifier on the left has wider scope than the one on the right. For this test, I made sure every example had at least one quantifier that was   14 a numeral; using numbers makes it easier to establish whether or not I have a distributive reading because I can use counting.  1.3 Methodology The data in this thesis were gathered through the author’s fieldwork, which took place from 2010/01-2011/09 and 2015/09-11, via the standard methods of controlled elicitation, adapted from Matthewson (1998, 2004) and Bruening (2008). These consist of i) translations from English to Kwak’wala, ii) grammaticality judgments on researcher-constructed Kwak’wala utterances, and iii) acceptability judgments of Kwak’wala sentences in discourse contexts enhanced by visual stimuli (pictures). The data presented throughout this thesis are categorized into five types: utterances that were volunteered by the speakers (√√), utterances that were accepted by the speakers (√), utterances that were sometimes accepted by the speakers (?), utterances that were judged as ungrammatical by the speakers (*), and utterances that were judged as grammatical but infelicitous (#). “-” is used where relevant data is missing. The discourse contexts were presented verbally or with visual representations such as Lego, drawings, or small origami dolls.  The elicitor asked whether certain constructed sentences were felicitous in the constructed contexts. In particular, the grammaticality and acceptability judgments were tested several times in different situations on different days, since they were not easy to get.  Two Kwak’wala speakers acted as my consultants. They did an excellent job, carefully considering my questions and giving many comments to help my understanding. I refer to them throughout this thesis as S1 and S2. The data are presented in the following format.    15 13 S1 Judgment  (√√, √, ?, *, #, or -)    /    S2 Judgment  (√√, √, ?, *, #, or -) Kwak’wala Example (in the practical orthography, explained in 2.1) Morphological Representation ‘English Translation’  All errors are my responsibility.    16 2. KWAK’WALA Kwak’wala, also known as Kwakiutl, is the indigenous language spoken in the northern part of Vancouver Island and the adjacent mainland of British Columbia in Canada. It belongs to the Northern Wakashan language group, which also includes Haisla, Heiltsuk, and Oowekyala. There are fewer than 200 fluent speakers and most speakers are elders, so documentation is urgent and important as the long-term viability of the language is in question. The people who speak this language are called Kwakwaka’wakw. There are five dialects (Kwak’wala, Liḵ’wala, T’ła̱t’łasik’wa̱la,’Nak’wala, and G̱uts’ala), one of which is called by the same name as the whole language. The map below shows the different territories and dialect areas of Kwak’wala (Pasco et al. 1998). In the following map, the dialect of my two language consultants (the Kwak’wala dialect) is coloured in orange. These two consultants are Ruby Dawson-Cranmer from Gwa’yi (Kingcome Inlet), and Freda Shaughnessy from Gilford Island. I call the former S1 and the latter S2 through this thesis. Most unfortunately and sadly, however, Freda Shaughnessy passed away before I could complete the data collection, so some data are missing.    17  Figure 2 Map of Kwakwaka’wakw Territories (Pasco et al., 1998)  The next sections discuss orthographic issues, and introduce some core properties of Kwak'wala syntax, including core word orders and determiner distribution.  2.1 Orthography The American Phonetic Alphabet (APA) and the practical orthography developed by the U’mista Centre in Alert Bay are both used to represent the Kwak’wala language. The latter,   18 which is used by several Kwak’wala communities, is the one used in this thesis. The phonological inventory of Kwak’wala is given below in the U’mista orthography.   Bilabial Alveolar Lateral Velar Labialized Velar Uvular Labialized Uvular Glottal Stops p    b t    d  k    g kw  gw ḵ   g̱ ḵw   g̱w ' Ejectives p̓ t̓  k̓ k̓w ḵ̓ ḵ̓w  Affricates  ts   dz tł   dł      Ejective Affricates  t̓s t̓ł      Fricatives  s ł x xw x̱ x̱w h Sonorants m n l y w    Glottalized Sonorants 'm 'n 'l 'y 'w    Table 5 Consonant Inventory   Front Central Back High i  u Mid e a̱ o Low  a  Table 6 Vowel Inventory  2.2 Core Word Order Kwak’wala is usually characterized as a VSO (Verb (or Predicate) Subject Object) language. The sentence in (14) shows the core clausal word order without an auxiliary. All open class elements, which correspond to nouns, verbs, and adjectives in English, can function as the main predicate of a sentence.    19 14 S1 √√    /    S2  √√ [Predicate      Subject      Object       Oblique]  ga̱ls=i         Mary=x̱a    gukw=sa    ga̱lyayu.2   paint=DIST    Mary=ACC  house=OBL brush  ‘Mary painted a house with the brush.’   Main predicates may be preceded by a series of auxiliaries (Anderson 1984). According to Anderson, auxiliaries include the semantically empty element la̱, negation, and motion predicates. I define auxiliaries as verbal elements or verbal supplementary elements, which precede the main predicate in a monoclausal structure and share a single set of subject agreement markers with the main verb3. (15a) shows the auxiliary la̱ in a sentence with the core word order. When a sentence has an auxiliary, such as la̱, then it is possible for the subject to sit between the auxiliary and the predicate, as shown in (15b). Both S1 and S2 accepted the sentences in (15) as grammatical.  15 (a)  S1  √    /    S2  √ (with surprise intonation)  [Auxiliary     Predicate      Subject      Accusative   Oblique] la̱-'m       ga̱ls=i        Mary=x̱a    gukw=sa    ga̱lyayu. AUX-DISC  paint=DIST    Mary=ACC  house=OBL brush ‘Mary painted a house with the brush.’                                                    2 Clitic boundaries are marked by an equals sign (=), and affix boundaries are marked by a hyphen (-). 3 In the configuration AUX=LOC V=LOC DP, the pre-predicative locative marker (usually =ox̱, but also =i) acts as agreement doubling the locative marker on the post-predicative subject (Littell 2016).   20 (b)  S1  √    /    S2  √ (with surprise intonation)  [Auxiliary           Subject   Predicate      Accusative    Oblique] la̱-'m=i            Mary     ga̱lsa=x̱a     gukw=sa     ga̱lyayu. AUX-DISC=DIST    Mary     paint=ACC    house=OBL  brush ‘Mary painted a house with the brush.’  For S2 these sentences were only acceptable in a certain context4, due to the discourse marker 'm (Littell 2016), which is always present with this word order. Another kind of auxiliary is the negative auxiliary k̓i's. (16a) shows this auxiliary in a sentence with the core word order.  (16b) shows that the subject can sit between the negation and the predicate.  16 (a)  S1  √    /    S2  √ [Negation    Predicate         Subject      Accusative    Oblique] k̓i's       ga̱ls=i          Mary=x̱a    gukw=sa     ga̱lyayu. NEG       paint=DIST      Mary=ACC  house=OBL  brush ‘Mary did not paint a house with the brush.’   (b)  S1  √√    /    S2  - [Negation       Subject    Predicate    Accusative    Oblique] k̓i's=i         Mary     ga̱lsa=x̱a    gukw=sa     ga̱lyayu. NEG=DIST    Mary     paint=ACC   house=OBL  brush ‘Mary did not paint a house with the brush.’                                                  4 S2 said that the sentence cannot be produced if it is just a statement. If she thought that Mary could paint the house and just tells me about it, she cannot put la̱ at the front of the sentence. However, if she did not think that Mary could, then she is surprised by the fact that Mary did it and tells me with special intonation.   21 It is possible for more than one auxiliary to occur in a sentence, although they then occur in a fixed order. For example, the negative auxiliary can occur with a modal auxiliary like weł (‘can’). In sentences with more than one auxiliary, there are a number of possible word orders. The subject can occur after the predicate (17a), before the predicate following both auxiliaries (17b), in between the two auxiliaries (17c), or before the auxiliaries (17d). These sentences were all acceptable to S1. S2 accepted (17b) but I do not have other relevant data for S2 for these sentences.  17 (a)  S1  √√    /    S2  -  [Negation  Auxiliary   Predicate        Subject       Accusative   Oblique] k̓i's      weł       ga̱ls=i          Mary=x̱a     gukw=sa     ga̱lyayu. NEG      can       paint=DIST       Mary =ACC   house=OBL  brush ‘Mary could not paint a house with the brush.’  (b)  S1  √    /    S2  √ [Negation    Auxiliary      Subject    Predicate   Accusative   Oblique] k̓i's       weł=i        Mary     ga̱lsa=x̱a    gukw=sa     ga̱lyayu. NEG       can=DIST     Mary     paint=ACC   house=OBL  brush ‘Mary could not paint a house with the brush.’   (c)  S1 √    /    S2  - [Negation       Subject Auxiliary  Predicate   Accusative    Oblique] k̓i's=i         Mary    weł       ga̱lsa=x̱a    gukw=sa     ga̱lyayu. NEG=DIST    Mary    can        paint=ACC   house=OBL  brush ‘Mary could not paint a house with the brush.’       22 (d)  S1  √    /    S2  -  [Subject   Negation     Auxiliary  Predicate     Accusative   Oblique] Mary    k̓i's=i         weł       ga̱lsa=x̱a    gukw=sa     ga̱lyayu. Mary    NEG=DIST    can        paint=ACC   house=OBL  brush ‘Mary could not paint a house with the brush.’   The order k̓i's weł (NEG can) is the only possible order of auxiliaries; the modal cannot precede the negation, giving an alternative scope (can > NEG). As mentioned above, the core clausal word order without an auxiliary is VSO, shown in (14), repeated in (18a). In some instances, it is also possible for the subject to precede the main predicate even when no auxiliary is present (18b-c).  18  (a) S1  √√    /    S2  √√ [Predicate       Subject    Object       Oblique] ga̱ls=i         Mary=x̱a    gukw=sa    ga̱lyayu.  paint=DIST     Mary=ACC  house=OBL brush ‘Mary painted a house with the brush.’  (b)  S1  √    /    S2  *  [Subject   Predicate     Object       Oblique] Mary    ga̱ls=x̱a     gukw=sa    ga̱lyayu. Mary    paint=ACC  house=OBL brush ‘Mary painted a house with the brush.’  (c)  S1  (√) (not explicitly mentioned)  /    S2  √ [Subject,    Predicate    Object       Oblique] Mary,     ga̱ls=x̱a     gukw=sa    ga̱lyayu.  Mary,     paint=ACC  house=OBL brush ‘Mary painted a house with the brush.’    23 S1 accepted (18b), but S2 did not. S2 said that the sentence would be acceptable if there was a comma after ‘Mary’, as shown in (18c). The comma corresponds to an intonation break. The examples in (18) involve a subject that is a proper noun.  If the subject is a non-quantified common noun, it usually cannot occur clause-initially. The examples in (19) show how this works. (19a) has the typical predicate-subject word order, and it is acceptable to both consultants. (19b) has a clause-initial subject, and it was rejected by both consultants. However, if there is a pause (shown by a comma) between the subject ba̱gwana̱m (‘man’) and the predicate, then subject-predicate word order was marginally accepted by S1 only.  19 (a)  S1  √√    /    S2  √√ [Predicate       Subject] t'sa̱k'w=i        ba̱gwana̱m short=3DIST     man ‘The man is short.’  (b)  S1  *    /    S2  * [Predicate     Subject] ba̱gwana̱m      t'sa̱k'w=i        man            short=3DIST       ‘The man is short.’      24 (c)  S1  ?    /    S2  * [Predicate,     Subject] ba̱gwana̱m,      t'sa̱k'wa        man            short        ‘The man is short.’ S1’s Comment: you can say it, by pointing to a man.5  These data show that S1 and S2 have slightly different grammars. Although they both accept predicate-initial sentences, they have different grammaticality judgments for subject-initial sentences. For subject-initial sentences where the subject is a proper noun, S1 always judges them as grammatical (although she prefers predicate-initial sentences). It makes no difference whether or not there is a pause after the subject. S2, on the other hand, only accepts these kinds of sentences if there is a pause after the subject. For subject-initial sentences where the subject is not a proper noun, S1 accepts them only if there is a pause after the subject. S2 does not accept them, even with a pause. So far the subjects I have examined have limited acceptability in pre-predicate position. Turning next to quantified subjects, I show in sections 3.3.1, 4.3.1, and 5.3.1 that these can precede the main predicate, with or without a pause. In fact, S1 and S2 not only accept, but also voluntarily offer these sentences. The following summarizes the generalizations about word order in Kwak’wala discussed above.                                                    5 These constituents might not form a sentence, saying ‘A man! He’s short!’ However, S2 does not accept them, even with a pause.   25 20 (a)  Kwak’wala is a VS(O) (Predicate Subject (Object)) language, and nouns, verbs, and      adjectives can function as the main predicate of a sentence.  (b)  Main predicates may be preceded by a series of auxiliaries; subjects may occur following auxiliaries and preceding the main predicate.  (c)  Subjects can appear clause-initially under restricted circumstances: if they are  quantificational, or with a pause for some speakers.  2.3 Determiner System The complex determiner system morphology of Kwak’wala has been discussed previously by a variety of researchers, including Anderson (1984, 2005), Bach (2006), Chung (2007), Nicolson and Werle (2009), and Black (2011). The features of a DP include case, location, definiteness, tense, and visibility, which are split into pre- and post-nominal positions. The basic structure of the DP is schematized in (21a) and exemplified in (21b) (Black 2011).6  21 (a)  [DP =Case=Locative=Demonstrative   [NP Noun=Temporal=Visibility]]      PRENOMINAL Determiner        POSTNOMINAL Determiner  (b)  S1  √    /    S2  - x̱=ox̱=da                         gukw=t̓ł=ox̱      CASE=LOCATIVE=DEMONSTRATIVE  noun=TEMPORAL=VISIBILITY ACC=MED=DET                  house=FUT=2VIS  ‘that future-to-be house (which is not built yet)’                                                  6 Littell (2016) argues that the temporal elements are clausal, not part of the D system.   26 (22) shows that pre-nominal determiners encliticize to the preceding words.  22 S1 √√ t'sa̱xḵ̓ [DP =ox̱=da     [NP pa̱lawas=ex̱]] sick=MED=DET         flower=MEDVIS   ‘The flower is sick.’  Table 7 shows the full set of determiner elements. The table is a modification of Boas (1947) and Black (2009, 2011).  Prenominal Postnominal Distance Case Loc Dem Temp Vis Proximate visible -∅   (Nom) -ga  -wa̱ł (Dist.Pst) -x Medial visible -x̱ (a) (Acc) -ox̱ -da (Det) -∅ (Pres) -ix̱, ex̱, -x̱ Distal invisible -s(a)  (Obl) -i  -t̓ł (Fut) -a/-e’ Table 7 Determiner System   For the purposes of this thesis, as the brackets in (21) indicate, I will treat the prenominal determiners as a set sitting in D, and the postnominal determiners as another set inside NP. I do this in spite of doubts about whether each element is actually in the same projection as the others. A finer-grained analysis might treat them individually as heads of functional projections, with the ones on the left higher in the tree than those on the right, as proposed by Chung (2007). I cannot tell where exactly the determiners sit, but the ones on the left are somewhere in the general vicinity of D, and the ones on the right are in the vicinity of NP. However, their exact positions in DP are not relevant to the discussion of quantifiers so I do not go further into this problem.    27 2.4 Reduplication There is another feature represented in nominal phrases: reduplication is used to indicate the plural in this language. Double reduplication is frequently seen on ‘men’ in S1’s utterances. This form is no different in meaning from single reduplication.  23 S1 √ ha̱'m-x'id  [DP =i=da         [NP bi-bi-ba̱gwana̱m]]=x̱a   ka̱leps. eat-PERF=DIST=DET            RED-RED-man=ACC    grape ‘Men ate grapes.’  S2 sometimes mentioned that, with quantifiers, a plural noun may appear without reduplication.  24 S1  √    /    S2  √√ t̓soya    Ruby=sa     ḵ̓ina̱m    buk    lax̱     Ayako.  give     Ruby=OBL   many     book   PREP   Ayako ‘Ruby gave many books to Ayako.’     28 3. Ḵ̓INA̱M (‘MANY’) IS A WEAK QUANTIFIER In this section, I describe the distribution of ḵ̓ina̱m (‘many’). I begin in section 3.1 by examining ḵ̓ina̱m in predicative contexts. In section 3.2, I look at the internal structure of quantified DPs containing ḵ̓ina̱m. In particular, in section 3.2.1, I discuss quantified DPs with ḵ̓ina̱m in each argument position, and in section 3.2.2 I focus on the interaction of adjectives with ḵ̓ina̱m inside DP. In section 3.3, I examine clause-initial quantifiers instances of ḵ̓ina̱m, including clause-initial quantifiers inside DPs and bare ḵ̓ina̱m, which are investigated in sections 3.3.1 and 3.3.2, respectively.  3.1 Predicative Status of Ḵ̓ina̱m The quantifier ḵ̓ina̱m (‘many’) can be the main predicate of a sentence. This is shown in (25). This case is equivalent to an existential sentence in English.  25 S1  √√    /    S2  √√ [Q      D             Sub      ] ḵ̓ina̱m=ox̱=da        gi-gukw. many=MED=DET     RED-house ‘There are many houses.’  Because it can be a main predicate, I assume that ḵ̓ina̱m is a weak quantifier.   29 3.2 The Internal Structure of Quantified DPs: Ḵ̓ina̱m In this section, I investigate the internal structure of quantified DPs7, specifically the relative ordering of quantifiers and pre-determiners (3.2.1), as well as that of quantifiers and adjectives (3.2.2).   3.2.1 Ḵ̓ina̱m in Pre- vs. Post-determiner Position The following examples show grammaticality judgments for sentences with quantified NPs in various positions. In the sentences in (26-29), ḵ̓ina̱m occurs in each possible position within an argument DP. Ḵ̓ina̱m normally appears after the prenominal determiner when it follows the main predicate. However, it can appear before the prenominal determiner when it modifies subjects. The examples in (26) illustrate a quantified object (with accusative case-marker). Both speakers reject ḵ̓ina̱m in pre-determiner position (a) but accept it in post-determiner position (b).  26 ḵ̓ina̱m (‘many’): pre- and post-determiner position: Accusative marker (a)  S1  *    /    S2  *   [Verb              Sub     [    Q                  D         Obj8       ]] ha̱'m-x'id=i       Ruby  [DP ḵ̓ina̱m [DP =x̱a  [NP  aya̱ndzis]]. eat-PERF=DIST    Ruby       many=ACC        orange ‘Ruby ate many oranges.’                                                      7 There is also a partitive construction in Kwak’wala, but the expressions are only attested for one of my consultants. This is discussed briefly in Appendix A. 8 ‘Obj’ here indicates that the whole DP functions as the direct object.   30  (b)  S1  √√    /    S2  √ [Verb               Sub  [      D     Q           Obj    ]] ha̱'m-x'id=i       Ruby[DP =x̱a [QP ḵ̓ina̱m [NP aya̱ndzis]]. eat-PERF=DIST    Ruby=ACC     many       oranges ‘Ruby ate many oranges.’  The examples in (27) illustrate a quantified object (with oblique marker). Both speakers reject ḵ̓ina̱m in pre-determiner position (a) but accept it in post-determiner position (b).   27 ḵ̓ina̱m (‘many’): pre- and post-determiner position: Oblique marker (a)  S1  *    /    S2  *         [Verb    Sub    [     Q                 D       Obl    ]     PP                    ] t̓soya    Ruby  [DP ḵ̓ina̱m [DP =sa [NP buk]]]   lax̱     Ayako. give     Ruby      many=OBL         book    PREP9  Ayako ‘Ruby gave many books to Ayako.’  (b)  S1  √    /    S2  √ [Verb     Sub  [    D            Q              Obl   ]      PP                   ] t̓soya    Ruby[DP =sa   [DP ḵ̓ina̱m [NP buk]]]    lax̱     Ayako.  give     Ruby=OBL       many     book     PREP   Ayako  ‘Ruby gave many books to Ayako.’   The examples in (28) contain a quantified subject (with Loc (Dist) and Det markers) following the predicate. S2 rejects, but S1 sometimes accepts, the quantifier in pre-determiner position (a). Both speakers accept ḵ̓ina̱m in post-determiner position (b).                                                  9 la is a preposition with an accusative marker x̱. They occur together so lax̱ is indicated as a preposition in this thesis.   31 28  (a) S1 ? [Verb                 [    Q                D            Sub             ]         Obj    ]    ha̱'m-x'id      [DP ḵ̓ina̱m [DP =i=da   [NP bi-bi-ba̱gwana̱m]]]=x̱a   ka̱leps.        eat-PERF          many=DIST=DET      RED-RED-man=ACC     grape ‘Many men ate grapes.’  S2 *  [Verb         [     Q         D        Sub     ]    Verb       ]    gwał       [DP ḵ̓ina̱m [DP =i=da [NP ts'a̱daḵ]]]  dza̱lxwa. finish         many=DIST=DET    woman    run ‘Many women just finished running.’  (b)  S1 √ [Verb        [     D                  Q          Sub             ]       Obj   ] ha̱'m-x'id  [DP =i=da        ḵ̓ina̱m [NP bi-bi-ba̱gwana̱m]]=x̱a   ka̱leps. eat-PERF=DIST=DET        many      RED-RED-man=ACC    grape ‘Many men ate grapes.’  S2 √ [Verb   [   D            Q          Sub     ]   Verb    ]  gwał  [DP =i=da     [DP ḵ̓ina̱m [NP ts'a̱daḵ]]]  dza̱lxwa.  finish= DIST=DET      many      woman    run ‘Many women just finished running.’  The examples in (29) show a quantified subject (with Loc and Det markers) preceding the predicate and following the auxiliary. Both speakers accept ḵ̓ina̱m in pre-determiner position (a) and post-determiner position (b).      32 29 (a)  S1  √ [Aux       [ Q      D           Sub           ]    Verb         Obj    ] la̱-'m    ḵ̓ina̱m=i=da        bi-bi-ba̱gwana̱m   ha̱'map=x̱a   ka̱leps. AUX    many=DIST=DET    RED-RED-man     eat=ACC     grape ‘Many men ate grapes.’  S2  √ [Aux       [ Q       D             Sub          ]    Verb    ] la̱-'m      ḵ̓ina̱m=i=da         ts'i-ts'i-ts'a̱daḵ    ne'nakwa. AUX       many=DIST=DET     RED-RED-woman go.home ‘Many women went home.’  (b)  S1  √ [Aux [D                  Q        Sub           ]    Verb         Obj     ]  la̱-'m=i=da              ḵ̓ina̱m   bi-bi-ba̱gwana̱m   ha̱'map=x̱a   ka̱leps. AUX-DISC=DIST=DET    many    RED-RED-man     eat=ACC     grape ‘Many men ate grapes.’   S2  √ [Aux  [D                  Q         Sub          ]     Verb    ] la̱-'m=i=da              ḵ̓ina̱m    ts'i-ts'i-ts'a̱daḵ     ne'nakwa. AUX-DISC=DIST=DET    many     RED-RED-woman  go.home ‘Many women went home.’ Consultant Comment: the sentence is possible; the fact (‘many women went home!’) surprised her so she has a special intonation10.     3.2.2 Interaction of Ḵ̓ina̱m and Adjectives This section explains the distribution of ḵ̓ina̱m when it occurs DP-internally with adjectives. S1 and S2 have different judgments for these types of sentences. For S1, a quantifier following a pre-nominal D in argument position may precede an adjective, as in (30a) and (31a), but it is not                                                 10 due to the discourse marker 'm (Littell 2016).   33 required to. Examples (30b) and (31b) show that the quantifier can also follow an adjective. S1 does not offer but accepts sentences with quantifiers that follow adjectives. The meaning difference between (30a) and (30b) has not been tested.  30 (a)  S1  √√ [Verb           Sub    [     D          Q           Adj        Obj   ]] dzam-x'id     Ayako[DP =x̱a   [NP [QP ḵ̓ina̱m]    dzi-dzastu  'abals.]] plant-PERF    Ayako=ACC          many      RED-blue  apple ‘Ayako grew many blue apples.’  (b)  S1  √ [Verb           Sub    [      D       Adj            Q        Obj  ]] dzam-x'id     Ayako[DP =x̱a   [NP dzi-dzastu  [QP ḵ̓ina̱m]  'abals]] plant-PERF    Ayako=ACC        RED-blue     many    apple  31 (a)   S1  √ [Neg               Sub     Verb  [     D            Q        Adj          Obj      ]] k̓i's  weł=i       Ayako  ha̱'map[DP =x̱a   [NP [QP ḵ̓ina̱m]  a̱m-a̱me'     ki-kegas.]] NEG  can=DIST    Ayako  eat=ACC              many    RED-small   RED-cake ‘Ayako cannot eat many small cakes.’  (b)  S1  √ [Neg                Sub     Verb  [    D        Adj           Q          Obj     ]] k̓i's   weł=i       Ayako  ha̱'map[DP =x̱a   [NP  a̱m-a̱me'   [QP ḵ̓ina̱m]    ki-kegas.]] NEG   can=DIST    Ayako  eat=ACC           RED-small   many      RED-cake ‘Ayako cannot eat many small cakes.’  For S2, on the other hand, ḵ̓ina̱m in post-determiner position must precede adjectives, as shown in (32). In (32a), volunteered by S2, ḵ̓ina̱m precedes the adjective dzi-dzastu (‘blue’). The   34 sentence in (32b), on the other hand, where ḵ̓ina̱m follows the adjective dzi-dzastu (‘blue’), is rejected by S2.  32 (a)  S2  √√ [Verb           Sub    [    D          Q        Adj         Obj        ]] dzam-x'id     Ayako[DP =x̱a   [NP [QP ḵ̓ina̱m]  dzi-dzastu   i-'abals.]] plant-PERF    Ayako=ACC           many   RED-blue   RED-apple ‘Ayako grew many blue apples.’  (b)  S2  *  [Verb           Sub    [    D       Adj            Q          Obj        ]] dzam-x'id     Ayako[DP =x̱a   [NP dzi-dzastu  [QP ḵ̓ina̱m]   i-'abals]] plant-PERF    Ayako=ACC       RED-blue     many     RED-apple ‘Ayako grew many blue apples.’  3.2.3 Summary of Predicative Status and Distribution of DP-Internal Qs: Ḵ̓ina̱m The following table summarizes the distribution of predicative ḵ̓ina̱m as exemplified in 3.1. Both speakers allow ḵ̓ina̱m as the main predicate of a clause.  S1 S2 Predicative Quantifier √√ √√ QPRED D Sub Table 8 Summary of Distribution: Predicative Ḵ̓ina̱m  The following tables show the distribution of DP-Internal QPs. Looking first at quantified non-subjects, as exemplified in (26-27), both speakers reject sentences in which ḵ̓ina̱m (‘many’) occurs in pre-determiner position.     35 S1 S2 Quantified Non-Subject * * Verb   Sub   [Q   D   Object] √√ √ Verb   Sub   [D   Q   Object] * * Verb   Sub   [Q   D   Oblique] √ √ Verb   Sub   [D   Q   Oblique] Table 9 Summary of Distribution: DP-Internal Ḵ̓ina̱m (Non-Subject)  When ḵ̓ina̱m (‘many’) is associated with the subject, as exemplified in (28-29), the speakers have different judgments. S1 sometimes allows subjects with ḵ̓ina̱m in the pre-determiner position of a post-predicative DP, while S2 does not, as shown in the first line of Table 10. However, S1 initially judged sentences with ḵ̓ina̱m in the pre-determiner position of a post-predicative DP as ungrammatical, but later sometimes accepted them.   S1 S2 Quantified Subject ? * Verb   [Q   D   Sub] √ √ Verb   [D   Q   Sub] √ √ Aux   [Q   D   Sub]   Verb  √ √ Aux   [D   Q   Sub]   Verb Table 10 Summary of Distribution: DP-Internal Ḵ̓ina̱m (Subject)  The clearest difference between S1 and S2 has to do with their judgments of sentences in which quantifiers interact with adjectives, as shown in (30-32): although both speakers allow ḵ̓ina̱m in pre-adjective position, S1 allows ḵ̓ina̱m in post-adjective position while S2 does not.       36 S1 S2 Interaction with Adjective √√ √√ Verb   Sub   [D   Q    Adj   Obj] √ * Verb   Sub   [D   Adj   Q   Obj] Table 11 Summary of Distribution: DP-Internal Ḵ̓ina̱m (interaction with adjectives)  The speakers’ judgments fall into a subset relation: everything that S2 accepts, S1 accepts, but not vice-versa.  3.3 Clause-Initial Quantified DPs and Bare Quantifiers: Ḵ̓ina̱m 3.3.1 Quantified DPs So far we have looked only at main clauses. I have shown in section 3.2.1 that quantified DP subjects with ḵ̓ina̱m can precede the main predicate, as also shown in (33). Similarly, in embedded clauses, quantified subjects can also precede the predicate of the embedded sentence, as shown in (34).  33 S1  √√ [[Q          D              Sub             ]    Verb   PP                      ]]   [ḵ̓ina̱m [DP =i=da        [NP bi-bi-ba̱gwana̱m]]]   la     lax̱    gukw=i's     ts'a̱daḵ.  many= DIST=DET          RED-RED-man       go     PREP  house=POSS  woman ‘Many men went to a woman’s house.’  S2  √ [[Q         D             Sub       ]    Verb             Obj]  [ḵ̓ina̱m [DP =i=da      [NP babagwa̱m]]]  k̓ak̓ada̱xwsila=x̱a   buk.  many=DIST=DET        boy          read=ACC         book ‘Many boys are reading books.’     37 34 S1  √√    /    S2  √√ [Verb           Sub   [[Q          D           Sub             ]   Verb            Obj  ] nik̓iḵal=i       Ken  [ḵ̓ina̱m [DP =i=da     [NP bi-bi-ba̱gwana̱m]]]  ix'ake=x̱       Mary. think=DIST     Ken   many=DIST=DET       RED-RED-man      find.cute=ACC  Mary ‘Ken thinks that many men like Mary.’  3.3.2 Bare Quantifiers So far we have looked at cases where the quantifier and an associated nominal phrase form a surface constituent (either i=da ḵ̓ina̱m bi-bi-ba̱gwana̱m or ḵ̓ina̱m=i=da bi-bi-ba̱gwana̱m (‘many men’)). However, there are also cases where the quantifier appears at the front of the sentence without the noun it quantifies over, and the noun appears in argument position without its quantifier. This section presents sentences where ḵ̓ina̱m is separated from its restriction. Bare quantifiers can appear either with or without prenominal determiner elements.  3.3.2.1 Bare Quantifiers with Determiners: Ḵ̓ina̱m Bare quantifiers can appear with determiners. In (35a), ḵ̓ina̱m occurs with the locative (medial) marker ox̱ and the determiner marker da. In (35b), ḵ̓ina̱m occurs with just the locative marker ox̱. However, it is not possible for the determiner da to occur without the locative marker ox̱, as shown in (35c), because there is a general constraint against da appearing without a locative element (Littell 2016)11.                                                     11 Although =a=da (without deictic specification) sometimes seems to be possible (Littell 2016).   38 35 (a)  S1  ? [[QSUB    DLOC+DDETi]    Verb         DLOC+DDETi       Sub                 Obj       ]  ḵ̓ina̱m=ox̱=da        ha̱'m-x'id-su=ox̱=da          bi-bi-ba̱gwana̱m=x̱a    aya̱ndzis. many=MED=DET     eat-PERF-PASS=MED=DET     RED-RED=man=ACC  orange ‘Many men ate an orange.’  (b)  S1  √ [[QSUB   DLOCi]      Verb       DLOCi   DDET      Sub                 Obj     ] ḵ̓ina̱m=ox̱      ha̱'m-x'id-su=ox̱=da        bi-bi-ba̱gwana̱m=x̱a    aya̱ndzis. many=MED     eat-PERF-PASS=MED=DET   RED-RED=man=ACC  orange ‘Many men ate an orange.’  (c)  S1 *  [[QSUB     DDETi]     Verb         DLOC   DDETi      Sub                 Obj     ] ḵ̓ina̱m=da     ha̱'m-x'id-su=ox̱=da         bi-bi-ba̱gwana̱m=x̱a    aya̱ndzis. many=DET     eat-PERF-PASS=MED=DET    RED-RED=man=ACC  orange ‘Many men ate an orange.’  In (35a, b), the locative marker =ox̱ is doubled; this constitutes a form of agreement, in which a predicative auxiliary and the main predicate are locative-marked. This type of agreement only occurs with subjects. In contrast, agreement is impossible with an object (36). However, it is not clear whether this is because of a failure of agreement or because the pre-predicative quantifier cannot be associated with an object.  36 S1  *   [QOBJ     DACCi   Verb                     Sub     [DACCi    Obj      ]] ḵ̓ina̱m=x̱a     ha̱'m-x'id=i=da          babagwa̱m=x̱a   i-'aya̱ndzis. many=ACC     eat-PERF=DIST=DET      boy=ACC       RED-orange ‘The boy ate many oranges.’    39 3.3.2.2 Bare Quantifiers without Determiners: Ḵ̓ina̱m When bare quantifiers occur without any determiners, it is possible for them to be associated with both subjects and non-subjects, but associating with non-subjects is less preferred than with subjects. (37) shows that S1 accepted a bare quantifier associated with the subject.  37  S1  √  [Q        Verb     [D            Sub            ]              Obj    ] ḵ̓ina̱m   ha̱'m-x'id=i=da         bi-bi-ba̱gwana̱m [DP =x̱a    [NP ka̱leps]]. many    eat-PERF=DIST=DET     RED-RED-man=ACC          grape ‘Many men ate a grape.’  (38) shows the non-subject cases. S1 volunteered (38a), in which the quantifier occurs with the noun it quantifies over in post-predicative object position. When I asked if she could put the quantifier at the front, as in (38b), she offered (38c), which is a passive construction with a bare quantifier. S1 sometimes accepts (38b).  38  (a) S1  √   [Verb                    Sub        [    D    Q          Obj   ]] ha̱'m-x'id=i=da          ba̱gwana̱m [DP =x̱a   ḵ̓ina̱m  [NP ka̱leps]]. eat-PERF=DIST=DET      man=ACC          many       grape ‘The man ate many grapes.’     40 (b)  S1  ?  [QOBJ      Verb                        Sub         [D    Obj   ]] ḵ̓ina̱m   ha̱'m-x'id=[DP i=da       [NP ba̱gwana̱m]] =x̱a  ka̱leps. many    eat-PERF=DIST=DET          man=ACC        grape ‘The man ate many grapes.’  (c)  S1  √√ [QSUB      Verb-PASS  [D            Sub   ]           Obj       ] ḵ̓ina̱m   ha̱'m-x'id-su=sa       [NP ka̱leps]=a̱=sa     ba̱gwana̱m]12. many    eat-PERF-PASS=OBL      grape=VIS=OBL  man ‘Many grapes are eaten by the man.’  When the quantifier corresponds to a passive subject, S2 does not require comma intonation (i.e., a pause) after the quantifier, as shown in (39a). However, when the quantifier corresponds to the object, S2 does require comma intonation, as shown in (39b). In (39b), there is no pause, and the utterance is rejected. If comma intonation is added as in (39c), then the utterance becomes sometimes grammatical.  39 (a)  S2  √ [QSUB        Verb-PASS   [D                Sub  ]]  ḵ̓ina̱m     ha̱'m-x'id-su=ox̱=da           kegas.  many      eat-PERF=PASS=MED=DET      cake  ‘Many cakes are eaten.’                                                    12 The sentence apparently lacks a subject.   41 (b)  S2  *  [QOBJ       Verb                 Sub      [D    Obj     ]] ḵ̓ina̱m    ha̱'m-x'id=i=da       ga̱nana̱m=x̱a  ki-kegas. many     eat-PERF=DIST=DET   child=ACC    RED-cake ‘The child ate many cakes.’  (c)  S2  ?  [QOBJ,      Verb                 Sub       [D     Obj  ]] ḵ̓ina̱m,   ha̱'m-x'id=i=da       ga̱nana̱m=x̱a  kegas. many,    eat-PERF=DIST=DET   child=ACC    cake ‘The child ate many cakes.’  The following tables summarize the distribution of the bare quantifier ḵ̓ina̱m. Bare quantifiers can appear with determiners when they correspond to subjects, as shown in Table 12. However, it is not possible for the determiner da to occur without the locative marker.    S1 S2 Bare Quantifier (a) ? - QSUB    DLOC+DDETi   Verb   DLOC+DDETi   Sub (b) √ - QSUB    DLOCi   Verb   DLOCi  DDET   Sub (c) * - QSUB    DDETi   Verb   DLOC   DDETi   Sub (d) * - QOBJ    DACCi    Verb   Sub   DACCi   Obj Table 12 Summary of Distribution: Bare Quantifier Ḵ̓ina̱m with Determiners  Table 13 shows that bare quantifiers can appear without determiners when they correspond to subjects. S1, who has a more liberal grammar, sometimes accepts fronted Qs associated with objects. On the other hand, S2 only allows them when dislocated.      42  S1 S2 Bare Quantifier (a) √ - QSUB   Verb   [D   Sub]   Obj (b) √√ √ QSUB   Verb-PASS   [D   Sub] (c) ?  * ? QOBJ   Verb   Sub   [D    Obj] QOBJ,   Verb   Sub    [D    Obj] Table 13 Summary of Distribution: Bare Quantifier Ḵ̓ina̱m without Determiners  3.4 Summary of Distribution: Ḵ̓ina̱m This chapter laid out the contrastive behavior of ḵ̓ina̱m (‘many’) in pre- versus post- predicative position.  Both speakers disallow ḵ̓ina̱m in pre-determiner position when it occurs on post-predicative non-subjects. When it is associated with subjects, however, the speakers have different judgments. S2 does not allow subjects with ḵ̓ina̱m in the pre-determiner position of post-predicative DPs, while S1 sometimes does.  Unlike with the pre-determiner cases, there is a clear difference between speakers regarding how quantifiers interact with adjectives: S1 allows ḵ̓ina̱m to follow an adjective, but S2 does not. When ḵ̓ina̱m occurs as a bare quantifier, both speakers allow it to appear at the front of the sentence. Bare ḵ̓ina̱m is usually related to the subject. For bare quantifiers which are associated with objects, S1 sometimes accepts the sentences, but S2 only accepts the sentences with comma intonation. Bare quantifiers can occur either with or without determiners. Although the speakers sometimes differ from each other, they follow a regular pattern, such that S1 is more liberal and S2 is more restrictive. In many instances, their judgments agree.   43 4. WI'LA (‘ALL’) IS A STRONG QUANTIFIER This chapter focuses on the distribution of wi'la (‘all’). In section 4.1, I show that wi'la  cannot function as a main predicate. Section 4.2 examines the internal structure of quantified DPs. Finally I investigate clause-initial quantifiers and bare quantifiers in section 4.3.  4.1 Predicative Status of Wi'la The quantifier wi'la (‘all’) differs from ḵ̓ina̱m (‘many’) in that it cannot be the main predicate of a sentence. This is shown in (40).  40 S1  *    /    S2  * [Q         D             Sub    ] wi'la(-m)=ox̱=da        gi-gukw. all(-DISC)=MED=DET    RED-house ‘all the houses’ S1 Comment: I do not know what you want to say. S2 Comment: It is not a sentence.  Because it cannot be the main predicate, I assume that wi'la is a strong quantifier.  4.2 The Internal Structure of Quantified DPs: Wi'la Unlike ḵ̓ina̱m (‘many’), wi'la (‘all’) only appears before the prenominal determiner.     44 4.2.1 Wi'la in Pre- vs. Post-determiner Position In this section, I test the grammaticality of sentences with quantified NPs in different positions. The sentences below in (41-44) have wi'la in each possible position within an argument DP. The judgments of S1 and S2 are provided for each sentence. The examples in (41) show a quantified object (with accusative marker). Both speakers accept wi'la in pre-determiner position (a) but reject it in post-determiner position (b).  41 wi'la (‘all’): pre- and post-determiner position: Accusative marker (a)  S1  √√    /   S2  √ [Verb              Sub    [   Q           D      Obj   ]] ha̱'m-x'id=i       Ruby  [DP wi'la [DP =x̱a [NP lemon]]. eat-PERF=DIST    Ruby       all=ACC        lemon ‘Ruby ate all the lemons.’  (b)  S1  *    /   S2  *  [Verb              Sub  [    D     Q         Obj]] ha̱'m-x'id=i       Ruby[DP =x̱a [QP wi'la  [NP lemon]]. eat-PERF=DIST    Ruby=ACC     all       lemon  ‘Ruby ate all the lemons.’  The examples in (42) show a quantified object introduced by an oblique maker. Both speakers accept wi'la in pre-determiner position (a) but reject it in post-determiner position (b).     45 42 wi'la (‘all’): pre- and post-determiner position: Oblique marker (a)  S1  √√ [Verb      Sub    [   Q         D     Obl       ]      PP           ] t̓soya    Ruby  [DP wi'la [DP =sa [NP bi-buk]]]     lax̱     Ayako. give     Ruby       all=OBL         RED-book    PREP   Ayako ‘Ruby gave all the books to Ayako.’   S2  √   [Verb      Sub    [   Q         D      Obl]      PP           ] t̓soya    Ruby  [DP wi'la  [DP =sa [NP buk]]]   lax̱      Ayako.  give     Ruby      all=OBL          book    PREP   Ayako ‘Ruby gave all the books to Ayako.’  (b)  S1  *  [Verb      Sub  [    D       Q        Obl]        PP           ] t̓soya    Ruby[DP =sa   [DP wi'la [NP bi-buk]]]   lax̱     Ayako.  give     Ruby=OBL        all       RED-book  PREP   Ayako  ‘Ruby gave all the books to Ayako.’   S2  *   [Verb      Sub  [    D       Q        Obl]      PP           ] t̓soya    Ruby[DP =sa   [DP wi'la [NP buk]]]   lax̱     Ayako. give     Ruby=OBL        all       book    PREP   Ayako    ‘Ruby gave all the books to Ayako.’  The examples in (43) show a quantified subject (with Loc (Dist) and Det markers) following the predicate. Both speakers accept wi'la in pre-determiner position (a) but reject it in post-determiner position (b).    46 43 wi'la (‘all’): pre- and post-determiner position: Locative and Determiner markers in post-predicative position  (a)  S1  √ [Verb          [   Q        D        Sub             ]         Obj       ] ha̱'m-x'id     [DP wi'l [DP =i=da [NP bi-bi-ba̱gwana̱m]]] =x̱a     i-'aya̱ndzis. eat-PERF          all=DIST=DET    RED- RED-man=ACC       RED-orange ‘All the men ate oranges.’  S2  √   [Verb    [Q      D           Sub   ]    PP                ] la      wi'la=i=da        ts'a̱daḵ    lax̱      gukw=a̱'s. go     all=DIST=DET     woman   PREP    home=POSS ‘All the women went to her house.’   (b)  S1  *  [Verb       [   D            Q         Sub            ]       Obj      ] ha̱'m-x'id [DP =i=da        wi'la  [NP bi-bi-ba̱gwana̱m]] =x̱a  i-'aya̱ndzis. eat-PERF=DIST=DET       all        RED- RED-man=ACC    RED-orange  ‘All the men ate oranges.’  S2  *   [Verb [D               Q       Sub    ]  PP              ] la-'m=i=da            wi'la   ts'a̱daḵ    lax̱     gukw=a̱'s. go-DISC=DIST=DET    all     woman   PREP   home=POSS ‘All the women went to her house.’  The examples in (44) show a quantified subject (with Loc and Det markers) preceding the predicate. Both speakers accept wi'la in pre-determiner position (a) but reject it in post-determiner position (b).    47 44 wi'la (‘all’): pre- and post-determiner position: Locative and Determiner markers in pre-predicative position (a)  S1  √  [Aux        [Q     D                Sub       ]    Verb     ] la̱-'m       wi'la-'m=i=da         ts'i-ts'a̱daḵ    dłax̱w-x'id. AUX-DISC  all-DISC=DIST=DET    RED-woman   stand.up-PERF ‘All the women stood up.’  S2  √   [Aux    [Q       D              Sub          ]    Verb    ] la̱     wi'la-'m=i=da          ts'i-ts'i-ts'a̱daḵ    ne'nakwa. AUX   all-DISC=DIST=DET     RED-RED-woman go.home ‘All the women went home.’  (b)  S1  * [Aux [D                   Q       Sub       ]    Verb    ] la̱-'m=i=da               wi'la   ts'i-ts'a̱daḵ    dłax̱w-x'id. AUX-DISC=DIST=DET     all     RED-woman   stand.up-PERF ‘All the women stood up.’  S2  *  [Aux [D                  Q       Sub          ]    Verb    ] la̱-'m=i=da              wi'la   ts'i-ts'i-ts'a̱daḵ    ne'nakwa. AUX-DISC=DIST=DET    all     RED-RED-woman go.home ‘All the women went home.’  4.2.2 Summary of Predicative Status and Distribution of DP-Internal Qs: Wi'la The following table shows the distribution of predicative wi'la (‘all’) as shown in 4.1. Unlike ḵ̓ina̱m (‘many’), neither of the speakers allow wi'la to be used as a predicate.      48 S1 S2 Predicative Quantifier * * QPRED D Sub Table 14 Summary of Distribution: Predicative Wi'la  The following tables summarize the distribution of DP-Internal QPs with wi'la. Looking first at quantified non-subjects, as shown in (41-42), neither of the speakers allow wi'la in post-determiner position, unlike ḵ̓ina̱m (‘many’).  S1 S2 Quantified Non-Subject √√ √ Verb   Sub   [Q   D   Object] * * Verb   Sub   [D   Q   Object] √√ √ Verb   Sub   [Q   D   Oblique] * * Verb   Sub   [D   Q   Oblique] Table 15 Summary of Distribution: DP-Internal Wi'la (Non-Subject)  When wi'la is associated with the subject, as shown in (43-44), both speakers allow subjects with wi'la in pre-determiner position, but reject it in post-determiner position, as shown in Table 16.   S1 S2 Quantified Subject √ √ Verb   [Q   D   Sub] * * Verb   [D   Q   Sub] √ √ Aux   [Q   D   Sub]   Verb  * * Aux   [D   Q   Sub]   Verb Table 16 Summary of Distribution: DP-Internal Wi'la (Subject)     49 4.3 Clause-Initial Quantified DPs and Bare Quantifiers: Wi'la 4.3.1 Quantified DPs So far we have looked only at main clauses. I have shown in section 4.2.1 that quantified DPs with wi'la (‘all’) also can precede the main predicate. This is illustrated in (45). Similarly, in embedded clauses, quantified subjects can also precede the predicate of the embedded sentence, as shown in (46).  45 S1  √√ [[Q           D            Sub            ]    Verb   PP                            ] [wi'la-'m[DP =ox̱=da    [NP bi-bi-ba̱gwana̱m]]]  la     lax̱    gukw=a̱s=sa       ts'a̱daḵ. all-DISC=MED=DET      RED-RED-man      go     PREP  house=POSS=OBL  woman ‘All the men went to a woman’s house.’   S2  √√ [[Q         D              Sub]             Verb           Obj     ]    [wi'l [DP =i=da        [NP bi-ba̱gwana̱m]]]   ha̱'m-x'i=x̱a     aya̱ndzis. all=DIST=DET         RED-man        eat-PERF=ACC   orange ‘All the men ate oranges.’  46 S1  √√ [Verb           Sub   [[Q           D           Sub]               Verb          Obj]] nik̓iḵal=ox̱    Ken   [wi'la-'m[DP =i=da     [NP bi-bi-ba̱gwana̱m]]]  ix'aka=x̱a      Mary. think=MED    Ken    all-DISC= DIST=DET     RED-RED-man      find.cute=Acc  Mary. ‘Ken thinks that all the men like Mary.’   4.3.2 Bare Quantifiers This section focuses on sentences in which the quantifier wi'la occurs at the front of the sentence without the noun it quantifies over. Wi'la in these constructions can appear either with   50 or without prenominal determiner elements.  4.3.2.1 Bare Quantifiers with Determiners: Wi'la Bare quantifier wi'la may appear with prenominal determiner elements, such as the distal locative marker i and the determiner da. This is shown in (47). Both these markers may occur together (47a), or the locative marker may occur without the determiner (47b). However, the determiner marker da appears to require a locative marker (47c), as it usually does.  47  (a)  S1  √ [QSUB  DLOC+DDETi    Verb       DLOC+DDETi      Sub                          ]  wi'l=i=da           ha̱'m-x'id=i=da         bi-bi-ba̱gwana̱m=x̱a   ki-ka̱leps.  all=DIST=DET       eat-PERF=DIST=DET     RED-RED-man=ACC   RED-grape   ‘All the men ate a grape.’  (b)  S1  √  [QSUB  DLOCi    Verb      DLOCi  DDET      Sub                          ]  wi'l=i        ha̱'m-x'id=i=da         bi-bi-ba̱gwana̱m=x̱a   ki-ka̱leps.  all=DIST     eat-PERF=DIST=DET     RED-RED-man=ACC   RED-grape   ‘All the men ate a grape.’  (c)  S1  *  [QSUB  DDETi    Verb      DLOC  DDETi      Sub                          ]  wi'l=da       ha̱'m-x'id=i=da         bi-bi-ba̱gwana̱m=x̱a   ki-ka̱leps.  all=DET       eat-PERF=DIST=DET     RED-RED-man=ACC   RED-grape   ‘All the men ate a grape.’  So far I have looked at examples in which wi'la is a bare quantifier corresponding to the subject of a clause. However, when wi'la is a bare quantifier corresponding to the object, it   51 cannot appear with a determiner (i.e., the accusative marker x̱a). (48) shows that it is ungrammatical for pre-predicative wi'la to be associated with the object when it co-occurs with determiners (but more generally, nothing marked with x̱a can ever appear in front of the predicate). In the next section I address the issue of bare wi'la when it appears without determiners.  48 S1  *  [QOBJ  DACCi    Verb                    Sub       DACCi   Obj     ] wi'la=x̱a      ha̱'m-x'id=i=da         ba̱gwana̱m=x̱a    ki-ka̱leps. all=ACC       eat-PERF=DIST=DET     man=ACC        RED-grape  ‘The men ate all the grapes.’  4.3.2.2 Bare Quantifiers without Determiners: Wi'la For sentences in which wi'la occurs as a bare quantifier with no determiner, S1 and S2 have different grammaticality judgments. For S1, the bare quantifier wi'la can associate with both subjects and non-subjects. For S2, on the other hand, the bare quantifier wi'la can only associate with subjects. In (49a), wi'la is associated with the subject, and it is grammatical for S1. S2’s judgment is not available for this example. In (49b), wi'la is again associated with the subject in the passive sentence, and it is grammatical for both S1 and S2. In (49c), wi'la is associated with the object. For S1, this sentence is grammatical, but for S2 it is not.      52 49 (a)  S1  √    /    S2  -    [QSUB     Verb    [D               Sub           ]       Obj    ] wi'la   ha̱'m-x'id=i=da          bi-bi-ba̱gwana̱m=x̱a   ka̱leps. all     eat-PERF=DIST=DET      RED- RED-man=ACC   grape ‘All the men ate a grape.’  (b)  S1  √    /    S2  √ [QSUB        Verb-PASS    D                    Sub     ] wi'la      ha̱'m-x'id-su=ox̱=da               ki-kegas. all        eat-PERF= PASS=MED=DET         RED-cake ‘All the cakes are eaten.’  (c)  S1  √    /    S2  *  [QOBJ   Verb            Sub   [D     Obj           ]   PP          ] wi'la  duḵwa̱l=i      John=x̱a    bi-bi-ba̱gwana̱m  lax̱     ikaslas. all    see=DIST      John=ACC  RED-RED-man    PREP   party ‘John saw all the men at the party.’                                                             The following tables summarize the distribution of the bare quantifier wi'la. First of all, I look at when it occurs with the determiners (Table 17). In this case, when it corresponds to the subject, it can occur either with both determiners, Loc and Det (Table 17a), or with just Loc (Table 17b), but not with just Det (Table 17c). When it corresponds to the object, it is not grammatical with the determiner Acc (Table 17d).      53  S1 S2 Bare Quantifier (a) √ - QSUB    DLOC+DDETi Verb  DLOC+DDETi   Sub (b) √ - QSUB    DLOCi   Verb   DLOCi+DDET   Sub (c) * - QSUB    DDETi   Verb   DLOC+DDETi   Sub (d) * - QOBJ    DACCi   Verb   Sub   DACCi  Obj Table 17 Summary of Distribution: Bare Quantifier Wi'la with Determiners  Next, I look at when bare quantifier wi'la occurs without determiners. In this case, it is grammatical as the subject for both S1 and S2 (Table 18a-b). However, as the object, it is grammatical for S1 but ungrammatical for S2.   S1 S2 Bare Quantifier (a) √ - QSUB   Verb   [D   Sub]   Obj (b) √ √ QSUB   Verb-PASS   [D   Sub]   Obj (c) √ * QOBJ   Verb   Sub   [D    Obj] Table 18 Summary of Distribution: Bare Quantifier Wi'la without Determiners  4.3.3 Summary of Distribution: Wi'la This chapter laid out the contrastive patterns of wi'la (‘all’) when it occurs in pre- and post- predicative position. Both speakers allow wi'la in pre-determiner position but not in post-determiner position. Neither of the speakers allow wi'la as a predicate. Both speakers allow bare quantifier wi'la at the front of the sentence. However, in this case, if it occurs with determiners, it is only allowed to be related to the subject but not to the object. Without determiners, on the other hand, bare quantifiers can associate both with subject and object for S1, but only with the subject for S2.   54 5. NUMERALS ARE WEAK QUANTIFIERS In this chapter, I describe the distribution of numerals. I begin in section 5.1 with the predicative status of numeral quantifiers. In section 5.2, I look at the internal structure of DPs with numeral quantifiers. In particular, in section 5.2.1, I show DPs with numeral quantifiers in each argument position, and in section 5.2.2 I focus on the interaction of numerals with adjectives inside DP. In section 5.3, clause-initial numeral quantifiers and bare numeral quantifiers are investigated. Many relevant data for S2 are missing in this chapter. I give data for S2 only when they are available. There are also tables in section 5.2.3 and 5.4 to confirm which data are missing.  5.1 Predicative Status of Numerals It is possible for numerals to be the main predicate of a sentence. For example, in (50), pre-determiners of the noun gi-gukw (‘houses’) are attached to the predicate sa̱k'(a) (‘five’). This is equivalent to an existential sentence in English.  50 S1  √√ [Q     D              Sub     ]   sa̱k'=i=da          gi-gukw.   five=DIST=DET     RED-house   ‘There are five houses.’  Because they can be main predicates, I assume that numerals are weak quantifiers.   55 5.2 The Internal Structure of Quantified DPs: Numerals In this section, I investigate the internal structure of quantified DPs, specifically the relative ordering of quantifiers and pre-determiners (5.2.1) as well as that of quantifiers and adjectives (5.2.2).  5.2.1 Numerals in Pre- vs. Post-determiner Position The following examples show grammaticality judgments for sentences with quantified NPs in various positions. In these sentences, numerals occur in each possible position within an argument DP in (51-54). Numerals normally appear after a prenominal determiner when they follow the main predicate. However, they can also appear before a prenominal determiner when modifying subjects. The examples in (51) show a quantified object sa̱k'a lemon (‘five lemons’) (with the accusative marker x̱a). Both speakers reject numerals in pre-determiner position (51a) but accept them in post-determiner position (51b).  51 Numerals: pre- and post-determiner position: Accusative marker (a)  S1  *    /    S2  * [Verb              Sub    [   Q          D       Obj  ]] ha̱'m-x'id =i       Ruby  [DP sa̱k'a [DP =x̱a [NP  lemon]]. eat-PERF=DIST    Ruby       five=ACC        lemon ‘Ruby ate five lemons.’      56 (b)  S1  √    /    S2  √ [Verb              Sub  [    D      Q           Obj   ]] ha̱'m-x'id =i       Ruby[DP =x̱a [DP sa̱k'a    [NP lemon]]. eat-PERF=DIST    Ruby=ACC     five         lemon ‘Ruby ate five lemons.’   The examples in (52) show a quantified object with the oblique marker sa. S1 rejects the numeral in pre-determiner position (52a). Both speakers accept the numeral in post-determiner position (52b).  52 Numerals: pre- and post-determiner position: Oblique marker (a)  S1  *  [Verb      Sub    [   Q        D      Obl  ]    PP           ] t̓soya    Ruby  [DP mu [DP =sa [NP buk]]]   lax̱     Ayako. give     Ruby      four=OBL     book    PREP   Ayako ‘Ruby gave four books to Ayako.’  (b)  S1  √√  [Verb      Sub  [    D       Q          Obl  ]     PP           ]  t̓soya    Ruby[DP =sa   [DP mu    [NP buk]]]    lax̱     Ayako.   give     Ruby=OBL       four        book     PREP   Ayako   ‘Ruby gave four books to Ayako.’       57 S2  √  [Verb    Sub                Obj    [     D          Q           Obl       ]] ga̱lsa   Mary=x̱a     mu 13   gi-gukw[DP =sa      [DP ma'ł ]   [NP gi-ga̱lyayu ]]].  paint   Mary=ACC   four   RED-house=OBL        two        RED-brush ‘Mary painted four houses with two brushes.’  Next I look at quantified subjects. The examples in (53) show a quantified subject (with Loc and Det markers) following the predicate. S2 rejects but S1 accepts it in pre-determiner position (a). Both speakers accept numerals in post-determiner position (b).  53 Numerals: pre- and post-determiner position: Locative and Determiner markers in post-predicative position (a)  S1  √ [Verb           [   Q            D        Sub        ]       Obj      ]    ha̱'m-x'id     [DP yuda̱xw [DP =i=da [NP ba̱gwana̱m]]] =x̱a  i-'aya̱ndzis. eat-PERF          three=DIST=DET       man=ACC          RED-orange ‘Three men ate oranges.’      S2  *  [Verb          [   Q          D              Sub   ]        Obj                           ]    dala         [DP mu[DP =i=da     [NP ts'a̱daḵ ]]]=x̱a  sa̱k'a   ga̱ldas  lax̱=is      gukw. carry          four=DIST=DET      woman=ACC    five    box    PREP=POSS  house. ‘Four women carried five boxes to her (another woman) house.’                                                    13 I did not have an opportunity to ask S2 for a judgment on sentences containing only quantified objects with the oblique marker sa. In (52b) for S2, although the sentence has two quantified objects, the target is only the relation between the quantifier and the pre-determiner of the oblique marked nominal phrase, as the bracketing indicates.   58 (b)  S1  √ [Verb        [    D               Q           Sub       ]         Obj      ] ha̱'m-x'id=[DP i=da        [DP yuda̱xw [NP ba̱gwana̱m]]] =x̱a   i-'aya̱ndzis. eat-PERF=DIST=DET       three       man=ACC                 RED-orange ‘Three men ate oranges.’       S2  √ [Verb [D             Q             Sub   ]        Obj                           ]     dala=[DP i=da       [DP mu      [NP ts'a̱daḵ]]] =x̱a  sa̱k'a   ga̱ldas   lax̱=is      gukw. carry=DIST=DET     four          woman=ACC  five    box    PREP=POSS  house. ‘Four women carried five boxes to her (another woman) house.’  The examples in (54) also show a quantified subject (with Loc and Det markers), but this time the subject precedes the predicate. In this case, it is possible for the quantifier - the numeral ma'ł (‘two’) - to either precede the determiners (54a) or follow them (54b). These are S1’s judgments.  54 Numerals: pre- and post-determiner position: Locative and Determiner markers in pre-predicative position (a)  S1  √  [Aux      Q     [D             Sub]          Verb   PP      ] la̱-'m    ma'ł=i=da          ts'i-ts'a̱daḵ    la     ne'nakwa. AUX    two=DIST=DET      RED-woman   go    home ‘Two women went home.’     59 (b)  S1  √√ [Aux   [D                Q      Sub       ]   Verb   PP      ] la̱-'m=i=da             ma'ł   ts'i-ts'a̱daḵ   la     ne'nakwa. AUX-DISC=DIST=DET   two   RED-woman  go    home ‘Two women went home.’  5.2.2 Interaction of Numerals and Adjectives This section shows the distribution of numerals when they occur with adjectives. For S1, a quantifier following a pre-nominal D in argument position may precede an adjective, as in (55a), but it is not required to. Example (55b) shows that the quantifier can also follow the adjective. There are no data from S2 on this point.  55 (a)  S1  √ [Neg               Sub     Verb  [    D          Q      Adj         Obj     ]] k̓i's  weł=i       Ayako  ha̱'map[DP =x̱a   [NP [QP mu]  a̱m-a̱me'    ki-kegas.]] NEG  can=DIST    Ayako  eat=ACC             four   RED-small  RED-cake ‘Ayako cannot eat four small cakes.’   (b)  S1  √ [Neg               Sub     Verb    [    D       Adj           Q       Obj     ]] k̓i's  weł=i       Ayako  ha̱'map [DP =x̱a  [NP  a̱m-a̱me'   [QP mu]    ki-kegas.]] NEG  can=DIST    Ayako  eat=ACC           RED-small   four    RED-cake ‘Ayako cannot eat four small cakes.’  5.2.3 Summary of Predicative Status and Distribution of DP-Internal Qs: Numerals The following table summarizes the distribution of predicative numerals shown in 5.1. S1 allows numerals as predicates.    60 S1 S2 Predicative Quantifier √√ - QPRED D Sub Table 19 Summary of Distribution: Predicative Numerals  The following tables show the distribution of DP-internal QPs. Looking first at quantified non-subjects, as shown in (51-52), both speakers reject sentences in which numerals occur in pre-determiner position (Table 20, lines 1 and 3). Both speakers accept sentences with numerals in post-determiner position (Table 20, lines 2 and 4).   S1 S2 Quantified Non-Subject * * Verb   Sub   [Q   D   Object] √ √ Verb   Sub   [D   Q   Object] * - Verb   Sub   [Q   D   Oblique] √√ √ Verb   Sub   [D   Q   Oblique] Table 20 Summary of Distribution: DP-Internal Numerals (Non-Subject)  When numerals are associated with the subject, as shown in (53-54), the speakers have different judgments. S1 allows subjects with numerals in pre-determiner post-predicative position, while S2 does not, as shown in the first line of Table 21.   S1 S2 Quantified Subject √ * Verb   [Q   D   Sub] √ √ Verb   [D   Q   Sub] √ - Aux   [Q   D   Sub]   Verb  √√ - Aux   [D   Q   Sub]   Verb Table 21 Summary of Distribution: DP-Internal Numerals (Subject)   61 The following table summarizes the distribution of numerals when they occur with adjectives. S1 allows numerals like sa̱k'a (‘five’) in pre- and post-adjective position.  S1 S2 Interaction with Adjective √ - Verb   Sub   [D   Q    Adj   Obj] √ - Verb   Sub   [D   Adj   Q   Obj] Table 22 Summary of Distribution: DP-Internal Numerals (interaction with adjectives)  5.3 Clause-Initial Quantified DPs and Bare Quantifiers: Numerals 5.3.1 Quantified DPs So far I have looked only at main clauses. I have shown in section 5.2.1 that quantified DP subjects with numerals can precede the main predicate, as shown in (56). Similarly, in embedded clauses, quantified subjects may precede the predicate, as shown in (57).  56 S1  √√ [Q               D             Sub            ]    Verb   PP                      ]]  ['ma̱'łgwa̱'nał [DP =ox̱=da     [NP bi-bi-ba̱gwana̱m]]]  la     lax̱    gukw=i's      ts'a̱daḵ. eight=MED=DET               RED-RED-man      go     PREP  house=POSS    woman ‘Eight men went to a woman’s house.’ S2  √√ [[Q        D                Sub             ]     Verb      ] [sa̱k' [DP =i=da         [NP bi-bi-ba̱gwana̱m]]]    dłax̱w-x'id. five=DIST=DET           RED-RED-man        stand-up-PERF ‘Five men stood up.’     62 57 S1  √√ [Verb          Sub   [[Q       D           Sub             ]      Verb            Obj]] nik̓iḵal=i      Ken  [mu [DP =i=da     [NP bi-bi-ba̱gwana̱m]]]     ix'ake=x̱a       Mary. think=DIST     Ken   four=DIST=DET     RED-RED-man         find.cute=Acc   Mary. ‘Ken thinks that four men like Mary.’  5.3.2 Bare Quantifiers So far we have looked at sequences of nominal phrases with quantifiers (e.g., either =i=da sa̱k'a bi-bi-ba̱gwana̱m or sa̱k'=i=da bi-bi-ba̱gwana̱m (‘five men’)). However, there are also cases where the quantifier appears at the front of sentence without the noun it quantifies over, and the noun appears in argument position without its quantifier. This section presents sentences where quantifiers appear at the front of the sentence without the nouns they quantify over. Bare quantifiers can appear either with or without prenominal determiner elements.  5.3.2.1 Bare Quantifiers with Determiners: Numerals  Bare quantifiers can appear with determiners. In (58a), a numeral occurs with the medial locative marker ox̱ and the determiner da. In (58b), a numeral occurs with just the locative marker ox̱. However, it is not possible for the determiner da to occur without the locative marker ox̱, as shown in (58c), because there is a general constraint against da appearing without a locative element.       63 58 (a)  S1  √ [QSUB  DLOC+DDETi   Verb      DLOC+DDETi     Sub                  Obj     ] mu =ox̱=da         ha̱'m-x'id=ox̱=da        bi-bi-ba̱gwana̱m=x̱a   aya̱ndzis. four=MED=DET     eat-PERF=MED=DET     RED-RED-man=ACC   orange ‘Four men ate the orange.’  (b)  S1  √    [QSUB  DLOCi     Verb    DLOCi DDET       Sub                  Obj    ] mu =ox̱       ha̱'m-x'id=ox̱=da        bi-bi-ba̱gwana̱m=x̱a   aya̱ndzis. four=MED    eat-PERF=MED=DET     RED-RED-man=ACC   orange ‘Four men ate the orange.’  (c)  S1  *     [QSUB  DDETi    Verb     DLOC DDETi      Sub                  Obj     ] mu =da      ha̱'m-x'id=ox̱=da        bi-bi-ba̱gwana̱m=x̱a   aya̱ndzis. four=DET    eat-PERF=MED=DET     RED-RED-man=ACC   orange ‘Four men ate the orange.’  In (59), yuda̱xw (‘three’) is associated with the object, and S1 judges the sentence as ungrammatical, since a x̱a-marked nominal can never precede the predictate.  59 S1  *  [QOBJ   DACCi     Verb                    Sub         DACCi   Obj      ] yuda̱xw=x̱a     ha̱'m-x'id=i=da         ba̱gwana̱m=x̱a     ki-ka̱leps. three=ACC      eat-PERF=DIST=DET     man=ACC         RED-grape  ‘The man ate three grapes.’  5.3.2.2 Bare Quantifiers without Determiners: Numerals As shown in section 3.3.2.2 and 4.3.2.2 respectively, the bare quantifiers ḵ̓ina̱m (‘many’) and wi'la (‘all’) can associate with both subjects and objects. Bare numeral quantifiers, however,   64 cannot appear without a determiner. The examples in (60) show that the bare quantifier mu (‘four’) without a determiner is ungrammatical for S1 whether the quantifier is associated with an active subject (60a), a passive subject (60b) or an object (60c).  60 (a)  S1  *  [QSUB   Verb       [D            Sub           ]       Obj     ] mu   ha̱'m-x'id=ox̱=da        bi-bi-ba̱gwana̱m=x̱a   aya̱ndzis. four   eat-PERF=MED=DET     RED-RED-man=ACC   orange ‘Four men ate the orange.’   (b)  S1  *  [QSUB   Verb-Pass   [DDET   DLOC       Sub     ]] mu   ha̱'m-x'id-su=i=da            ki-kegas. four   eat-PERF-PASS=DIST=DET     RED-cake ‘Four cakes are eaten.’    (c) S1  * [QOBJ   Verb                    Sub        D     Obj     ] mu   ha̱'m-x'id=ox̱=da        ba̱gwana̱m=x̱a   aya̱ndzis. four   eat-PERF=MED=DET     man=ACC       orange ‘The men ate four oranges.’  The following tables summarize the distribution of bare numerals. Bare numerals can appear only when they correspond to subjects (Table 23), and only with determiners (Table 24). However, it is not possible for the determiner da to occur without the locative marker, as shown in Table 23 line 3.      65 S1 S2 Bare Quantifier √ - QSUB   DLOC+DDETi   Verb   DLOC+DDETi   Sub   Obj √ - QSUB   DLOCi   Verb   DLOCi   DDET   Sub   Obj * - QSUB   DDETi   Verb   DLOC   DDETi   Sub   Obj * - QOBJ   DACCi    Verb   Sub   DACCi   Obj Table 23 Summary of Distribution: Bare Quantifier Numerals with Determiners  S1 S2 Bare Quantifier * - QSUB   Verb   [D   Sub]   Obj * - QSUB   Verb-PASS   [D   Sub] * - QOBJ   Verb   Sub   [D    Obj] Table 24 Summary of Distribution: Bare Quantifier Numerals without Determiners  5.4 Summary of Distribution: Numerals This chapter laid out the contrastive behavior of numerals in pre- versus post-predicative position. Both speakers disallow numerals in pre-determiner position when they are associated with non-subjects. When they are associated with subjects, however, the speakers have different judgments. S1 allows numerals to precede D in post-predicative subject DPs, while S2 does not. S1 allows numerals to precede or follow an adjective, but there is no data for S2. Numerals occur as a bare quantifier, but while wi'la (‘all’) and ḵ̓ina̱m (‘many’) can occur with and without determiners, S1 only allows bare numeral quantifiers that occur with determiners, not without. With determiners, S1 allows a bare numeral quantifier to appear at the front of the sentence, but only if it is related to the subject, not to the object, like wi'la (‘all’) and ḵ̓ina̱m (‘many’).   66  I assume that numerals are weak quantifiers, like ḵ̓ina̱m, since they can be main predicates, but the main difference between ḵ̓ina̱m and numerals is that ḵ̓ina̱m can occur with and without determiners, while a bare numeral quantifier cannot occur without them.   67 6. THE SYNTAX OF KWAK’WALA QUANTIFIERS This chapter provides a syntactic account of Kwak’wala quantifiers. From their predicative status and their DP-internal position, shown in chapters 3, 4, and 5, I assume that ḵ̓ina̱m (‘many’) and numerals are weak quantifiers; wi'la (‘all’), on the other hand, is a strong quantifier. The following table summarizes the distribution.   wi'la (‘all’) ḵ̓ina̱m (‘many’) Numerals Syntactic Distinction Strong Weak Weak Predicate Status * √ √ DP-Internal Pred [D Q N] * √ √ Pred [Q D N]   ‘non-subject’ √ * * Pred [Q D N]    ‘subject’ √ S1 ? / S2 * S1 √ / S2 * [D Q N] Pred * √ √ [Q D N] Pred √ √ √ Table 25 Summary of the Distribution of Quantificational Elements  While wi'la cannot be the main predicate, ḵ̓ina̱m and numerals can. Wi'la cannot occur in post-determiner position in any argument position, but ḵ̓ina̱m and numerals vary in their acceptability in this position. In objects, they can only occur in post-determiner position. In subjects, they occur in pre- and post-determiner position in pre-predicative position. For post-determiner post-predicative position, S2 rejects the sentences, while S1 accepts the constructions for numerals but gives variable judgments for ḵ̓ina̱m.  6.1 Clausal Syntactic Structure Although the core word order is Verb (Predicate) Subject Object in Kwak’wala, we have already seen that DPs can precede the main predicate in some cases. There are asymmetries   68 between subjects and non-subjects, and only a subject can move up to pre-predicative position, since only subjects are accessible to A’-movement in this language (Anderson 1984, Sherer 2014, Littell 2016), and in order for an object to move, it must therefore be promoted to subject first, as illustrated in (61) (it tests more like an A-position. although it is the launching pad for A’-movement.)  61 (a)  S1  * [[Q      D   Obj     ]   Verb                 Sub      ]  ḵ̓ina̱m=x̱a   ki-kegas    ha̱'m-x'id=i=da        ba̱gwana̱m.  many=ACC   RED-cake    eat-PERF=DIST=DET    man  ‘The man ate many cakes. ’  (b)  S1  √√ [[Q      D           Sub     ]   Verb                  Obl       ]  ḵ̓ina̱m=ox̱=da       ki-kegas    ha̱'m-x'id-su=sa         ba̱gwana̱m.  many=DIST=DET      RED-cake    eat-PERF-PASS=OBL     man  ‘Many cakes were eaten by the man.’   In accordance with the analyses of Anderson, Sherer, and Littell and the characteristics of Kwak’wala syntax described in 2.2, repeated in (62), I describe the syntactic structure I assume in this chapter.  62 (a)  VSO (Verb (or Predicate) Subject Object) language. (b)  All open class elements, which correspond to nouns, verbs, and adjectives in English,  can function as the main predicate of a sentence. (c)  Main predicates may be preceded by a series of auxiliaries.   69 For the rest of this chapter, I follow Sherer’s (2014) analysis of the basic sentence structures.  6.1.1 Core Sentence Structure (Predicate Subject Object) In this section I show the structure of the core word order, (Aux) Predicate Subject Object, adapting Sherer’s (2014) analysis. Since adjectives, nouns, and verbs can function as the main predicate of a sentence, as shown in (63), VP/AP/NP are in turn selected by vP (i.e., base-generated in [Comp, vP])14 and the subject is base-generated in [Spec, vP].  63 (a)  S1  √√ [Adj       D         Subject] t'sa̱k'wa=da          ba̱gwana̱m short=DET           man ‘The man is short.’  (b)  S1  √√ [Noun   D     Subject] gig̱ame'=i      James. chief=DIST      James ‘James is a chief.’  (c)  S1  √√ [Verb      D              Subject] dłax̱w-x'id=i=da            ba̱gwana̱m.  stand.up-PERF=DIST=DET    man         ‘The man stood up.’                                                  14 In Sherer’s analysis, the Voice head takes Part(icipial) Phrase as its complement (cf. Legate 2010) and nominalizers (i.e., voice morphemes) are generated as the head of a Part(icipial) Phrase (cf. Collins 2005). However, what is relevant to this thesis is only the voice distinction (active/passive) so I will have vP instead of PartP as the complement of the Voice head.   70 In each case, a predicative element (N, V, or Adj) first moves up to v, then to Voice so that we can get VSO order (the solid lines in Figure 3).                   TP                                AuxP                                            Aux’                             Aux                     VoiceP                                                                    Voice’                                                                   -PASS                 vP                                                                          SubDP              v’                                                                                                    v                NP/VP/AP                                                                                                                          A’/N’/V’                                                                                                                A/N/V                                                                                                                                                                Figure 3 Basic Sentence Structure  Next, consider the sentences in (64), repeated from (15). (64a) is the core word order with an auxiliary, and (64b) shows that the speakers also accept (Aux)SVO order. In this case the subject moves up to [Spec, VoiceP] (the dotted lines in Figure 3).      71 64 (a)  S1  √    /    S2  √ (with surprise intonation)  [Auxiliary     Predicate      Subject      Object       Oblique] la̱-'m       ga̱ls=i        Mary=x̱a    gukw=sa    ga̱lyayu. AUX-DISC  paint=DIST    Mary=ACC  house=OBL brush ‘Mary painted a house with the brush.’   (b)  S1  √    /    S2  √ (with surprise intonation) [Auxiliary           Subject   Predicate      Object      Oblique] la̱-'m=i            Mary     ga̱lsa=x̱a     gukw=sa    ga̱lyayu. AUX-DISC=DIST    Mary     paint=ACC    house=OBL brush ‘Mary painted a house with the brush.’  The examples in (65) show a two auxiliaries case, repeated from (17c) and (17d). As explained in section 2.2, negation is categorized as an auxiliary in this language. In these sentences, the subject either appears between the auxiliaries (a) or precedes the auxiliaries (b). For (65a), the subject DP in [Spec, VP] will move up to the second [Spec, AuxP]. For (65b), the subject moves further up to the first [Spec, AuxP], as shown in Figure 4.   65  (a)  S1  √ [Negation       Subject   Auxiliary   Predicate    Accusative   Oblique] k̓i's=i         Mary    weł        ga̱lsa=x̱a    gukw=sa    ga̱lyayu. NEG=DIST    Mary    can         paint=ACC   house=OBL brush ‘Mary could not paint a house with the brush.’   (b)  S1  √  [Subject   Negation     Auxiliary  Predicate     Accusative   Oblique] Mary    k̓i's=i         weł       ga̱lsa=x̱a    gukw=sa     ga̱lyayu. Mary    NEG=DIST    can        paint=ACC   house=OBL  brush ‘Mary could not paint a house with the brush.’   72                 TP                                AuxP                   (b)                        Aux’                             Aux                     AuxP                                             (a)                    Aux’                                                                   Aux                     VoiceP                                                                     SubDP                 Voice’                                                                                                 Voice                   vP      Figure 4 Relationship between Pre-Predicative Subject and Auxiliaries  6.2 Syntactic Structure of Clause-Initial Quantified DPs In this section, I will describe the possible syntactic structures for each quantifier in clause-initial quantified DPs. Generally, there are two main possibilities for clause-initial quantifiers: they could either form a counstituent with the DP, or not. For those clause-initial quantifiers which are not DP-internal, there are two possible structural configurations: the quantifier may either be a predicate, or an auxiliary which precedes the main predicate as an ‘A-type’ (adverbial) quantifier. A-type quantifiers are discussed in Partee et al. (1987), and by Evans (1995) for Mayali, and Jelinek (1995) for Straits Salish.     73 6.2.1 Ḵ̓ina̱m (‘many’) There are three possible analyses for sequences of clause-initial ḵ̓ina̱m plus DP (66), as shown in (67). In the first (67a), the quantifier forms a constituent with the DP. In the second (67b), the quantifier does not form a constituent with its DP, but is a main predicate followed by a relative clause. In the third (67c), the quantifier is an auxiliary which precedes the main predicate as an ‘A-type’ (adverbial) quantifier.   66 S1  √√ ḵ̓ina̱m=i=da          bi-bi-ba̱gwana̱m  ha̱'m-x'i=x̱a     i-'aya̱ndzis. many=DIST=DET      RED-RED-man    eat-PERF=ACC   RED-orange ‘Many men ate oranges.’  67 Hypothesized Structures (a)  [DP ḵ̓ina̱m [DP =i=da       [NP bi-bi-ba̱gwana̱m]]]   ha̱'m-x'i=x̱a     i-'aya̱ndzis many=DIST=DET            RED-RED-man       eat-PERF=ACC   RED-orange  (b)  [Pred ḵ̓ina̱m [DP =i=da       [[NP bi-bi-ba̱gwana̱m]    [CP ha̱'m-x'i=x̱a    i-'aya̱ndzis]]]] many=DIST=DET              RED-RED-man          eat-PERF=ACC  RED-orange  (c)  [[Aux ḵ̓ina̱m] [[DP =i=da       bi-bi-ba̱gwana̱m]   [VP  ha̱'m-x'i=x̱a     i-'aya̱ndzis]] many=DIST=DET            RED-RED-man         eat-PERF=ACC   RED-orange  The structure (67a) is shown in Figure 5.    74                 TP                                VoiceP                                            Voice’                             -PASS                     vP                                             SubDP (Q)          v’                                                                      v                       VP                                                                           V                    ObjDP                                                                                                                                                    Figure 5 Hypothesized Structure of (67a)  The verb ha̱'map (‘eat’) is base-generated in V and moves up to v. Then it moves up to Voice. The subject ḵ̓ina̱m=i=da bi-bi-ba̱gwana̱m  (‘many men’), which is base-generated in [Spec, vP], precedes the verb so that we get the SVO order. For the non DP-internal quantifiers (67b) and (67c), the difference is in whether ḵ̓ina̱m is originally generated as an element functioning as the main predicate, or an auxiliary preceding the subject and main predicate. First, Figure 6 shows the structure for (67b).    75                 TP                                VoiceP                                             Voice’                                Voice                   vP                                              SubDP               v’                                                                      v                     PredP (QP)                                                                           Q                                                                          Figure 6 Hypothesized Structure of (67b)  The predicate ḵ̓ina̱m is base-generated in [Comp, vP], like in Figure 3, and moves up to v and to Voice. The rest of the sentence is a subject which has a relative clause headed by bi-bi-ba̱gwana̱m (‘men’).  (68) shows the comparison between an adjective and a quantifier in clause-initial position. The parallelism between an ordinary predicate and ḵ̓ina̱m is shown in (68a) and (68b).   68 S1  √√    /    S2  √√ (a)  t'sa̱k'w=i=da        ba̱gwana̱m  dłax̱wała    lax̱a    ts'a̱ya.   short=DIST=DET    man        stand       PREP   chair ‘The man who is standing on the chair is short.’      76 (b)  S1  √√ ḵ̓ina̱m=i=da         bi-bi-ba̱gwana̱m dłax̱wała    lax̱a    ts'a̱ya. many=DIST=DET     RED-RED-man   stand       PREP   chair ‘The men who are standing on the chair are many.’     S2  √√                               ḵ̓ina̱m=i=da         ba̱gwana̱m  dłax̱wała    lax̱a    ts'a̱ya. many=DIST=DET     man        stand       PREP   chair. ‘The men who are standing on the chair are many.’   Figure 7 shows the structure of (67c), where the subject sits between the auxiliary and the verb. The verb, which is base-generated in V, is moved to v and then Voice. Here, ḵ̓ina̱m is a base-generated auxiliary, and the subject, which is base-generated in [Spec, vP], moves up to [Spec, VoiceP] to get AuxSV(O) order.                  TP                                AuxP(QP)                                            Aux’                             Aux(Q)                 VoiceP                                                                    Voice’                                                                    Voice                   vP                                                                           SubDP              v’                                                                                                   v                       VP    Figure 7 AuxSV structure (67c)   77 Figure 7 is not correct. As shown in (74-75) below, wi'la can be freely ordered in a string of auxiliaries, so if ḵ̓ina̱m behaves the same way, the ungrammaticality of (69-70) is unexpected. Ḵ̓ina̱m in (69) stands in an Aux position [Aux DP Aux V], and the consultant judges it as ungrammatical. Likewise, ḵ̓ina̱m in (70) stands in an Aux position with other two auxiliaries [Aux DP Aux Aux V] or [Aux DP Aux Aux V], and the consultant judges these as ungrammatical.  69  S1  *  [Aux   [ D                Sub           ]  Q         V          ] la̱-'m=i=da              bi-bi-ba̱gwana̱m  ḵ̓ina̱m      dłax̱w-x'ida.  AUX-DISC=DIST=DET    RED-RED-man    many     stand.up-PERF   ‘Many men stood up.’   70 (a)  S1   * [Aux [ D                   Sub           ]  Q        Aux    V             Obj   ] la̱-'m=i=da               bi-babagwa̱m   ḵ̓ina̱m    k̓i's    ha̱'m-x'i=x̱a    ka̱leps. AUX-DISC=DIST=DET      RED-boy        many    NEG   eat-PERF=ACC  grape. ‘Many boys did not eat a grape.’  (b)  S1   * [Aux  [D                Sub        ]  Aux    Q          V              Obj    ] la̱-'m=i=da             bi-babagwa̱m  k̓i's     ḵ̓ina̱m       ha̱'m-x'i=x̱a     ka̱leps. AUX-DISC=DIST=DET   RED-boy      NEG     many      eat-PERF=ACC   grape. ‘Many boys did not eat a grape.’  However, there is no evidence to decide whether the clause-initial ḵ̓ina̱m is a predicate or forms a DP constituent with a nominal restriction. I will remain agnostic as to whether initial ḵ̓ina̱m is a main predicate or part of a fronted DP.   78 6.2.2 Wi'la (‘all’) In this section I show the possible syntactic structures for sequences of clause-initial wi'la plus DP (71). The hypothesized structures are shown in (72). In the first (72a), the quantifier forms a constituent with the DP. In the second (72b), the quantifier does not form a constituent with its DP, but is a main predicate followed by a relative clause. In the third (72c), the quantifier is an auxiliary which precedes the main predicate. The tree structures corresponding to these three analyses are shown in Figure 5, Figure 6, and Figure 7 above, respectively.  71 S1  √√ wi'la-'m=i=da           bi-bi-ba̱gwana̱m  ha̱'m-x'i=x̱a     i-'aya̱ndzis. all-DISC=DIST=DET     RED-RED-man    eat-PERF=ACC   RED-orange ‘All the men ate oranges.’  72 Hypothesized Structures (a)  [DP wi'la-'m [DP =i=da     [NP bi-bi-ba̱gwana̱m]]]   ha̱'m-x'i=x̱a     i-'aya̱ndzis     all-DISC=DIST=DET         RED-RED-man       eat-PERF=ACC   RED-orange  (b)  [Pred wi'la-'m [DP =i=da     [[NP bi-bi-ba̱gwana̱m]   [CP ha̱'m-x'i=x̱a     i-'aya̱ndzis]]]] all-DISC=DIST=DET          RED-RED-man        eat-PERF=ACC   RED-orange  (c)  [[Aux wi'la-'m ] [[DP =i=da     bi-bi-ba̱gwana̱m]     [VP ha̱'m-x'i=x̱a      i-'aya̱ndzis]] all-DISC=DIST=DET        RED-RED-man          eat-PERF=ACC   RED-orange  However, I have already shown that wi'la cannot be predicative as shown in (73) (repeated from (40)). Therefore, the structure in (72b) can be ruled out.      79 73 The constituents form a DP, not a complete sentence. wi'la(-'m)=ox̱=da         gi-gukw. all(-DISC)=MED=DET      RED-house ‘all the houses’  The sentence in (73) must be analyzed such that =ox̱=da gi-gukw (‘the houses’) is the subject and the clause-initial wi'la is the predicate; however, we already know that wi'la cannot be predicative, and therefore (73) cannot be analyzed as a clause, but only as a DP fragment. (74) and (75) show that wi'la can be analyzed as an auxiliary, because it can be freely ordered in a string of auxiliaries. In (74a), it can be analyzed as the second auxiliary ([Aux Aux DP V]), although it could also be forming part of DP here ([Aux [wi'la DP] V]). In (74b), wi'la must be analyzed as a second auxiliary [Aux DP Aux V]. (74c) and (74d) show the core order with two auxiliaries preceding the predicate [Aux Aux V DP] and [Aux Aux V DP].  74 (a)  S1  √    [Aux         Q       [D            Sub           ]  V         ] la̱-'m       wi'la-'m=i=da         bi-bi-ba̱gwana̱m  dłax̱w-x'id.  AUX-DISC   all-DISC=DIST=DET    RED-RED-man    stand.up-PERF   ‘All the men stood up.’  (b)  S1  √   [Aux [D                 Sub         ]  Q       V         ] la̱-'m=i=da              bi-ba̱gwana̱m   wi'la     dłax̱w-x'id.  AUX-DISC=DIST=DET    RED-man      all     stand.up-PERF   ‘All the men stood up.’      80 (c)  S1  √ [Aux         Q       V         [D                 Sub         ] la̱-'m       wi'la     dłax̱w-x'=i=da               bi-ba̱gwana̱m.  AUX-DISC   all     stand.up-PERF=DIST=DET     RED-man    ‘All the men stood up.’  (d)  S1  √ [Q       Aux          V        [D                 Sub         ] wi'la       la̱-'m        dłax̱w-x'=i=da               bi-ba̱gwana̱m.  all      AUX-DISC    stand.up-PERF=DIST=DET     RED-man ‘All the men stood up.’  Likewise, wi'la can co-occur with two other auxiliaries: (75a) and (75b) can be analyzed as [Aux DP Aux Aux V].  75 (a)  S1  √ [Aux  [D                 Sub        ]  Q      Aux   V             Obj       ] la̱-'m=i=da              bi-babagwa̱m  wi'la    k̓i's    ha̱'m-x'i=x̱a    aya̱ndzis. AUX-DISC=DIST=DET    RED-boy      all    NEG   eat-PERF=ACC  orange ‘All the boys did not eat an orange.’  (b)  S1  √ [Aux [D                Sub       ]   Aux    Q        V              Obj      ] la̱-'m=i=da             bi-babagwa̱m  k̓i's     wi'la      ha̱'m-x'i=x̱a     aya̱ndzis. AUX-DISC=DIST=DET   RED-boy      NEG     all       eat-PERF=ACC   orange ‘All the boys did not eat an orange.’   Therefore, I assume that the clause-initial wi'la is structurally ambiguous; it can be either an auxiliary, or form part of a DP constituent with its nominal.     81 6.2.3 Numerals The same hypothesized structures as outlined above are appropriate for sequences of clause-initial numerals plus DP (76), as shown in (77). In the first hypothesized structure (77a), the quantifier forms a constituent with the DP. In the second (77b), the quantifier does not form a constituent with its DP, but is a main predicate followed by a relative clause. In the third (77c), the quantifier is an auxiliary which precedes the main predicate. The structures associated with these analyses are as shown above in Figure 5, Figure 6, and Figure 7 respectively.  76 S1  √√ mu=i=da            bi-bi-ba̱gwana̱m  ha̱'m-x'i=x̱a     i-'aya̱ndzis. four=DIST=DET      RED-RED-man    eat-PERF=ACC   RED-orange ‘Four men ate oranges.’  77 Hypothesized Structures (a)  [DP mu [DP =i=da        [NP bi-bi-ba̱gwana̱m]]]   ha̱'m-x'i=x̱a     i-'aya̱ndzis four=DIST=DET        RED-RED-man       eat-PERF=ACC   RED-orange  (b)  [Pred mu [DP =i=da       [[NP bi-bi-ba̱gwana̱m]    [CP ha̱'m-x'i=x̱a     i-'aya̱ndzis]]]]  four=DIST=DET        RED-RED-man         eat-PERF=ACC   RED-orange  (c)  [[Aux mu] [[DP =i=da       bi-bi-ba̱gwana̱m]   [VP  ha̱'m-x'i=x̱a     i-'aya̱ndzis]] four=DIST=DET      RED-RED-man         eat-PERF=ACC   RED-orange  However, (78) and (79) suggest that like ḵ̓ina̱m, numerals cannot be auxiliaries since they cannot appear in the full range of Aux positions. Mu (‘four’) in (78) stands in an Aux position: [Aux DP Aux V] (a), [Aux Aux V DP] (b), and [Aux Aux V DP] (c), and the consultant judges the utterances as ungrammatical.    82 78 (a)   S1  *   [Aux  [D                Sub             Q      V        ] la̱-'m=i=da              bi-bi-ba̱gwana̱m  mu      dłax̱w-x'id.  AUX-DISC=DIST=DET    RED-RED-man    four   stand.up-PERF   ‘Four men stood up.’  (b)  S1  * [Aux        Q       V        [D                 Sub            ] la̱-'m      mu       dłax̱w-x'=i=da               bi-bi-ba̱gwana̱m.  AUX-DISC  four    stand.up-PERF=DIST=DET     RED-RED-man    ‘Four men stood up.’  (c)  S1  * [Q        Aux         V        [D                 Sub            ] mu       la̱-'m       dłax̱w-x'=i=da               bi-bi-ba̱gwana̱m.  four     AUX-DISC   stand.up-PERF=DIST=DET     RED-RED-man ‘Four men stood up.’  Likewise, mu (‘four’) in (79) stands in an Aux position along with two other auxiliaries in the structures [Aux DP Aux Aux V] and [Aux DP Aux Aux V], and the consultant judges these as ungrammatical.  79 (a)  S1   * [Aux  [D                 Sub       ]   Q     Aux    V             Obj   ] la̱-'m=i=da              bi-babagwa̱m  mu    k̓i's    ha̱'m-x'i=x̱a    ka̱leps. AUX-DISC=DIST=DET    RED-boy      four  NEG   eat-PERF=ACC  grape. ‘Four boys did not eat a grape.’      83 (b)  S1   * [Aux [D               Sub           Aux    Q       V              Obj     ] la̱-'m=i=da             bi-babagwa̱m   k̓i's     mu       ha̱'m-x'i=x̱a     ka̱leps. AUX-DISC=DIST=DET   RED-boy       NEG     four    eat-PERF=ACC   grape. ‘Four boys did not eat a grape.’  However, as in the ḵ̓ina̱m case, there is no further evidence to decide whether a clause-initial numeral is a predicate or a part of a DP. Therefore, I remain agnostic as to which analysis is correct.  6.3 DP-Internal Structure with Quantifiers As I have shown in earlier chapters, there is a distributional difference among wi'la, ḵ̓ina̱m, and numerals inside DP: wi'la obligatorily appears in pre-determiner position, while the distribution of ḵ̓ina̱m and numerals varies. Figure 8 shows the structures of wi'la (a) and ḵ̓ina̱m and numerals (b).  (a)          DP      QP                   DP       Q               D          NP   wi'la   (b)          DP       D                    NP                        QP          NP                           Q             ḵ̓ina̱m/numerals  Figure 8 DP Internal Structure with Wi'la and Ḵ̓ina̱m/Numerals    84 I assume that wi'la is generated in pre-determiner position. As for ḵ̓ina̱m and numerals, they can appear in post-determiner position in any argument position, like ordinary adjectives, as shown in (80). In (80a), ḵ̓ina̱m precedes the noun, and in (80b), the adjective dzastu (‘blue’) precedes the noun.  80 (a)  S1  √√ [Verb           Sub    [ D       Q          Obj     ]] dzam-x'id     Ayako=x̱a      ḵ̓ina̱m]    i-'abals.]] plant-PERF    Ayako=ACC    many      RED-apple ‘Ayako grew many apples.’  (b)  S1  √√ [Verb           Sub    [ D     Adj        Obj   ]] dzam-x'id     Ayako=x̱a    dzi-dzastu  abals.]] plant-PERF    Ayako=ACC  RED-blue  apple ‘Ayako grew blue apples.’  In the case that ḵ̓ina̱m and numerals appear in pre-determiner position, there are two possibilities: they are either base-generated in post-determiner position and move up to adjoin to the DP, as shown in Figure 9, or they are base-generated in pre-determiner position, like wi'la, as shown in Figure 8a.     85                DP                                DP                      D                    NP                   =i=da          QP         NP                                         Q                               ḵ̓ina̱m/numerals  Figure 9 Possibility for Pre-determiner Ḵ̓ina̱m and Numerals  In later sections, however, I will show that semantic evidence is neutral on this question, since there is no interpretive difference between pre- and post-determiner ḵ̓ina̱m. Moreover, although in this respect ḵ̓ina̱m/numerals behave like ordinary adjectives, in other respects, their behavior is not exactly the same as an adjective, because ḵ̓ina̱m/numerals can occur in pre-determiner position whereas adjectives cannot, as shown in (81). Therefore, I label the phrase headed by ḵ̓ina̱m/numerals QP.  81 S1  * [Verb           Sub        [Adj       D    Obj   ]] dzam-x'id     Ayako    dzi-dzastu=x̱a  abals.]] plant-PERF    Ayako    RED-blue=ACC apple ‘Ayako grew blue apples.’  6.4 Active and Passive Sentence Structures This section describes how quantified non-subjects are blocked from appearing in pre-predicate position.   86 6.4.1 Active and Passive Sentence Structure (82a) shows the core word order (VSO), and (82b) shows SVO order. The sentences in (82) can have the structures in Figure 10.  82 (a)  S1  √ [Verb         [Q       D              Sub      ]        Obj  ]      ha̱'m-x'id     wi'la-'m=i=da          ba̱gwana̱m=x̱a    kegas.    eat-PERF     all-DISC=DIST=DET     man=ACC        cake ‘All the men ate a cake.’  (b)  S1  √√ [[Q         D             Sub            ]   Verb           Obj  ]      wi'la-'m=i=da         bi-bi-ba̱gwana̱m  ha̱'m-x'i=x̱a     kegas.    all-DISC=DIST=DET    RED-RED-man    eat-PERF=ACC  cake ‘All the men ate a cake.’               VoiceP                              Voice’                  -PASS                 vP                             SubDP                   v’                                               v                      VP                                                                       V                      DP                                                                                   ObjDP              Figure 10 Subject can move up    87 For the core word order VSO (82a), the verb first moves up to v, then to Voice to generate VSO order (solid lines).  To get the SVO order (82b), the subject moves up to [Spec, VoiceP] (dotted lines). On the other hand, when the object is quantified, the quantified DP can either appear in its base-generated post-predicative argument position (83a), or the passive construction is used to promote it to subject position, as in (83b). However, a quantified object DP cannot appear in pre-predicative position, as shown in (83c).  Following Sherer (2014), I assume, in this structure, the verb moves up to v, then to Voice. However, due to the presence of the subject, the object that occurs below the subject is blocked from A'-extracting due to minimality violation (Sherer 2014). Thus, (83a) is possible but (83c) is not.  83  (a) S1  √√  [Verb                    Sub        [      D       Q          Obj    ]] ha̱'m-x'id=i=da        ba̱gwana̱m [DP =x̱a  [QP ḵ̓ina̱m [NP ki-kegas]]. eat-PERF=DIST=DET    man=ACC           many        RED-cake ‘The man ate many cakes.’  (b)  S1  √√ [Q       D          Sub     ]   Verb                  Obl       ] ḵ̓ina̱m=ox̱=da       ki-kegas    ha̱'m-x'id-su=sa         ba̱gwana̱m. many=MED=DET     RED-cake    eat-PERF-PASS=OBL     man ‘Many cakes were eaten by the man.’   (c)  S1 * [Q       D   Obj     ]   Verb                 Sub       ] ḵ̓ina̱m=x̱a   ki-kegas    ha̱'m-x'id=i=da        ba̱gwana̱m. many=ACC   RED-cake    eat-PERF=DIST=DET    man ‘The man ate many cakes.’   88                VoiceP                              Voice’                  -PASS                 vP                             SubDP                   v’                                               v                      VP                                                                       V                      DP                                                                                   ObjDP              Figure 11 Object cannot move up  6.4.2 Passive Sentence Structure This section explains how the passive structure is obtained. Sherer (2014) states that case-marked DPs cannot be extracted without being promoted to subject, while non-case-marked DPs can. She claims that this is due to a minimality effect, where the highest DP is the one allowed to extract, and subjects and non-case marked adjuncts are equally high in the tree, whereas case-marked elements are lower. In Figure 12, the Voice head, which has a +PASS feature, blocks the subject DP in [Spec, vP] from moving up to [Spec, Voice], but because the subject position is empty, the non-case-marked object moves to the subject position.    89              VoiceP                              Voice’                  +PASS                 vP                                 vP                  	  DP                 SubDP                v’                                            v                        Figure 12 Object fills in the subject position  6.5 Structure of Bare Quantifiers 6.5.1 Bare Quantifiers without Determiners For sentences with bare quantifiers (i.e., sentences in which a quantifier is associated with a non-contiguous DP), grammaticality judgments depend on three factors: 1) which quantifier is being used, 2) whether the quantifier is associated with the subject or the object, 3) whether it is S1 or S2 making the judgment. S1 accepts sentences with bare wi'la, no matter if wi'la is associated with the subject or the object. She also accepts sentences with bare ḵ̓ina̱m if ḵ̓ina̱m is associated with the subject, but she usually rejects sentences with bare ḵ̓ina̱m if it is associated with the object. S1 rejects all sentences with bare numerals. On the other hand, S2 accepts sentences with bare wi'la only when it is associated with the subject, but not when it is associated with the object. She accepts sentences with bare ḵ̓ina̱m when it is associated with the subject. She also occasionally accepts sentences with bare ḵ̓ina̱m when it is associated with the object, but only with a following pause, or ‘comma intonation’   90 after ḵ̓ina̱m. Usually, however, she rejects these types of sentences. I did not test these types of sentences with S1 (i.e., sentences with bare ḵ̓ina̱m associated with the object with comma intonation). Likewise, I was not able to test the sentences with bare wi'la and numerals for S2. The data for bare quantifiers is given in sections 3.3.2.2, 4.3.2.2, and 5.3.2.2. The information from those sections is summarized in the table below. The first line shows speaker judgments for bare quantifiers associated with the subject. The second line shows speaker judgments for bare quantifiers associated with the object. The third line shows speaker judgments for bare quantifiers associated with the object, with a pause after the bare quantifier.  wi'la ḵ̓ina̱m Numerals Bare Quantifier S1 S2 S1 S2 S1 S2 √ - √ - * - QSUB   Verb   [D   Sub]   Obj √ * ? * * - QOBJ   Verb   Sub   [D    Obj]    ?   QOBJ,   Verb   Sub    [D    Obj] Table 26 Summary of Distribution: Bare Quantifiers without Determiners  Since some data are missing for S2, I discuss only S1’s data in the rest of this section. One putative difference between wi'la and ḵ̓ina̱m is whether the quantifiers can be auxiliaries or not. In section 6.1, I explained that only wi'la can be analyzed as an auxiliary, and auxiliaries can appear at the beginning of the sentence. So I assume that bare quantifier wi'la can be an auxiliary. However, the status of ḵ̓ina̱m as an auxiliary is more difficult to establish. Though in section 6.2.1 above I claimed that it coud not function as an auxiliary, examples such as (84) below, repeated from (38b), seem to argue the opposite: here, ḵ̓ina̱m can neither be analysed as part of a fronted DP (since its nominal restriction is stranded in post-predicative position) nor a main   91 predicate (since it is not followed by a DP complement); this leaves auxiliary status as the only remaining analytical potion. More research is clearly required here.  84  S1  ? [QOBJ      Verb                        Sub         [D    Obj   ]] ḵ̓ina̱m   ha̱'m-x'id=[DP i=da       [NP ba̱gwana̱m]] =x̱a  ka̱leps. many    eat-PERF=DIST=DET          man=ACC        grape ‘The man ate many grapes.’               AuxP                                Aux’                  Aux                   VoiceP              ḵ̓ina̱m                               SubDP                 vP  Figure 13 Bare Ḵ̓ina̱m at Aux   92 7. STRONG VS. WEAK QUANTIFIER DISTINCTIONS This chapter explores the semantic characterizations of strong and weak quantifiers, focusing in particular on the distinction between cardinal and proportional readings, as characterized in the third line of Table 27. This test is only available for ḵ̓ina̱m, as I will explain in the next section.  Property Strong Quantifiers Weak Quantifiers Can be one-place main predicates No Yes DP-Internal Position Pre-Determiner Post-Determiner Cardinal or Proportional Readings Proportional Cardinal  Table 27 Properties of Strong/Weak Quantifiers  This chapter also analyzes the syntax-semantics interface of quantifiers in Kwak’wala, and in particular whether the semantic properties of quantificational expressions are correlated with their structure and/or syntactic position.  7.1 Theoretical Background The first property shown in Table 27 is a syntactic as well as a semantic property, as shown in sections 3.1, 4.1, and 5.1. Strong quantifiers, such as all or most, take two arguments, a restriction and a nuclear scope; in contrast, weak quantifiers like no, many, some, or few take a single argument and are interpreted intersectively (Keenan (1996), Milsark (1979), Partee (1988)). The examples in (85) demonstrate this.      93 85 (a)  All women danced. (b)  Many women danced.  In (85a), the two arguments of the quantifier are women and danced, and all relates these two, making the set of women a subset of the set of dancers, as shown in Figure 14.              the set of women is a subset of the set of dancers (if a person is a woman, this person is a dancer)  Figure 14 Set Theoretic Representation of “All women danced.”  On the other hand, many is a one-place predicate which can be interpreted intersectively. In (85b), there is a set of people who both danced and are women, and its cardinality is large, as shown in Figure 15.       danced women   94  women     danced           a set of people who both danced and are women (and its cardinality is large, say 50)  Figure 15 Set Theoretic Representation of “Many women danced.”  What Figure 15 shows is that the interpretation is that the set of women who danced is large. This does not require looking at the proportion of women dancers compared to the size of the set of women, unlike a strong quantifier would do. We can interpret many as a one-place predicate such that there is a set X, and X consists of women, and X has a large cardinality, and the members of X danced15. This is the weak reading of many. While in English, weak quantifiers do not usually function syntactically as main predicates, they may be distinguished by there-insertion (Milsark 1979) which permits weak but not strong quantifiers (86).  86 (a)  There are many cats. (b)  * There are all the cats.                                                  15 The notion of “large cardinality” is relative to a standard set by the speaker: hence the alternative term “value judgment quantifier” sometimes employed for elements such as many.   95 Since (the) cats is an argument, but there is not, (86a) is felicitous as there is only one argument which the quantifier many takes. (86b), on the other hand, is infelicitous since the quantifier all needs to take two arguments. The there-insertion test also distinguishes strong or proportional readings of weak quantifiers from weak or cardinal readings (the second property in Table 27) The example in (87) shows that there-insertion is infelicitous with proportional many, just as it is with all.  87 * There were many of the dogs.  Partee (1988) argues that many is ambiguous, having both cardinal and proportional readings, as briefly explained in 1.2.2.1. On the other hand, strong quantifiers, such as all or most, have only a proportional reading since their truth always depends on the relation between two arguments (e.g., women and danced in Figure 14). Consider the following sentences, repeated from (8): (88a) has a cardinal reading, and (88b) has a proportional reading.  88 (a)  There are many dogs who barked. (b)  Many of the dogs barked.  Reading Truth Condition Cardinal | dogs ∩ barked| ≥ n, context-dependent n Proportional | dogs ∩ barked| ≥ k, context-dependent k         | dogs | Table 28 Truth Conditions of Cardinal/Proportional Readings (Partee 1988)    96 The truth conditions for the cardinal and proportional readings are listed in Table 28. The cardinal reading is satisfied if there is a large number, say 100, or even as low as 10 in certain contexts, but crucially independent of the proportion. So if 10 contextually counts as a large number, it will count as large even if it is a small proportion, e.g., if there were 50 dogs total and 10 barked. On the proportional reading, there has to be a large proportion, so 10 out of 15 will work, and even 4 out of 5 will work, although 4 is a small number. But 10 out of 50 will not satisfy the proportional reading. Turning to the meaning of the sentences in (88), (88a) has only a weak reading and requires that the context involves a large number of dogs cardinally, but does not require a large proportion, so the sentence is felicitous if ten dogs barked. The sentence would be infelicitous if only three dogs barked. (88b), on the other hand, has a proportional reading, so the sentence is infelicitous if there are 500 dogs but only 10 dogs barked, although ‘10’ is a large number. In the next section, I describe the cardinal and proportional readings of weak Qs in Kwak’wala. I examined only sentences with ḵ̓ina̱m since universals (wi'la in Kwak’wala) and numerals are never ambiguous with respect to the cardinal versus proportional distinction, like in English. In terms of universals, the sentence in (89) is felicitous only when all the women in the given context went home (e.g. there are 20 women in the classroom and all of them went home). Numerals have only a cardinal reading.  89 S1  √√    /    S2  √√ wi'la-'m=i=da          ts'i-ts'i-ts'a̱daḵ     ne'nakwa. all-DISC=DIST=DET     RED-RED-woman  go.home ‘All the women went home.’    97 7.2 Cardinal and Proportional Readings in Kwak’wala I tested cardinal vs. proportional readings for ḵ̓ina̱m in each argument position. I list cardinal and proportional readings for S1 in section 7.2.1 and for S2 in section 7.2.2. The elicitation scenarios were conducted like the ones illustrated for English sentences in section 1.2.2.1; each time, pictures, models constructed from Legos, or origami paper dolls were provided to give consultants clear contexts. The prediction here is that, in all the cases, the consultants will reject context (i) because the number of the individuals which satisfy the intersection of the nominal predicate and the VP is small both cardinally and proportionally. On the other hand, I predict that the consultants will accept context (ii) if the sentences have proportional readings, and accept context (iii) if the sentences have cardinal readings. Also, it is predicted that both consultants will accept context (iv), in which the number of individuals which satisfy the intersection of the nominal predicate and the VP is big both cardinally and proportionally.  7.2.1 S1’s Judgments In terms of semantic judgments, S1 sometimes gives a negative judgment when the context supports a proportional reading with a small number of individuals corresponding to the nominal predicate of the quantifier: the sentences are dispreferred but still sometimes acceptable. In (90), ḵ̓ina̱m quantifies over the subject in pre-predicative position.       98 90 Q-Subject   Predicate ḵ̓ina̱m=i=da        bi-bi-ba̱gwana̱m  dłax̱w-x'id.  many=DIST=DET    RED-RED-man    stand.up-PERF     ‘Many men stood up.’  Context (i): showed 2 LEGO people standing up and 3 LEGO people lying down [Judgment: REJECTED]  Context (ii): showed 4 LEGO people standing up and 1 LEGO person lying down [Judgment: Sometimes ACCEPTED].  Context (iii): showed 14 LEGO people standing up and 3 LEGO people lying down [Judgment: ACCEPTED]  Context (iv): showed 10 LEGO people standing up and 30 LEGO people lying down [Judgment: ACCEPTED]  In (91), ḵ̓ina̱m quantifies over the subject, which sits between the auxiliary and the predicate.   91 Aux   Q-Subject   Predicate la'm=i=da           ḵ̓ina̱m    bi-bi-ba̱gwana̱m   dłax̱w-x'id. AUX=DIST=DET     many     RED-RED-man     stand.up-PERF ‘Many men stood up.’  Context (i): showed 2 LEGO people standing up and 5 LEGO people sitting down. [Judgment: REJECTED]  Context (ii): showed 8 LEGO people standing up and 1 LEGO person sitting down. [Judgment: ACCEPTED]  Context (iii): showed the diagram to explain that there were 100 men but only these 9 LEGO people stood up. [Judgment: ACCEPTED]  Context (iv): showed 20 LEGO people standing up and 30 LEGO people lying down [Judgment: ACCEPTED].  In (92), ḵ̓ina̱m quantifies over the oblique.   99 92 Predicate   Subject   Q-Oblique   Accusative t̓soya    Ruby=sa    ḵ̓ina̱m    buk=lax̱a    Ayako. give     Ruby=OBL  many     book=PREP   Ayako ‘Ruby gave many books to Ayako.’   Context (i): showed 5 books to her and said that she gave me 2 books and kept 3 books. [Judgment: REJECTED]  Context (ii): showed 5 books to her and said that she gave me 4 books but she kept 1 book. [Judgment: Sometimes ACCEPTED]  Context (iii): showed 30 books to her and said that she gave me 6 books but she kept the others. [Judgment: ACCEPTED]  Context (iv): showed 30 books and said that she gave me 28 books but kept 2 books. [Judgment: ACCEPTED]  In (93), ḵ̓ina̱m quantifies over the object (with the accusative marker).  93 Predicate   Subject   Q-Accusative duḵwa̱l=an=x̱a      ḵ̓ina̱m      gi-ga̱lga̱'omas. see=1PRON=ACC    many       RED-animal ‘I saw many animals.’  Context (i): showed 5 animals to her and said that she saw 2 animals and missed 3 animals. [Judgment: REJECTED]  Context (ii): showed 6 LEGO animals to her and said that she saw 5 animals and she missed the other one. [Judgment: Sometimes ACCEPTED]  Context (iii): showed a diagram to explain that there were 100 animals in the zoo but she only saw these 11 LEGO animals. [Judgment: ACCEPTED]  Context (iv): showed 16 animals and said that she saw 13 animals but missed the others. [Judgment: ACCEPTED]  In (94), ḵ̓ina̱m quantifies over a prepositional phrase (with the preposition marker).    100 94 Predicate   Subject   Accusative   Q-Preposition duḵwa̱l=i      John=x̱a    Peggy  la  lax̱a   ḵ̓ina̱m    paati. see=DIST      John=ACC  Peggy  go PREP  many     party  ‘John saw Peggy at many parties.’  Context (i): used 1 LEGO person and 3 houses (party places) to explain that John went to three parties in 1 week and saw Peggy at two of them [Judgment: REJECTED].  Context (ii): used 1 LEGO person and 5 houses (party places) to explain that John went to 5 parties in 1 week and saw Peggy at 4 of them [Judgment: Sometimes ACCEPTED].  Context (iii): explained that John is a party animal so he went to 50 parties in one month and he saw Peggy at 10 parties. [Judgment: ACCEPTED]  Context (iv): showed 15 party houses to explain that John is a party animal so he went to 15 parties in a week and Peggy also went to 12 parties. [Judgment: ACCEPTED]    The following table summarizes S1’s judgments. S1 accepts the sentences in the situations (ii-iv), at least sometimes; but the sentences are infelicitous when the context involves a small number of individuals and a small proportion like in context (i). In these contexts, the speaker told me to use the actual number instead of ḵ̓ina̱m. Likewise, even when the context involved a large proportion of individuals, the consultant sometimes told me to use the actual number when I used a small number of individuals in the situation (ii) cases. However, I suspect that this variable judgment is caused by using the wrong methodology, since every time I did the readings test I first asked her the basic cardinal test, e.g. “how many animals are ‘many’ in this situation”. Therefore, after S1 had said that “6 animals” are required to say ḵ̓ina̱m =i=da gi-ga̱lga̱'omas (‘there are many animals’), it sounded odd to her to be asked if the sentence duḵwa̱l= a̱n=x̱a ḵ̓ina̱m gi-ga̱lga̱'omas (‘I saw many animals’) is accepted in the situation “she saw 5 animals and she missed the other one” because she has already answered that ‘less than 6’ is not ḵ̓ina̱m   101 (‘many’). Therefore, I predict that she would more consistently accept the sentence with a small number and a large proportion if the test were appropriately conducted.   Situation Cardinal Prediction Proportional Prediction Result for S1 (i) small No., small proportion # # # (ii) small No., large proportion # √ ? (iii) large No., small proportion √ # √ (iv) large No., large proportion √ √ √ Table 29 S1’s Judgments on Cardinal/Proportional Readings  7.2.2 S2’s Judgments In terms of semantic judgments, S2 has the constant judgment that ḵ̓ina̱m has a cardinal reading, and gives a negative judgment when the context involves a large proportion of individuals but not a large cardinality. In (95), ḵ̓ina̱m quantifies over the subject in pre-predicative position. The consultant rejected context (ii), unlike S1. Note, however, that even if it is a large proportion, 2 probably cannot count as ‘many’ (just as it cannot in English, where ‘many’ clearly has a proportional reading).  95 Aux   Q-Subject    Predicate la̱-'m=i=da              ḵ̓ina̱m    ts'i-ts'i-ts'a̱daḵ      ne'nakwa. AUX-DISC=DIST=DET    many     RED-RED-woman   go.home ‘Many women went home.’                                       Context (ii):  explained that there are 3 female students in the university, and 2 women went home. [Judgment: REJECTED]   Context (iii): explained that there are 10,000 female students in the school and 200 students went home in the middle of the day, since we had a typhoon today and some trains are not running. [Judgment: ACCEPTED]    102 In (96), ḵ̓ina̱m also quantifies over the subject in pre-predicative position. (95) and (96) show a syntactic difference, in that ḵ̓ina̱m is before and after the determiner respectively. Unlike with English ‘many’, the judgment for ḵ̓ina̱m is not affected by its position. The consultant rejected context (ii), but again even if it is a large proportion, 2 probably cannot count as ‘many’.  96 Aux   Q-Subject   Predicate la̱-'m        ḵ̓ina̱m=i=da        ts'i-ts'i-ts'a̱daḵ       ne'nakwa. Aux-Disc    many=DIST=DET    RED-RED-woman    go.home ‘Many women went home.’  Context (ii):  explained that there are 3 female students in the university, and 2 women went home. [Judgment: REJECTED]  Context (iii):  explained that there are 10,000 female students in the school and 200 students went home in the middle of the day, since we had a typhoon today and some trains stop easily. [Judgment: ACCEPTED]  In (97), ḵ̓ina̱m quantifies over the subject in a passive sentence. The consultant accepted context (ii) but judging from her comments, the judgment depended on its cardinality, not proportionality.  97 Passivized Sentence (Q-Theme(subject)    Predicate   Agent) ḵ̓ina̱m=i      ki-kegas,    ha̱'m-x'id-su-wa̱ł      Jessie. many=DIST   RED-cake  eat-PERF-PASS-PST    Jessie. ‘Jessie ate many cakes.’   Context (ii):  explained that there are 3 cakes and Jessie ate 2 of them. [Judgment: ACCEPTED. However, she mentioned that two is still a lot for one person. She did not care that it is proportionally big.]      103 Context (iii): explained that Jessie tried to continue to eat cakes for 24 hours, on a TV program, and she ate 70 cakes. She was actually able to eat 500 cakes because she did it before, but this time she ended up getting sick so she could only eat 70 cakes. [Judgment: ACCEPTED]  In (98), ḵ̓ina̱m quantifies over a prepositional phrase (with the preposition marker). The consultant rejected context (ii) but even if it is a large proportion, 2 probably cannot count as ‘many’.  98 Predicate   Subject   Accusative   Q-Preposition duḵwa̱l=i     John=x̱a     Peggy  lax̱a   ḵ̓ina̱m   ikaslas. see=DIST     John=ACC   Peggy  PREP  many    party. ‘John saw Peggy at many parties.’  Context (ii):  explained that John went to 3 parties this year and saw Peggy at 2 parties. [Judgment: REJECTED]  Context (iii): explained that John went to 200 parties this year and saw Peggy at 50 parties. [Judgment: ACCEPTED]  The following table summarizes S2’s judgments. Since she has the constant judgment that ḵ̓ina̱m has a cardinal reading, and gives a negative judgment when the context involves a large proportion of individuals but not a large cardinality, ‘#’ is put in the box for context (ii), unlike S1. While S2 definitely has the cardinal reading, we do not know for certain if S2 lacks the proportional reading because the “2 out of 3” does not tell us much: even if it is a large proportion, 2 probably cannot count as ‘many’ (as in English), except in (97) (which is analyzable as being accepted under its cardinal reading rather than its proportional one). But since S1 also sometimes rejected the sentences when the context involved a large proportion of individuals but not a large cardinality, the speakers are not so different in judgments.   104  Situation Cardinal Prediction Proportional Prediction Result for S2 (i) small No., small proportion # # - (ii) small No., large proportion # √ # (iii) large No., small proportion √ # √ (iv) large No., large proportion √ √ - Table 30 S2’s Judgments on Cardinal/Proportional Readings  7.3 Analysis This section explores a semantic account of the strong/weak distinction in Kwak’wala.  7.3.1 Strong/Weak Distinction and Predicative Status of Kwak’wala Quantifiers (25) and (40), repeated below as (99), show that the quantifier wi'la (‘all’) cannot be the main predicate, unlike ḵ̓ina̱m (‘many’). Since there is only one argument ((=ox̱=da) gi-gukw), whether the sentences are felicitous depends on the number of arguments the quantifiers ḵ̓ina̱m and wi'la need.  99 (a)  S1  √√    /    S2  √√ [Q       D            Sub     ] ḵ̓ina̱m=ox̱=da        gi-gukw. many=MED=DET     RED-house ‘There are many houses.’  (b)  S1  *    /    S2  * [Q          D               Sub     ] wi'la(-'m)=ox̱=da          gi-gukw. all(-DISC)=MED=DET       RED-house        ‘all the houses’ S2 Comment: It is not a sentence.   105 There is a plausible semantic explanation for the fact that only ḵ̓ina̱m, not wi'la, can function as a main predicate: wi'la cannot be analyzed as a predicate due to a type mismatch (according to an analysis of ḵ̓ina̱m as <e, t> and wi'la as <<e, t>, t>) (see Milsark (1977), Barwise & Cooper (1981), Keenan (1987), Partee (1995)). The different internal structure of quantified DPs, shown in section 3.2 and 4.2, also supports the idea that only ḵ̓ina̱m acts like an adjective whose type is <e, t >.  7.3.2 Strong/Weak Distinction and Cardinal and Proportional Readings in Kwak’wala As explained in (89) and repeated in (100), wi'la is never ambiguous with respect to the cardinal versus proportional distinction: the sentence is felicitous only when all the women in the given context went home.  100 S1  √√    /    S2  √√ wi'la-'m=i=da          ts'i-ts'i-ts'a̱daḵ     ne'nakwa. all-Disc=DIST=DET     RED-RED-woman  go.home ‘All the women went home.’ Context: There are 100 girls in the garden, and 100 girls went home.  On the other hand, S1 has cardinal and proportional readings for ḵ̓ina̱m, although she disprefers the latter reading (there is not enough data to decide whether S2 has only a cardinal reading for ḵ̓ina̱m).  This means that S1 has strong and weak readings of ḵ̓ina̱m and S2 has at least a weak quantifier ḵ̓ina̱m.  Therefore, S1 has a strong wi'la, a strong ḵ̓ina̱m, and a weak ḵ̓ina̱m, and S2 at least has a strong wi'la and a weak ḵ̓ina̱m. The following table is a revised version of Table 25, showing the weak versus strong distinction.     106  wi'la (‘all’) ḵ̓ina̱m (‘many’) ḵ̓ina̱m (‘many’) numerals  Strong Strong Weak Weak Predicate Status * * √ √ DP-Internal Pred [D Q N] * √ √ Pred [Q D N]  ‘non-subject’ √ * * Pred [Q D N]   ‘subject’ √ S1 ? / S2 * S1 √ / S2 * [D Q N] Pred * √ √ [Q D N] Pred √ √ √ Semantics Cardinal N/A * √ N/A Proportional N/A S1 √ / S2  - * N/A Table 31 Strong and Weak Distinction in Kwak’wala Quantifiers  7.3.3 Position Mismatches Now I return to the DP-internal positioning of the quantifier. In sections 3.2 and 4.2, we saw that although S1 dispreferred the proportional reading, ḵ̓ina̱m has both cardinal and proportional readings. However, there is no correlation between pre- vs. post-determiner positions for the strong vs. weak readings of ḵ̓ina̱m. On the other hand, since it is not clear if S2 lacks the proportional reading, we do not know if there is any correlation between pre- vs. post-determiner positions for the strong vs. weak readings of ḵ̓ina̱m for S2. However, there is one case in which there is a correlation between the readings and syntactic position, namely, existential sentences.  101  (a) S1  √√    /    S2  √√ [Q       D              Sub     ] ḵ̓ina̱m=ox̱=da        bi-busi. many=MED=DET     RED-cat ‘There are many cats.’    107 I did not test the readings of this sentence with S2 but S1 rejects the sentence with a proportional reading, where I know there are 6 cats in the village and I saw 5 of them over there when I visited (she said that the sentence is felicitous when there are more than 5 cats). In section 6.3, based on a distributive difference among wi'la, ḵ̓ina̱m, and numerals, I argued that the quantifiers are base-generated in different places. Wi'la, which obligatorily appears in pre-determiner position, is generated in pre-determiner position, as in Figure 16a. On the other hand, ḵ̓ina̱m and numerals, whose distribution varies, can appear in pre- or post-determiner position, as shown in Figure 16b, rather than that they are base-generated there. But the readings of sentences with ḵ̓ina̱m are not affected by the position of the quantifier. Likewise, numerals, which have only cardinal readings, are generated either in pre- or post-determiner position. However, I have no actual evidence as to whether they are base-generated or moved into pre-determiner position.  (a)          DP      QP                   DP       Q               D          NP  wi'la ḵ̓ina̱m  numerals   (b)          DP       D                    NP                        QP          NP                           Q             ḵ̓ina̱m/numerals  Figure 16 DP Internal Structure with Wi'la and Ḵ̓ina̱m/Numerals   108 8. QUANTIFIER SCOPE This section describes the ambiguity between collective, distributive, and cumulative readings of quantifiers, and how it is related to quantifier scope.  8.1 Theoretical Background It is broadly recognized that expressions like negation and quantification are scopal expressions, and a sentence with two or more scopal expressions is commonly ambiguous between different readings. In the case of two quantified phrases, they will commonly be ambiguous between collective, distributive, and cumulative readings. These readings have been explored in many languages in order to investigate the different possible scopes, and I do so for Kwak’wala in section 8.2. The phenomenon of scope can be illustrated by looking first at the English sentences in (102).  102 (a)  Mary loves Ken. (b)  Someone loves everyone.  The sentence in (102a) is true only when there are referents called Mary and Ken and when it is true that Mary loves Ken. But the meaning of (102b) is ambiguous and has two readings. For example, in (102b) if someone has wide scope over everyone, then there is one person who loves everyone. On the other hand, if everyone has wide scope over someone, then everyone is loved by a potentially different person.   109 A common way to analyze scope ambiguities is to reduce them to structural ambiguities (May 1985). The different structures are shown in Figure 17. The universal quantifier everyone is required to move up at Logical Form in order to get the (b) structure.  (a)               someone                        everyone    (b)        everyone                someone                             everyone Figure 17 Logical Form: Relations of Quantifiers in English  Someone c-commands everyone on the surface, so the wide scope reading for the universal quantifier is derived as shown in (b). That requires moving up the universal quantifier everyone at Logical Form, since the movement does not change the word order at surface form. This movement is called quantifier raising (May 1985). In order to investigate quantifier scope in Kwak’wala, I will lay out 3 kinds of readings (collective, distributive, cumulative) of quantified sentences in the next section.   8.2 Collective, Distributive, and Cumulative Readings in Kwak’wala This section lays out the collective, distributive, inverse-scope distributive, and cumulative readings for S1 and S2; both speakers have the same judgments about scope. The data were provided to the speakers, in different orders, in different sessions. It can be seen that in Kwak’wala, all the readings are attested in all of the following templates. The number after the   110 template shows where the data are included in this thesis. The last two sets are given in the Appendix, since their quantified elements (like nominative subjects or accusative elements) are the same as those of (103), with different word order, and the word order does not interfere with speakers’ judgments. For each case, S2’s judgment is listed only when available.  103 (a)  NomQP  Predicate  AccQP  (107) (b)  NomQP  Predicate  OblQP   (108) (c)  NomQP  Predicate  PrepQP   (109)  104 (a)  Predicate  Nom  AccQP  OblQP  (110) (b)  Predicate  Nom  AccQP  PrepQP  (111) (c)  Predicate  Nom  OblQP  PrepQP  (112)  105 in Appendix B.1 (a)  Predicate  NomQP  AccQP  (118) (b)  Predicate  NomQP  OblQP  (119) (c)  Predicate  NomQP  PrepQP  (120)  106 in Appendix B.2 (a)  Aux  NomQP Predicate AccQP  (121) (b)  Aux  NomQP Predicate OblQP  (122) (c)  Aux  NomQP Predicate PrepQP  (123)   111 The elicitation scenarios were conducted like the ones illustrated for English sentences in section 1.2.2.3, using pictures, models constructed from Lego, and origami paper dolls.  107 Nominative Marker and Accusative Marker (1)  S1  √√ wi'la-'m=ox̱=da        ts'i-ts'a̱daḵ   dala=x̱a     sa̱k'a  gi-ga̱ldas   lax̱=i's      gukw. all-DISC=MED=DET    RED-woman  carry=ACC  five   RED-box   PREP=POSS  house ‘All the women carried five boxes to her (another woman’s) house.’   [Collective] used 4 LEGO people and five big box drawings, and explained that Ann, Jessica, Kate, and Beth helped their teacher to move but because the boxes are big, they carried the boxes together (4 women, 5 boxes).  [Judgment: ACCEPTED]  [Distributive] used 4 LEGO people and 20 box drawings, and explained that Ann, Jessica, Kate, and Beth helped their teacher to move, and each of the four women carried five boxes (4 women, 20 boxes).  [Judgment: ACCEPTED]  [Cumulative] used 4 LEGO people and 5 box drawings, and explained that they helped their teacher to move by carrying five boxes to her house, but we do not know who carried how many boxes (4 women, 5 boxes).  [Judgment: ACCEPTED]  [Inverse Distributive] used 20 people created by origami and 5 big box drawings, and explained that the teacher asked her students to help her to move and 20 female students helped her. There are five boxes and each box is carried by four women to her house (20 women, 5 boxes).  [Judgment: ACCEPTED]   The Number of Boxes Possible Reading Scopal Configuration: wi'la (‘all’) / sa̱k'a (‘five’) Ann Jessica Kate Beth 5 (Together) Collective √ [wi'la, sa̱k'a] (scopeless) 5 5 5 5 Distributive √ [wi'la > sa̱k'a] 5 (Separately) Cumulative √ [wi'la, sa̱k'a]  (scopeless)  The Number of Women Possible Reading Scopal Configuration Box A Box B Box C Box D Box E 4 4 4 4 4 Inverse Distributive √ [sa̱k'a  > wi'la] Table 32 Readings with Nominative Marker and Accusative Marker for S1     112 (2)  S2  √√ mu=i=da          ts'a̱daḵ    duḵwa̱la=x̱a    sa̱k'a  bi-ba̱gwana̱m. four=MED=DET     woman    see=ACC       five   RED-man ‘Four women saw five men.’   [Collective] showed a picture to explain that four woman together saw five men (4 women, 5 men).  [Judgment: ACCEPTED]  [Distributive] showed a picture to explain that each woman saw five men (4 women, 20 men).  [Judgment: ACCEPTED]  [Cumulative] showed a picture to explain that she knew four women in total saw five men. The speaker is looking for five men, John, Ken, Tom, James, and Jim. Mary saw John and Ken, but three other women each saw one man. She does not know who saw who, but you know these four women saw five men in total (4 women, 5 men).  [Judgment: ACCEPTED]  [Inverse Distributive] -  The Number of Men Possible Reading Scopal Configuration: mu (‘four’) / sa̱k'a (‘five’) Woman1 Woman2 Woman3 Woman4 5 (Together) Collective √ [mu, sa̱k'a] (scopeless) 5 5 5 5 Distributive √ [mu > sa̱k'a] 5 (Separately) Cumulative √ [mu, sa̱k'a] (scopeless) Table 33 Readings with Nominative Marker and Accusative Marker for S2  108 Nominative Marker and Oblique Marker (1)  S1  √√ ma'ł=i=da         buk-wa̱ł   k̓ak̓ada̱xwsila-su-wa̱ł =sa  sa̱k'a   gi-ga̱n-ga̱nana̱m. two=DIST=DET     book-PST  read-PASS-PST=OBL      five    RED-RED-child. ‘Two books were read by five children.’   [Collective] used 5 LEGO people and 2 book drawings, and explained that Tom, Lisa, Ken, Mike, and Sally sat together to read two books (2 books, 5 children).  [Judgment: ACCEPTED]  [Distributive] used 10 people created by origami and 2 book drawings, and explained that each book was read by a different five children (2 books, 10 children).  [Judgment: ACCEPTED]    113 [Cumulative] used 5 LEGO people and 2 book drawings, and explained that Tom, Lisa, Ken, Mike, and Sally read two books. Some children read Book A and some children read Book B, but we do not know who read which books (2 books, 5 children).  [Judgment: ACCEPTED]  [Inverse Distributive]  used 5 LEGO people and 10 book drawings, and explained that Tom, Lisa, Ken, Mike, and Sally each read two books (10 books, 5 children).  [Judgment: ACCEPTED]  The Number of Children Possible Reading Scopal Configuration: ma'ł  (‘two’) / sa̱k'a (‘five’) Book A Book B 5 (Together) Collective √ [ma'ł, sa̱k'a] (scopeless) 5 5 Distributive √ [ma'ł  > sa̱k'a] 5 (Separately) Cumulative √ [ma'ł, sa̱k'a] (scopeless)  The Number of Books Possible Reading Scopal Configuration Tom Lisa Ken Mike Sally 2 2 2 2 2 Inverse Distributive √ [sa̱k'a  > ma'ł] Table 34 Readings with Nominative Marker and Oblique Marker for S1  (2)  S2  √√ ma'ł=i=da         buk-wa̱ł   k̓ak̓ada̱xwsila-su-wa̱ł=sa  sa̱k'a   gi-ga̱n-ga̱nana̱m. two=DIST=DET     book-PST  read-PASS-PST=OBL      five    RED-RED-child. ‘Two books were read by five children.’  [Collective] used 5 people and 2 book drawings, and explained that Tom, Lisa, Ken, Mike, and Sally sat together to read two books (2 books, 5 children).  [Judgment: ACCEPTED]  [Distributive] -  [Cumulative] used 5 people and 2 book drawings, and explained that Tom, Lisa, Ken, Mike, and Sally read two books. Some children read Book A and some children read Book B, but we do not know who read which books (2 books, 5 children).  [Judgment: ACCEPTED]  [Inverse Distributive] used 5 people and 10 book drawings, and explained that Tom, Lisa, Ken, Mike, and Sally each read two books (10 books, 5 children).  [Judgment: ACCEPTED]      114 The Number of Children Possible Reading Scopal Configuration: ma'ł  (‘two’) / sa̱k'a (‘five’) Book A Book B 5 (Together) Collective √ [ma'ł, sa̱k'a] (scopeless) 5 5 Distributive - [ma'ł  > sa̱k'a] 5 (Separately) Cumulative √ [ma'ł, sa̱k'a] (scopeless)  The Number of Books Possible Reading Scopal Configuration Tom Lisa Ken Mike Sally 2 2 2 2 2 Inverse Distributive √ [sa̱k'a  > ma'ł] Table 35 Readings with Nominative Marker and Oblique Marker for S2  109 Nominative Marker and Preposition Marker S1  √√    /    S2  √√ sa̱k'=i=da          gi-ga̱n-ga̱nana̱m    la   lax̱    mu   gi-gukw. five=3rd.LOC=DET   RED-RED-child    go  PREP  four   RED-house. ‘Five children went to four houses.’  [Collective] used 5 LEGO/origami people and 4 house drawings, and explained that Tom, Lisa, Ken, Mike, and Sally went to 4 houses together (5 children, 4 houses).  [Judgment: ACCEPTED]  [Distributive] used 5 LEGO/origami people and 20 house drawings, and explained that Tom, Lisa, Ken, Mike, and Sally each went to 4 houses (5 children, 20 houses).  [Judgment: ACCEPTED]  [Cumulative] used 5 LEGO/origami people and 4 house drawings, and explained that Tom, Lisa, Ken, Mike, and Sally went to these houses. Some of them went to more houses than others, and all the houses had someone but no houses had all the children. We do not know who went to which house (5 children, 4 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] used 20 people created by origami and 4 house drawings, and explained that there are four groups which consist of five children and each group went to a different house (20 children, 4 houses).  [Judgment: ACCEPTED]       115 The Number of Children Possible Reading Scopal Configuration: mu (‘four’) / sa̱k'a (‘five’) House A House B House C House D 5 (Together) Collective √ [mu, sa̱k'a] (scopeless) 5 5 5 5 Distributive √ [mu > sa̱k'a] 5 (Separately) Cumulative √ [mu, sa̱k'a] (scopeless)  The Number of Houses Possible Reading Scopal Configuration  Tom Lisa Ken Mike Sally 4 4 4 4 4 Inverse Distributive √ [sa̱k'a > mu] Table 36 Readings with Nominative Marker and Preposition Marker  110 Accusative Marker and Oblique Marker (1)  S1  √√ ga̱lsa   Mary=x̱a     mu   gi-gukw=sa       ma'ł    gi-ga̱lyayu. paint   Mary=ACC   four   RED-house=OBL   two    RED-brush ‘Mary painted four houses with two brushes.’   [Collective] used 1 LEGO person, two brushes (with different colors) and 4 house drawings, and explained that Mary painted four houses with two colors (4 houses, 2 brushes).  [Judgment: ACCEPTED]  [Distributive] used 1 LEGO person and 8 brushes (with different colors) and 4 house drawings and explained that she painted 4 houses with 2 different colors (4 houses, 8 brushes).  [Judgment: ACCEPTED]  [Cumulative] used 1 LEGO person and two brushes (with different colors) and 4 house drawings, and explained that we heard that Mary painted 4 houses using two brushes but we do not know how she painted (4 houses, 2 brushes).  [Judgment: ACCEPTED]  [Inverse Distributive] used 1 LEGO person and two brushes (with different colors) and 4 house drawings, and explained that Mary painted 4 houses with one color and the other 4 houses with the other color (8 houses, 2 brushes).  [Judgment: ACCEPTED]      116 The Number of Brushes Possible Reading Scopal Configuration: mu (‘four’) / ma'ł (‘two’) House1 House2 House3 House4 2 (Together) Collective √ [mu, ma'ł] (scopeless) 2 2 2 2 Distributive √ [mu > ma'ł] 2 (Separately) Cumulative √ [mu, ma'ł] (scopeless)  The Number of Houses Possible Reading Scopal Configuration Brush A Brush B 4 4 Inverse Distributive  √ [ma'ł  > mu] Table 37 Readings with Accusative Marker and Oblique Marker for S1   (2)  S2  √√ ga̱lsa   Mary=x̱a     mu   gi-gukw=sa       ma'ł   gi-ga̱lyayu. paint   Mary=ACC   four   RED-house=OBL   two    RED-brush ‘Mary painted four houses with two brushes.’   [Collective] -  [Distributive] -   [Cumulative] showed drawings to explain that the speaker asked Mary to paint four houses and gave her two brushes, but she does not know how Mary painted. What she knows is that she used two brushes and painted four houses. (2 brushes, 4 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] showed drawings to explain that Mary painted four houses with one brush and another four houses with the other brush (8 houses, 2 brushes).  [Judgment: ACCEPTED]   The Number of Brushes Possible Reading Scopal Configuration: mu (‘four’) / ma’ł (‘two’) House1 House2 House3 House4 2 (Together) Collective - [mu, ma'ł] (scopeless) 2 2 2 2 Distributive - [mu > ma'ł] 2 (Separately) Cumulative √ [mu, ma'ł] (scopeless)  The Number of Houses Possible Reading Scopal Configuration Brush A Brush B 4 4 Inverse Distributive √ [ma'ł  > mu] Table 38 Readings with Accusative Marker and Oblique Marker for S2    117 111 Accusative Marker and Preposition Marker (1)  S1  √ duḵwatł=an=x̱a       k'a̱t'ła   ni-'na̱mukw  lax̱   yuda̱xw   gi-gukw. see=1st.PRON=ACC   six      RED-friend   PREP three     RED-house ‘I saw six friends at three houses.’   [Collective] used 6 LEGO people and three houses to explain that I went to three houses, and met the same six friends in each house (6 friends, 3 houses).  [Judgment: ACCEPTED]  [Distributive] used 6 LEGO people and 18 houses to explain that I have 6 close friends, and went to 18 parties in one month. I met each friend in three houses (6 friends, 18 houses).  [Judgment: ACCEPTED]  [Cumulative] used 6 LEGO people and 3 houses to explain that I went to 3 parties and met 6 friends between the three parties. I met more friends in some parties than the others, so I do not give you the details but just want to say that I went to three houses for parties and met six friends in total (6 friends, 3 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] used 18 LEGO people and 3 houses to explain that I have many friends and met six different friends at three houses (18 friends, 3 houses).  [Judgment: ACCEPTED]  The Number of Houses Possible Reading Scopal Configuration: k'a̱t'ła (‘six’) / yuda̱xw (‘three’) Friend1 Friend2 Friend3 Friend4 Friend5 Friend6 3 (Together) Collective √ [k'a̱t'ła, yuda̱xw] (scopeless) 3 3 3 3 3 3 Distributive √ [k'a̱t'ła > yuda̱xw] 3 (Separately) Cumulative √ [k'a̱t'ła, yuda̱xw] (scopeless)  The Number of Friends Possible Reading Scopal Configuration House A House B House C 6 6 6 Inverse Distributive √ [yuda̱xw  > k'a̱t'ła] Table 39 Readings with Accusative Marker and Preposition Marker for S1  (2)  S2  √√ duḵwatł=ox̱    Ayako=x̱a    sa̱k'a   ni-'na̱mukw  lax̱   yuda̱xw   gi-gukw. see=MED     Ayako=ACC  five    RED-friend   PREP three     RED-house ‘Ayako saw five friends at three houses.’   [Collective] showed the picture to explain that Ayako went to three houses, and met the same five friends in each house (5 friends, 3 houses).  [Judgment: ACCEPTED]   118 [Distributive] showed the picture to explain that Ayako has five friends and met each friend in three houses (5 friends, 15 houses).  [Judgment: ACCEPTED]   [Cumulative] showed the picture to explain that Ayako went to three houses, and met five friends in total. I did not tell you who Ayako met at which house. You just know that Ayako went to three houses and met five friends in total (5 friends, 3 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] showed the picture to explain that Ayako met five different friends at three houses (15 friends, 3 houses).  [Judgment: ACCEPTED]  The Number of Houses Possible Reading Scopal Configuration: sa̱k'a (five) / yuda̱xw (‘three’) Friend A Friend B Friend C Friend D Friend E 3 (Together) Collective √ [sa̱k'a, yuda̱xw] (scopeless) 3 3 3 3 3 Distributive √ [sa̱k'a > yuda̱xw] 3 (Separately) Cumulative √ [sa̱k'a, yuda̱xw] (scopeless)  The Number of Friends Possible Reading Scopal Configuration House A House B House C 5 5 5 Inverse Distributive √ [yuda̱xw > sa̱k'a] Table 40 Readings with Accusative Marker and Preposition Marker for S2  112 Oblique Marker and Preposition Marker (1)  S1  √√ t̓soya    Ayako=sa    sa̱k'a    bi-buk      lax̱     yuda̱xw  bi-babagwa̱m. give     Ayako=OBL  five     RED-book   PREP  three     RED-boy ‘Ayako gave five books to three boys.’  [Collective] used 3 LEGO people and 5 books to explain that Ayako got 5 books for the boys and gave them to them to share (5 books, 3 boys).  [Judgment: ACCEPTED]  [Distributive] used 15 LEGO people and 5 books to explain that I have 15 students so I divided them into five groups. Then I got 5 books and gave one to each group (5 books, 15 boys).  [Judgment: ACCEPTED]  [Cumulative] used 3 LEGO people and 5 books to explain that Ayako mistakenly got 5 books, instead of 6 books so I gave the books to the boys and told them that they should decide which books they would keep. I do not know who kept which books (5 books, 3 boys).  [Judgment: ACCEPTED]   119 [Inverse Distributive] used 3 LEGO people and 15 books to explain that I got 15 books for the boys and gave five books to each of them (15 books, 3 boys).  [Judgment: ACCEPTED]  The Number of Boys Possible Reading Scopal Configuration: sa̱k'a (‘five’) / yuda̱xw (‘three’) Book 1 Book 2 Book 3 Book 4 Book 5 3 (Together) Collective √ [sa̱k'a, yuda̱xw] (scopeless) 3 3 3 3 3 Distributive √ [sa̱k'a > yuda̱xw] 3 (Separately) Cumulative √ [sa̱k'a, yuda̱xw] (scopeless)  The Number of Books Possible Reading Scopal Configuration Boy A Boy B Boy C 5 5 5 Inverse Distributive √ [yuda̱xw > sa̱k'a] Table 41 Readings with Oblique Marker and Preposition Marker for S1   (2)  S2  √√ t̓so=i         Hannah=sa    ma̱'łgwa̱'nał   bi-buk      lax̱     yuda̱xw  babagwa̱m. give=OBL     Hannah=OBL  eight          RED-book   PREP  three     boy ‘Hannah gave eight books to three boys.’  [Collective] showed the picture to explain that Hannah got eight books and gave them to three boys to share (8 books, 3 boys).  [Judgment: ACCEPTED]  [Distributive] -  [Cumulative] showed the picture to explain that Hannah got eight books and gave them to three boys, and they decided themselves who owns which books (8 books, 3 boys).  [Judgment: ACCEPTED]   [Inverse Distributive] showed the picture to explain that Hannah got 24 books and gave them to 3 boys. Each boy got eight books. (24 books, 3 boys).  [Judgment: ACCEPTED]     120 The Number of Boys Possible Reading Scopal Configuration: 'ma̱'łgwa̱'nał (‘eight’) / yuda̱xw (‘three’) B1 B2 B3 B4 B5 B6 B7 B8 3	 (Together) Collective √ ['ma̱'łgwa̱'nał, yuda̱xw ] (scopeless) 3 3 3 3 3 3 3 3 Distributive - ['ma̱'łgwa̱'nał > yuda̱xw] 3   (Separately) Cumulative √ ['ma̱'łgwa̱'nał, yuda̱xw] (scopeless)  The Number of Books Possible Reading Scopal Configuration Boy A Boy B Boy C 8 8 8 Inverse Distributive √ [yuda̱xw > 'ma̱'łgwa̱'nał] Table 42 Readings with Oblique Marker and Preposition Marker for S2  The data show that all readings are available in Kwak’wala, so each sentence has an ambiguous meaning. The inverse distributive reading is usually hard to get, so the speakers sometimes took some time to consider if the sentence is felicitous in the situation, but they accepted the reading as well. It is important to note that although I do not have any case where speakers provided a judgment of infelicity for this section, informal evidence can be provided that the speakers would have been willing to reject the sentences if the readings were in fact unavailable. This is shown by the fact that the speakers did point out wrong words, wrong word orders and/or wrong pronounciations when these occurred during the elicitation sessions.  8.3 Analysis As seen in section 8.2, Kwak’wala has all of the four readings (collective, distributive, inverse distributive, and cumulative). In particular, the inverse distributive reading indicates that the second quantified expression can have wide scope over the first one. This means that the second one is required to move up at Logical Form to get the reading. Therefore, I assume that the relations between two quantified DPs in the sentence with the readings shown in (107ii,iv) (repeated in (113ii,iv)) are as in Figure 18.   121 113 Nominative Marker and Accusative Marker S1  √√ wi'la-'m=ox̱=da       ts'i-ts'a̱daḵ   dala=x̱a     sa̱k'a  gi-ga̱ldas   lax̱=i's      gukw. all-DISC=MED=DET   RED-woman  carry=ACC  five   RED-box   PREP=POSS  house ‘All the women carried five boxes to her (another woman’s) house.’   [Distributive] used 4 LEGO people and 20 box drawings, and explained that Ann, Jessica, Kate, and Beth helped their teacher to move, and four women each carried five boxes (4 women, 20 boxes).  [Judgment: ACCEPTED]  [Inverse Distributive] used 20 people created by origami and 5 big box drawings, and explained that the teacher asked her students to help her to move and 20 female students helped her. There are five boxes and each box is carried by four women to her house (20 women, 5 boxes).  [Judgment: ACCEPTED]  (ii)   wi'la-'m=ox̱=da ts'i-ts'a̱daḵ                     sa̱k'a gi-gildas     (iv)   sa̱k'a gi-ga̱ldas    wi'la-'m=ox̱=da ts'i-ts'a̱daḵ                                 sa̱k'a gi-ga̱ldas  Figure 18 Logical Form: Relations of Kwak’wala Quantifiers  Both in (ii) and in (iv) the subject DP wi'la-'m=ox̱=da ts'i-ts'i-ts'a̱daḵ (‘all the women’) is generated higher than the object DP sa̱k'a gi-ga̱ldas (‘five boxes’). In (ii), what needs to be considered here is the situation in this reading: there are 4 women and 20 boxes. This means that the first DP scopes over the second one at surface structure, so there is no need for Quantifier Raising. In (iv), on the other hand, because the situation is that there are 20 women and 5 boxes due to the inverse distributive reading, the second DP sa̱k'a gi-ga̱ldas (‘five boxes’) needs to   122 scope over the subject wi'la-'m=ox̱=da ts'i-ts'i-ts'a̱daḵ (‘all the women’) to get the reading. Thus, the second DP moves up higher than the subject. Notice that the surface form in (113ii) is the same as that of (113iv) but they are different in Logical Forms, as shown in Figure 18. There are two interpretations for one surface form, which means that the surface ordering does not reflect the scopal interpretation. I also assume that for the cumulative reading, there is no Quantifier Raising and the structure at Logical Form is as in Figure 18 (ii).   123 9. IMPLICATIONS AND CONCLUSION 9.1 Differences in Quantifier Status between Salish and Kwak’wala This thesis investigated the existence of generalized quantifiers (GQs): i.e., whether Kwak’wala quantifiers can be analyzed as ‘GQ-creating (strong/proportional) quantifiers’. It appears that Kwak’wala does have real generalized quantifiers, as shown in chapter 7. This follows the NP-Quantifier universal proposed by Barwise and Cooper (1981) that all natural languages have NPs whose semantic function is to express generalized quantifiers. However, this issue has been much debated for Salish (cf. Jelinek 1995, Matthewson 1998, 2001, Davis 2010), with regard to the scopal interpretation.  Davis proposed that although the [DP Q D NP] structure predicts a distributive reading, in reality it has only a cumulative reading.  Neither DP-adjoined strong Qs nor DP-internal weak Qs with strong readings show any scopal asymmetries, so Salish lacks generalized quantifiers. On the other hand, in Kwak’wala, quantifiers have all of the collective, cumulative, distributive, and inverse distributive readings, and the inverse distributive reading indicates scopal interpretation. Therefore, although this implementation still requires further research to investigate if there is Quantifier Raising in Kwak’wala, this thesis showed that Kwak’wala and Salish languages differ in this respect.  9.2 Conclusion I have described and analyzed the syntax and semantics of quantification in Kwak’wala. I have argued that Kwak’wala has a strong quantifier wi'la (‘all’) and weak quantifiers like ḵ̓ina̱m (‘many’), which differ in distribution. I showed that one crucial difference between   124 ḵ̓ina̱m/numerals and wi'la is their predicative status; ḵ̓ina̱m/numerals can be main predicates but wi'la cannot, which parallels the situation in other languages like English. In argument positions, on the other hand, the distribution varies. wi'la appears only in pre-determiner positions, but judgments with respect to the positions of ḵ̓ina̱m and numerals vary. As objects, they can occur in post-determiner position. As subjects, they occur in pre- or post-determiner position in pre-predicative position. S2 rejects ḵ̓ina̱m and the numerals in post-determiner post-predicative position, but S1 varies in her judgments. However, there is no correlation between pre- vs. post-determiner positions for the strong vs. weak readings of ḵ̓ina̱m, unlike in the English case. Although quantified DPs can usually appear clause-initially in active and passive sentences, their syntactic status is unclear. Possibly wi'la can be analyzed either as an auxiliary or a part of DP, and ḵ̓ina̱m and numerals can be analyzed either as a predicate or a part of DP. Because there is no correlation between their positions and their interpretation as either cardinal or proportional, except in the sentences which are equivalent to existential sentences in English, further elicitation is required to investigate this issue. Scopal ambiguity with quantifiers is attested in Kwak’wala, which implies there is quantifier raising for inverse distributive readings. Although this thesis documents Kwak’wala quantifiers, the focus is limited to three kinds of quantifiers. Other Kwak’wala quantifiers are unexplored. This is a very small step, but it is my hope that this thesis could be of assistance in documenting the language of the Kwakwaka’wakw, with the ultimate aim of preventing its loss, and in gratitude to my wonderful consultants and their people.  125 References Anderson, S. R. (1984). Kwakwala Syntax and the Government-Binding Theory. In E.-D. Cook, & D. B. Gerdts, Syntax and semantics. Volume 16: The syntax of Native American Languages, pp. 21-75. New York: Academic Press.  Anderson, S. R. (2005). Aspects of The Theory of Clitics. Oxford: Oxford University Press.  Bach, E. (2006). Paradigm Regained: Deixis in Northern Wakashan. SOAS Working Papers in Linguistics 14, 267-281.  Barwise, J., & Cooper, R. (1981). Generalized Quantifiers and Natural Language. Linguistics and Philosophy 4, 159-219.  Black, A. (2009). The Nominal Phrase in Kwak'wala. Phonological and Semantic Interactions with Syntax. Ms., University of British Columbia.  Black, A. (2011). Definiteness, deixis, and demonstratives in Kwak’wala. UBC QP Defense Presentation, April 21, 2011.  Boas, F. (1947). Kwakiutl grammar: with a glossary of the suffixes. (Z. B. Harris., Ed.) Philadelphia: American Philosophical Society.  Bruening, B. (2008). Quantification in Passamaquoddy. In Matthewson, L. North-Holland Linguistic Series: Linguistics Variations. Volume 64: Quantification: a cross-linguistic perspective (pp. 67-103). Emerald Group Publishing Limited.  Chung, Y. (2007). The internal structure of the Kwak'wala nominal domain. The 42nd International Conference on Salish and Neighbouring Languages. 20, pp. 101-118. University of British Columbia Working Papers in Linguistics.  Collins, C. (2005). A smuggling approach to the passive in English. Syntax 8, 2, pp. 81-120. Davis, H. (2010). Salish languages lack generalized quantifiers after all! SALT 20. Evans, N. (1995) A-Quantifiers and Scope in Mayali. In E. Bach, E. Jelinek, A. Kratzer and B. H. Partee (eds.), Quantification in Natural Languages, pp. 207-270. Dordrecht, Kluwer.  Jelinek, E. (1995). Quantification in Straits Salish. In E. Bach, E. Jelinek, A. Kratzer and B. H. Partee (eds.), Quantification in Natural Languages, pp. 487-540, Dordrecht. Kluwer.  Janzen, J. (2015). Tsitsa̱ḵala̱m sa Ruby. Ms.  Legate, J. A. (2010). The Structure of Implicit Agents in Passives. North East Linguistics Society Annual Meeting.   126 Littell, P. (2016). Focus, Predication, and Polarity in Kwak’wala. Ph.D Dissertation. University of British Columbia.  Keenan, E. (1987). A semantic definition of "Indefinite NP". In E. Reuland and A. Meulen (eds), The Representation of (In)definiteness, 286-317. Cambridge, Massachusetts: MIT Press.   Keenan, E. (1996). The Semantics of Determiners. In S. Lappin, The Semantics of Determiners, pp. 41-65. Oxford: Basil Blackwell.  Matthewson, L. (1998). Determiner Systems and Quantificational Strategies: Evidence from Salish. The Hague: Holland Academic Graphics.  Matthewson, L. (2001). Quantification and the Nature of Cross-linguistic Variation. Natural Language Semantics, 9, 145-189.  Matthewson, L. (2004). On the Methodology of Semantic Fieldwork. International Journal of American Linguistics, 70, 369-415.  May, R. (1985) Logical form. Cambridge, Massachusetts: MIT Press.  Milsark, G. (1977). Toward an explanation of certain peculiarities of the existential construction in English. Linguistic Analysis, 3, 1-29.   Milsark, G. (1979). Existential sentences in English. New York: Garland Pub.  Nicolson, M., & Werle, A. (2009). An investigation of modern Kwak'wala determiner systems. Ms, University of Victoria.  Partee, B. H., Bach, E., and Kratzer, A. (1987) Quantification: A Cross-Linguistic Perspective. NSF proposal, UMass, Amherst.  Partee, B. (1995). Quantificational structures and compositionality. In E. Bach, E. Jelinek, A. Kratzer and B. H. Partee (eds.), Quantification in Natural Languages, pp. 541-601. Dordrecht, Kluwer.  Partee, B. (1988). Many quantifiers. The ESCOL 5. The Ohio State University.  Pasco, J., Compton, B.D., & Hunt, L. (1998). The Living World: Plants and Animals of the Kwakwaka’wakw. U’mista Cultural Society.  Sherer, L. (2014). Nominalization and Voice in Kwak’wala. MA thesis. University of British Columbia.   127 Appendices Appendix A Partitive Expressions There is a partitive construction in Kwak’wala. These expressions are only attested in S2’s speech.  114 (a)  S2  √√ ḵ̓ina̱m=dax̱=i=da            babagwa̱m   k̓ak̓ada̱xwsila=x̱a   buk. many=PART=DIST=DET       boy         read=ACC         book ‘Many of the boys are reading the book.’   (b)  S2  √√ k̓ak̓ada̱xwsil=i=da    babagwa̱m=x̱a   ḵ̓ina̱m=dax̱      buk. read=DIST=DET      boy=ACC       many=PART     book. ‘The boys are reading many of the books’   115 (a)  S2  √√ sa̱k'a=dax̱=i=da           babagwa̱m   k̓ak̓ada̱xwsila=x̱a   buk. five=PART=DIST=DET      boy         read=DET         book ‘(There are 30 boys at the café.) Five of the boys are reading the book.’  (b)  S2  √√ k̓i's=dax̱=i=da            babagwa̱m   k̓ak̓ada̱xwsila=x̱a   buk.  none=PART=DIST=DET     boy         read=DET         book ‘None of the boys are reading the book.’  However, the partitive suffix does not need to be employed in discourse contexts which allow a partitive interpretation, where the quantified phrase picks out a proper subset of the individuals satisfying the noun predicate, as shown in (116) with ḵ̓ina̱m (‘many’) and (117) with a number.    128 116 S2  √√ ḵ̓ina̱m=i=da          babagwa̱m   k̓ak̓ada̱xwsila=x̱a   buk. many=DIST=DET      boy         read=ACC         book Scenario: There are 30 boys at the café and many of them are reading the book. ‘Many of the boys are reading the book.’   117 S2  √√ sa̱k'=i=da          babagwa̱m,  mu=i=da         babagwa̱m  k̓ak̓ada̱xwsila=x̱a   buk. five=DIST=DET     boy        four=DIST=DET    boy        read=ACC         book ‘Four (of five) boys are reading the book.’ Literally: ‘There are five boys. Four (of the) boys are reading the book.’      129 Appendix B Collective, Distributive, and Cumulative Readings (Additional) This section lists the semantic judgments for S1 and S2 not listed in section 8.  B.1 Nominative Marker and Accusative/Oblique/Preposition Marker (Subjective in Post-predicative Position) 118 Nominative Marker and Accusative Marker (Predicate NomQP AccQP) S1  √√    /    S2  √√ dal=i=da            mu          ts'a̱daḵ=x̱a    sa̱k'a   ga̱ldas   lax̱=i's      gukw. carry=DIST=DET     four          woman=ACC  five    box    PREP=POSS  house. ‘Four woman carried five boxes to her (another woman) house.’   [Collective] used four people created by origami, and five big box drawings, and explained Sally, Lisa, Kate, and Mary helped their teacher to move but because the boxes are big, they carried the boxes together (4 women, 5 boxes).  [Judgment: ACCEPTED]  [Distributive] used four people created by origami, and 20 box drawings, and explained Sally, Lisa, Kate, and Mary helped their teacher to move, and each of the four women carried five boxes (4 women, 20 boxes).  [Judgment: ACCEPTED]  [Cumulative] used four people created by origami, and 5 box drawings, and explained that they helped their teacher to move by carrying five boxes to her house, but we do not know who carried how many boxes (4 women, 5 boxes).  [Judgment: ACCEPTED]  [Inverse Distributive] used 20 people created by origami and 5 big box drawings, and explained that the teacher asked her students to help her move and 20 female students helped her. There are five boxes and each box is carried by four women to her house (20 women, 5 boxes).  [Judgment: ACCEPTED]      130 The Number of Boxes Possible Reading Scopal Configuration: mu (‘four’) / sa̱k'a (‘five’) Sally Lisa Kate Mary 5 (Together) Collective √ [mu, sa̱k'a] (scopeless) 5 5 5 5 Distributive √ [mu > sa̱k'a] 5 (Separately) Cumulative √ [mu, sa̱k'a] (scopeless)  The Number of Women Possible Reading Scopal Configuration Box A Box B Box C Box D Box E 4 4 4 4 4 Inverse Distributive √ [sa̱k'a  > mu] Table 43 Readings with Nominative Marker and Accusative Marker  119 Nominative Marker and Oblique Marker (Predicate NomQP OblQP) (1)  S1  √√ k̓ak̓ada̱xwsila-su=i=da   ma'ł       bi-buk=sa       sa̱k'a    bi-babagwa̱m. read-PASS=DIST=DET    two        RED-book=OBL  five     RED-boy. ‘Five boys read two books.’  [Collective] used 5 Lego people and 2 books to explain that five boys sat together and read the two books together (5 boys, 2 books).  [Judgment: ACCEPTED]  [Distributive] used 5 Lego people and 10 books to explain that each boy read two different books (5 boys, 10 books).  [Judgment: ACCEPTED]  [Cumulative] used 5 Lego people and 2 books and explained that I gave two books to the boys and knew that the books were read by them but I do not know who read which books (5 boys, 2 books).  [Judgment: ACCEPTED]  [Inverse Distributive] used 10 Lego people and 2 books and explained that I am a teacher and got two books for my students (10 students) and told them that they should read either book. So five students read one book and the others read the other (10 boys, 2 books).  [Judgment: ACCEPTED]      131 The Number of Books Possible Reading Scopal Configuration sa̱k'a (‘five’) / ma'ł (‘two’) Boy1 Boy2 Boy3 Boy4 Boy5 2 (Together) Collective √ [sa̱k'a, ma'ł] (scopeless) 2 2 2 2 2 Distributive √ [sa̱k'a > ma'ł] 2 (Separately) Cumulative √ [sa̱k'a, ma'ł] (scopeless)  The Number of Boys Possible Reading Scopal Configuration Book A Book B 5 5 Inverse Distributive √ [ma'ł > sa̱k'a] Table 44 Readings with Nominative Marker and Oblique Marker for S1   (2)  S2  √√ k̓ak̓ada̱xwsila-su=i=da    ma'ł       bi-buk=sa       sa̱k'a    gi-ga̱n-ga̱nana̱m. read-PASS=DIST=DET     two        RED-book-OBL  five     RED-RED-child. ‘Five children read two books.’  [Collective] used a picture to explain that five children sat together and read the two books together (5 children, 2 books).  [Judgment: ACCEPTED]  [Distributive] used a picture to explain that 5 children each read two different books (5 children, 10 books). [Judgment: ACCEPTED]  [Cumulative]  -  [Inverse Distributive] used a picture to explain that there are two books and each book was read by a group of five different children (10 children, 2 books).  [Judgment: ACCEPTED]   The Number of Books Possible Reading Scopal Configuration sa̱k'a (‘five’), ma'ł (‘two’) Tom Lisa Ken Mike Sally 2 (Together) Collective √ [sa̱k'a, ma'ł] (scopeless) 2 2 2 2 2 Distributive √ [sa̱k'a > ma'ł] 2 (Separately) Cumulative - [sa̱k'a, ma'ł] (scopeless)  The Number of Children Possible Reading Scopal Configuration Book A Book B 5 5 Inverse Distributive √ [ma'ł > sa̱k'a] Table 45 Readings with Nominative Marker and Oblique Marker for S2     132 120 Nominative Marker and Preposition Marker (Predicate NomQP PrepQP) S1  √√ la̱'etł   sa̱k'=i=da           bi-bi-ba̱gwana̱m   lax̱    yuda̱xw    gi-gukw. enter   five=DIST=DET      RED-RED-man     PREP  three       RED-house. ‘Five men entered into three houses’  [Collective] used 5 LEGO people and 3 houses to explain that 5 men visited three houses together (5 men, 3 houses).  [Judgment: ACCEPTED]  [Distributive] used 5 LEGO people and 15 houses to explain that 5 men each visited three houses (5 men, 15 houses).  [Judgment: ACCEPTED]  [Cumulative] used 5 LEGO people and 3 houses to explain that they have a bird but the bird flew away. There are only 3 houses in this area so they went to look for it. We do not know who went to which house (5 men, 3 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] used 15 LEGO people and 3 houses to explain that there are 3 groups which consist of 5 men, and each group visited 1 house (15 men, 3 houses).  [Judgment: ACCEPTED]  The Number of Houses Possible Reading Scopal Configuration: sa̱k'a (‘five’) / yuda̱xw (‘three’) Tom John Ken Mike Jim 3 (Together) Collective √ [sa̱k'a, yuda̱xw] (scopeless) 3 3 3 3 3 Distributive √ [sa̱k'a > yuda̱xw] 3 (Separately) Cumulative √ [sa̱k'a, yuda̱xw] (scopeless)  The Number of Men Possible Reading Scopal Configuration House A House B House C 5 5 5 Inverse Distributive √ [yuda̱xw > sa̱k'a] Table 46 Readings with Nominative Marker and Preposition Marker for S1  S2  √√ la   sa̱k'=i=da           gi-ga̱n-ga̱nana̱m   lax̱    mu   gi-gukw. go  five=DIST=DET      RED-RED-child   PREP  four   RED-house. ‘Five children went to four houses’  [Collective] showed the picture to explain that there are five children, and everyone went to the same houses (5 children, 4 houses).  [Judgment: ACCEPTED]    133 [Distributive] showed the picture to explain that there are five children, and each child went to four different houses (5 children, 20 houses).  [Judgment: ACCEPTED]  [Cumulative] showed the picture to explain that there are five children and four houses. Some children went to more houses than others, and all the houses had someone but no houses had all the children (the speaker does not know who went to which house) (5 children, 4 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] showed the picture to explain that there are twenty children and four houses. A different five children went to each house (20 children, 4 houses).  [Judgment: ACCEPTED]  The Number of Houses Possible Reading Scopal Configuration: sa̱k'a (‘five’) / mu (‘four’) Tom Lisa Ken Mike Sally 4 (Together) Collective √ [sa̱k'a, mu] (scopeless) 4 4 4 4 4 Distributive √ [sa̱k'a > mu] 4 (Separately) Cumulative √ [sa̱k'a, mu] (scopeless)  The Number of Children Possible Reading Scopal Configuration House A House B House C House D 5 5 5 5 Inverse Distributive √ [mu > sa̱k'a] Table 47 Readings with Nominative Marker and Preposition Marker for S2  B.2 Nominative-Marked Subject and Accusative/Oblique/Preposition Marker (Nominative in Post-Auxiliary Position) 121 Nominative Marker and Accusative Marker (Aux NomQP Predicate AccQP) S1  √ la̱'m  mu=i=da         bi-babagwa̱m  dala=x̱a     sa̱k'a   gi-ga̱ldas  lax̱=i's      gukw. AUX  four=DIST=DET    RED-boy       carry=ACC  five    RED-box  PREP=POSS  house ‘Four boys carried five boxes to her (another woman’s) house.’   [Collective] used 4 Lego people and 5 boxes to explain that four boys carried five boxes together (4 boys, 5 boxes).  [Judgment: ACCEPTED]  [Distributive] used 4 Lego people and 20 boxes to explain that they helped their teacher to move, and each of the four boys carried five boxes (4 boys, 20 boxes).  [Judgment: ACCEPTED]    134 [Cumulative] used 4 Lego people and 5 boxes and explained that they helped their teacher to move by carrying five boxes to her house, but we do not know who carried how many boxes (4 boys, 5 boxes).  [Judgment: ACCEPTED]  [Inverse Distributive] used 20 Lego people and 5 big box drawings, and explained that the teacher asked her students to help her to move and 20 boys helped her. There are five boxes and each box is carried by four boys to her house (20 boys, 5 boxes).  [Judgment: ACCEPTED]  The Number of Boxes Possible Reading Scopal Configuration: mu (‘four’) / sa̱k'a (‘five’) Boy1 Boy2 Boy3 Boy4 5 (Together) Collective √ [mu, sa̱k'a] (scopeless) 5 5 5 5 Distributive √ [mu > sa̱k'a] 5 (Separately) Cumulative √ [mu, sa̱k'a] (scopeless)  The Number of Boys Possible Reading Scopal Configuration Box A Box B Box C Box D Box E 4 4 4 4 4 Inverse Distributive √ [sa̱k'a  > mu] Table 48 Readings with Nominative Marker and Accusative Marker for S1  122 Nominative Marker and Oblique Marker (Aux NomQP Predicate OblQP) S1  √ la̱'m   yuda̱xw=i=da     buk-wa̱ł   k̓ak̓ada̱xwsila-su-wa̱ł=sa  a̱dła̱bu  gi-ga̱n-ga̱nana̱m. AUX   three=DISC=DET   book-PST  read-PASS-PAST=OBL     seven   RED-RED-child. ‘Three books were read by seven children.’  [Collective] used 7 LEGO people and 3 books to explain that there are seven children, and everyone read the same books (3 books, 7 children).  [Judgment: ACCEPTED]  [Distributive] used 21 LEGO people and 3 books to explain that there are three groups of seven children, and every group read a different book (3 books, 21 children).  [Judgment: ACCEPTED]  [Cumulative] used 7 LEGO people and 3 books to explain that there are seven children and three books, and the speaker does not know who read which book, but seven children read three books (3 books, 7 children).  [Judgment: ACCEPTED]  [Inverse Distributive] used 7 LEGO people and 21 books to explain that each book was read by a different seven children (21 books, 7 children).  [Judgment: ACCEPTED]      135 The Number of Children Possible Reading Scopal Configuration: yuda̱xw (‘three’) / adła̱bu (‘seven’) BookA BookB BookC 7 (Together) Collective √ [yuda̱xw, adła̱bu] (scopeless) 7 7 7 Distributive √ [yuda̱xw > adła̱bu] 7 (Separately) Cumulative √ [yuda̱xw, adła̱bu] (scopeless)  The Number of Books Possible Reading Scopal Configuration C1 C2 C3 C4 C5 C6 C7 3 3 3 3 3 3 3 Inverse Distributive √ [adłabu > yuda̱xw]  Table 49 Readings with Nominative Marker and Oblique Marker for S1  123 Nominative Marker and Preposition Marker (Aux NomQP Predicate PrepQP) S1  √ la̱'m=i=da       sa̱k'a     bi-bi-ba̱gwana̱m   la̱'etł   lax̱     yuda̱xw    gi-gukw. AUX=DIST=DET  five       RED-RED-man    enter   PREP   three       RED-house. ‘Five men entered into three houses.’  [Collective] used 5 LEGO people and 3 houses to explain that 5 men visited three houses together (5 men, 3 houses).  [Judgment: ACCEPTED]  [Distributive] used 5 LEGO people and 15 houses to explain that 5 men each visited three houses (5 men, 15 houses).  [Judgment: ACCEPTED]  [Cumulative] used 5 LEGO people and 3 houses to explain that they have a bird but the bird flew away. There are only 3 houses in this area so they went to look for it. We do not know who went to which house (5 men, 3 houses).  [Judgment: ACCEPTED]  [Inverse Distributive] used 15 LEGO people and 3 houses to explain that there are 3 groups which consist of 5 men, and each group visited 1 house (15 men, 3 houses).  [Judgment: ACCEPTED]      136 The Number of Houses Possible Reading Scopal Configuration: yuda̱xw (‘three’) / sa̱k'a (‘five’) Man1 Man2 Man3 Man4 Man5 3 (Together) Collective √ [sa̱k'a, yuda̱xw] (scopeless) 3 3 3 3 3 Distributive √ [sa̱k'a > yuda̱xw] 3 (Separately) Cumulative √ [sa̱k'a, yuda̱xw] (scopeless)  The Number of Men Possible Reading Scopal Configuration House A House B House C 5 5 5 Inverse Distributive √ [yuda̱xw > sa̱k'a] Table 50 Readings with Nominative Marker and Preposition Marker for S1      137 Appendix C Tsitsa̱ḵala̱m sa Ruby This section lists the occurrence of quantifiers in an unpublished book “Tsitsa̱ḵala̱m sa Ruby” provided by Jonathan Janzen.  124 Wildman / Ba̱k̓wa̱s (10) g̱eła  lax̱a  ma̱kola  o'a̱'m  la  ḵ̓ina̱mi   sa̱'ya's  lax̱i's  uk̓wine'  ḵ̓ina̱mox̱  g̱eła   lax̱=a        ma̱kola  o'a̱'m  la     ḵ̓ina̱m=i   sa̱'ya=a's  lax̱=i's            dwell  to=D3.ACC  island   just   PREP   lots=D3   hair=3PS  to=3PSG.POSS      uk̓wine'   ḵ̓ina̱m=ox̱  body      lots=D2  ‘He was on the island a long time, and got lots of hair all over his body. Lots of it’  125 Wildman / Ba̱k̓wa̱s (25) la'a̱'m  yuda̱xwba̱'naxwas  hela  da  ma̱kola   la-'a̱'m      yuda̱xw  ba̱'naxwas   hela     da  ma̱kola  PREP-DISC  three    days       be.there  the  island  ‘Then he was there for three days on the island’  126 Wildwoman / Dzunuḵwa (6) 'na̱m ukw'mi  da  ga̱nana̱m  ḵ̓og̱igax̱tła  'na̱m -ukw-'m=i       da   ga̱nana̱m   ḵ̓og̱iga=x̱=xtł  one-person-DISC=D3  the   child      hunchback=D2=name  ‘One child was called Hunchback’      138 127 Wildwoman / Dzunuḵwa (14) k̓a̱yo's  la  giga̱nga̱nana̱m  lax̱a  la̱x̱e'a̱  sa Dzunuḵwa  k̓a̱yo's  la     gi-ga̱n-ga̱nana̱m  lax̱=a       la̱x̱e'=a̱         s=a          Dzunuḵwa NEG    PREP  PL-PL-child     to=D3.ACC  basket= D3.ACC OBL=D3.ACC  Wild Woman  ‘No children were in Wild Woman’s basket anymore’  128 Taken to the House of Horrors / Dax'ida'eł lax̱a Yax̱sa̱m Gukwdzi (47) La̱'mis  wi'la  lawe  x̱a̱n  gwiłgwela  La̱-'m=is         wi'la  lawe    x̱=a̱n         gwiłgwela PREP-DISC=and   all    take.off  ACC=1PPOSS  clothing  ‘She took off all my clothes’   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0314095/manifest

Comment

Related Items