UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The prosodic structure of Finnish and the theory of phonological government Millard, David 1991

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

Item Metadata


831-ubc_1992_spring_millard_david.pdf [ 2.86MB ]
JSON: 831-1.0086597.json
JSON-LD: 831-1.0086597-ld.json
RDF/XML (Pretty): 831-1.0086597-rdf.xml
RDF/JSON: 831-1.0086597-rdf.json
Turtle: 831-1.0086597-turtle.txt
N-Triples: 831-1.0086597-rdf-ntriples.txt
Original Record: 831-1.0086597-source.json
Full Text

Full Text

THE PROSODIC STRUCTURE OF FINNISHAND THE THEORY OF PHONOLOGICAL GOVERNMENTbyDAVID MILLARDBMus, The University of Victoria, 1982Diploma in Applied Linguistics,The University of British Columbia, 1985A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF ARTSinTHE FACULTY OF GRADUATE STUDIES(Department of Linguistics)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIADecember 1991© David Millard, 1991In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of  LinguisticsThe University of British ColumbiaVancouver, CanadaDate	 December 30, 1991DE-6 (2/88)The Prosodic Structure of Finnishand the Theory of Phonological GovernmentbyDavid MillardSubmitted September 1991 to the Department of Linguistics, UBC,in partial fulfillment of the requirements for the degree ofMaster of Arts.ABSTRACTThe goal of this thesis is to examine same of the claims of theTheory of Phonological Government in the light of data taken from theFinnish language. The Theory of Phonological Government describes theoraganization of segments into constituent groupings on the basis ofgovernment relations established at the level of Underlying Represen-tation. These prosodic constituents are related to each other, again interms of government. While these prosodic constituents are functionallysimilar to traditional notions of the syllable, division of a surfacestring into syllables does not necessarily yield the relations predictedby the theory. Government Theory, then, makes predictions which aredifferent from those of other syllable theories and these predictions canbe tested. The Finnish language exhibits a number of phonological proces-ses that are sensitive to prosodic structure and thus offers an excellenttest case for the theory.Chapter One presents an overview of the theory and the predictions itmakes. Chapter Two examines same general phonological characteristics ofFinnish. In Chapter Three I examine two prosodically-influenced segmentaliideletion processes that interact with each other and shows how GovernmentTheory accounts for than in a principled fashion with minimal appeal tospecific 'rules' of deletion.In Chapter Four I examine the process of Consonant Gradation, amutation process that is driven by syllable structure. In Chapter Five Idemonstrate that certain surface strings of segments that appear to beCVVC and CVCC syllables are best treated as being other than syllables andshow how the theory not only accounts for the data but also predicts theexistence of such non-syllabic strings. Finally, in Chapter Six I resumethe discussion of Consonant Gradation and examine the nature of someexceptional gradation forms, showing how the theory accounts for thesemore unusual forms.Thesis supervisor: Professor P. A. ShawiiiTABLE OF CONTENTSABSTRACT 	 iiLIST OF ABBREVIATIONS 	 viACKNOWLEDGEMENTS 	  viiINTRODUCTION 	 1Notes to the Introduction 	  3CHAPTER ONE - The Theory of Phonological Government 	  41.1 The Structure of Segments 	 41.2 Representations of Phonological Structure 	  61.3 Prosodic Constituents 	  71.4 The Governing Properties of Segments 	  101.5 Government between Neutral Segments 	  111.6 The Relation of Segments to Prosodic Categories . . . 	  121.7 The Generation of Prosodic Structure 	  14Notes to Chapter One 	  16CHAPTER TWO - The Sound Pattern of Finnish 	  182.1 Consonant Sequences 	  182.2 Stress 	  22Notes to Chapter Two 	  24ivCHAPTER THREE - The Deletion of Segments 	  263.1 Two Segmental Deletion Processes 	  263.2 e-Deletion 	  273.3 Empty Categories and Phonological Government 	  283.4 t-Deletion 	  35Notes to Chapter Three 	  40CHAPTER FOUR - Consonant Gradation 	 414.1 General Features of Consonant Gradation 	  414.2 The Nature of Consonant Gradation 	  42Notes to Chapter Four 	  48CHAPTER FIVE - WC and VCC Groupings 	  505.1 CVVC and CVCC Non-syllables 	  505.2 CVCC Sequences 	  515.3 Monanoraic Sequences 	  535.4 Monomoraic Vowel-Sonorant Sequences 	  565.5 CVVC Sequences 	  595.6 The Derivation of Doubling Forms 	  63Notes to Chapter Five 	  72CHAPTER SIX - Further Aspects of Consonant Gradation 	  766.1 Same Exceptions to Consonant Gradation 	  766.2 Some Remarks on Glide Formation 	  79Notes to Chapter Six 	  82CONCLUSION 	 83LIST OF ABBREVIATIONSWhile I have endeavoured to avoid the use of abbreviations it hasbeen necessary for the sake of space to use some abbreviations in theexamples and derivations. Standard linguistic abbreviations have beenincluded here for the sake of completeness.ant	 anteriorATR	 Advanced Tongue RootC	 consonantcont	 continuantcor	 coronalECP	 Empty Category Principleess.	 essive caseF	 metrical footgen.	 genitive caseICG	 Interconstituent Governmentillative caselab	 labialM	 metrical wordm	 moraN Nucleus nodenam.	 nominative caseO Onset nodepart.	 partitive casepl.	 pluralR	 Rime nodes	 strong (branch of a metrical structure)d syllableSCG	 Subconstituent Governmentsg.	 singularson	 sonoranttransl.	 translative case3 vowelV:	 long vowelvoi	 voicedw	 weak (branch of a metrical structure)viACKNOWLEDGEMENTSIt is difficult to assess, after four years, which of the many peoplewho have influenced the writing of this thesis made which particular con-tribution. I shall have to offer most of my friends and fellow students a'global' statement of thanks--otherwise, the list of names would be pro-hibitively long.Among the instructors and faculty members outside my thesis commit-tee, I must single out Drs. Glyne Piggott, Zita MbRobbie, Ewa Czaykowska-Higgins, and Charles Ulrich. To them I owe much of my grounding inphonology and not a few warnings about the pitfalls of working in a non-standard framework. If I chose to ignore those warnings, it is no reflec-tion on their powers of persuasion. I would also like to thank Dr. GuyCarden who more than once banished an administrative difficulty for me andwho, incidentally, first introduced me to the intricacies of Finnishphonology.To the non-supervisory members of my committee, Drs. M. D. Kinkadeand Juta Kitching, I offer my heartfelt thanks not only for their commentsand suggestions, but also for their patience in putting up with mysomewhat unorthodox work habits. Needless to say, they do not agree witheverything I have to say, but this thesis would nonetheless be poorer wereit not for their influence.To my supervisor, Dr. Patricia Shaw, I cannot offer enough thanks forher instruction, her encouragement, her patience and her energy on mybehalf. I owe any right to call myself a phonologist to her. Again,while she does not agree with everything I say, this thesis simply couldnot have been written without her.Finally, I must acknowledge my family, whose lives have all beentouched in one way or another by the writing of this thesis, and all ofwham offered 'gentle reminders' to me to finish writing it.This thesis is dedicated to the menory of my mother, Peggy Millard,who never even knew I had taken up linguistics but who always supported mein my past endeavours.viiINTRODUCTIONIn a paper entitled "Generative Phonology vs. Finnish Phonology" LyleCampbell states that "[t]heory and data are in a trading relation (the so-called hermeneutic circle or spiral). That is, the theory explains thedata, but a particular set of data (a language) described in terms of thetheory may reveal inadequacies in the theory which lead to revisions."(Campbell 1981; 147). He then goes on to argue that Finnish is a languagerich in theoretical test cases and has a history of exposing inadequaciesin generative phonological theory. Generative Phonology has seen manychanges since the publication of that paper, but the somewhat joculartitle still carries same weight. Finnish remains a good language fortesting phonological theories and while advances in Generative Phonology,particularly in the areas of autosegmental representation and syllablestructure, have made Finnish more amenable to phonological explanation,the problems set by Finnish have by no means all been solved.This thesis will focus on some of the problems associated with theprosodic structure of Finnish. Finnish has a number of processes that aresensitive to prosodic structure, ranging from a series of deletions andtheir various effects to the weakening process of Consonant Gradationwhich is sensitive to the structure of the Rime. I cannot claim to haveresolved all the issues of all these processes, but I am able to sketchthe form that such a solution must take. As these processes are sensitiveto what may be termed informally 'syllable structure', any account of themmust, therefore, be undertaken in the context of a well-articulated theoryof syllable structure. I will investigate the contribution that onesyllable theory, the theory of Phonological Government presented by Kaye,Lowenstamr and Vergnaud (1989) has to make toward the understanding of1a number of phonological problems in Finnishl. I shall also argue thatthe theory of Phonological Government makes a number of non-standardpredictions and assumptions that account for the Finnish data in an ele-gant fashion.In the first chapter of this thesis I shall outline the theoreticalassumptions I shall be using. The second chapter will present some of thephonological characteristics of Finnish and demonstrate how the theory canbe applied to describe them. In these two chapters I shall assess somedifficulties the theory faces in its standard form. Specifically, I shalldemonstrate the need of a subtheory dealing with the erection of prosodicstructure and demonstrate the inability of Charm Theory to account for allthe observed licit and illicit consonant sequences in Finnish.In the remaining chapters I shall examine particular instances ofprosodically sensitive processes. Chapter Three deals with two inter-acting deletion processes. Chapter Four presents the basic details ofConsonant Gradation. Chapter Five examines the variety of observed 'syl-lable' types in Finnish and presents a Government-based account of them.Chapter Six returns to the subject of Consonant Gradation, focussing onsane apparently exceptional aspects of the process in the light of theaccount developed in preceding chapters.2Notes to the Introduction1. Kaye, Lowenstanm and Vergnaud actually exclude the notion 'syllable'from their theory in favour of a sequence of alternating onsets and rimeswithout any additional hierarchy. However, as Consonant Gradation invol-ves a relationship within easily defined Onset-Rime (OR) pairs, it will beconvenient to speak of these OR pairs informally as syllables. Thisconvention will also facilitate comparisons with alternative representa-tions in models that include the syllable as a legitimate category withintheir framework. I shall therefore continue to use terms such as 'syl-lable structure', 'onset of a closed syllable' etc. as the most economicalway of describing these notions in informal discussion.I shall further note here that Kaye, Lowenstamm and Vergnaud (1989)is a published version of an earlier manuscript (Kaye, Lowenstamm andVergnaud 1987). The 1989 version is a translation into German of a paperoriginally written in English. As the German version shows minimal re-vision over the English version (and, indeed, maintains some uncorrectederrors of the English version) I have chosen, when actually quoting thepaper, to quote the English text rather than to retranslate the German.In these instances, I have cited the location of the material in both theEnglish and German versions. Where I simply cite the location of an ideawithout direct quotation, I provide references to the published versiononly.3CHAPTER ONE - The Theory of Phonological GovernmentIn this chapter I shall outline the basic aspects of GovernmentTheory (cf. Kaye, Lowenstamm and Vergnaud 1989) and sketch out an analysisof the problems to be addressed in this thesis.1.1 The Structure of Segments'In the theory of Phonological Government, segments comprise neitherthe structureless feature matrices of Chamsky and Halle (1968) nor thehierarchy of features found in e.g. Clements (1985). Instead, the repre-sentation of segments involves a marriage of the theory of distinctivefeatures with the unitary segmental components of Dependency Phonology(see Anderson, Ewen and Staun 1985). Each segment is composed of one ormore 'elements', which are themselves bundles of a subset of the totalnumber of features available (see Kaye, Lowenstamm and Vergnaud, (1985)for a detailed discussion). Each element represents a fully specifiedsegmental matrix which is capable of functioning independently as a pro-nounceable speech sound. One feature of each element acts as a head forthe element (in Kaye, Lowenstamm and Vergnaud's terminology, this is the'hot' feature) and this feature alone can interact with other elements (byspreading, for example). Within a segment elements are organized in head-operator relations with the head contributing its entire matrix and theoperators adding only their 'hot' features to change the value of thecorresponding features within the head. To give a more concrete demon-stration I shall use the elements A and I which, in the absence of any4operators, define the sounds [a] and [I], respectively. The full repre-sentation of these elements is presented in (1):(1)	 (fram Kaye, Lowenstanm and Vergnaud 1985: 309)A- ROUND+ BACK:LILO_- ATR+ LCrelI- ROUND- BACK1 + HIGH- ATR-LOW (Hot features are underlined)When combined in a fusion operation (designated by the operator '.'),these elements define two new sounds the identity of which is determinedby which element is the head:(2) (from Kaye, Lowenstamm and Vergnaud 1985: 309-10)head	 operator	: - ROUND :	: - ROUND :	 : - ROUND	: + BACK '	' - BACK '	  - HIGH :	 •	 1 + HIGH :	1 -  g? I	 	 i	1	+I - ATR	 '	1 - ATR	I	 I	: + LOW	 '	1	 : - LOW	 :	 11A	 .	 I	 = []	1 - ROUND '	 : - ROUND '	 1 - ROUND	: - BACK '	 : + BACK :	 1 - BACK	+ HIGH :	 .	 1 - HIGH. '	 : - HIGH' 	- ATR	 1	  - ATR	 1 	: - ATR	 :	- LOW	 :	 : + LOW	 1	 : - LOW	 1I	 A	 = [e]By way of shorthand notation, the elemental structure of a segmentalrepresentation may be written as:(3) X	 XA	[m]	 [E]5where, following the conventions of Dependency Phonology (see Anderson,Ewen and Staun 1985), the head is placed above the operator(s). However,where it is necessary only to identify a segment without referring to itsconstituent structure, I shall employ conventional phonetic symbols:(4) X	 XBe1.2 Representations of Phonological StructureThe representations of phonological structures within this theory arebased on the principles of Autosegmental Phonology. Segments (representedas clusters of elements) are grouped on a melodic tier and associated topositions (designated 'X') on a skeletal tier. Skeletal positions arefurther grouped into prosodic constituents ('syllables') which are in turnorganized into metrical groupings. A basic schema is presented in (5):(5) Metrical Structure: 	 Word/ \Foot	 s wI	 IIProsodic Constituents:	 0 R 0 R1 1 :	 (see discussionNIN immediately below)III#Skeletal Tier:	 X X X XIII![IIIMelodic Tier:	 s s s s (s=segment)In practice, it will not usually be necessary to specify every level ineach derivation.61.3 Prosodic ConstituentsThe notion 'prosodic constituent' is intimately tied to the notion of'government' in this theory. Government is defined (Kaye, Lowenstamm andVergnaud 1989, 37; 1987, 8) as "a binary, asymmetric relation holding...between two skeletal positions". The relation is strictly local: thegovernor and governee must be adjacent; and unidirectional: at a givenprosodic level, governors govern in one direction only. In definingprosodic constituents two types of Government are required: governmentwithin a constituent (Subconstituent Government or SCG) and governmentbetween constituents (Interconstituent Government or ICG) 2 . Subconsti-tuent Government holds between the head of a constituent and its comple-ment, and is strictly head-initial. As a consequence of this, allprosodic constituents are maximally binary (a head cannot govern in twodirections at once) and left-headed. Interconstituent Government holds(for the present discussion) between constituent heads and skeletal posi-tions to their left (i.e. ICG is head-final).In the present theory, there are three constituents: the Onset,Nucleus and Rime (abbreviated 0, N and R). The head of the Onset is theskeletal position under the left branch of the Onset node (should itbranch) while the head of the Rime is the skeletal position under the leftbranch of the Nucleus node. Onset and Rime are maximal projections of7their heads while the Nucleus is a medial projection of the head of theRime. The camplete inventory of well-formed constituent types istherefore:(6) 0	 0	 R	 R	 RI1\	 I	 I	 1\I'I	 1 \	 N	 N	 N \1	 1	 \	 II	 1\	 1	 \X	 X X	 X	 X X	 X XBecause heads must always be adjacent to their complements, the headof a Rime may govern only the right-hand branch of the Nucleus (where theNucleus branches and the Rime does not) or the right-hand branch of theRime (where the Rime branches and the Nucleus does not). A structurewhere both the Nucleus and Rime branch is ill-formed, as the segmentdominated by the right branch of the Rime would be separated from the headby the right branch of the Nucleus and would therefore be ungoverned.Such a structure is illustrated in (7):(7)where z is ungovernable because it is not adjacent to the head x (Kaye,Lowenstamm and Vergnaud 1989, 39).Returning to the discussion of Interconstituent Government, it is nowpossible to define the ICG configurations available. Kaye, Lowenstamm andVergnaud (1989, 53; 1987, 24) present two principles of Government thatlimit the possible IOG structures:(8) a. Only the head of a constituent may govern.b. Only the nucleus (or a projection thereof) may governa constituent head.8(8a) is a general condition on Government. (8b) applies specifically toICG, for while the head of an Onset may govern a preceding Rimal orNuclear complement, it cannot govern the head of a non-branching Rime,whereas the head of a Rime can govern the head of a non-branching Onset.There are, therefore, five possible ICG configurations:(9) a. Government of a Rimal complement by an OnsetR	 01\	 1N\X X <-- Xb. Government of a Nuclear complement by an OnsetR	 01\X X <-- Xc. Government between contiguous NucleiX < 	 Xd. Government between a Rime and an Onset complement0\	 N \\	 i	 \X X < 	 X (X)e. Government of the head of an Onset by the head of a Rime0	 R1\N \\X <-- X (X)(9c) demonstrates one form of government between two Nuclei.Another possible government structure involving Nuclei is Projection9Government (Kaye 1987: 6). The theory allows for the projection of Nucleionto a separate autosegmental tier. On the level of the nuclear projec-tion, Nuclei are adjacent and may govern each other. The direction ofProjection Government is subject to parametric variation and in Finnishthe relevant direction is left-to-right3 . Projection Government willbecome significant in the discussion of Pseudo-Campensatory Lengthening inChapter Five.1.4 The Governing Properties of SegmentsIn any theory of syllabification, there must be same designation ofwhat segments may occur in an Onset position or in the Nucleus or Rime.In the theory of Government this is handled largely by a property called'Charm' (see Kaye, Lowenstamm and Vergnaud 1985 for a full discussion ofCharm) 4 . Charm is the property of certain elements and may be positive(Xi), negative (X- ) or neutral--i.e. charmless (X0). Like the hot fea-tures, elements impart their charm value to the segment they are associa-ted with. The distribution of charmed segments is determined by thefollowing principles (Kaye, Lowenstamm and Vergnaud 1989, 42; 1987, 13):(10) a. Charmed segments may be associated with governingpositions (may be governors); charmless segmentsmay be associated with governed positions.Charmed segments may not be governed. [But see (8b)]b. Positively charmed segments may not occur in non-nuclearpositions; negatively charmed segments may not occur innuclear positions.As a working hypothesis (and one which I will adopt), Kaye, Lowenstamm andVergnaud (1985, 311; 1989, 42) assign charm values in this fashion:vowels headed by the element W (which includes [+low] in its specifica-tion) or associated with the element ki ([+ATR]) are positively charmed;10stops, affricates and non-strident fricatives are negatively charmed; allother segments are neutral. It should be noted that although the terms'consonant' and 'vowel' have no primal status in the theory (consonantali-ty and vocalicity are considered byproducts of prosodic structure) I shalluse them as a convenient designation for the class of sounds that occupynon-nuclear positions and the class of sounds that occupy nuclear posi-tions respectively. This designation correlates quite directly with thetraditional classification of consonants and vowels.1.5 Government Between Neutral SegmentsNot all consonantal contacts are between negatively charmed andneutral segments. For example, in the English word armour, the [r] andthe [m] are both neutral segments. Same neutral segments must be able togovern others. The governing properties of neutral segments are restric-ted by their complexity. A segment is said to be more complex thananother if it is composed of a greater number of elements. Thus, giventhe structures for [m] and Er]:(11)	 [r]No	 11°?oUo(The meaning of these symbolsis discussed in Chapter Two)it is perspicuous in the representation itself that [m] is more complex11than [r]. Kaye, Lowenstamn and Vergnaud (1989, 63; 1987, 34) stipulatethat:(12)A neutral segment may govern if it has a complexitygreater than its governee.The order of complexity among neutral segments is given as (Kaye, Lowen-stamm and Vergnaud 1989, 63; 1987, 34):(13) {glides, r} < 1 < nasals [< s] 	 (s added - -DM)The theory also allows for the possibility of empty positions (skeletalpositions with no melodic content). An empty position has a complexity ofzero, of course, and may (indeed, must) be governed by any segment withphonetic content (Kaye, Lowenstamm and Vergnaud 1989, 62). The existenceof empty positions and their distribution will be argued for in ChapterThree.1.6 The Relation of Segments to Prosodic CategoriesAs presented in Kaye, Lowenstamm and Vergnaud (1989), the theorylacks a principled means of erecting prosodic structure. It is impliedthat there is a pre-existent prosodic tier consisting of a string of Onsetand Rime nodes which associate via the skeletal tier to the charm-bearingelements in a top-down fashion (the Nucleus appears to be a projectionfrom positively charmed elements). This position is not tenable, as shallsoon be made clear. A better approach would be to project prosodic struc-ture, bottom-up, according to the charm properties of the elements, buteven this approach has its problems.In the strongest version of the theory (Kaye, Lowenstamn and Vergnaud1989, 42 ff.) vowels are positively charmed, obstruents negatively charmed12and sonorants charmless. These charm assignments make it easy to stipu-late that the distinction between consonants and vowels that allowed oneto assign vowels to Nuclei and consonants to Onsets is their charm value.This idyllic scenario cannot be maintained, however, for Kaye, Lowenstammand Vergnaud (1989, 47-48; 59-62) are forced to allow for the possibilityof both charmless vowels and obstruents 5 .In the theory, vowel tenseness and positive charm are properties ofthe ATR element I+. In a language such as English which exhibits ClosedSyllable Laxing, the element I* is not lexically significant. Vowels areunderlyingly charmless and receive a charm value (became tensed) in non-branching Rimes later in the derivation.Charmless obstruents must be postulated in order to allow for ob-struent sequences (e.g. English cactus). Government of one negativelycharmed segment by another is not permitted (10a), so the obstruentsequence must have the form [kot- ].Given that there are both charmless vowels and obstruents at thelevel of Underlying Representation, charm is no longer sufficient todistinguish consonants fran vowels. Given the further fact that vowelscan remain charmless through the course of a derivation and still beassociated with a Nucleus one must conclude that it is not only the charmvalue of a segment that determines the prosodic categories with which itmay be associated.I see no easy way out of this dilemma. I feel canpelled to concludethat vowels are distinguished from consonants on the basis of sane proper-ty P (perhaps the feature [vocalic] or some other feature) which vowelspossess and consonants do not. This same property also defines vowels asable to project Nuclei and ensures that when they receive a charm value,it is positive. This solution is admittedly ad hoc but will serve for the13time being6 .1.7 The Generation of Prosodic StructureWe are now in a position to derive prosodic structure fram underlyingrepresentations. To take a simple case, that of ranta 'shore' nam. sg.the lexical representation consists of the segments of the lexeme and theskeletal positions to which they may associate:(14) X X X X Xr° e no t- eUpon mapping the segments to the skeletal positions, the P-bearingelements (which in this case are also charmed) project Nuclei:(15) N	 NX X X X X111111r° e n° t-Next, the following right-headed (Interconstituent) Government configura-tions are established (Nuclei govern non-nuclei; charmed segments governcharmless segments):(16)XX XX X1111111111r° a+ n° t- a+1_1 It14Left-headed (Subconstituent) Government configurations are established:I__II	 IN '1	 NI	 4r 	11XXXXX1	 1	 I	 I	 II	 I	 I	 I	 I3 antat I	 t	 I	 IT	 II	 IA Rime node is now projected from each Nucleus and with each Rime there isan ancillary Onset that precedes it7 :	(18)	 0 R	 0 RIN 11 *XXX XX3 antaTIT!I-	I -... IT	 II	 IFinally, the Onset and Rime nodes are maximally filled as their governmentconfigurations allows:	(19)	 0 R	 0 RI  II \	 I	 1I	 I	 I1 N \	 1 N1	 nI	 \	 I	 II	 I	 1XX XXX1	 1	 I	 I	 II	 I	 I	 I	 I3 a n t aThe erection of prosodic structure is not always so simple as thisexample implies, but I will treat the more problematic cases as theyarise.(17)15Notes to Chapter One1. I present the structure of phonological elements in the manner ofKaye, Lowenstamm and Vergnaud (1989) primarily because I am utilizingtheir theory of prosodic structure. Analogues to Charm and other govern-ing properties could be proposed for representations of segments using,for example, feature hierarchies. For the most part, my discussion ofsegemental interactions will have little to say about the nature of thesegments involved other than their governing properties.2. I use the term 'Subconstituent Government' rather than 'ConstituentGovernment' to stress the fact that the Goverment domain in question iscontained within a constituent.3. The direction of projection Government in Finnish is revealed by thedirection of Vowel Harmony which is left-to-right.4. The term 'Charm' has been borrowed from the physical sciences whereit is a property attributed to the hypothetical subatomic particlesquarks. It is not a wholly appropriate borrowing, for Charm and Govern-ment theory is distinct from Particle Phonology (see Schane (1987) andreferences therein). The usage is a manifestation of what Dr. Jay Powell(personal communication) calls 'Physics Envy' in linguistics.5. Both versions of the Kaye, Lowenstamm and Vergnaud paper (1987 and1989) in the discussion of neutral segments within Nuclei state that"negatively charmed segments may appear in such positions [i.e. a governedposition within N]" (Kaye, Lowenstamm and Vergnaud 1989, 48; 1987, 18).Clearly, as negatively charmed segments are absolutely prohibited withinthe Nucleus (Kaye, Lowenstamm and Vergnaud 1989, 42), this is a misreadingfor 'charmless segments'. Possibly, this is an uncorrected reference totheir earlier version of the theory (Kaye, Lowenstamm and Vergnaud 1985)where there were only two charm values--positive and negative. In the1989 version the category 'neutral' or 'charmless' has been introduced andapplies to all the elements designated 'negative' in the 1985 paper. Theterm 'negative' has been given a new meaning in the 1989 paper (see Kaye,Lowenstamm and Vergnaud 1989, 42).6. Difficulty in providing a principled distinction between consonantsand vowels is hardly restricted to Government Theory. See Selkirk (1984)for a arguments against basing the distinction on major class features.Selkirk bases the distinction between potential heads of a Nucleus andnon-heads of a Nucleus in a given language on the basis of sonority.Under this view, segments of a certain degree of sonority or higher couldbe nuclear heads and project an N node. Sonority correlates roughly withKaye, Lowenstamm and Vergnaud's notion of 'complexity'. Unfortunately,complexity could not be used in this fashion, for some neutral segments,such as to are less complex than other, more sonorous segments, such as s°(t° contains two elements, so has three). This is an area that requiresfurther research.7. I assume that the generation of ancillary Onsets is a universalprinciple of phonology. Glyne Piggott (UBC lectures, 1988) has arguedthat there are no truly vowel-initial syllables. In the present frame-work, Onsets and Rimes are in an Interconstituent Government16configuration. One could formulate a principle (analogous to the 8-criterion of syntax) that all Onset nodes must be governed by a Rime nodeand all Rime nodes must govern an Onset node. Where these conditions arenot met, the missing constituent must be inserted to render the expressionwell-formed.8. It will be seen in Chapters Three and Five that failure to satisfythe available Onset and Rime nodes or conditions on govetueent will resultin syncope or epenthesis of segments.17CHAPTER TWO - The Sound Pattern of FinnishThis chapter will examine the sound inventory of Finnish (withparticular reference to the consonants) in terms of the Theory of Phono-logical Government. The theory predicts, on the basis of a particularinventory, what clusters are permissible in a language. It will be seenthat for Finnish, the fit between theory and fact is close, but there aresome surprising discrepancies, not all of which are easily explained.2.1 Consonant SequencesFinnish has the following inventory of phonemes (after Lehtinen 1962,xv)i:(1) p t k	 i yv s h	 e 8 o1 r j	 a	 am nIn the theory of segmental representations the available elements andtheir identifying 'hot' features include (Kaye, Lowenstamm and Vergnaud1989, 74; 1985, passim.):(2) RP	 [+ coronal]	L- 	[+ slack vocal cords]	H-	[+ stiff vocal cords]	?a	 [+ constricted]	CP	 [+ continuant]	U°	 [+ round]	Io	 [- back][ - high]v° the cold vowel--no hot feature	1+	 [+ ATR]18The segmental structures of the Finnish sounds are thus:(3) p- 	t-	k- 	v- 	so	 10	 ro	 jo	 mo	 no	 ho 2?°	 R°	 v°	 ?o	 ?o	 ?°	 R°	 I°	 ?o 	R°	 CD°I	 II	 I	 II	 I	 I	 I	 I	 I	 I';()U° ?o ?° u° R° R°	 U°I	 I	 II	 II	 I	 1	 I	 I	 I	 IIi- 	H-	H- 	I..- 	0°	 No	 NoCD°i+	 e+	 yf	 o+	 a+	 e	 0+	 u+I°	 io	 I°	 io	 A+	 A+	 uo	 uoI	 I	 I	 I	 II	 I	 I	 I	 I	 II+	 A+	 U°	 U°	 I°	 A+	 i+I	 I	 I	 II	 I	 I	 II+	 I+	 I+	 I+Of the consonants (which form the subject of this section), [p,t,k,v]are negatively charmed, the remainder are neutral, having the order ofcomplexity [n,m,s] > [1] > [r,h] 3 . Finnish syllables have no branchingonsets (Karlsson 1983, 19; Whitney 1956, 15). Consonant sequences inFinnish words are thus restricted to two types: CPC - (InterconstituentGovernment of a neutral segment by a negatively charmed segment) and CPC0(Interconstituent Government of a neutral segment by a neutral segment ofgreater complexity). The third remaining possibility, C - 00 would definethe Subconstituent Government relationship of a branching Onset which is19unavailable in Finnish. Based on this inventory, Government Theory pre-dicts the following distribution of sequences:(4) permitted sequences 	 non-permitted sequencesCPC- 	hp 1p rp sp mp4 *C- Co	 *ph *pl *pr *ps *pm *pn *pvht It rt st nt 	 *th *tl *tr *ts *tm *tn *tvhk lk rk sk 1)k	 *kh *kl *kr *ks *km *kn *kvhv lv ry sv my nv	 *vh *v1 *vr *vs *vm *vn *vp *vt *vkCP<CP	 hn rm hm	 *CP>cP *ml #mr *Ins	 *Inhin rn hn	 *nl *nr *ns	*nhis rs hs	 *sl *sr *sm *sn *shrl hl	 *lr	 *lh*hr	*rh*C- C-	*pt *pk *pv*tp *tk *tv*kp *kt *kv*vp *vt *vkThese predictions accord fairly well with the facts of Finnish, but notcompletelys. Incorrect predictions are listed below:(5)attested sequences 	 non-attested sequencespredicted impermissible	 predicted permissibleC- Caps ts ks	 CPC- *hp *sp #mv . *nvC° >C° nh rh ns ms	 C"<C' *ln *rn *rl *hsC- C- tk6The distribution of these unexpectedly good sequences varies. The C-s se-quences ps, ts, ks, ns are quite common, (e.g. lapsi 'child', katsoa 'tolook', maksa 'liver', kansa 'people'), as is the sequence tk, (e.g. jatko'continuation'). The sequences nh and rh are rare, but they do occur ine.g. vanha 'old', harha 'delusion'. Interestingly, the unexpectedly goodclusters involve either [s] in the second position or another coronalsound in the first position. [s] is well-attested for clustering inlanguages where no other sounds may and in ways contrary to its position20in sonority hierarchies. In English, for example, CCCC sequences arepossible in words such as demonstrate. If another sound is substitutedfor the s, the result is invariably bad (*demonktrate, *demomptrate).Similar effects are observed in Finnish (e.g. pyrstO 'tail of a bird', but*pyrtk(5). To my knowledge, this special status of [s] remains unexplainednotand I willA pursue the subject here beyond noting the problem.The sequence as is rare (but occurs in e.g. jamsi 'rasp'). This com-parative rarity is not so remarkable as one might first think. In Fin-nish, [m] appears in syllable-final position only before [p] and is oftenderived by assimilation of /n/. Word-final Intl surfaces as [n] (e.g.koditon 'homeless' nam. sg . versus kodittomat 'homeless' nam. pl.) 7 .The sequence tk is not so easy to deal with, as C - C- sequences arenot permitted by the theory. Kaye, Lowenstamm and Vergnaud (1989, 59-62;1987 30-32) allow for the possibility of a series of charmless (and there-fore governable) stops since stop-stop sequences obviously do occur inmany languages. I hesitate to suggest that Finnish is such a language forthe presence of a charmless stop series which included, e.g., the sound towould predict the sequences *tp, *tv in addition to the acceptable se-quence tk. I shall leave the question of the occurrence of tk in Finnishunresolved, pending further research.In addition to the CsC clusters mentioned above, Finnish exhibitsanother type of CCC cluster in which a sonorant is followed by a geminateobstruent. Examples include helppo 'easy', kynttilä 'candle' and poltta'burn'. I shall deal with the structure of these forms in Chapter Five,for they offer a problem for the theory. However, the segmental contentof these clusters never violates the sequencing restrictions discussedabove.21The unexpectedly bad clusters present some curiosities. The lack ofclusters ending in p is a puzzling but well-known idiosyncracy of Finnish(Prince 1984, 239). The absence of *nv, Amy, *ln, *rn, *rl may be asonority effects. As my purpose in this section is to provide a list ofpossible consonant sequences in Finnish rather than to explain the distri-bution of observed sequences, I will not pursue the issue further.2.2 StressGovernment Theory does not deal directly with stress but it will beseen that stress interacts with other prosodic processes in Finnish. Iwill deal with these cases as they arise and present the basic facts ofFinnish stress here in terms of the metrical theory of Hayes (1980).In Finnish, primary stress falls on the first syllable of a word;secondary stress (which is optional in nouns) falls on subsequent odd-numbered syllables (Whitney (1956, 16), Lehtinen (1962, xvii), Keyser andKiparsky (1984, 23-25)). In terms of Metrical Phonology, a left-headedbinary Foot is erected at the left margin of a word (the application isreiterated if secondary stress is applied). The stress patterns for thetwo pronunciations of sammalena 'moss' essive singular are represented:(6) without secondary stress	 with secondary stress	w w w	 w s w	sammalena	 sammal enaBoth Whitney and Lehtinen (op. and loc. cit. above) mention that inwords of five or more syllables, secondary stress is shifted to the fourth22syllable if the third syllable is short. The example Lehtinen gives isOppettamatomon (gloss not given). I will not be dealing with any examplesof this type which are, in any case, more a consideration for MetricalTheory than for Government Theory.23Notes to Chapter Two1. As Finnish spelling is very nearly identical to phonetic transcrip-tion, I shall be citing Finnish forms in standard Finnish orthography. Imust, therefore, note some differences. Among vowels, the IPA equivalentsof a and 6 are m and m respectively. The finnish alphabet also includesthe letters <d> and <g>. Outside of their occurrence in loanwords, <d>represents a voiced dental stop, but <g>, where it occurs in the sequence<ng> represents part of a geminate velar nasal. Thus, the weak grade formHelsingin (gen. sg . of Helsinki is pronounced [helsiuuin]. Lehtinen in-cludes /d/ and /u/ among the phonemes of Finnish, but these two sounds arederived by Consonant Gradation and are of limited distribution.2. The representation for [h] is not included among the consonantalrepresentations of Kaye, Lowenstamm and Vergnaud (1989). On the basis ofits continuancy being its most salient feature, I have chosen to representit as O. The sound of [h] varies according to position. Intervocalical-ly it can be quite lightly articulated (Karlsson 1983, 17). Beforeconsonants I have heard it came close in sound to a velar fricative.Acoustic studies reported in Shuken (1984, 124) suggest that in the Fin-nish [h] the glottis is open wider than for whispered vowels.See the last section of Chapter Five for same interesting hCinteractions.3. I have excluded j from the consonant clusters discussed for tworeasons. First, j does not occur before other consonants (one finds iinstead) and second, j is often derived from i by glide formation. Glideformation which occurs 1) where the plural morpheme /-i-/ occurs inter-vocalically (talo 'house' + -i- pl. + -(e)n gen. ---> taloja) and 2) wherea non-low front vowel spreads onto an Onset left empty by deletion (e.g.arki 'weekday' + -n gen. ---> arjen--see Chapter Six).Not all Cj sequences are derived by glide formation. Sane examplesinclude kirja 'book', kalju 'bald' and tyhjyys 'emptiness'. With theexception of loanwords such as anjovis 'anchovy', j is restricted toappearing after 1, r and h--the least camplex sounds in Finnish.4. Nasals regularly assimilate to following stops, so I have not in-cluded sequences such as mk, mt, np etc.5. See Suomi (1985, 220-1) and Prince (1984, 239) for partial lists ofillicit Finnish consonant sequences. A dictionary search has failed toturn up evidence for certain clusters and I have provisionally includedthem among the list of non-permitted sequences. These include *hr, *hl,*hs, *mh, *sh, *lh,	 *sl, *tv, *kv, *vt, *vk. The theory predictsthat the absence of these clusters is a systematic gap, but I do not haveenough evidence to evaluate this claim fully.Certain illicit sequences do appear in compounds (where governmentrelations do not hold) and in unassimilated loanwords (cf. Suomi 1985,220-1).6. The sequence kt also occurs, but only in underlying forms and in afairly limited distribution. The underlying /kt/ sequence is always pre-vented from surfacing either by lenition of the /k/ or palatalization ofthe /t/. For example, the form /kakte/ 'two' appears as kaksi in thenominative singular, as kanden in the genitive singular and as kahtena in24the essive singular.7. Word-final consonants must be coronal in Finnish (Whitney 1956, 15).In the modern language, only the nim alternation is attested; word-finalnon-coronals simply do not occur. An approach such as ItO's Coda Con-dition (It8 1986, 26 and passim) might be co-opted, pending a satisfactoryresolution of the question of non-coronal stop+s clusters. In GovernmentTheory, it might take the form of a prohibition against non-coronals(segments lacking the element RP) occupying the right branch of a Rimewhen they are not Interconstituent Governed by some segment.8. Sonority and complexity are very nearly analogous. In the presentframework, a difference in complexity is all that is required to permit asequence. In some sonority analyses (e.g. Selkirk 1984) a sufficient dif-ference may be required to permit a sequence. Sonority differences leaveunexplained the impermissibility of of the nasal+v sequences since v,being negatively charmed, does not require a complexity difference togovern a charmless segment. The labiality of v may be involved in thisimpermissibility in same, as yet, unclear fashion. If this is so, p and vare most likely patterning together.25CHAPTER THREE - The Deletion of SegmentsThis chapter examines the constituent structure of Finnish lexicalforms and provides a Government-based account of certain deletionprocesses that are sensitive to constituent structure.3.1 Two Segmental Deletion ProcessesKeyser and Kiparsky (1984) present an interesting and insightful ac-count of some camplex facts of Finnish syllable structure. Two processesthat receive particular attention are what the authors term e-Deletion andt-Deletion. The set of data in (1) presents the processes of e-Deletionand t-Deletion (Keyser and Kiparsky 1984, 9 & 11):(1)'house"glass"fire"water"clay " snow " moss "root /talo/ /lasi/ /tule/ /vete/ /save/ /lure/ /sammale/nom. sg . talo lasi tuli vesi savi lumi sarrmalpart. sg . taloa lasia tulta vett& savea lunta sammaltanam.pl. talot lasit tulet vedet savet lumet sammaletvesidpart. pl. taloja laseja tulia savia lumia sammaliagen. sg. talon lasin tulen veden saven lumen samnaleness. sg . talona lasina tulena veten& savena lumena sammalenaThe effects of t-Deletion may be seen by camparing the partitivesingular with the partitive plural. The partitive suffix is /-tA/, butwhen it is affixed to a polysyllabic vowel-final stem, the t does notsurface (e.g. talo + -tA --> taloa) 1 .The rule of e-Deletion syncopates syllable-final /e/ at morphologicalboundaries (except in word-final position in disyllables). Where e-Deletion has created a consonant-final allomorph, t-Deletion does notapply and the /t/ surfaces, as in the partitive singular forms tulta,vettd, lunta and samnalta. 2 .26In Keyser and Kiparsky's analysis, t-Deletion and e-Deletion areformulated as in (2):(2)a.	 t-Deletion	 b.	 e-DeletionC	 oi#	 V	 cr	/ \	 I	 \--> 0 / c v v	 1--> 0 /	 \. 	 :t	 eWhere 0;1 denotes anunstressed syllable.As the rules are applied cyclically, intervocalic t and syllable-final eare not deleted within stems. Furthermore, it must be stipulated that e-Deletion is blocked in the nominative singular of disyllabic e-stems, andwhere an illicit cluster would result. For example, /sammale + -nA/ --->sammalena 'moss' ess. sg . (*sammalna); /save + -tA/ ---> savea 'clay'part. sg . (*savta). The rules must be ordered so that the application ofe-Deletion bleeds the potential application of t-Deletion (without thisordering, the derivation would proceed: /saveta/ ---> /savea/ ---> *sava).3.2 e-DeletionIn Keyser and Kiparsky's analysis, the rule of e-Deletion is inter-preted as the breaking of an association line from the segment /e/ when itis found in syllable-final position, even when doubly linked3 :(3) a) V	 b)	 V \C VT4e	 e27Such rule applications were explicitly argued against by Hayes (1986)where the Linking Constraint was introduced:(4) Linking Constraint [Hayes 1986, 331]. Association linesin structural descriptions are interpreted as exhaustive.If we accept the validity of the Linking Constraint (and Hayes presents astrong case) then the e-Deletion rule cannot be made to do double duty inthis fashion. Furthermore, the number of conditions on the e-Deletionrule (it cannot apply word-finally in disyllables or before the essivesuffix /-nA/, it may not create illicit clusters) make it a suspicioussort of rule. I will therefore argue in the following section that theeffects of e-Deletion are achieved through the action of a specific dele-tion rule, but fall out from the interaction of the prosodic theory andthe structure of these roots.3.3 EMpty Categories and Phonological GovernmentIn the theory of Phonological Government, deletions of segmental ma-terial do not affect the government relations within the skeleton. If amelody element (segment) is deleted, its skeletal slot remains. The dis-tribution of empty skeletal positions is restricted by the Empty CategoryPrinciple:(5) Empty Category Principle (ECP) (Kaye 1987, 6)Empty skeletal positions are subject to a special form ofgovernment (they must be properly governed).28Proper government is defined:(6)	 Proper Government (adapted from Kaye 1987, 11)A skeletal position A is properly governed by anotherskeletal position B iff:i. B governs Aii. B has phonetic contentiii. No other government relation intervenesbetween A and B.The proper government relation that is relevant to the deletion ofvowels is that which holds between adjacent Nuclei (Projection Govern-ment). If in a language L, the direction of projection government isright to left, then the Nucleus dominating the skeletal position of adeleted vowel must be preceded immediately on the Nucleus projection by aNucleus that dominates a phonetically realized vowel:(7) N ----> N	 N1	 1	 ao	 1	 1XXXXX X11	1	 1 I1	 I	 1	C V C	 C VThe underlying form of this hypothetical example would be /CVCVCV/ and thesurface form would be [CVCCV]. In just these circumstances, the twoconsonants that came together as a result of the vowel deletion are not ina government relation. It would therefore be possible to create a clusterby vowel deletion that would not otherwise be tolerated in the language 4 .This cannot be the strategy of Finnish e-Deletion. If the final e ofe-stems remains unrealized because its skeletal position is properly29governed, then the representation of e.g. tulta 'fire', part. sg . wouldbe:(8)(Projection	 N---->N	 NGovernment)	 :	 1 1XXXXXX1 1 1 1 1 11 1 1 1 1 1tulOtaNotice that the 1 of the root and the t of the suffix are separated by anempty skeletal position; therefore, no government relation holds betweenthem. In exactly this case, it is to be expected that superficiallyillicit clusters may occur--that *savta (from /save+tA/) should be well-formed. Since *savta is in fact ill-formed, it must be the case thatthere is no empty Nucleus separating the two consonants. The structure oftulta must be:(9) 0 R	 0 R1	 1\	 1I	 1N \	 : N1 : \ : IIX X X X XI : 1 II IIt u 1 t aIn order to generate this structure, I propose that the root-final eis not associated with a skeletal position at the level of UnderlyingRepresentation:(10)	 X X XI	 I	 II	 I	 Itule30The affixation of the partitive suffix /-tA/ takes the form:(11) X X X	 X XI	 II	 I	 I1	 I	 Itule	 t aaffixationN	 N	 projection of N andestablishment of ICGX X X	 X X	 relationsI	 I	 II I I I ItuletaI1_1 I+1_10 R	 0 R	 projection of 0 and R1	: \	 1	 ,11 	 (first R picks upN \	 ,1 N	 governed segment)I1:	 \,	 .	 1	 IX X X	 X Xi	 I	 I	 I	 1I	 I	 I	 I	 It u 1 e t a	 tultaThe e being excluded from the prosodic structure fails to surface and isultimately deleted by stray erasure.All things being equal, it might be expected that the surface form of(10) (the nominative singular) would be *tul rather than tuli. However,Finnish nouns have a minimal word length requirement of a metrical foot(see McCarthy and Prince, 1986, 2). In terms of the present theory, thismeans that the prosodic structure of nouns must conform (minimally) to thetemplate OROR5 . In erecting prosodic structure according to this template,the skeletal slots will map onto the OR tier one to one (according to thecharm values of their segments):(12) 0 R 0 R11 N11 11	 1X X X1 11tule31In order for the noun to be well-formed, the final Rime must came to domi-nate a segment. The necessary skeletal slot is inserted will be insertedto satisfy the OR template, the floating e associates to it, projects aNucleus and is incorporated into the prosodic structure:(13) O R ORX XI X X <ItuleO R, ORININX X X XtuleO RORX X X Xtule epenthesis of skeletal slot> tul iassociation of /e/ andprojection of Nincorporation of N into RForms to which the essive suffix /-nA/ is attached offer a peculiarproblem. The stem-final /e/ always surfaces. Thus, the essive form of/tule/ is tulena and not *tulna. This comes as no surprise, for it hasbeen established that Finnish does not allow any Cn clusters (see ChapterTwo).However, the behaviour of the essive suffix cannot simply be ascribedto an aversion to generating Cn sequences. Finnish has a class of nominalroots such as perhe 'family' with the underlying form:(14)	 X X X X X XI	 I	 I	 I	 1I	 I	 I	 Iperhe32When affixed to these roots (which are discussed in Chapter Five), CVsuffixes such as the partitive /-tA/ normally geminate. For example, thepartitive form of perhe is perhetta. The essive form is not *perhenna butperheeng. The /n/ of the essive suffix fails to spread backwards to fillthe empty skeletal position and the root vowel lengthens instead:	(15)XXXXXXXX	 *XXXXXXXX:	 1	 1/	 1	 1	:	 1	 \I	perhe+na	 p e r h e+ n aIf the suffixal /n/ were in a government relation with the root-finalempty position, it would be a case of proper government and the emptyposition should remain empty. As the vowel spreads onto it, the emptyposition must be governed by the vowel. Normally, head-final governmentrelations are established before head-initial relations (see the ChapterOne discussion of the generation of prosodic structure). The /n/ of theessive suffix therefore systematically fails to govern a precedingposition6 .The derivation of tulena 'fire' ess. sg . proceeds as in (16). Thereis a crucial difference between this derivation and that in (11). In (11)the /1/ of the root was Interconstituent Governed by the suffix-initialconsonant. That government relation established the /1/ as a non-head andtherefore not within an Onset. In order to satisfy the OR template, thenon-head had to be incorporated into a Rime. In the derivation in (16),the /1/ of the root is not Interconstituent Governed and so is eligible tobecome a head. It therefore associates to an Onset node. Additional Rime33and Onset nodes must be inserted to accommodate the governed /n/ of thesuffix.(16)	 X X X	 X XI	 III	 It u 1 e	n at	 I	 tN N1	 11	 1	X X X	 X XI	 1I	 II	 I	 I	 :	 It u l enaO R 0	 RIII	 I	 I	 :N1	N,I I	 I	 1I 	 I	 I	X X X	 X XI	 I	 II I	 I	 :	 Itulenaaffixation and establishmentof ICG relationsprojection of Nmapping (one to one) ontoOR templateO R 0 R 0 R	 IC governed /n/ mapped into1	 1	 1	 11 1 i	 1	 an Onset; additional R in-N 1	 I N	 serted to satisfy well-11	 1	 i	 1  1	formedness constraint onX X X	 X X	OR tier1	 11	 I	 I 	 ItulenaO R 0 R 0 R	 floating /e/ incorporated1	 1 I I I	 into free Rime (by insertion1	 1	N1 N 1, N	 of X and projection of N)1	 I	 1	 1	 1	 1I JimXX X X XXIIII '1 IIIIIt u l enaThe derivation of forms such as savea 'clay' part. sg . (underlyingly/saveta/) proceeds in similar fashion. The v - t- sequence is not a pos-sible government configuration, so the /v/ associates to an Onset as the/1/ of tulena does. The /t/ undergoes 't-Deletion'.Stems longer than two syllables which end in /e/ (e.g. /sammale/'moss') corroborate the assumptions I have made above. In the nominativesammal, the minimal word restriction is met without having to realize the34/e/. In the partitive sammalta, the government relation between the /t/and the /1/ defines the /1/ as a non-head. The /1/ thus becomes part of aRime. All government conditions are satisfied without realizing the /e/.In the essive form sammalena the inability of the /n/ to govern the /1/allows the /1/ to associate to an Onset. The /e/ surfaces as it does inthe derivation of tulena, above: as a Rime separating two Onsets.3.4 t -DeletionThe above analysis eliminates the need of a rule of e-Deletion in thegrammar of Finnish and attributes the observed effects to the structure ofthe roots involved.In the case of t -Deletion we have a genuine rule of Finnish thatdeletes intervocalic t. Keyser and Kiparsky's (1984) rule of t-Deletionis repeated in (17):(17)=(2a)	 t-Deletion/ \--> 0 /CVVtWhere alq, denotes anunstressed syllable.This rule makes crucial reference to the notion 'unstressed syllable'.This weak syllable does not actually dominate the target segment of therule but dominates the segment adjacent to it. This is a curious condi-tion for a rule. What the condition really states is that the syllable of35which the t is in the onset must not be dominated by the weak member of ametrical Foot--i.e. it must not be in this configuration:(18) *	 F/\/ \s	 wOR ORI	 1	 11	 1CV tVViewed this way, the condition makes more sense, for the target segment isnow within an identifiable domain. The course of investigation must nowbe to determine the process whereby t deletes and what there is about theconfiguration in (18) that blocks that process.Recall the discussion of the Empty Category Principle and ProperGovernment above ((5) and (6), repeated here):(19)=(5) Empty Category Principle (ECP) (Kaye 1987, 6)Empty skeletal positions are subject to a special form ofgovernment (they must be properly governed).(20)=(6) Proper Government (adapted from Kaye 1987, 11)A skeletal position A is properly governed by anotherskeletal position B iff:i. B governs Aii. B has phonetic contentiii. No other government relation intervenes between A and B.Formulated another way, these principles may be regarded as stating thatproperly governed skeletal positions may be left phonetically unrealized.Obviously it cannot be that any properly governed skeletal position willbe left empty, but it does imply that if a language specific rule doesdelink segmental information froth a skeletal position, the rule will notapply if the position is not properly governed.36Given the ECP, the formulation of the t-Deletion rule is quitesimple:(21) t-Deletion0X4t	 condition: t is not dominated by the weakbranch of a metrical FootThe ECP and Proper Government define most conditions on this rule.Only intervocalic t will be affected. In (22a) the government relationsof an intervocalic t are shown. In (22b), the Interconstituent Governmentrelations of t preceded by a Rime-final sonorant are shown:(22)a. R 0 R	 b. R	0 RI	 1	 I	 I	 II	 I	 I	 :\	 I	 I	N 1 N	 N \	 1 N1	 11	:	 \	 I	 iI	 I	 I	 I	 I	X X X	 X X X XI	 I	 I	 I	 I	 1I	 I	 I	 I	 I	 I	 1	V t V	 V r t Vt	 I	 .1.	 I	 1	'—'	1	 1	 1t	 II—,In (22b) the government relation between the t and the r violates condi-tion (iii) of the definition of Proper Government. The InterconstituentGovernment relation between the t, and the r precludes Proper Governmentof the t by the following vowel.The condition on the stress configuration must, unfortunately, bestipulated as a condition on this rule alone. While it is tempting tosuggest that the Foot defines a government domain so that consonantsfalling within that domain are not Properly Governed and therefore cannotbe deleted, this cannot be the case. If consonants within metrical Feetwere never Properly Governed, or if this condition were generalized beyond37the context of this rule, the deletion of k in the weak grade forms ofdisyllabic nouns such as joen ('river' gen. sg . from /joke+n/--see thediscussion of Consonant Gradation in the next chapter) would not bepossible.Keyser and Kiparsky (1984, 23-25) demonstrate an interesting option-ality to t-Deletion that makes it clear that it is stress (rather, say ,than the number of syllables in the word) that interacts with t-Deletion.The Finnish nominal system has two options for stress assignment. Primarystress may be assigned to the first syllable of a word with weak stress onthe remaining syllables or primary stress may fall on the first syllableand secondary stress on the third. If the optional secondary stressprecedes a t-initial syllable, the t is not deleted. Thus, the partitivesingular of e.g. lokero 'pigeonhole' nay be either 16keroa or 16kerata:(23) a.	 18keroa 'pigeonhole' part. sg .	 b. 16kerOta	 idem/ \	 \	 \/	 \	 \	 \/ \	 / \/	 \	 / \S	 w	 w	 w	 s	 w	 s	 wOROROROR	 OROROROR11111111	 1111111111111111	X X X X X X X X	 X X X X X X X X11111111 111111111 ok e r o fta 1 r o t aIn both (23a) and (23b), the position occupied by the suffixal t isproperly governed by the following a, but in (23b) it is in the domain ofa Foot and may not delete whereas in (23a) it is not within a Foot and maydelete. The interaction of stress and Government is an area where little38research has been done. Further investigation may reveal the details ofthis interaction, but for the time being, the stress pattern must remain astipulation on the t-Deletion rule.39Notes to Chapter Three1. Finnish has a vowel harmony system in which suffix vowelsmonize with root vowels for backness. Accordingly, I have usedsegment in representations of suffixes that include harmonizing2. The raising of final /e/ to [i] in non-doubling roots is afeature of Finnish. There is only one non-doubling root ending(nukke 'doll') that I have found not to do this.must har-an archi-vowels.regularin e3. In an interesting parallel to the behaviour of e-final roots (seepreceding note), virtually all of what Keyser and Kiparsky call 'contrac-ted stems' (in my Chapter Five discussion, I call them 'doubling roots')surface with a word-final short [e]. The behaviour of these forms revealsa different underlying structure, however. In Keyser and Kiparsky'sanalysis, e-raising roots end in a short /e/ while contracted stems end ina long /e/ which is geminated around an empty C-slot (Chapter Three,(3b)--the C-slot is required for certain consonantal doublings associatedwith these roots). They therefore attempt to unify the behaviour of alle-final roots by having both undergo e-Deletion. They are certainly rightin seeing a connection between these two types of root and it must be nocoincidence that virtually all e-final roots in Finnish behave in one ofthese two ways.4. Kaye (1987, 17 ff.) demonstrates that vowel deletions of this sort docreate otherwise impermissible clusters in Moroccan Arabic.5. Monosyllables with long vowels (e.g. luu 'bone') vowels comprise afoot as they are bimoraic. It may also be that all superficially longvowels in Finnish are heteronucleic having the form (see Chapter Five):0 R 0 RI	 IIX X X\ /1The nouns with light diphthongs such as tie 'road' present more of aproblem as they are arguably monomoraic and may constitute a genuineexception. Campbell (1981, 175), however, demonstrates that these diph-thongs (which are derived historically from long vowels) pattern with longvowels in the language game kontti kieli. For example, the disguised formof the kieli 'language' is kooli kientti rather than *koli kientti. But,see my discussion of CVVC sequences in Chapter Five for evidence of amonamoraic interpretation.6. Whether the non-governing behaviour of this suffix represents a mor-phological aspect of the suffix or a phonological aspect of the Finnish/n/, I cannot say. Geminate n does arise through the gradation of /nt/sequences (e.g. ranta 'shore' nam. sg .	 rannan gen. sg .) but these gemi-nates are derived by left to right assimilation and not by right to leftspreading.40CHAPTER FOUR - Consonant GradationIn this chapter, I shall examine the phenomenon of Consonant Grada-tion in Finnish and present arguments for analyzing Gradation as a proso-dically determined process.4.1 General Features of Consonant GradationConsonant Gradation is a process of what is traditionally describedas 'weakening'. Because the effects of weakening processes have a widevariety of surface manifestations it is impossible to characterize Conso-nant Gradation as the addition of any specific set of elements (or fea-tures) to the phonological representation of the affected segments. InFinnish, these weakenings range from degemination, to voicing (with orwithout spirantization), to total assimilation, to deletion. In short,weakening affects both the internal structure of segments and their proso-dic relationships with other segments or positions. Examples (fram Karls-son 1983,38) of gradation in the nominal system triggered by case suffixesof the form -C or -CCV are given in (1):strong grade weak grade glosskauppa kaupassa 'shop' nam. sg./iness. sgkatto katolla 'roof' nan. sg./adess. sg .takki takissa 'coat' nan. sg./iness. sg .kipu kivussa 'pain' nan. sg./iness. sg .aiti aidille 'mother' nan. sg./allat. sgjoki joesta 'river' nam. sg./elat. sg .kampa kammalla 'comb' nam. sg./adess. sg .ranta rannalla 'shore' nan. sg./adess. sg .kenka kengasta 'shoe' nam. sg./elat. sg .kulta kullaksi 'gold' nan. sg./transl. sg .kerta kerran 'time' nan. sg./gen. sg .(1)pp ---> ptt ---> tkk ---> kp ---> vt ---> dk ---> 0mp ---> mmnt ---> nnt)k ---> 4nIt ---> 11rt ---> rr414.2 The Nature of Consonant GradationFinnish Consonant Gradation is superficially similar to a number ofprocesses attested in other languages and generally subsumed under thename 'mutation'. Lieber (1987; 72) defines mutations as "phenomena inwhich lexical stems exhibit two or more allomorphs that differ in only asingle marginal segment (for example, an initial or final C or a vowelclosest to the end of a word) and which appear in distinct morphological,syntactic, or phonological environments." One well-known mutation thatshares strong similarities with Finnish Consonant Gradation is the Welshlenition mutation. Same of the alternations of these two processes arecompared below:(2) Finnish Gradation	 Welsh LenitionPP - Ptt tkk kp vt dk 0p bt dk gb vdg - 0In (2), a series of 'strong' stops (geminate in Finnish, voiceless inWelsh) alternate with a series of 'weaker' stops (single in Finnish,voiced in Welsh). A second series comprising the weaker segments of thefirst series of alternations alternates with what must historically havebeen a series of fricatives'. In both languages the modern reflex of thevelar fricative is 0. Lieber (1987) analyzes mutation as the addition ofa floating autosegment to the specification of the affected consonant.42Thus, in a Welsh example such as barged [basged] 'a basket' - y fasged [@vasged] 'the basket' (the symbol '@' represents schwa), Lieber's analysiswould (using her notation) take the form:(3) V C	 VCCVC11:+antl	 a s g e d-cor+lab-cont:Mutation	 :+voiTier	 :+cont:The above account will derive voiced fricatives fram their correspondingstops (allowing, of course, for the overwriting of the prespecified[-cont] feature of the stem-initial consonant) 2 .Such an account cannot, however, be transferred to Finnish. Considerthe following examples of Finnish Consonant Gradation:(4)strong grade	 ratti 'wheel'	 rata 'track'	 nominative sg.rattina	 ratana	 essive sg.rattiin	 rataan	 illative sg.weak grade	 ratin	 radan	 genitive sg.ratiksi	 radaksi	 translative sg.The first difficulty encountered is that the target consonant isseparated from the case suffix by a vowel. A Lieber-style analysis wouldassume that the first component of the suffix is the mutation autosegment.All of the consonantal mutations Lieber discusses involve consonants adja-cent to the mutating morpheme. She allows (p. 96) for the possiblity ofnon-local mutations by placing the mutating autosegment on its own tier.Such a tactic is no longer considered viable in light of the developmentof the feature-hierarchy model (see, for example, Clements 1985).43The degeminating paradigm (ratti -ratin etc.) offers more seriousproblems. Degemination is not a process normally triggered by the addi-tion of a feature to a segmental representation. Indeed, in the light ofstandard autosegmental representations of geminates, it is unprincipled toposit a feature that deletes one of the association lines of a doublylinked segment:(5) X X	 X X	/	 /	A 	 A[ 4-7 ]A third problem with the Lieberian approach is that if one assumesthat the putative autosegment can spread across short vowels, why can itnot also spread across long vowels. The illative case morphololgy gene-rates long vowels and Gradation fails to apply (see (4)). This failure ofGradation in illative forms is not simply a property of the illative casesuffix, for Gradation systematically fails before long vowels. Comparerata 'track' with taide 'art' 3 :(6) rata 'track'	 taide 'art'	 nominative sg.radan	 taiteen	 genitive sg.radaksi	 taiteeksi	 translative sg.ratana	 taiteena	 essive sg.In taide and in all nouns that lengthen their final vowels, Gradation isnot observed even when the suffix involved triggers Gradation after shortvowels. Since in autosegmental representations there is no distinctionbetween long and short vowels on the melodic tier, it cannot be the casethat a putative mutating autosegment is blocked from association by a longvowel but not by a short one.If the trigger of Consonant Gradation is not to be found in the44segmental structure, then where is it? A descriptive answer may be foundin almost any discussion of Finnish Gradation, whether technical or peda-gogical. Here are sane examples:(7)i) In polysyllabic stems z t, k are subject to ConsonantGradation if they are followed by an ending whicha) consists of only one consonant orb) begins with two consonants,and also on condition thatc) between Et, t, k and the ending there is onlya short vowel or a diphthong (not consonants or asyllable boundary)(Karlsson 1983: 31)ii) If a short syllable beginning with k, p or t is closed bythe addition of a consonant, then the k, p or t undergoescertain changes called 'mutation' or 'softening'.(Whitney 1956: 21-22)iii)The strong consonant grade occurs in open and/or longsyllables; the weak consonant grade occurs in shortclosed syllables.(Lehtinen 1962: 523)iv) . . . consonant gradation . . . voices t (and weakens p, kin various ways) and degeminates double stops before a branch-ing rime, that is, in the onset of a closed syllable.(Keyser and Kiparsky 1984: 15)The generalization common to all four of these descriptions is that Conso-nant Gradation applies to a consonant in the onset of a short closedsyllable. In other words, what is relevant to the action of Consonant,Gradation is not the concatenation of certain morphemes, but the syllablestructure that results fram the addition of suffixes to a stem 4 . If thisgeneralization is valid (and I believe it is), then any analysis of Conso-nant Gradation must be expressed (at least in part) in terms of syllablestructure. Finnish Consonant Gradation may thus be defined as a process45of phonological weakening conditioned by certain relationships among syl-labic constituents.'Phonological weakening' is an elusive concept. I have argued abovethat the kinds of weakening observed in Finnish are not easily accountedfor in a model (such as Lieber's) that treats mutations as the addition ofa set of features to the represenation of segments. The elemental theoryof Kaye, Lowenstamm and Vergnaud (1989) fares little better for much thesame reasons. Several proposals for dealing with weakening have been made,including, for example, Foley (1977), Vennemann (1988). These theorieshave not received widespread acceptance, due partly to their preferencefor treating segments as holistic units rather than as feature complexes.I am not prepared at this time to propose a theory of weakening, butI will make a few remarks on how I see weakening being handled within theframework I am using. Weakening occurs when a susceptible segment appearsin a weakening environment. In Finnish, such an environment is (informal-ly) the onset of a closed syllable. More properly, weakening will affectthe segment dominated by an Onset which is Interconstituent Governed by aheavy (branching) Rime:(8)0:\N \X X Xy zAn appropriate segment in this position is subject to a number of restric-tions. If it is doubly linked (i.e. geminate) it will delink one of itsassociation lines (e.g. ratti ratin) 6 . If it is singly linked to a46skeletal position it becomes transparent to spreading of voicing and con-tinuancy from the vowel which heads the Rime that governs it (e.g. kipukivun). If it is linked to another segment (e.g. if the preceding segmentshares elements with it by way of assimilation) then it will assimilate tothat segment totally (e.g. ranta rannan).The preceding is admittedly a tentative proposal which awaits theelaboration of a fully developed theory of weakening. It is sufficientfor present purposes to say that certain consonants in Finnish weaken in agiven context (before a branching Rime) and when they do weaken, theyweaken in the ways described. Of more significance is the distribution ofthis context for Gradation and the question of what constitutes a branch-ing Rime in Finnish. It is to these questions that I turn in the follow-ing chapter.47Notes to Chapter Four1. The Finnish /d/ does not pattern completely with the other weak gradestops. It would be expected (in Foley's (1977) strength hierarchy, forexample), that if the /p/ had spirantized, so would the /t/. In standardFinnish, as described by Karlsson (1983) Whitney (1956) and Lehtinen(1962) the /t/ is voiced but not spirantized. However, Whitney (1956; 13)reports that [d] is "sometimes so soft that it is hardly heard".2. A word must be said about velar fricatives. Lieber treats thissubject in a footnote (Lieber 1987; 129, n. 36): "The segment [g] must betreated as utterly exceptional in any framework. Here we might assume aspecial rule which deletes the [y] that has been created [by the lenitionprocess] 	 " I confess to being suspicious of special rules thatdelete derived sounds that do not otherwise occur in a language, particu-larly as velar fricatives seem to be the most frequent victims of suchrules. Foley (1977; 30-33) discusses exactly this propensity of velarsacross languages to be the weakest of segments in strength hierarchies, soit would appear that more is happening than the generation of a segmentthat violates the phonotactic constraints of a particular language.Just what is happening is not clear and Foley's calculus of strengthvalues does not translate into terms of distinctive features or phonologi-cal elements (in Foley's framework, numerical values are derived for thevarious segments and rewrite rules assign than phonetic values). Foleydoes have same useful insights, however, for the concept of 'weakening'implies a change in the relative strength of a sound, not in its makeup.The difference between the strong and weak reflex of a sound is expres-sible in terms of their difference in distinctive features, but as Finnishdemonstrates, the difference between all the strong reflexes and all theweak reflexes in a language do not necessarily boil down to a differencein the same set of features.Since Finnish deletes derived [y] I cannot ignore the problem entire-ly, particularly as the site of deletion has significant phonologicaleffects. However, my discussions will be confined to the licensing ofthese deletions and their phonological effects. I will leave it toanother to develop a full analysis of phonological weakening.3. taide is a member of a class of noun roots that lengthen their finalvowel in certain instances which will be discussed in Chapter Five. Atthe risk of getting ahead of my presentation, I must point out that thegraded form taide conforms to the general pattern of Finnish Gradation.Such nouns derive historically from consonant-final forms and it is theremnant of that lost consonant that accounts for both the vowel length-ening and the unusual gradation pattern, as will be seen.4. This generalization provides a fourth argument against a Lieberianapproach (and I, in all fairness, must point out that Lieber makes noattempt to account for Finnish data in this fashion and almost certainlyrecognizes that this is not the way to go about it). It is curious,indeed, that the set of suffixes which have the floating autosegment is48exactly the set of suffixes that generate stem-final short closed syl-lables. There is, however, a set of exceptions to this generalization:the possessive suffixes -nsA (3 sg.), -mme (1 pl.), -nne (2 pl.) and -nsA(3 pl.). These suffixes do not trigger Gradation:rata 'track' nom. sg .radaksi	 transl. sg .	 ratansa nam. 3 sg./pl. poss.radalla	 adess. sg .	 ratanne nam. 1 pl. poss.radassa	 iness. sg .	 ratanne nam. 2 pl. poss.The likeliest explanation for these exceptional suffixes comes from Lexi-cal Phonology. If Consonant Gradation is restricted to level I (wherecase suffixation can be expected to occur) then possessive suffixes (whichgo outside case and number suffixes) belong to level II phonology whereGradation does not apply.5. This is not to say that I think Lieber's approach is necessarily thewrong one to take in the cases she discusses. It would be useful toprovide a unified account of mutations that included the Finnish type withthe Welsh type, however.6. The singly linked segment thus created would not be subject to fur-ther weakening for there is a crucial difference between a singly linkedsegment derived form UR and one that has undergone degemination. In thelatter case, the position of the segment remains a governor for the emptyposition created by degemination. This government relation will resistfurther weakening:a) singly linked	 b) degeminatedR 0 R	 R	 0 R\	 \	 i	 i\N	 [	 N \	 N \	 ,	 \	\	 1	 \	 i	 i	 \X X X X	 X X X X XI1	 1	 1	 I	 I	 II	 I	 I	 I	 IVCVC	 V	 C V Ct	 1governmentdomain49CHAPTER FIVE - WC and VCC GroupingsIn this chapter I will examine some putative syllable types in Fin-nish and demonstrate that superficial structure in Finnish is not repre-sentative of underlying structure.5.1 CVVC and CVCC Non-syllablesIn the discussion of the theory of Phonological Government (ChapterOne, above) I presented a list of possible constituents from which thefollowing inventory of 'syllables' (Onset-Rime sequences) can be derived:(1) Inventory of OR pairsa .	 b.	 c.	 d.	 e.	 f.0 R	 0 R	 0 R	 0	 R	 0	 R	 0	 Ri	 I	 I	 I	:	 :\	 :\	 1	 1I	 I	 I	 I	 I	 :\	 1\	 1\I: N	 : N	 : N \	 : \	 N	 : \	 N	 I \	 N \I	 II	 I	 I	 1\	 1	 :	 \	 :	 \	 11	 :	 \	 :\	 :	 \	 1	 \X X	 X X x	 X X x	 X x X	 X xXx	 XxXxS'OG	 I t1_1	 1	 t1	 11	 t1	 11	 t	 1	 t	 11	 t	 11	 1	 t1	ICG t__:	t	 1	 t	 1	 t	 1	 1	 1_1t	 1	 t	 1i_i	 1_1	 I_I	 1._..1	 1_1CV	 CV:	 CVC	 CCV	 CCV:	 CCVCCVV	 CCVVAs Finnish lacks branching Onsets (Karlsson 1983, 19; Whitney 1956, 15),a'H only (la-c) are relevant to the present discussion. Nevertheless, theschemata CV, CVV and CVC do not at first glance appear to suffice for an50analysis of Finnish. Specifically, Finnish exhibits what seem to be CVCCand CVVC syllables':(2) a.	 CVCC 'syllables'salskeaPyrstiikonsti'slender''fish- or bird-tail''trick' Polttaakynttilamonarkkilammuhelm)kanssa'to burn''candle''monarch''lamp''easy''with'b.	 CVVC 'syllables'muuttaamuuntaakiillekiiltosaastataloon kaupunkiinkuntaan'to move''to transform''mica''lustre''filth''house'	 illative sg.'town'	 ill. sg .'commune' ill. sg .ranteeksi	 'wrist' translative sg.ranteiksi	 'wrist'	 transl. pl.rannoiksi	 'shore'	 transl. pl.taiteet	 'art'	 nom. pl.terveen	 'healthy' gen. sg .perkeleettá	 'devil'	 partitive pl.5.2 CVCC SequencesThe motivation for claiming the existence of CVCC syllables in Fin-nish is the presence (on the surface) of CCC clusters (Prince 1984, 239;Ito 1986, 40-47). Both of these authors recognize the restrictions onclusters of this type: they are always either sonorant+s+stop orsonorant+geminate stop. Prince and Ito both propose syllable structureconstraints that effectively preclude sequences of C+[-cont]+C, relying on(i) sonority sequencing constraints to guarantee that the first C is a51sonorant and (ii) the double linking of geminates to allow them whereother [-cont]+C sequences are disallowed. While their analyses do accountfor the data in terms of descriptive adequacy, they fail to make somesignificant observations.A great many (if not most) of the sonorant+geminate forms in Finnishare loanwords. Examples in addition to the forms for 'candle', 'lamp' and'monarch' in (2a) include: polkka 'polka', korkki 'cork', =Aid 'monk'and temppeli 'temple'. None of these forms has a geminate in the sourcelanguage; gemination is therefore a Finnish development perhaps reflectinga strategy particularly favoured in borrowing 2 . This observation doesnot, of course, explain why Finnish should develop a series of CCC clus-ters when they are otherwise disfavoured, but it does shed light on why somany of the CCC clusters that do occur involve geminates.As for CCC sequences where the middle member is s and only s, thegovernment relations which determine segmental sequencing allow s as theonly non-sonorant consonant to be governed by a following obstruent (seeChapter Two, (4), above). It follows that there could be no form such as*kalpka since there could not be a form *kapka in the first place. It isless clear in the face of forms such as halm6 'fool' and kampi 'crank'where there cannot be, for example, a word such as *heilmpli as no sequen-cing constraints are violated per se. I will thus concede the operationof a constraint that limits the middle member of a CCC cluster to s.525.3 Monomoraic SequencesPrince and Ito assume the possibility of Rimes of the form:(3) * R1\\N \\ \X y zOrIt is here that the problem of CCC clusters is revealed, for such struc-tures are precluded by the Government framework. If the claims of thetheory are to be maintained, the obvious fact of CCC clusters in Finnish(regardless of their limited distribution) must be accommodated within theframework. Herein lies the true reason why the first member of thesesequences is a sonorant. I propose that the sonorant in a Finnish CCCcluster does not depend directly from either the Rime node or the Nucleusnode at all, but shares the skeletal position of the nuclear vowel crea-ting a monamoraic sequence3 :(4):\N \X X/ \V son CThus, it is not simply a preference for left to right decreasing ofsonority in Rimes that assures that the first segment of a CCC cluster isa sonorant, but rather the fact that in Finnish (as in many other lan-guages) only vowels and sonorants may appear in the Nucleus of a Rime.The monamoraic structure is not wholly unmotivated for Finnish. There isa set of light or 'rising' diphthongs (ie, uo, yo) in Finnish (Karlsson1983, 19; Whitney 1956, 14). They consist of a charmless vowel preceding53a charmed vowel (which acts as a head) attached to the same skeletal slot:(5)/ \Vo VtThe direction of government is not fixed and monomoraic vowel sequences ofthe reverse order do occur in Finnish. Such diphthongs arise, for example,in the suffixation of the plural morpheme -i-, which, having no skeletalposition of its own, associates parasitically to a position in the root.If affixed to a root ending in a long vowel, the root vowel is shortened.If affixed to a root ending in a short vowel, the root vowel may mutate orsyncopate4 . If affixed to a root ending in a light diphthong, the first(non-head) element of the diphthong is deleted:(6) a. affixation of [—i—]plural+[-11A]ess. to a long vowelmaa + i + nA ---> mainluu + i + nA ---> luina'country''bone'XXX XX11111mainab. X X X	 X X ma +i+na(7) a. affixation of [ —i — ]plura1 4 [ -11A]ess. to a short vowelmykka + i + nA ---> mykkina 'mute'kala + i + nA ---> kaloina 'fish'lasi + i + nA ---> laseina	 'glass'save + i + nA ---> savina	 'clay'b. XX XX	 X X	 XXX X XX1111	 / \111111	 11	 11k a 1 a+i+na	 kalo ina54(8) a. affixation of [-i- ]plura141-nMess. to a light diphthongtie + i + nA ---> teinasuo + i + nA ---> soinayo + i + nA ---> Oina'road''swamp''night'b. X X	 X X XXXX>	 / \so inasu o+i+na0There is a further piece of evidence for the monanoraic nature of thediphthongs created by plural affixation. Disyllabic roots which doubletheir final vowel behave like the forms with underlying long vowels as in(6). The behaviour of disyllabic roots is demonstrated in (7). I shallmotivate the structure of these forms in the next section, but a distinc-tion may be drawn on the basis of their surface behaviour. If one can-pares, for example, lasi 'glass' with perhe 'family' the main differencebetween them is the presence of the long vowel in certain forms of perhe:(9) lasi	 perhe	 nan. sg .lasing	 perheena	 ess. sg .laseina	 perheina	 ess. pl .The true distinction between long and short-vowel roots appears in formsthat undergo Consonant Gradation:(10) synti 'sin'	 ranne 'wrist'	 nan. sg .syntina	 ranteena	 ess. sg .synniksi	 ranteeksi	 transl. sg .synneiksi	 ranteiksi	 transl. pl.In the translative singular forms in (11) the basic gradation pattern isestablished (weak grade with short vowels before a CC cluster; stronggrade with long vowels before a CC cluster). In the translative pluralforms, the diphthongs pattern with the vowel lengths--a diphthong derivedfrom a short vowel is accompanied by the weak grade. A diphthong derived55from a long vowel is accompanied by the strong grade. The difference inrepresentation between these two forms (which I shall argue for in thenext section) is:(11)	 synneiksi0 R	 0 R	 0 Rranteiksi0 R	 0 R 0 R	 0 RI	 :\	 :	 :\	 I	I	I	I	 :	 I\	 I	 II	 I	 1\	 :	 IN \	 : N \	 I N	 : N \	 : N	 N \	 , N\	 :	 1	 \	 I	 :	 1I	 :	 \	 II	 II	 :	 \	 II	 II	X X X X X X X X	 X X X X X X X X XI	 I	\ /	 / \	I	 I	 I	 I	 I	 I	 I	 I	 I	 I	 II	 I	 I	 I	 I	 I	 I	 I	 I	 I	 I	 I	 I	 IS y	 ne iksi	 rante i ksi5.4 Monomoraic Vowel-Sonorant SequencesHaving established the presence of monamoraic vowel sequences inFinnish, I will suggest that there are also monamoraic vowel-sonorantsequences. Thus, for example, the representation of helppo 'easy' is:(12) 0 R	 0 RN	 NI	 \X X XXX/ \ \/helpoThere is further evidence of the nuclear nature of these sonorants.A few forms (e.g. /lapse/ 'child', /kiitokse/ 'gratitude') which undergoe-Deletion (see Chapter Three, above) simplify their clusters in syllable-final positions before consonants:(13) lapse + tA --->	 lastakiitokse + tA ---> kiitostalapse + n --->	 lapsenkiitokse + n ---> kiitoksen'child' part. sg.'gratitude' part. sg'child' gen. sg.'gratitude' gen. sg.•56Forms with nasals behave the same way. Thus tuhat (/tuhante/) 'thousand'patterns with /lapse/ and not with /helppo/ 5 :(14) tuhante + to ---> tuhatta *tuhantta	 'thousand' part. sg .Although the geminate in tuhatta is a false geminate, being derived byconcatenation of the /t/ of the root and the /t/ of the suffix, from thepoint of view of prosodic structure, there is essentially no differencebetween a true and a false geminate. If (as in other frameworks) one cansyllabify e.g. kentta 'field' as:(15) 0	 R	 0	 Rt	I\\	 :	iI	 I1	 N \ \	 1 N1	 II	:	 \	 \	I	 II	 I	 tX X XX X X1\ /	 1k1e n	 t	 athere should be no reason why one cannot derive *tuhantta in the same way,whether the tt collapses to form a true geminate (16a) or remains a falsegeminate (16b):(16) a.	 0	 R	 0	 R	 0	 R1	 1 \\	1	 11	 I	 1	 : 	 1	 1	N 	 1	i 	 N \ \	 1	 NI	 I	 II	 1	 I	 1	 \	 \	 :	 iX X X X X X X X1	 1I	1	 :	 \ i	It	 1	 I	 tt	 uhant	 ab.	 0	 R	 0	 R	 0	 R	1\\1	 1\\	 1	 11	 1	 1	 1	 11	 N	 1	 N\\	 :N	I 	 I	 1	 \	 \ I1X X X X X X X X1	 1	 1	 1	 1	 1	 11	 1	 1	 1	 1	 1	 1	 1tuhantt	 aHowever, as this kind of structure is not available in the present frame-work, there is no way to derive the hypothetical forms in (16). Assuming57that the underlying forms of kentt& and tuhat map onto skeletal slots as:(17)	 X X X X X/ \ \ /	 i	 andkent	 aX X X X X XI	 I	 I	 1	 1	 II	 I	 I	 I	 I	 Ituhant ethen one expects a different prosodic structure. Finnish will not tole-rate the triconsonantal cluster that would result from syllabifying/tuhante+tA/ as *tuhantta so same other tactic must prevail. Intercon-stiuent Government relations are established between the root and theaffix (the fate of the /e/ is discussed in Chapter Three, above) 6 :(18) 1	 I	 *---I1	 1IX X X X X X X XI	 I	 Ituhantet aThe mapping onto 0 and R nodes can be satisfied by the segments in govern-ment relations, so the ungoverned /n/ remains unexpressed 7 :(19) 0 R 0 R	 0 R,I1	 :	 :	 :\	 1	 IN	 :	 N \	 1	 N1I	 1:	 1	 I	 I	 \	 IX X X XXX X Xt u h anteta 	 > tuhatta	1	 :$ *0 0585.5 CVVC SequencesGovernment Theory excludes the representation of WC Rimes:(20)	 * R:\N \I\ \X X XV CNevertheless, Finnish seems to attest CVVC syllables (cf. (2b) repeatedhere):(21)=(2b) CVVC 'syllables'a. muuttaa	 'to move'muuntaa	 'to transform'kiille	 'mica'kiilto	 'lustre'saasta	 'filth'b. taloon 	'house'	 illative sg.kaupunkiin	 'town'	 ill. sg .kuntaan 	'commune' ill. sg.c. ranteeksi	 'wrist'	 translative sg.ranteiksi	 'wrist'	 transl. pl.rannoiksi	 'shore'	 transl. pl.taiteet	 'art'	 nom. pl.terveen	 'healthy' gen. sg .perkeleett&	 'devil'	 partitive pl.There is evidence, however, that most of the above V:C and WC sequencesdo not define constituents. Consider first the illative forms in (21b).59The presentation of more illative forms will reveal the nature of thissuffix:(22) nam. sg	 ill. sg .	 ill. pl .maa	 maahan	 maihin	 'country'tie	 tiehen	 teihin	 'road'luu	 luuhun	 luihin	 'bone'talo	 taloon	 taloihin	 'house'kala	 kalaan	 kaloihin	 'fish'After monosyllables and diphthongs, the illative suffix takes the form-hVn with the suffix vowel being a copy of the last stem vowel. Thespreading of the stem-final vowel is probably related to the regularprocess of Vowel Harmony in Finnish. In Vowel Harmony, suffix vowels gaintheir expression of backness fram the value for backness in the root orstem to which they are affixed. In the case of the illative, the entireidentity of the vowel is so determined. The spreading of a Harmony pro-cess is different fram that which creates long vowels. Vowel Harmonyspreads elements on the Nuclear projection and so is not blocked by thepresence of intervening consonants:(23) /kapu/ + /-nA/e ssi v .	 'cone'I°-,-A+	 U	_o	 A*	I	 I	 I	I	 I	 I	N	 N	 N	1	 1	 1	I	 I	 IX X X X X XI	 II	 n 	 Ivo	 ?o	 No1I	 :	 1?o	 uo	 Ro..	 ..	 I	 IH- 	H-	 ?ok a p y n a60A linearization or 'Tier Conflation' operation then produces the represen-tation:(24) X X Xi X X X1	 II	 1	 1	 I	 :v° A'' ?° U° R° AlI	 I	 II	 1I	 I	 I	 1	 I?o II, uo 10 ?o 10I	 11	 1H-	I NokapyndThe illative morpheme has the form:(25) 0	 R1 	:\1N \11	 1	 \X X X1h	 nand when affixed to e.g. /talo/ 'house' creates the following structure:(26) OROR	 0 R	 0 R 0 R 0 R1	 1	 1	 I	 1\ 	 	 I	 I	 1	 1	 1\I	 I	 I	 I	 I	 1	 I	 I	:aN loN	 ' N\	 laNloN:N\:	 \1	 1	 \:	 +	 11	 1	 \	 1	 \i 	\T.-.„1 .	 \....._.	X x x x	 x x x	 x x x x x x x.:	 I	 I.	 I	 It	 1	 h	 n	 t	 1	 h	 nAfter tier conflation this structure is derived:(27) O R O R O RI,'	 I	 1	 1	 ,'	 1\N 1 N 1 N \I1	 I	 1	 I	 I	 I	 \XXXX XX X1	 1	 1	 :	 I	I	 I	 II	 Italohon61The intervocalic h is deleted by a rule:(28) 0h	 condition: /h/ may not be preceded by avowel sequence or long vowelThe condition precludes the derivation of e.g. *tieen rather than tiehen'road' ill. sg . or *maaan rather than maahan 'country' ill. sg. 8 .A second indication that V:C and WC sequences in Finnish do not forma constituent is the failure of Consonant Gradation before these struc-tures:(29)nam. sg .	 transl. sg .	 transl. pl .ranne	 ranteeksi	 ranteiksi	 'wrist'taide	 taiteeksi	 taiteiksi	 'art'aie	 aikeiksi	 aikeiksi	 'intention'laite	 laitteeksi	 laitteiksi	 'appliance'In Chapter Four, I argued that the trigger for Gradation is a branch-ing Rime node which governs the target for Gradation. Under the as-sumption that V:C and WC sequences constitute branching Rimes (as in(30)), the expectation would be that these branching Rimes also triggerGradation. The data in (29) show that this is not so.(30) V:C and WC as branching RimesR	 R\	 :\N \	 N \\	 :\X X X	 X X X:/	 1	 I	 I	 II	 I	 IV C	 V V COne must therefore conclude either that the specification of 'branchingRime' is not sufficient to define the trigger for Gradation (for example,Gradation may be sensitive to the number of morae in the Rime) or that the62sequences in question are not branching Rimes. The Theory of PhonologicalGovernment explicitly disallows the structures in (30) (see Chapter One,(6)). On the basis of the structures derived in the discussion of illa-tive case morphology, above, I conclude that V:C and VVC sequences arerepresented:(31)	 R 0 RN N \\X X X\/V	 CandR 0 R:\N N \\IX X X1	 1	 1I	 1	 IVi V2 CA form such as ranteeksi 'wrist' transl. sg . is thus represented:(32)	 0 R	 0 R OR	 0 RI	IX	 I\I	 I	 II	 1	 1	  N\	 INN\	\ 	 \X X XX X X XX1	 1	 I	1\	I	 11	 I	  	 1	 1rant	 a	 k s i5.6 The Derivation of Doubling FormsThe doubling of root-final vowels is not the only interesting proper-ty of these noun (and adjective) roots. The term for these forms thatKeyser and Kiparsky (1984) use is 'contracted stems'. The reason for thisappellation is the fact that they behave as though they had a finalconsonant in the nominative and either lengthen their final vowel ordouble a following consonant is other cases. Thus, ranne 'wrist' exhibitsthe weak grade in the nominative even though the final syllble seems to be63(33)	 C V C C V C VI	 I	 I 	 Iranteopen and thus not a trigger for Consonant Gradation. Keyser and Kiparsky(1984, 19) assign these roots the form:and rely on their rule of e-Deletion to degeminate the /e/ and leave theempty C to trigger Consonant Gradation:(34) a/ \	 / \/	 / /\CVCCVCraneI have argued above against e-Deletion and the use of a rule that affectsgeminates as well as single segments. Furthermore, in a theory thattreats the skeleton as undifferentiated timing slots, there is no need topostulate two extra positions in these roots. Keyser and Kiparsky arecorrect in their interpretation of the data, that an empty position inthese roots affects their surface behaviour, but I shall propose an alter-native analysis that does not require two additional skeletal positions.I postulate that after association to the skeleton, these roots havethe form:(35) X XX X XXI	 I	 1 :	 1I	 I	 I	 Irante64The empty position triggers several effects on the surface.(36) a. Consonant Gradation. Forms such as ranne 'wrist'exhibit gradation before an 'open' syllable.b. Phrase and sentence level sandhi:terve lapsi 'healthy child' is pronounced [tervellapsi]terve poika 'healthy boy' is pronounced [terveppoika]c. affixation of partitive case suffix /-tA/ shows gemination:tervett4 'healthy' part. sg .rannetta 'wrist' part. sg .compare: tulta <--- /tule/ 'fire' + /-tA/ ('e-Deletion')taloa <--- /talo/ 'house' + /-ta/ ('t-Deletion')d. Gemination in ca pounds:ranne 'wrist' + kello 'clock' ---> [rannekkello] 'wristwatch'e. Vowel lengthening where the following consonant cannotgeminate:i) final consonants:terve 'healthy' + -t (nom. pl.) ---> terveetperhe 'family' + -n (gen. sg .) ---> perheenii) geminates and clusters:paine 'pressure' + -ssA (inessive sg.) ---> paineessaterve 'healthy' + -1tA (ablative sg.) ---> terveeltaiii) the essive suffix /-nA/ (see above, section 5.2):perkele 'devil' + -nA (essive sg.) ---> perkeleen5The gemination of consonants either in affixation or sandhi I willrefer to simply as 'doubling'. I have designated this process of vowellengthening 'Pseudo-Compensatory Lengthening' because the lengthening ob-served, while similar to true compensatory lengthening, does not fill aposition voided by a morphological or phonological process but fills aposition present, but unrealized, in the underlying structure of the mor-pheme. If compensatory lengthening were a general feature of Finnish65phonology, one might expect lengthening to occur where Consonant Gradationhas resulted in degemination (in Government Theory, the skeletal slot isnot deleted, but remains empty). This is not the case; for example, thegenitive sg. form of kukka 'flower' is kukan not *kuukan.As I suggested above, the underlying form of a root like /terve/'healthy' is:(37)	 XXXXXX1t e r v eAs a convenient shorthand, I will represent these roots in text as e.g./tervex/.The status of the empty position at the end of the root is ambiguous.It is in a position that is potentially properly governed by the precedingvowel, in which case the vowel could spread onto it (i.e. Pseudo-Compen-satory Lengthening would apply). It is also potentially properly governedby a following consonant, in which case the consonant could spread onto it(doubling). The correct government configuration (as determined by suf-fixation) will determine the prosodic structure realized.66In the unmarked case (derivation of the nominative) there is no overtsuffix attached to the root. The derivation would proceed as follows:(38)	 X X X X X X	 mapping onto skeleton andestablishment of ICG	t e r v e	 configurationsTIt	 III	 I	 I	 I	 I1• 	 II	 II__I	 1._...1I	 I	 I	 II	 I 	 I	 1N 1	 N1	 1 r	 1	 4rX X X Xi X XI	 I	 II	 I	 I	 It e r v e0 R	0 RI	:\	 I	 :\I	 II N \	 I N \I	 II I	 \	 I	 1	 \I	 IXX XXX XI	 I	 I	 I	 II	 I	 I	 I	 It e r v eprojection of N and establish-ment of SCG configurationsmapping onto OR tierThe question that immediately comes to mind is: 'Why is the emptyslot, Subconstituent Governed by a Nucleus, incorporated into the Rimerather than into the Nucleus?' It is clear from the behaviour of thesewords (doubling of following consonants) that this is so. If the Nucleuscame to dominate the position, the result would be a long vowel (*tervee).To further complicate matters, the empty position does sometimes came tobe part of a Nucleus (in the cases where Pseudo-Compensatory Lengtheningapplies). Keyser and Kiparsky (1984, 22) suggest that "the Finnishlexicon permits no V sequences, whether long vowels or diphthongs, inunstressed syllables". This assertion is corroborated by Whitney (1956,14): "Diphthongs, with the exception of those in which -i- is the secondelement occur only in the first syllable of a word [the locus of primarystress]. Elsewhere, such combinations of vowels are not diphthongs , but67belong to separate syllables, and this is noticeable in the pronuncia-tion"9 . The import of these observations is that the structure of FinnishW sequences outside of stressed (or initial) syllables is either:(39) XVI V2or(40) ROR	 RORI	 I	 I	 II	 I	 I	 IN N	 and N NI	 I	 I	 II	 I	 I	 IX X	 X XI	 I	\ /I	 IV1 V2	 VI have already observed that Finnish has no branching Onsets. Ishall therefore further assert that there is a restriction on the branch-ing of Nuclei: outside of stressed syllables N may not branch in Finnish.This being the case, an empty position governed by a Nucleus must bedominated by a Rime and not by a Nucleuslo.68In a derivation where doubling occurs, such as painetta 'pressure'part. sg ., the empty position is Interconstituent Governed by the suffix-inital consonant:(41)	 X X X X X X	 X X	1t	 +	 1	1	 1	 1	 [	 1	 1	 :	p a' io n e 1 	t at	 1 I	1_1	 t	i_i	 I	 :	 1t	 II—I	I__I	 I__I	 	 I	 I	 I	I	 I	 I	I	 I	 I	 I	N	 1	i	 N :	 N	II I	 I	I	 I	 S	 IX X X X X X X X: 	 1	 I	I	 II	 I	I	 I	 II	 I	paine	 t aassociation and ICGprojection of N and SCGO R 0 R	 0 R	 mapping onto OR andI	 I	 I	 II	 I	 I ;\	 I 1	 gemination of t ontoN	 1 N \ 1 N	 governed empty position11	: \	1	 1	 \	 :	 1I  1XXXX X XXX1I I I 1 \I II	 I	 I	 I	 1paine	 t aHere, the empty position is identified as non-nuclear by virtue of Inter-constituent Government by a non-vocalic segment. It therefore isincorporated into R rather than N. The spreading of the /t/ does notnecessarily follow automatically, but may be governed by rule--the evi-dence is as yet unclear on this point.69The final case involves the derivation of a long vowel as inperheeksi 'family' transi. sg.:(42)	 X X X X X X	 X X XI	 : 	 1	 I	 I	 4.	1	 1	 11	 1	 1	perhe	k s it	 I	 t	 I	 t	 1	 11_1	 1_1	1_1 1t	 II-- 11_1	 1 _11	 1	 I	 I1	 1I	 :	 1	 IN :	 N N :	 N1 	 I	1	 :	 *	 1IXXXXXXXXX: II	 II	 II	 1	 II	 II	 II	P e r h e	 k s iassociation and ICGprojection of N and SCGO R	 0 R 0 R	0 R	 mapping onto OR tier:\1\	 1	 :	 I	 1	,	 and spreading of vowel1	 ,1 N \	 1 N	 N \	 1 N1	 1	 1I	 I	 I	 \	 I	 II	 :	 \	 I	 : I	 IXX XXX X XXX1	 1	 1	 111	 1	 I	 1 	 /	 I	 1	 1perhe	 k s iHere, the /k/ of the suffix is not in an ICG relationship and must enterinto a government relationship with the position preceding it. In orderfor this to happen, the position must project a Nucleus so as to become agoverning category. The vowel must spread into the empty position so thatit can govern (for empty positions cannot govern). It may be that thespreading of the vowel precedes and facilitates the projection of N, theprohibition on branching Nuclei preventing the two positions beingdominated by the same N node.That the structure derived in (42) is the correct one is supported bythe fact (mentioned above) that forms such as ranteeksi 'wrist' transl.sg . do not gradate in spite of the fact that superficially (and in sometheories of syllable structure) they appear to have a susceptible segmentin the onset of a closed syllable. Further corroboration of this70structure can be found in the behaviour of the plural morpheme /-i-/.When affixed to a doubling root (e.g. /rantex/ 'wrist'), the plural mor-pheme (which is a segment without a skeletal position) associates to theroot-final position. When affixed to a non-doubling root (e.g. /ranta/'shore') the plural morpheme associates parasitically to the root-finalposition (and incidentally triggers a vowel mutation). With the furtheraddition of a CCV case affix, the surface forms are superficially similar:ranteiksi 'wrist' transl. sg . and rannoiksi 'shore' transl. sg .. Thedifference between them is that /ranta/ undergoes Consonant Gradationwhile /rantex/ does not. In the one case, the ViC sequence constitutes abranching Rime and in the other it does not. The structure of these formsmust therefore be:(43)	 a. rannoiksi	 b. ranteiksi/ranta + i + ksi/	 /rantex + i + ksi/gradation	 no gradationOR	 OR	 0 R OR1	 1OROR OR1 1\	 1 1\ 1 1\ i i :\N\ N\	 I	 N IN\ N\ INIII \	 I I\ II11\11II I1\11IIX X XX X	 X X X X X X X X X X X X1 1	 1/ /	 1	 1	 1r a	 n	 o iksi r a n t e i k s i71Notes to Chapter Five1. The Finnish lexicon contains a number of unassimilated loans thatappear to have branching onsets: trigonametria 'trigonametry', mikrofilmi'microfilm', platina 'platinum', Skotlanti, 'Scotland', strategia 'strate-gy'. Word-initial clusters most likely involve extrametricality (seeKaye, Lowenstamm and Vergnaud 1989, 44-47). Word-internal clusters thatviolate Government sequencing constraints may involve empty positions (seeabove, Chapter Three) that separate the offending segments, e.g.mikOrofilOmi.This second approach might account for the CSC clusters, but toappeal to empty positions in this case would allow for the possibility ofCCC clusters in which the second sound is an obstruent. Such forms arenot attested (*pyrt0k45). On the whole, it seems safer to postulate thatCsC forms pattern with sonorant+geminate sequences like polttaa 'toburn'--especially as the first segment of CsC clusters are always sono-rants.2. Dr. Zita McRobbie (personal communication) informs me that thegemination of native forms such as polttaa 'to burn' and kanssa 'with' isa relatively recent development in Finnish. It is not, however, an acrossthe board phenomenon. Finnish has many words of the form CV[+son]CV:jalka 'foot', pentu 'cub', kumpu 'hill'.One byproduct of this gemination in loanwords is that the the resem-blance of the word to its form in the lending language is not lost whenthe word appears in a weak grade form. A comparison (using the genitivesuffix -n) of the effects of Consonant Gradation on these forms and theeffects Gradation would have in non-geminated versions is given below:a) actual Gradation pattern b) Gradation pattern without geminatepolkka-polkan	 *polka-polan	 'polka'lamppu-lampun	 *lampu-lammun	 'lamp'monarkki -monarkin	 *monarki -monarjin 'monarch'munkki -munkin	 *munki -mungin	 'monk'This may be just a fortuitous coincidence of the borrowing of these formsat the same time Finnish was undergoing the historical change mentionedabove, but in support of the idea is the fact that while 'monarch' wasborrowed as monarkki, 'monarch' became monarkia. The lack of geminationmay result from from a metrical effect of the extra syllable, but it istempting to suggest that the extra syllable protects the k from Gradation(the genitive is monarkian) and so the k did not need to double. Olderborrowings such as ranta 'shore' (from German Strand) undergo Gradationnormally (i.e. ranta -rannan). Newer borrowings fail to undergo Gradationeven without gemination to preserve the original sound. Thus auto -auton'car' (although Dr. McRobbie informs me that she has heard *audon collo-quially).On the question of quantity in Finnish, see Lehiste (1965).3.	 I owe this suggestion to Glyne Piggott (personal communication).Much of the ensuing discussion might be more clearly presented in the72notation of Moraic Phonology (see Hayes, 1988). However, as long as it isunderstood that the representation:0 R	 0 RI	 I\	 I	 II	 I1	 N \\	NX X X XX\la m p	 uis equivalent to:there should be no difficulty in following the X-slot notation.4. The details of these mutations and syncopations are too complex andnumerous to present here. See Karlsson (1983, 39-43) for discussion.5. That the underlying form is /tuhante/ may be seen from the illativeand essive forms tuhanteen and tuhantena. Forms with CC-initial suffixesexhibit the expected weak grade pattern; e.g. tuhanneksi translative sg.6. The fact that tuhat derives fran the same Germanic form as does theEnglish 'thousand' does not settle the question of whether the [e] thatsurfaces in longer forms which exhibit 'e-deletion' is underlying orepenthetic. Much depends on when this word was borrowed fram Germanic.If the borrowing was long enough ago, the borrowed form may have been*thusundl rather than *thusund.7. In contrast, the underlying /n/ will be picked up if the OR mappingcannot be satisfied on the basis of government relations alone. Thederivation of the essive form tuhantena wherein the /nj of the suffixcannot Interconstituent Govern the root-final /t/ proceeds:I.,..II	 I	 I	 IN	 t	 ___N	 II	 I NI	 I	 I	 III	 *	II	 *	I	I	 *	IIX X X X X X X XI	 I	 I	 I	 I	 I	 I	 II	 I	 I	 I	 I	 I	 I	 It u h a n ten a(derivation continues)73OR OR	 0*ORI	 I	 III	 I\	 11I	 N	 iI	 N\	 1I	 1I	 NI	 I	 I	 II	 I	 1I	 I	 I	 I\	 I	 I	 IX X X X XX X XIIIIIIII1 $	 1	 11t u h a n ten aOR OR	 OR OR1111\\1111:IIN::IIN\1 \,IINI1,IINI1X X X X X X X X XII IIII I1 I1 I1 I1 11 IIt u h a n t e n a8. The effect of the condition on the rule likely falls out of a generalconstraint on vowel sequences. In these cases, the [h] is required tobreak the hiatus between vowels (see Schane 1987 for a discussion ofhiatus-breaking).9. Lehtinen (1962, xvii) treats Vu and Vy sequences outside of theinitial syllable as tautosyllabic diphthongs as well. She also assertsthat long vowels are always tautosyllabic. As she is dealing with pronun-ciation and not with phonological structure, her assertions are not toodamning a criticism of the present proposal. In fact, where long vowelsderived by morphological processes such as the affixation of the illativecase suffix have the structure:R 0 RN NX X X\it is to be expected that in pronunciation the heterosyllabic geminatevowel will be surface as a single (phonetically) long vowel.10. The stipulation that N may not branch in unstressed Rimes raises thepossibility that N may branch where secondary stress is assigned. Thispossibility will not generate counterexamples to doubling where longvowels surface instead of an empty Rime-final consonant. Finnish has nonominal roots of sufficient length to create this situation. A threesyllable form such as perkele 'devil' is not subject to secondary stressassignment. A four syllable form (if such exist) would have the Nucleusdominating the last vowel in an unstressed position. The possibility ofsecondary stress assignment in an affixed form exists, e.g. prkelettA,but the ICG relation between the suffixal /t/ and the root-final emptyposition precludes the interpretation of that position as Nuclear (ICG hasprecedence over SCG).Evidence for branching of N is found in the stress pattern of some74trisyllabic nouns. Keyser and Kiparsky (1984, 24-25) offer the examplemellakka	 If secondary stress is not applied, the essive pluralform is mellakkoina. If secondary stress is applied, then the essiveplural is ngllakOina. One possible interpretation of these data is thatthe application of secondary stress allows the structure:0 R	 0 R	 0 R 0 R1	 I	 I1\	 1N\	N\	N	 I N\	 1	 i	 \	 i	 :\X X XXX XXXXXX\:	 I	 II	 I	 I	 I1	 1	 1	 1	 1m e	 1 a	 k oin awhere the branching of N triggers Consonant Gradation (in this case,degemination of k). This does not seem likely to me for the behaviour ofthe plural marker /-i-/ (for example, the mutation of /a/ to [o] here andthe cases where /-i-/ shortens or deletes root-final vowels--see abovesection 5.2) indicate that the plural marker does not have its own skele-tal slot in UR. It is not clear how assigning secondary stress couldinsert a position to which the plural marker could associate.Another possibility is that the stress pattern itself influencesgradation, but this cannot be the whole story, for one would expect to seenon-plural forms with secondary stress in the weak grade and this is notthe case (*mellakana).A third possibilty is that branching of a Rime always triggers Grada-tion and branching below the Rime in conjunction with stress will triggerGradation. In this case, the branching of the Nucleus or of the skeletalposition itself would count as the trigger for Gradation:0 R	 0 R	 0 R OR1	 1\	 I	 :\	 I	 1	 1	 1N \	 N \	 N	 N\	 I	 1	 \X X XXX XX X X X\:	 / \	I	 Im e	 1 a	 k o i n aClearly, further investigation is required to resolve this issue.75CHAPTER SIX - Further Aspects of Consonant GradationIn this chapter I shall return to the subject of Consonant Gradationand examine same exceptional and unusual manifestations of this process.6.1 Same Exceptions to Consonant Gradationstrong grade weak gradekauppa kaupassa 'shop' nam. sg./iness. sgkatto katolla 'roof' nan. sg./adess. sg.takki takissa 'coat' nam. sg./iness. sg .kipu kivussa 'pain' nam. sg./iness. sg .aiti Aldine 'mother' nam. sg./allat. sgjoki joesta 'river' nam. sg./elat. sg .kampa kammalla 'comb' nam. sg./adess. sg .ranta rannalla 'shore' nam. sg./adess. sg .kenkd kengast& 'shoe' nam. sg./elat. sg .kulta kullaksi 'gold' nam. sg./transl. sg .kerta kerran 'time' nan. sg./gen. sg .There are certain exceptions to the above list and a number of ef-fects which interact with the deletion of k. Gradation fails when t and kare preceded by non-sonorantsl:(2) posti	 postin	 *posdin	 'post' nam. sg./gen. sg.rasti	 rastin	 *rasdin	 'checkmark'	 ,,poski	 poskin	 *posin	 'cheek'	 .,matka	 matkan	 *matan	 'journey'	 ,,karahka karahkan *karahan	 'stick'but:	 Lahti	 landen	 *lahten	 'gulf'	 ItTurku	 Turun	 *Turkun	 city name	 IIjalka	 jalan	 *jalkan	 'foot'	 IInahka	 nahan	 *nahkan	 'skin'	 ,,The failure of t to voice after s is not surprising--proximity to thevoiceless sibilant is sufficient reason for voicing not to occur. Thefailure of k to be deleted after s is another matter. Deletion of theIn my earlier discussion of Consonant Gradation I presented the basicfacts, repeated here:(1)pp ---> ptt ---> tkk ---> kp ---> vt ---> dk ---> 0mp ---> mmnt ---> nnnk ---> qgIt ---> 11rt ---> rr76segment would leave an empty skeletal position (the glide-formation evi-dence below substantiates this). Empty positions must be properly govern-ed, that is, they must be governed by a phonetically expressed segment andno other government relation may overlap with this first relation. In thestructure:(3)	X X X X X X11	 1	 1	 11	 1	 1	 1	 1	 1poskin1	 11	 1	 1I1_ 1the k is governed by the i, but is not properly governed as the k is alsoin a government relation with the s. Deletion of the k would also leavethe s to be governed by an empty position. The k cannot, therefore, bedeleted.In the hC cases, the t is allowed to voice--proximity to the glottalfricative is not a sufficient deterrant to the spreading of voice to the tfrom the following vowel. Again, the case with h is not so tidy. In themajority of instances, k will not be deleted after h. Again this would bebecause the k governs the h and is itself not properly governed. Thisleads one to suppose that in the few examples like nahka/nahan 'skin' thek does not govern the h. Such non-government is possible if the h is (aswere the sonorants in e.g. helppo 'easy') linked to the same skeletal77position as the vowel. The representations of karahka 'stick' and nahka'skin' would be:(4)	 a.ORORI	 '\0IIRIIb.	 0IR	 0I	 IRN N I N NI I I 1 \ I I I I	 I II I I I \ I I I I	 I IX X X X X X X X X	 X XiIIIII II II II II I	 / 1\	 I 1Ik a r a h k a n a h k aIn (4a) the k governs the h and so is not itself properly governed andcannot be deleted. In (4b) the h and k are not in a government relation(Onsets cannot govern segments within the Nucleus--KLV 1989, 53) and sothe k is properly governed by the following vowel and can be deleted.This analysis is strained by examples such as jalka jalan 'foot'nam. sg ./gen. sg .. In order to maintain that the deleted k is not in agovernment relation with the preceding sonorant and is therefore properlygoverned, I must posit the structure:(5)	0 R 0 R	 0 R 0 RI	 II	 I	 :	 I\	 :	 I	 :	 I\:	 N	 : N \	 1	 N	 1 N \I	 : 	 \	I	 \X X XX X 	 > X X XX XI	 / \	 :	 :	 :	 :	 / \	 I	 1I	 1j a 1 k a n	 j a 1	 a nI have not found any instances of a root with the form CV[son]kV inwhich the k is not deleted in the weak grade. Accordingly, every root ofthis form must involve a monamoraic vowel-sonorant sequence. This is anunorthodox conclusion, to say the least. However, given the existence ofroots like /helppo/ 'easy' on the basis of which I originally argued formonamoraic vowel-sonorant sequences, it is not unreasonable to expect suchsequences to turn up elsewhere. Indeed, the simplest assumption a lan-guage learner would make is that once forms like helppo established the78existence of monamoraic vowel-sonorant sequences, all such sequences aremonamoraic unless there is evidence to the contrary.6.2 Same Remarks on Glide-FormationThe deletion of k also leads to other interesting effects. Of par-ticular interest is a process I have designated Glide Formation. Incertain circumstances, where k is deleted fran a position preceding a non-low non-round front vowel or a high round vowel, the vowel spreads to fillthe Onset of the syllable.strated in (6):(6)	 strong gradeThe distribution of Glide Formation is demon-weak grade/joke/ joki joen 'river' nan. sg ./gen. sg ./arke/ arki arjen 'weekday'/polkime/ polkimen poljin 'treadle' gen. sg./nam. sg ./jalka/ jalka jalan 'foot' nam. sg ./gen. sg./turku/ Turku Torun city name VT/luku/ luku luvun 'number'/kyky/ kyky kyvyn 'Ability' IV/huoku/ huoku huoun 'breath' VI/ruoko/ ruoko [ruo?on] 'reed' VIThe effects of Glide Formation (e.g. arki -arjen, luku -luvun make itclear that only the segmental structure of the /k/ is deleted and not itsskeletal position, for if the skeletal position of the Onset were deletedalong with the /k/, such a spreading would not likely occur. However, itis not simply a matter of spreading the appropriate vowel whenever apreceding /k/ has been deleted. The I0 element of the vowels /i/ and /e/spreads to form [j] when the deleted /k/ is preceded by a liquid, but notwhen the /k/ is preceded by a vowel. /y/ and /u/ (but not /0) form [v]when the deleted /k/ is surrounded by identical vowels 2 . Round (labial)79and non-round Glide Formation are in complementary distribution--labiali-zation occurs intervocalically but not between a liquid and a vowel; jod-formation occurs between a liquid and a vowel, but not in intervocalicposition.The derivation of Glide Formation proceeds as follows:(7) a. derivation of arjen 'weekday' gen. sg .0 R 0 RI	 I	 :\N	 N \X XX X/ \	 1	 :	 1a r k I0 n:*0 :gradation of /k/	O R 0 R	 spread of I° (colons:	 I I\	indicate that the elementsN	 N \	 below do not spread)I	 I	 I	 \X X X X/ \	 \a r I0 n  	 arjenA+I+80(7) b.	 derivation of luvun 'number' gen. sg .OROR\N	 N \I1111\XXXXX1111	 11111	 11 u k U° n0O R OR1	 :\1 N 1 N \ \1111xxXXgradation of /k/spread of Uo\:1 u	 U0 n 	 >	 luvunThe above derivations demonstrate the operation of the glide-formation processes. I shall leave the question of distribution of theprocesses--why they occur where they do or fail in other circumstances forfurther research.81Notes to Chapter Six1. The lack of native Cp clusters involving non-sonorants makes itimpossible to examine the gradation pattern of forms such as the hypo-thetical *kaspa or *kahpa. When preceded by a non-assimilated sonorant, palternates with v: kalpa -kalvan 'sword' nam. sg ./gen. sg ., arpi -arven'scar'.2. According to Collinder (1957, 6) the sound represented by <v> whereit has been derived by Glide Formation is [A] rather than the [v] foundelsewhere in Finnish. This sound is derived phonetically by the spread ofthe labial element UP into the position left empty by the deletion of k.Certainly, it is not the result of sate freakish velar/labial alternation.82CONCLUSIONThe purpose of this thesis has been to investigate the adequacy ofthe theory of Phonological Government with respect to Finnish. In themain, this performance has been good. The theory predicted with reason-able accuracy the kinds of heterosyllabic clusters one finds in Finnish.The most notable exception, the lack of clusters ending in p, represents agenuinely exceptional feature of Finnish and falls outside the context ofthe Principles and Parameters approach of the theory.The theory also provided an account of the interaction e-Deletion andt-Deletion, being able to treat the variations in root-final as as anartifact of structure rather than as a rule of grammar.The theory provided the beginnings of a full account of FinnishConsonant Gradation. The virtue of the theory of Government in thisregard is the description of the context of Consonant Gradation in anactive manner (an Onset Governed by a Branching Rime) rather than in apassive manner (the onset of a closed syllable). The theory's insistencethat there are no syllables with the structure CVVC also offered an ele-gant account of how these superficially closed syllables fail to triggerGradation. Although there is much work yet to be done with regard toConsonant Gradation (particularly in the verbal system, which I have notexamined), it is clear that Government is a valuable notion in thisregard.The theory of Phonological Government is still in its infancy. As itbecomes more fully developed and articulated--particularly in terms of howconstituent structure is erected--and as it is brought to bear on agreater number of languages and linguistic problems we shall see how wellit fares against rival theories. This single case study cannot answer any83of the big questions--a large cross-linguistic survey would be needed forthat. Even so, in the ongoing battle of "Generative Phonology vs. FinnishPhonology", Government theory emerges as an able contender.84REFERENCESAnderson, John, Colin Ewen and Jorgen Staun (1985). "Phonological Struc-ture: Segmental, Suprasegmental and Extrasegmental." in: Colin J.Ewen and John M. Anderson (eds.) Phonology Yearbook 2. Cambridge:Cambridge University Press. 201-222.Campbell, Lyle (1981). "Generative Phonology vs. Finnish Phonology:Retrospect and Prospect". in: D. L. Goyvaerts, ed. Phonology  in the1980's. Ghent: E. Story-Scientia. 147-182.Chamsky, Noam and Morris Halle (1968). The Sound Pattern of English.New York: Harper and Row.Collinder, Bjorn (1957). Survey of the Uralic Languages. Uppsala: Almqvistand Wiksell.Foley, James (1977). Foundatio  of Theoretical Phonology. Cambridge:Cambridge University Press.Ito, Junko (1986). Syllable Theory in Prosodic Phonology. UMAss.PhD Dissertation.Hayes, Bruce (1980). A Metrical Theory of Stress Rules. MIT PhDDissertation.	  (1986). "Inalterability in CV Phonology." Lanauaae 62. 321-351.	 (1988). "Compensatory Lengthening in Moraic Phonology". MS.Karlsson, Fred (1983). Finnish Grammar. trs. Andrew Chesterman.Juva: Werner SOderstrOm Osakeyhtiti.Kaye, Jonathan (1987). "Government in Phonology: The Case of MoroccanArabic". MS. 24 pp.Kaye, Jonathan, Jean Lowenstamm and Jean-Roger Vergnaud (1985). "TheInternal Structure of Phonological Elements: a Theory of Charm andGovernment." in: Colin J. Ewen and John M. Anderson (eds.) PhonologyYearbook 2. Cambridge: Cambridge University Press. 305-328.	 . (1987). "Constituent Structure and Government inPhonology".	 MS. 49 pp.	. (1989). "Konstituentenstruktur and Rektion in derPhonologie". Linauistiche Berichte, Sonderheft 2/1989: Phonologie.31-74. (Published version of the preceding).Keyser, S. J. and Paul Kiparsky (1984). "Syllable Structure in FinnishPhonology." in: Mark Aronoff and Richard T. Oehrle (eds.) Lanauaae Sound Structure. Cambridge, MA: MIT Press. 7-31.85Lehiste, use (1965). "The Function of Quantity in Finnish and Estonian."Language 41: 447-483.Lehtinen, Meri (1962). Basic Course in Finnish. American Council ofLearned	 Societies. Research and Studies in Uralic and AltaicLanguages,	 Project No. 57. New York: Humanities Press Inc.Lieber, Rochelle (1987). A Integrated Theory of Autosegmental Processes.Albany: SUNY Press.McCarthy, John J. and Alan S. Prince (1986). Prosodic Morphology.MS. 108 pp.Prince, Alan S. (1984). "Phonology with Tiers." in: Mark Aronoff andRichard T. Oehrle (eds.) Language Sound Structure. Cambridge MA: MITPress. 234-244.Schane, Sanford (1987). "The Resolution of Hiatus." in: Anna Bosch,Barbara Need and Eric Schiller (eds.) Papers from the 23rd Annual Regional Meeting ot the Chicago Linguistic Society--Part Two:Parasession on Autosegmental and Metrical Phonology. Chicago: CLS.279-290.Selkirk, Elisabeth (1984). "On the Major Class Features and SyllableTheory." in: Mark Aronoff and Richard T. Oehrle (eds.) Language SoundStructure. Cambridge MA: MIT Press. 107-136.Shuken, Cynthia (1984). "[?], [h], and Parametric Phonetics." in: Jo-AnnW. Higgs and Robin Thelwall, eds., Topics in Linguistic Phonetics inHonour ofEms. T. Uldall, Ulster: The New University of Ulster 111-139.Suomi, Kari (1985). "On Detecting Words and Word Boundaries in Finnish: ASurvey of Potential Word Boundary Signals". Nordic Journal ofLinguistics 8. 211-231.Vennemann, Theo (1988). Preference Laws , for Syllable Structure and theExplanation of Sound Ltwitsm. New York: Mouton de Gruyter.Whitney, Arthur H. (1956). Finnish. Teach Yourself Books, Bungay: Hodderand Stoughton.86


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items