YAQUI COORDINATION by Constantino Martínez Fabián

YAQUI COORDINATION by Constantino Martínez Fabián
YAQUI COORDINATION
by
Constantino Martínez Fabián
______________________________
Copyright © Constantino Martínez Fabián
A Dissertation submitted to the Faculty of
THE DEPARTMENT OF LINGUISTICS
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
2006
2
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
As members of the Dissertation Committee, we certify that we have read the
dissertation prepared by Constantino Martínez Fabián entitled Yaqui Coordination and
recommend that it be accepted as fulfilling the dissertation requirement for the Degree of
Doctor of Philosophy.
______________________________________________________________________
Date: December 8, 2005
D. Terence Langendoen
______________________________________________________________________
Date: December 8, 2005
Heidi Harley
______________________________________________________________________
Date: December 8, 2005
Andrew Carnie
______________________________________________________________________
Date: December 8, 2005
Simin Karimi
______________________________________________________________________
Date: December 8, 2005
Sheila Dooley
Final approval and acceptance of this dissertation is contingent upon the candidate’s
submission of the final copies of the dissertation to the Graduate College.
I hereby certify that I have read this dissertation prepared under my direction and
recommend that it be accepted as fulfilling the dissertation requirement.
________________________________________________ Date: December 8, 2005
Dissertation Director: D. Terence Langendoen
3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of requirements for an
advanced degree at The University of Arizona and is deposited in the University Library
to be made available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission,
provided that accurate acknowledgment of source is made. Request for permission for
extended quotation from or reproduction of this manuscript in whole or in part may be
granted by the copyright holder.
Constantino Martínez Fabián
4
ACKNOWLEDGEMENTS
I would not have been able to complete this journey without the aid and support of
countless people. I must first express my gratitude towards the members of my
committee: Heidi Harley, Andrew Carnie, Simin Karimi, Sheila Dooley, and especially to
Terry Langendoen who served as the dissertation director and advisor during my time as
student. Under his tutelage I developed a focus and became interested in Yaqui
coordination. He provided me with direction, technical support and became more of a
mentor and friend, than a professor. I doubt that I would ever have been able to finish this
project without his support. I owe him my eternal gratitude.
I am grateful also to Crescencio Buitimea Valenzuela and Melquiades Bejipone Cruz
who helped me by sharing their native knowledge of the Yaqui language, and to
Rosemary Emery whose administrative efficiency allow me to survive like a graduate
student.
I am deeply indebted to the University of Arizona Linguistics Department faculty
members Dick Oehrle, Diana Archangeli, Mike Hammond, Dick Demers, Andy Barss
and Susan Steele, whose expertise, understanding, and patience added considerably to my
graduate experience.
I must also acknowledge those in my linguistics graduate-student generation whose
motivation and encouragement were invaluable over the years: Keichiiro Suzuki, Amy
Fountain, Tom Craig and Ana Lidia Munguía Duarte. They each helped make my time in
the PhD program more fun and interesting.
5
Thanks to people at the Universidad de Sonora, especially to Mirna Castro Llamas for
her unconditional support of my linguistic project and to Fermín González Gaxiola,
Francisco Zaragoza Ortega and Carmen Velarde for their support and friendship.
Special mention is required for Jason Haugen whose friendship and patience in
reading and correcting the writing style improved the final version of this work. Of
course, any remaining mistakes are mine.
Finally, I thank María del Refugio Romero Telles, my wife, for her continuous
support and my children for their believing in me. Thanks to the people who contributed,
directly or indirectly, to this project and who were not mentioned here.
6
DEDICATION
To my mother Felicitas Fabián García, and (in memoriam) to my father Fructuoso
Martínez Benítez.
To the special children in my world: Rubén, Andrea, Fabián, Paulina, Alejandra,
Paola.
To my brothers Ciro, Guadalupe, Felícitas, Francisco, Edilburga, J. Carmen.
In short: to my family.
7
TABLE OF CONTENTS
LIST OF ABBREVIATIONS............................................................................................11
ABSTRACT.......................................................................................................................12
1
INTRODUCTION....................................................................................................13
1.1
Presentation .........................................................................................................16
1.2
Empirical goals....................................................................................................20
1.3
Theoretical goals .................................................................................................20
1.4
Background information of the Yaqui language .................................................22
1.4.1 Yaqui word order............................................................................................23
1.4.2 Introduction to Yaqui Coordination................................................................24
1.5
Optimality Theory ...............................................................................................33
1.5.1 OT basics ........................................................................................................34
1.5.2 The different OT approaches ..........................................................................36
1.5.3 Some OT constraints.......................................................................................38
1.5.4 A Syntactic Model ..........................................................................................40
1.5.5 The nature of the input for coordination.........................................................40
1.6
Yaqui phrase structure flexibility in OT..............................................................45
1.7
What is a coordinator?.........................................................................................47
1.8
Summary..............................................................................................................49
2
LITERATURE REVIEW.........................................................................................51
2.1
Disagreement in the Literature ............................................................................51
2.1.1 An HPSG approach (Abeillé 2003) ................................................................52
8
TABLE OF CONTENTS-Continued
2.1.2 A Minimalist approach (Camacho 2003) .......................................................57
2.1.3 An OT approach (Gáspár 1999) .....................................................................65
2.1.4 A LFG approach (Peterson 2004)...................................................................69
2.1.5 An Autolexical Approach (Yuasa and Sadock 2002).....................................74
2.1.6 A P&P approach (Borsley (2005)) .................................................................79
2.1.7 Summary.........................................................................................................82
3
THE STRUCTURE OF COORDINATION ............................................................85
3.1
Sentence Coordination.........................................................................................85
3.1.1 Distribution of into ‘and’ ................................................................................85
3.1.2 Other uses of the particle into.........................................................................97
3.1.3 Other particles that indicate ‘and’ coordination .............................................98
3.1.4 Setting the problem.........................................................................................99
3.2
Proposal about the structure of coordination.....................................................104
3.2.1 Background...................................................................................................104
3.2.2 Alternatives for the structure of coordination...............................................107
3.2.3 The coordinator into ‘and’ is not a head.......................................................115
3.2.4 The structure of coordination: A proposal....................................................123
3.3
Analysis in OT...................................................................................................136
3.3.1 Into in second position..................................................................................136
3.3.2 Into in first position ......................................................................................142
3.3.3 Analysis of two coordinators ........................................................................143
9
TABLE OF CONTENTS-Continued
3.3.4 Into in last position. ......................................................................................146
3.4
4
Summary of Chapter 3.......................................................................................152
OBC AND UBC IN YAQUI..................................................................................154
4.1
Verbal coordination ...........................................................................................154
4.1.1 Verbal balanced coordination .......................................................................155
4.1.2 OT Constraints for explaining Balancedness ...............................................157
4.1.3 The Coordinate Structure Constraint (CSC).................................................159
4.1.4 Verbal unbalanced coordination ...................................................................163
4.1.5 Reflection about pseudo-coordination, -subordination, and coordination....201
4.2
5
Conclusion.........................................................................................................215
NOMINAL COORDINATION .............................................................................216
5.1
Background on Nominal and verbal classes......................................................216
5.1.1 Number in nouns and in verbs ......................................................................216
5.1.2 Interactions between nouns and verbs ..........................................................222
5.2
Noun coordination and verbal agreement..........................................................225
5.2.1 Noun coordination and intransitive suppletive verbs ...................................225
5.2.2 Summary.......................................................................................................228
5.2.3 Analysis ........................................................................................................229
5.2.4 Analysis of the interaction between coordinate nouns and verbs.................229
5.2.5 Noun coordination and transitive suppletive verbs ......................................236
5.2.6 Interaction between pronouns and coordination...........................................239
10
TABLE OF CONTENTS-Continued
5.2.7 Summary.......................................................................................................240
5.3
Analysis of transitive verbs ...............................................................................242
5.3.1 The problems ................................................................................................244
5.3.2 Solving the problem in OT terms .................................................................247
5.4
NP conjunction and separateness of the events.................................................260
5.4.1 Observations about the Relative order of conjoined NPs.............................260
5.4.2 Summary.......................................................................................................263
5.4.3 OT analysis of pragmatic constraints ...........................................................263
5.5
Noun coordination and case marking ................................................................265
5.6
Summary of chapter 5 .......................................................................................269
6
CONCLUSIONS AND TOPICS FOR FUTURE RESEARCH ............................271
6.1
Conclusions .......................................................................................................271
6.2
Topics for future research..................................................................................275
REFERENCES ................................................................................................................279
11
LIST OF ABBREVIATIONS
1
first person
2
second person
3
third person
ACC
accusative
ADJ
adjective
ADV
adverb(ial)
AGR
agreement
APPL applicative
BENEF benefactive
CAUS causative
CONT continuative
COM commitative
COMP complementizer
COMPL completive
COND conditional
COORD coordination
COUNT counterfactual
DAT
dative
DECL declarative
DEM
demonstrative
DES
desiderative
DET
determiner
DIR
directional
DISTR distributive
FOC
focus
FUT
future
GEN
genitive
GER
gerundive
HAB
habitual
HO
human object
IMP
imperative
INC
inceptive
IND
Indicative
INSTR instrument
INT
interrogative
INTR
INTT
LOC
M
NNEG
NMLZ
NOM
OBL
OBJ
PASS
PST
PL
POSS
PPL
PRS
PROG
PROM
PROP
PST
PTCP
RECP
RED
REFL
REL
S
SBJ
SBJV
SG
SUB
TNS
TEMP
TERM
TOP
TR
intransitive
Intentive
locative
masculine
non- (e.g. NNOM non-nominative)
negation, negative
nominalizer/nominalization
nominative
oblique
object
passive
past
Plural
possessive
participle
present
progressive
prominent
propositional
past
participle
reciprocal
reduplication
reflexive
relative
sentence
subject
subjunctive
singular
subordinator
tense
temporal
terminative
topic
transitive
12
ABSTRACT
This research describes and explains in the OT framework the Yaqui coordination. It
is assumed that coordinate structures are asymmetric and, based in the Yaqui data, I
propose that the coordination is the result of an adjunct-host relation. This work shows
that the ConjP is inappropriate for explaining the place that the Yaqui coordinator into
‘and’ occupies in overt syntax. It demonstrates that the proposal which suggests that
coordinators in second position are clitics (Agbayani and Goldston 2002) can not be
maintained in Yaqui because such position is generated by fronting a topicalized
constituent. If we depart from the idea that clitics and topics move to different positions,
then a different explanation is required.
The proposal is extended to the analysis of unbalanced verbal chaining structures. It is
shown that some --kai constructions are marked syntactically as subordinated but actually
they are coordinate structures. In the final part of this work I describe and analyze the
agreement between coordinate nominals and verbs. The analysis indicates that Yaqui
responds partially to the system of CONCORD and INDEX features proposed by Halloway
King and Dalrymple (2004). However, its whole explanation requires the use of
constraints in order to explain the coordinate patterns of the language.
13
1
INTRODUCTION
This investigation is twofold: first, it describes the Yaqui coordination patterns.
Second, within the framework of Optimality Theory (OT) it explains some of the most
salient characteristics of this phenomenon: the structure of coordination, chaining
structures, patterns of agreement and coordination of maximal projections. I have
selected those topics because a literature review indicates that in spite of the fact that
coordination has been a motivation for research for a long time, there is still considerable
debate on these issues.
With respect to the structure of coordination, some researchers consider that it is a
headed construction. In other words, they consider that coordinate structures are
Conjunction Phrases where the coordinator is the head, the first conjunct the specifier,
and the second conjunct the complement. This conception is accepted by researchers like
Rebuschi, (2005), Abeillé (2003), Camacho (2003), Gáspár (1999), Johannessen (1998),
among others. As pointed out by Borsley (2005) this conception is widely accepted
within Principles and Parameters (P&P) theories, but it is rejected within other
frameworks. So, Borsley (2005) himself rejects the idea that coordinate structures are
Conjunction Phrases. This different conception of coordination is held by such
researchers as Cormack and Smith (2005), Peterson (2004), Yuasa and Sadock (2002),
Dalrymple and Kaplan (2000), and Bresnan (2000), among others. Given, in general,
those two alternative positions and based in the Yaqui data, I propose that coordination is
produced by adjunction structures as in (1):
14
(1)
CP[coord]
CP
CP[coord]
and
S1
CP
S2
The representation shows that a coordinator is an adjunct which attaches to a
maximal projection and introduces a feature [COORD] which licenses the further
adjunction of another maximal projection (the first conjunct). The proposal emerges from
the analysis of the coordination in the Yaqui language. The proposal is presented in
Chapter Three of this work, and it is done in the sense of Langendoen (2003). I consider
that the coordinator into ‘and’ is neither a head (Johannesen (998), Camacho (2003),
among others) nor a clitic (Agbayani & Goldston, (2002)).
The idea that coordination involves adjunction is held, for example, by Cormack and
Smith (2005). These researchers claim that the grammar is only capable of providing
asymmetric structures and that there are not devices in the grammar specific to
coordination. Therefore, the grammar will only provide adjunct-host structures and headcomplement structures. They give arguments in favor of an adjunct-host approach
combining a simplified version of Minimalism, with the addition of Combinators from
Combinatorial Grammar. In this work and within an OT approach, I suggest that
coordination must be restricted to adjunction structures too. This proposal predicts that if
coordination is adjunction and subordination is adjunction as well, then we would expect
15
some cases where it would be hard to tell if we have coordination or subordination1. This
is what we have when we consider such notions as pseudo-coordination and pseudosubordination. These concepts are explored in Chapter Four. As a hypothesis not
developed here, we can say that the double life of coordinators which sometimes behave
as subordinators is due to the fact that adjunction is taking place in both coordinate and
subordinate structures. If so, then the constraints involved will make the difference.
My analysis of Yaqui coordination is presented within the framework of Optimality
Theory (OT) (which essentially began with Prince and Smolensky 1993, McCarthy and
Prince 1993). This theory suggests that there are a set of universal, violable and rankable
constraints which explain the nature of linguistic data.
OT is a versatile framework which gives us a formal apparatus to handle and account
for variability of several types; in this case, the several positions that a coordinator like
into ‘and’ can occupy in sentence coordination. Any theory with strict rules cannot
accommodate syntactic variation without resource to hedges in the principles, as
demonstrated by Speas (1997). However, using violable constraints, the Yaqui
coordination patterns are easily explained within OT.
The work does not appeal to diachronic or comparative data; however, it is valuable
because it gives us a description of coordinated structures of Yaqui that were not
described before. In that sense, we have a set of data as those which a Yaqui learner is
1
To distinguish between these concepts is really an issue that requires further research. For example,
Asher and Vieu (2005), within the Segmented Discourse Representation Theory (SDRT), provide some
linguistic tests to clarify which relations are subordinated and which are coordinated at discourse level. On
the other hand, Verstraete (2005), within a constructional approach, uses the notion of illocutionary force to
distinguish coordinate constructions from subordinate ones.
16
faced with. Theoretically, the analysis shows the interaction between several modules of
the grammar which traditionally are considered to be separated: Phonology, Morphology,
Syntax, Semantics and Pragmatics. So, the reader will find in the tableaux, for example,
the interaction of syntactic and pragmatic constraints.
1.1
Presentation
Although Yaqui has been studied by many researchers (Lindenfeld (1973), Escalante
(1990), Dedrick and Casad (1999), among others) there are many areas which have not
been explored in detail, and one of them is coordination. This work describes and
analyzes several Yaqui coordination patterns. This research focuses on the description
and account of several Yaqui coordination patterns using the Optimality Theory (OT).
The work focuses in three main aspects: first, the structure of coordination; second,
coordinated chaining structures (unbalanced coordination); and third, problematic
agreement patterns of the language. Subsequent chapters present the data in that order.
The kind of data that the reader will find is exemplified below:
The structure of Coordination. Most proposals about the structure of coordination
are challenged by Yaqui sentence coordination. In this construction, the coordinator can
appear in three basic positions: first, second and last. The positions are defined (in the
following examples) in relation to the second conjunct: first position means at the
beginning of the second conjunct, second position means after some element of the
second conjunct, and last position means at last in the second conjunct or at last in a
single sentence. They are exemplified as follows.
17
First position:
(2)
[Joan bwika]
into [Peo into
[John sing.PRS]
and
[Peter and
‘John sings and Peter and Mary dance.’
Maria ye’e].
Maria dance.PRS]
In (2) the coordinator into ‘and’ both follows the first conjunct in brackets and it
precedes the second conjunct in brackets too. This is the way in that languages like
English and Spanish coordinate. These types of sentences are easily accommodated in
any account that takes the relation head-complement as central in the explanation of
coordinate structures: the first conjunct is the specifier, the coordinator is the head and
the second conjunct the complement (Johannessen (1998), Camacho (2003), a.o.). Now,
let’s see example (3):
Second position:
(3)
[Diana a=tu’ure-k]
[Peo into a
[Diana 3NNOM.SG=like-PST] [Peter and 3NNOM.SG
‘Diana liked it and Peter bought it.’
jinu-k].
buy-PST]
The sentence (3) contains the subject Peo ‘Peter’ of the second conjunct before the
coordinator into ‘and’, and for that reason we can say that it is in second position.
Therefore, the proposal that the first conjunct is in specifier position is not easy to
accommodate. Agbayani and Goldston (2002) suggest that languages with coordinators
in such position move the first element from the second conjunct and adjoin it to the
coordinator. That movement is triggered by its status as a clitic: it is assumed that those
coordinators are prosodically deficient and need to have a host. In chapter three I show
that into ‘and’ is not a clitic and that movement is triggered by topicality.
The following example shows the third possibility where into(ko) ‘and’ can appear in
open syntax:
18
Last position: (Crumrine 1961:22)
(4)
[ju’u o’ou kia
a-u=
‘omtemta
benasi]
[DET man just
3NNOM.SG-DIR=
angry
like]
[amau a’a=to’o
simlataka], [káa a-u= bitchu
intoko].
[back 3NNOM.SG=leave
went]
[not 3NNOM.SG-DIR=look and just]
‘The man looks as though he is angry with her, so he is leaving her behind and
does not even look at her.’
As example (4) indicates, into(ko) appears after the second conjunct. Again, the
specifier-head-complement structure is not easy to accommodate.
Coordinated chaining structures. Yaqui has what has been called Unbalanced
Coordination (Johannessen (1998)) or Pseudosubordination (Yuasa and Sadock (2002)).
From a typological perspective, Givon (2001), Yaqui must be classified as a SOV-type
chaining. The most salient syntactic feature of this type of clause chaining is the
assignment of most finite grammatical marking only to the final clause. However, the
entire chaining gets the tense indicated by the final clause. The next example shows three
clauses: the first two are marked with the suffix --kai which is a subordinator and the last
one is marked with --k which indicates past tense. However, all the clauses are
understood as past tense. The coordinator into ‘and’ can only optionally appear between
the last --kai clause and the tensed one, as indicated in (5).
(5)
[ili jamut yepsa-kai], [jichikia-ta
nu’u-kai], [jichik-taite-k].
[small woman arrive-SUB], [broom-NNOM.SG take-SUB], [sweep-INC-PST]
‘The young woman arrived, she took the broom (and) she began to sweep.’
This kind of data is treated in Chapter Four. We will see that these structures are
syntactically subordinated but semantically coordinated. I describe and analyze within the
OT framework these chaining structures.
19
Problematic agreement patterns. In Yaqui there are some verbs which agree with the
object. Under coordination when a verb which requires a singular object takes two
coordinated singular nouns, the plural verb can not be used in that case. However, with
intransitive verbs a coordinate subject must agree with a plural verb. This asymmetry is
analyzed in Chapter Five after a previous description of nominal and verbal classes in the
Yaqui language. The following contrast shows that the singular verb mea-k ‘to
kill.SG.OBJ-PST’ is used with one singular object (ex. (6) vs. (7)), or with the coordination
of two (or more) singular nouns (ex. (8) vs (9)).
(6)
Alejandra
maso-ta
Alejandra
deer-NNOM.SG
‘Alejandra killed a deer.’
mea-k.
kill.SG.OBJ-PST
(7)
*Alejandra maso-ta
Alejandra
deer-NNOM.SG
(‘Alejandra killed a deer.’)
sua-k.
kill.PL.OBJ-PST
(8)
Alejandra
[maso-ta
into
Alejandra
[deer-NNOM.SG
and
‘Alejandra killed a deer and a pig.’
kowi-ta]
mea-k.
[pig-NNOM.SG] kill.SG.OBJ-PST
(9)
*Alejandra [maso-ta
into
and
Alejandra
[deer-NNOM.SG
(‘Alejandra killed a deer and a pig.’)
kowi-ta]
sua-k.
pig-NNOM.SG] killed.PL.OBJ-PST
It is shown that Halloway King & Dalrymple’s (2004) system, which uses two types
of number features (CONCORD and INDEX features), cannot explain some of the agreement
patterns found in Yaqui. For that reason, the analysis in this work uses a set of constraints
which explain the alternations on agreement found in Yaqui.
20
1.2
Empirical goals
The main empirical goal of this work is to analyze and describe the relatively
unknown patterns of Yaqui coordination. As almost usual in every language and in every
topic that linguists explore, Yaqui presents very particular patterns of coordination that a
good theory of language should be able to predict and explain. As we can see through this
research, there are some challenging patterns that do not fit into traditional accounts. In
order to achieve this goal, I investigate several types of constructions: sentence
coordination, verbal chaining structures and agreement between nouns and verbs. In
short, the empirical goal of this research is to describe the most salient coordination
patterns of the language.
1.3
Theoretical goals
The aim of this work is to analyze Yaqui coordination within the framework of
Optimality Theory (OT). This theory of grammar has been (mostly) used to explain
phonological and morphological properties of languages, but not much work has been
devoted to the explanation of their syntactic properties. So, this dissertation intends to be
a contribution to the OT literature on syntax. The patterns of Yaqui coordination have
neither been described nor accounted for. The only work which describes some aspects
about coordinated structures is that of Dedrick and Casad (1999), but many facts have
been left untouched. Therefore, it is useful to look at and explain them. In order to do the
analysis, I use several constraints well-motivated in the literature, such as alignment
constraints, markedness constraints, and faithfulness constraints.
21
The theoretical contribution of this work relates to two aspects: it shows how OT can
be applied to syntax, an area where many scholars refuse to accept it, and where the idea
that there are a set of universal, violable and rankable constraints introduces enough
flexibility into the model that phenomena that are highly problematic in derivational
linguistic models are accounted for.
This work gives evidence that the Yaqui coordinator into(ko) ‘and’ cannot be
considered as a clitic (as suggested by Agbayani and Goldston (2002) for other
languages). It is demonstrated that the coordinator occupies several positions in sentence
coordination because it shares properties with adverbials in the language and, like those
elements, it has to be considered an adjunct. This conception is opposed to the idea that
coordinators are heads which project their own projection, with a specifier and a
complement, as suggested by researchers like Johannessen (1998, 2005), Camacho
(2003), Aoun Benmamoun and Sportiche (1994), among others.
It is suggested that a coordinated phrase (nominal in the following example) has the
following structure.
(10)
NP[coord]
NP
oranges and
NP [coord]
NP
apples
In the above structure the coordinator is adjoined to a phrase. This process of
adjunction leaves open the possibility of a new adjunction process, where another NP is
22
adjoined to the first one resulting in a coordinated structure. In this sense, I follow
Langendoen’s proposal (2003) in which to coordinate is to adjoin a coordinator.
This work intends to prove that that chaining structures of Yaqui are coordinate and
that pseudo coordination, pseudo subordination and coordination must be integrated in
the explanation of a theory of coordination. It is suggested that the OT approach can be
useful in the explanation of these phenomena because the constraints are rankable. The
Coordinate Structure Constraint proposed by Ross (1967) is taken, in OT terms, as a
violable constraint: Do not extract from a coordinate structure. So we do not need to use
the hedge of the Across the Board Extraction principle (ATB principle), which allows
extraction in some specific cases.
The last part of this research focuses on the analysis and explanation of some patterns
of agreement between nouns and verbs. I propose that the system used by Halloway King
and Dalrymple (2004) is unable to explain some facts about Yaqui agreement, but we can
recast some of their insights into OT constraints in order to explain the Yaqui data.
Finally, the empirical and theoretical goals of this research are valuable because there
was not an accurate description of the coordination patterns in the language and because
these patterns require an adequate theoretical account which the head-complement
conception of coordination is unable to give.
1.4
Background information of the Yaqui language
This section gives the reader background information about some of the
characteristics of Yaqui, such as word order and a brief description of Yaqui
coordinators.
23
1.4.1
Yaqui word order
Yaqui is an SOV language, and it does not tend to have a lot of variation on that
order. However, variation exists and it is possible to find general patterns of it. For
example, the object can move to final position of the sentence, leaving behind a
coreferential pronoun: S CL=V O (where CL= must be understood as a clitic object
pronoun).
(11)
Ruben ejkuela-po
ji’osia-m
to’o-siika.
leave-go.PST
Ruben school-LOC book-PL
‘Ruben left the books in the school and left.’
(12)
Ruben ejkuela-po
am=
to’o-siika
jume ji’osia-m.
Ruben school-LOC 3NNOM.PL= leave-go.PST DET.PL book-PL
‘Ruben left the books in the school and left.’
Adjuncts can appear before or after the verb, as for example in the following
comitative phrase:
(13)
inepo joan-ta-mak
teo-po
1SG John-NNOM.SG-COM church-LOC
‘I sang in the church with John.’
(14)
inepo teo-po
bwiika-k
1SG church-LOC sing-PST
‘I sang in the church with John.’
bwiika-k.
sing-PST
joan-ta-mak.
John-NNOM.SG-COM
Similar variation can be found in relative constructions: the relative clause may be
close to its head (the example (15) shows a post-nominal relative) or extraposed to final
position (ex. (16)).
Post-nominal relative:
(15)
Simon [uka
jamu-ta
[a=bép-su-ka-u]]
waata.
Simon DET.NNOM.SG woman-NNOM.SG NNOM.SG=hit-COMPL-PST-REL love.PRS
‘Simon loves the woman that hit him/that he hit.’
24
Extraposed relative:
(16)
Simon [uka
jamu-ta]
waata
Simon DET.NNOM.SG woman-NNOM.SG love.PRS
‘Simon loves the woman that hit him/that he hit.’
[a=bépsuka-u].
3SG.OBJ=hit-REL
As the above examples indicate, Yaqui does not always follow its canonical order
within the clause; there is some variation. These types of variation find natural accounts
in the OT model with different weights given to interacting factors from different
structures in the grammar.
1.4.2
Introduction to Yaqui Coordination
This section is a background on Yaqui coordination, it establishes the basic concepts
used in this work. It exemplifies the logical coordinators of Yaqui and presents some of
the most relevant characteristics.
1.4.2.1 Basic concepts
In this section I introduce some terms used in the description of Yaqui coordination.
Let us begin with the following terms found in Haspelmath (2004): “A coordinating
construction consists of two or more coordinands, i.e. coordinated phrases. Their
coordinate status may be indicated by coordinators, i.e. particles like and, and but. If one
or more coordinators occur in a coordinating construction, it is called syndetic.
Asyndetic coordination consists of simple juxtaposition of the coordinands” (Haspelmath
2004: 4). In this work the words coordinand and conjunct are used synonymously, as are
coordinator and conjunction.
Both types occur in Yaqui:
25
Syndetic:
(17) María tuuka
[Peo-ta-u
into Pablo-ta-u]
nooka- k.
Mary yesterday [Peter-NNOM.SG-DIR and Pablo-NNOM.SG-DIR] speak-PST
‘Mary spoke to Peter and Pablo yesterday.’
Asyndetic:
(18) [Joan], [Peo], [María], [Carlos], (into) [Tiibu]
si’ime bwiika-k.
[John] [Peter] [Maria] [Carlos]
(and) [Tiburcio]
all
sing-PST
‘John, Peter, María, Carlos, (and) Tiburcio, all of them sang.’
It is usual to distinguish two types of syndesis: monosyndetic coordination, which
involves only a single coordinator, and bisyndetic coordination, which involves two
similar coordinators.
Yaqui only has the first one:
(19)
Wiikit into taawe
ne’e.
bird and
sparrowhawk fly.PRS
‘The bird and the sparrowhawk are flying.’
The second one is illustrated by Kibrik (2004:538), in the Upper Kuskokwim
Athabaskan language:
(20)
[dineje]
‘il
[midzish]
moose
with caribou
‘moose and caribou’
‘il
with
1.4.2.2 Coordinated categories
Yaqui has the coordination of various grammatical categories. This work explores the
coordination with the particle into ‘and’. The categories that can be established are the
following ones:
26
1.4.2.2.1 Coordination of likes
Yaqui has the coordination of the following grammatical categories. As we can see in
the examples, all the examples can be categorized as the coordination of likes2.
DPs
(21) María Peo-ta
[juka
lapis-ta]
Maria Peter-NNOM.SG DET.NNOM.SG
pencil-NNOM.SG
[juka
yokia-ta]
mik-bae.
DET.NNOM.SG pen-NNOM.SG gift-INT
‘María will give a pencil and a pen to Peter.’
into
and
N(P)s
(22) [Kaba’i]
into [buuru]
ousi bwe-bwere-m.
Horse
and
donkeyvery RED-big-PL
‘The horse and the donkey are really big.’
AdjPs
(23) Joan [beme-k]
into [tutuli-k]
John young-NNOM and
beautiful-NNOM
‘John saw the young and beautiful woman.’
(24)
jamut-ta
woman-NNOM.SG
Joan [beme-k]
into [tutuli-k]
John young-NNOM.SG
and
beautiful-NNOM.SG
‘John saw the young (one) and beautiful one.’
AdvPs
(25) Aapo [junak]
into [ketun ian]
3SG then
and
still
‘(S)he was and is today still a teacher.’
bicha-k.
see-PST
bicha-k.
see-PST
maestro.
today teacher
2
Yaqui does not have coordination of single postpositions, which are all bound morphemes. In other
words, it is not possible to have a construction like ‘Mary planted corn behind and in front of her house: So
this English sentence is translated to Yaqui like the following one:
(i) Joan
amau jo’ara-po
bachi-ta
e’echa-k into
John
behind house-LOC
corn-NNOM.SG plant-PST and
‘John planted corn behind the house and in front too.’
bicha-po
in.front-LOC
ketchia.
too
27
PostPs
(26) Joan [torim-po]
into [bicam-po]
John Torim-LOC and
Vicam-LOC
‘John works in Torim and in Vicam.’
V(P)s
(27) Joan [bicha]
into
John see.PRS
and
‘Juan sees and hears.’
tekipanoa
work.PRS
[jikkaja].
hear.PRS
However, two transitive verbs cannot be coordinated as in (27) above. Each verb
requires its own object in overt syntax. The coordinate sentence (28) has two conjuncts
where each verb has in overt syntax its object3; so the sentence (29) is ungrammatical:
(28)
Peo [jita
jinu]
Peter something
buy.PRS
‘Pedro buys and sells something.’
into
and
(29)
*Peo [jita
[jinu
into
Peter something
buy.PRS
and
(‘Pedro buys and sells something.’)
[jita
something
nenka].
sell.PRS
nenka]].
sell.PRS
Finally, we have the coordination of two sentences as illustrated in (30) and (31):
Ss
(30) Joan ji’osiam
maria-ta
maaka-k Peo into
John book
Maria-NNOM.SG
give-PST Peter and
am=
nenka-k.
3NNOM.SG= sell-PST
‘John gave a book to Maria and Peter sold it to her.’
(31)
3
a-u
him-to
U
cu’u [wakas-ta
batte
ke’e-ka]
into
DET
dog cow-NNOM.SG almost
bite-PST
and
[uka
paros-ta
batte
bwiise-k].
DET.NNOM.SG hare-NNOM.SG almost
catch-PST
‘The dog almost bites the cow and it almost caught the hare.’
These kinds of examples ((29) and (32)) in this work are taken as sentential coordination where the
subject of the second conjunct is null. However an alternative analysis is possible where we can postulate
the coordination of two VPs. In chapter four I use the constraint Drop-Topic (Blutner and Zeevat (2004) for
explaining Yaqui verbal chains. So this explanation can be extended to cover examples like ((29) and (32)).
See the analysis of example (108) in Chapter Four.
28
1.4.2.2.2 Lack of coordination of unlikes
Contrary to languages like English and Spanish, it is hard to find coordination of
unlikes in Yaqui. As it is well known, coordination of unlikes is very common in
predicate position, as in the following examples: The children are awake and asking for
you, Paul is stupid and a liar (Peterson 2004:647-648). However, there are several
restrictions to this kind of coordination. It is shown in examples like *John sang
beautifully and a carol (Peterson 2004:647).
The lack of these kinds of structures in Yaqui seems to be related to the fact that the
language does not have a copulative verb and to the fact that adjectives (as well as nouns)
can be used as predicates, and they take the verbal suffixes. So if we try to coordinate
different categories (adjective and noun for example) they are derived into verbs and
emerge as coordination of likes. The following coordination indicates that the conjuncts
get optionally the same ending, showing that we have a coordination of likes.
(32)
Ume usi-m [bu-busala(mme)]
into [enchi
DET.PL boy-PL RED-awake.PRS (PL) and
2PL.OBL
‘The children are awake and asking for you.’
nattemai(mme)].
ask.PRS (PL)
The constraints that avoid coordination of unlikes must be highly ranked in Yaqui. If
we depart from Peterson’s idea that a main requisite for coordination of unlikes is that the
conjuncts must have the same syntactic function, examples like the following indicate
that there must be other constraints playing a role for the ungrammaticality of the
coordination in Yaqui, English and Spanish. We can see in the next examples that the
elements of the intended coordination are adjuncts and that each one can occur in the
same context. However, when we try to coordinate them, the coordination fails.
29
(33)
Nee jo’ara-u
1SG house-DIR
‘I went home.’
siika.
go.PST
(34)
Nee lunes-tu-k
siika.
1SG Monday-VERB-when go.PST
‘I left Monday.’
(35)
*Nee [jo’ara-u
(36)
Nee jo’ara-u
lunes-tu-k
siika.
1SG house-DIR
Monday-VERB-when go.PST
‘I went to the home Monday.’
into
lunes-tu-k]
siika.
A challenge for any theory of coordination is to explain why coordination can put
together different categories in some languages and why it cannot in other languages,
such as Yaqui. Schachter (1977), among others, has observed that the conjuncts must
share the same theta-roles. Givón (2001) mentions that coordination must cover the
constraint equi-case-role, which takes care of both: theta-roles and case (equi-case-role).
This work does not explore further the conditions why coordination of unlikes was not
attested.
1.4.2.3 The logical coordinators of Yaqui
The Yaqui logical coordinators presented in this section are the following: bweta
‘but’, o ‘or’ and into(ko) ‘and’. Between them, only into(ko) ‘and’ occupies several
positions, as exemplified below. Because this work only analyzes the constructions where
this coordinator appears, the exploration of the syntactic characteristics is centered in the
coordinator into ‘and’. (37) illustrates an example of coordination with bweta ‘but’. It
always occurs in middle of coordinate sentences.
30
(37)
[Joan bwite-k]
bweta [Peo e’e].
John run-PST
but
Peter not
‘John ran but Peter did not.’
The following example illustrates the use of the particle o ‘or’, which is a loan from
Spanish. It only can occur too in the middle of coordinated elements:
(38)
Ruben tekipanoa
o
matematika-m
Ruben work.PRS
or
mathematics-PL
‘Ruben works or studies mathematics.’
emo majta.
3REFL teach.PRS
The coordinator into(ko) ‘and’ occurs in several positions: first position, second
position or last position, as was indicated in (2), (3) and (4) previously. Here the example
shows into in second position.
(39)
[Dalia bwika-k]
[Peo into
Dalia sing-PST
Peter and
‘Dalia sang and Peter ate.’
ji’ibwa-k].
eat-PST
Asyntetic coordination is very common in Yaqui. A case of coordination of two
subordinated clauses is shown below. Being asyntetic, the coordinator does not need to
occur between the two bracketed clauses:
(40)
[Joan bwite-ka]
[po’o-po’oti-sime-ka]
RDP-bend.down-go.SG-GER
John run-GER
‘John running (and) bending down won.’
[yo’o-k].
win-PST
In Yaqui it is easy to find examples where two coordinators can co-occur, like a
compound coordinator, specially ta and into, when these coordinators co-occur, the
(bwe)ta ‘but’ goes first and into ‘and’ second, the coordinated sentence acquires an
adversative meaning. A co-occurring coordination is illustrated in what follows.
(41)
[kaa
ta-ta]
ta
into
NEG
RED-hot
but
and
‘It is not hot but it is not cold either.’
[kaa
NEG
seebe juni].
cold either
31
In what follows we are going to see some relevant aspects of the Yaqui coordinators.
1.4.2.4 Observations about Yaqui coordinators
There are some tests for checking if we are faced with logical coordinators.
According to Van Oirsow (1987:109), “one clear characteristic which is particular to
coordinating conjunctions as contrasted with, say subordinating conjunctions is that the
former have to occur in between the clauses which they coordinate and latter need not”.
From this point of view, a Yaqui sentence with bweytuk ‘because’ is a subordinated one:
(42)
[Joan kot-pea]
bweytuk
[aapo kaa
because
[3SGP not
[John sleep-DES]
‘John wants to sleep because he is not happy.’
allea].
happy]
(43)
bweytuk
[aapo kaa
allea]
[Joan kot-pea].
because
[3SGP not
happy]
[John sleep-DES]
‘John wants to sleep because he is not happy.’
But now contrast the following sentences. The coordinated sentence can not appear in
first position:
(44)
[Joan kot-pea]
into/bweta/o/
[áapo kaa
[John sleep-DES]
and/but/or/
[SG
not
‘John wants to sleep and/but/or/ he is not happy.’
allea].
happy]
(45)
*into/bweta/o/
[aapo kaa
allea] [Joan kot-pea].
and/but/or/
[3SG not
happy] [John sleep-DES]
(John wants to sleep and/but/or/ he is not happy.’)
The same author describes a second characteristic of coordination: “Coordinating
conjunctions are mutually exclusive and subordinating conjunctions are not” (Oirsow
1987:106). According to this observation the co-occurrence of into ‘and’ and bweytuk
‘because’ is expected, but not the co-occurrence of (bwe)ta ‘but’ and into ‘and’:
32
(46)
[Jorge tuuka
namuken] into
[Jorge yesterday drunk]
and
bweytuk
because
[neé= chae-k,
[1NNOM.SG=crie-PST
née
a=
tetemu-k].
1SG 3NNOM.SG
kick-PST]
‘Jorge was drunk yesterday and because he cried to me, I kicked him.’
(47)
[ini’i chu’u ousi junera] ta into [in
maala a=tu’ule].
[this dog
very ugly] but and [3SG.POSS mother 3NNOM.SG=like.PRS]
‘This dog is very ugly but my mother loves it.’
However, as we can see in the translation, the sentence has an adversative meaning
and not a conjunctive one. This fact suggests that into ‘and’ is functioning in these cases
more like and adverbial than like a logical conjunction. Actually, into can be better
translated in this situation like ‘in addition, moreover’.
From these facts we can conclude that into ‘and’ has at least two characteristics, it
can function as a conjunction or as an adverb.
1.4.2.5 Coordination of maximal projections
Verb coordination shows several properties, some of them are the following: it is
possible to have the coordination of two intransitive verbs, but it is not possible to have
the coordination of two transitive verbs sharing a single object. The second transitive
verb always requires an object pronoun, suggesting that it is not possible to have the
coordination of heads (Kayne 1994, Takano 2004).
(48)
Andrea
[bwika]
Andrea
[sing.PRS]
‘Andrea sings and plays.’
(49)
Fabian
caro-ta
[jinu-k]
Fabián
car-NNOM.SG [buy-PST]
‘Fabian bougth and sold the car.’
into
and
[yeewe].
[play.PRS]
into
and
[a=nenkak].
[3NNOM.SG=sell-PST]
33
(50)
*Fabián
caro-ta
[jinu-k]
Fabián
car-NNOM.SG [buy-PST]
(‘Fabian bougth and sold the car.’)
into
and
[nenka-k].
[sell-PST]
Related facts to the previous ones are the following: the coordination structure must
be able to explain the properties of Noun coordination: it can be continuous or it can be
discontinuous:
(51)
Paola [wepul
na’aso-ta]
into [wepul mansana-ta]
Paola one
orange-NNOM.SG and one apple-NNOM.SG
‘Paola ate one orange and one apple.’
(52)
Paola [wepul
na’aso-ta]
Paola one
orange-NNOM.SG
‘Paola ate one orange and one apple.’
bwa-ka into
eat-PST and
bwa-ka.
eat-PST
[wepul mansana-ta].
one apple-NNOM.SG
Adjective coordination can be continuous or discontinuous too, but it requires a
different case marker, the suffix --k ‘NNOM.SG’:
(53)
Paulina
[bemela-k
into teebe-k]
bicha-k.
Paulina
young-NNOM.SG
and
tall-NNOM.SG see-PST
‘Paulina saw the young and (the) tall (one).’
(54)
Paulina
[bemela-k]
bicha-k
Paulina
young-NNOM.SG
see-PST
‘Paulina saw the young and the tall.’
into
and
[teebe-k].
tall-PST
The examples (53) and (54) have a different case marker than nouns. This
characteristic is not treated in this work and remains a topic for further research.
1.5
Optimality Theory
In this section I introduce the formal mechanisms of Optimality Theory. I illustrate
the OT principles using examples from Yaqui.
34
1.5.1
OT basics
Optimality Theory (OT), as other linguistic models, proposes an input form and an
output form, as well as a relation between those two forms. In OT, the relation between
the input and the output form is mediated by two formal mechanisms: GEN and EVAL
(Archangeli 1997:14). For example, let’s take the process of fusion of into ‘and’ and
juchia ‘again’. If the input is composed of those two items, then we have to explain how
we reach the output form intuchia ‘and again’, and why we don’t get other logically
possible output forms:
A model schema for OT, based on Archangeli (1997:14), is shown below:
(55)
Input Form:
/into+juchia/
GEN
Candidate Set:
intochia intuchia intouchia intojchia intjuchia etc.
EVAL
(constraints)
Optimal output:
[intuchia]
Archangeli (1997), following McCarthy and Prince (1993), Prince and Smolensky
(1993), among others, establishes that every input form is composed from a universal
vocabulary which is given by Universal Grammar. Universal Grammar provides a
vocabulary for language representation; as a result, the inputs are well formed linguistic
objects, in the sense that the input forms do not contain non linguistic objects. These are
the only restrictions over the input forms.
35
GEN must be understood as a function which generates an infinite set of candidates,
it only has the restriction that the generated objects have to be linguistic objects,
composed from the universal vocabulary which restricts the input itself. From this point
of view the theory allows for GEN to be so creative, it is able to introduce and elide
material, it is able to re-arrange input material without any restriction. This characteristic
avoids appealing to any type of rule within the OT model. Another task for GEN is to
point out correspondences within the input and output forms. These correspondences are
crucial in the evaluation of faithfulness constraints. These constraints preserve the quality
of the input forms in relation with the output form. If we take the above example, the
constraints Vowel Faithfulness and Consonant Faithfulness require that each vowel and
each consonant in the input form be the same in the output form.
The constraint set (CON) is considered to be part of our innate knowledge of
language. From this point of view, each language uses the same constraint set, and each
constraint is thought to be universal. This approach allows us to conceive that languages
vary according to the constraint ranking. An important fact about OT is that constraints
can be violated, this possibility hinges on the position that constraints occupy in the
hierarchy of particular languages.
Eval(uation) is a mechanism that selects the optimal candidate from the candidate set
created by Gen. This mechanism uses a ranking of violable constraints. The optimal
output is the one that best satisfies those constraints. This satisfaction can be achieved in
two ways: violation from lower constraints in the ranking are tolerated in the optimal
form if that violation helps to avoid the violation of another constraint ranked higher in
36
the hierarchy. The lower-ranked constraints decide the optimal candidate when all
candidates tie over some constraint that is higher-ranked, either because all candidates
satisfied it or because all them violate it (Archangeli 1997).
1.5.2
The different OT approaches
Syntax, as other areas of language, is plagued by challenges that sometimes survive
across time. Coordination is a special topic that has been treated for many years, but if we
look at the state of the art, we will find that few agreements are reached on the
explanation of it. OT is a framework which emerges formally with the pioneering works
of McCarthy and Prince (1993) and Prince and Smolensky (1993). It proved to be useful
for explaining phonological and morphological facts. Since then it has been applied to
several disciplines of linguistics, including syntax. However, the phenomenon of
coordination has not been treated so much within this framework. Two specific works
about coordination and OT are the one of Gáspár (1999) and that of Hoeks & Hendricks
(2005). Therefore, this topic deserves further attention within this theory. We need to test
it against the data: and Yaqui coordination has properties that are challenging to any
theory of coordination.
On the other hand, as Beaver and Lee (2004) point out, there are many ways that OT
approaches have evolved. They analyzed the ones in the following tableau. They use the
seven phenomena mentioned there in order to see what can be solved by those specific
OT approaches. Their conclusions are the following: an X mark indicates that OT fails to
explain that phenomenon, the √ symbol indicates that it can explain such phenomenon.
37
OT Approach
Optionality
Ineffability
Uninterpretability
Total blocking
Partial Blocking
Freezing
Beaver and Lee (2004:150):
Ambiguity
(56)
Naïve production
Naïve interpretation
Back-and-forth
Strong
Weak
Asymmetric (IP)
√
X
X
X
X
X
X
√
X
X
X
X
X
√
X
√
X
√
√
X
X
√
X
X
X
X
X
√
X
√
X
X
X
X
√
√?
X
X
X
√
√
√
Asymmetric (PI)
√
X
√?
√
√
X
X
We can perceive that OT is still a theory which requires to be tested in most fields of
the language. If we look at coordination, there are relatively few works using this
framework. We need to know if this theory is able to solve some of the most recalcitrant
problems in the area of coordination.
It was shown by Speas (1997) that most syntactic principles in the Principles and
Parameters framework have a hedge that covers the fact that they are violable. For
example, the principle Satisfy essentially requires that all features be satisfied, but it has a
hedge that allows it to be violated: in overt syntax if they are “strong”, in covert syntax if
they are “weak” (Speas, 1997:184). In OT each constraint is potentially violable.
OT approaches seem to capture better language facts, than those approaches that use
rules; for example, Van Rooy (2004) shows that Centering theory (Grosz, Joshi and
Weinstein (1983)), designed to make predictions about anaphoric resolution and the
38
interpretational coherence in discourses, was better captured by Beaver’s (2004) OT
reformulation (called COT) and that an OT account is superior to any account that use
rules.
In Chapter Three it is shown how OT can explain the several positions that the
coordinator into ‘and’ can occupy in coordinate structure. In Chapter Four an explanation
based on constraints accounts for one type of unbalance coordination (called here -kai
constructions). Finally, in Chapter Five, it is shown how the OT approach is superior to a
system of rules proposed by Halloway King and Dalrymple (2004) in the account of
noun-verb agreement in Yaqui.
1.5.3
Some OT constraints
1.5.3.1 Alignment Constraints
In OT, there are alignment constraints that are in charge of allocating items in the
places where they appear in the sentence. Lee (2001:81), following Choi (1999), uses the
Canonical Phrase Structure Constraints, CANON, which are the following.
Canonical Phrase Structure Constraints CANON:
(57)
CANONGF (f-s/c-s correspondence): Grammatical Functions align with their
canonical argument positions in c-structucture according to the function
hierarchy (SUB> D. OBJ > I. OBJ > OBL > ADJUNCT).
(58)
CANONθ (a-s/c-s correspondence): non-verbal arguments at c-s align according
to the thematic hierarchy (Agent > Beneficiary > Experiencer/Goal >
Instrumental > Patient/Theme > Locative).
39
Other alignment constraints used by Lee (2001:81) are the Informational Structuring
constraints, described below:
Information Structuring Constraints:
(59)
TOP-L:
Topic aligns left in the clause.
(60)
FOC-L:
Focus aligns left in the clause.
(61)
BCK-R:
Background information aligns right in the clause.
(62)
COMPL-R: Completive information aligns right in the clause.
The analysis of Yaqui sentence coordination indicates that the Information
Structuring Constraints TOP-L and COMPL-R are very active in the language and that
these constraints are centrally responsible for the patters of into ‘and’, and others
coordinators, in first, second and last position.
1.5.3.2 Faithfulness Constraints
A faithfulness constraint requires identity between the input and output forms. I
follow Lee (2001:81) in the use of the following constraints:
(63)
IDENT-IO (P-ROLE):
The value of the proto-role features in the input (e.g.,
[VOL], [CAUS], [SENT], etc.) is preserved in the output.
(64)
DEP-IO (PROM): The
(65)
DEP-IO (NEW):
feature [PROM] in the output is present in the input.
The feature [NEW] in the output is present in the input.
As we are going to see in the analysis of sentence coordination, the faithfulness
constraints are highly ranked in Yaqui because a [PROM] and a [NEW] feature in the input,
is preserved in the output form.
40
1.5.4
A Syntactic Model
Following Sells (2001) and Choi (2001), among others, I assume that Gen is
responsible for all structure building and (hence that) the input is an unstructured set of
lexical items. The legitimacy of inputs could be defined on the basis of the possible
satisfaction of the selectional requirements of lexical elements contained in them; next,
we can see that the input contains several types of information, such as the proto-role
which they are going to play in the sentence as well as informational content, crucial in
the explanation of the behavior of lexical items within the sentence.
1.5.5
The nature of the input for coordination
A fundamental aspect in an OT approach to syntax is to establish the nature of the
input form. There are several proposals for the input for coordination. As an example: for
Oirsow (1987) the input of coordination is composed of full well-formed sentences, then
coordinated structures are produced by an optional rule in the language. For Camacho
(2003), the input is the numeration in the sense of a minimalist approach (Chomsky 1995,
2001). For Johannessen (1998) (within a minimalist framework too) the input is
composed of full CPs (derived or underived). My proposal is that the input consists of a
set of lexical items with more specifications than the numeration. My conception is
closely related to the LFG approach in the sense that each element in the input carries
information about the functions that each element must cover in the sentence. The input
is like that proposed by Gáspár (1999) in an OT approach to coordination, but it differs of
it because in my proposal the coordinator is in the input.
41
The nature of the input in coordination is problematic: it has been observed that there
are several alternatives in order to present information which is clearly related
implicationally (Winter, 2001). The case is illustrated with the next example. According
to Givón (2001:5), this kind of alternation tends to occur cross-linguistically. This is
supported by Yaqui and the glosses in English and Spanish. However, as the examples in
those languages show, the implicational relations do not always holds.
(66)
Maria teebe Peo into teebe. ↔
Mary tall
Peter and
tall
‘Mary is tall and Peter is tall.’
↔
‘María es alta y Pedro es alto’
María into Peo te-teebe.
Mary and
Peter RED-tall.
‘Mary and Peter are tall.’
‘María y Pedro son altos’
(67)
María into Peo nau
saja-k.
Mary and
Peter together
go.PL-PST
‘Mary and Peter left together.’
‘María y Pedro se fueron juntos’
↔
*Maríanau
siika
into Peo nau
Mary together go.SG.PST and
Peter together
(*Mary went together and Peter went together.’)
(*‘María se fue junta y Pedro se fue junto’)
siika.
go.SG.PST
The question to be answered is: Do we have the same input form for the previous
examples? The answer is no. Although those sentences are implicationally related, they
have different inputs. This idea is related to Peterson’s (2004:188) proposal. This
research mentions that in spite of the contrast seen in that type of sentences, and that in
spite of the fact that arguments are represented by the same f-structure, such arguments
represent different instantiations (or tokens) of the lexical features of the arguments
corresponding to different NP’s object and NP’s subject. In that framework, this
difference in f-structure has semantic consequences. The value of the
PRED
feature is a
semantic form. Each instantiation of a semantic form creates a unique object for
42
functional uniqueness and also for semantic interpretation. Thus, two objects in the fstructures can be interpreted differently.
Taking this idea within OT, the input will be different for each sentence in relation to
the number of tokens of a lexical item. The advantage of this would be that the semantic
interpretation will be potentially different as well for each winner candidate. Let us take
the following contrast. We can see in examples from (68) through (71) that there are
logical relations between those sentences. However, there are differences in the
interpretation of them. First, the coordinate structure (68) tends to be interpreted as
containing two events. Even the subject is interpreted as different in each sentence. The
second coordinated structure (69) is interpreted as containing two events and a
correferential subject. The third coordinated structure (70) tends to be interpreted as
containing two events of one action each one. An emphasis appears over the object
tajkaim ‘tortillas’. The structure must be interpreted as containing a single subject who
does those actions. Finally, the fourth coordinated sentence in (71) must be interpreted as
a single (continuous) event which contains a single subject who realizes the two actions
in a temporal sequence.
(68)
Maria tajta’im
ya’ak
into María
make-PST
and
María *i/j
María i tortillas
‘Maríai made tortillas and María j ate tortillas.’
tajka’im bwa-ka
tortillas eat-PST
(69)
Maria tajta’im
ya’ak
into María
María i tortillas
make-PST
and
María i/*j
‘Maríai made tortillas and María i ate tortillas.’
tajka’im bwa-ka
tortillas eat-PST
(70)
Maria tajta’im
ya’ak
into
María tortillas
make-PST
and
‘Maria made tortillas and ate tortillas.’
Ø
tajka’im
tortillas
bwa-ka
at-PST
43
(71)
Maria tajta’im
ya’ak
into
María tortillas
make-PST
and
‘María made tortillas and ate them.’
Ø
am=
3NNOM.PL=
bwa-ka
eat-PST
This kind of contrast indicates that we do not need to postulate a conjunction
reduction.
This kind of contrast indicates that we do not need to postulate a conjunction
reduction. It is not necessary to apply optional rules (Oirsow, 1987) or try to derive one
sentence from the other (Gáspár, 1999). In this way, in OT a fidelity constraint will force
the elements in the input to appear in the output. The input for a coordinated sentence like
(70) is represented as follows:
(72)
Input: [ ya’ak [1], [2] [1]=mariai-, [2]=tajkaim, bwaka[3], [4] [3]=mariaj, [4]=am, into]
It is considered that the input contains all the information related to lexical items.
Therefore, the inputs are taken to be a feature structure representing even discourse
functions, such as the features that indicate new [NEW] and prominent [PROM]
information.
Following Lee (2001), I assume a four way distinction of discourse functions based
on the features [NEW] and [PROM], as indicated below. This conception is related to that
of Choi (2001), Vallduví (1992), Lambrecht (1994) who consider that “information
structure is a domain of a grammar where the discourse-contextual information is
reflected at the sentence level. It shows how a sentence is partitioned or structured
according to the information coming from the discourse context such as ‘what the
sentence is about’ ‘what the new or informative part of the sentence is” (Choi, 2001:21).
Like those authors, in this work, I use two discourse information features: [PROM] and
44
[NEW]. The features are related to discourse newness [NEW] and discourse prominence
[PROM]. The feature [NEW] distinguishes what is ‘new’ or ‘informative’ from what is
‘given’, whereas the feature [PROM] picks out what is ‘important’ or ‘urgent’ in the
sentence. These two binary features may crosscut some of the existing notions of topic
and focus. The following table indicates the way in that the concepts of topic, focus,
background information and completive information are taken.
(73)
-NEW
+NEW
+PROM
Topic
Focus
-PROM
Background
Completive information
In this work, I give evidence that in Yaqui topics tend to occur in clause initial
position, the constraint requiring that topic be aligned to the left of the sentence wins the
battle against the constraint requiring that a coordinator be aligned to the left of a
sentence. There being only one left edge, only one of them can be satisfied. That explains
the occurrence of into ‘and’ in second position. When into(ko) is in last position, I argue
that it occupies that position because it is completive information, we will see later that
less prominent information must to be aligned to the right edge of a sentence.
The analysis of Yaqui coordination shows that the interaction between syntax and
information structure is well pursued in terms of ranking between syntactic constraints
and information-structural constraints. It shows that discourse-contextual information
plays a significant role in the explanation of Yaqui coordination. Discourse-pragmatic
information such as topic and completive information interacts directly with syntax,
especially in word order variation.
45
With this kind of input, Gen generates a candidate set which is evaluated by set of
ranked constraints over c(onstituent)-structure and i(nformational)-structure, which select
the optimal output.
1.6
Yaqui phrase structure flexibility in OT
Yaqui is an SOV language; however, it is not a rigidly structured language. It allows
certain flexibility in phrase-structural descriptions, however, as usual in these relatively
flexible languages, there is usually a particular phrase structural description which is
considered to be unmarked, default, or canonical, while others are regarded as more
marked or non-canonical. As Choi (2001:18) mentions, “the marked or non-canonical
structures are often associated with certain discourse functions so that the structures
appear only in certain discourse contexts”. The same researcher (Choi (2001:19) specifies
that the default or canonical order is the one that is preferred when no discourse context is
provided or when the context demands that the whole sentence be focused, i.e. of new
information, for instance, when the sentence is uttered out of context or when it is an
answer to a question like what’s up?, any news?, what happened?
According to this conception of canonical word order, the answer to the next question
illustrates the canonical word order of a coordinated sentence, on it the coordinator
appears in second position:
(74)
¿jitá yeu
siika?
What out
going.SG?
What’s going on?
(75)
Peo juka
wikit-ta
bwise-k Diana into
Peter DET.NNOM
bird-NNOM grasp-PST Diana and
‘Peter grasped the bird and Diana sold it.’
a=
3NNOM.SG=
nenka-k.
sell-PST
46
Whereas the discontinuous noun coordination is non-canonical because the answer to
a question like the following gives a continuous noun coordination, where both
coordinated items have the same status in relation with the feature [PROM]:
(76)
¿jitá empo ya’a-k?
What 2SG do-PST
What did you do?
(77)
née
[bocham
into
[+NEW
+NEW
[+PROM
+PROM
1SG shoes
and
‘I bought shoes and shirts.’
supem]
+NEW]
+PROM]
shirt
jinuk.
bought
Discontinuous coordination results when the speaker takes the coordinated noun to
the right of the sentence as completive information, i.e. information considered as [+NEW,
-PROMINENT]. An interesting fact about discontinuous coordination is that it can be taken
as an example where an i-structure domain is broken: the coordinated object in the
example is [+NEW], but it is not continuous.
(78)
Née
[bocham]
jinuk
[+NEW]
[+PROM]
1SG shoes
bought
‘I bought shoes and shirts.’
[into
[+NEW
[-PROM
and
supem].
+NEW]
-PROM]
shirt.
The two previous examples indicate that, as many other languages, Yaqui word order
reacts to discourse context. As Choi (2001:23) notes, the information structuring does not
always match the syntactic structuring. The principles that tell us how grammatical
features or functions are to be realized in the surface phrase structure, i.e. C(onstituent)structure, and the principles that tell us how the I(nformational)-structure is to be realized
in the c-structure may impose conflicting requirements on the c-structure. These
47
potentially conflicting requirements are proposed to be OT constraints, which are violable
and ranked. It is predicted that the constraints, having these characteristics, would give
rise to languages that are more sensitive to constituency and other languages more
sensitive to the prominence hierarchy. Further, some could be more sensitive to [PROM]
and other more sensitive to [NEW].
1.7
What is a coordinator?
Under my view, a coordinator is a lexical item that is adjoined to a maximal
projection and it introduces the feature [COORD] (Langendoen 2003). It is in the input
form (i.e. I assume that it is not introduced by constraints, as proposed by Gáspár 1999).
This feature licences the adjunction of additional material. The ulterior adjunction of
material will depend on the nature of the input. So, within an interaction like the
following, the input for speaker’s (79b) production would consist of just the items in
(79b). I assume that sentence (79a) is background for the production of (79b)). In other
words, I assume that the speaker in (79b), after interpreting sentence (79a), selects the
required items for producing fragment (79b).
(79)
a) Joan
Pesio-u
‘John
Hermosillo-DIR
‘John went to Hermosillo.’
siika.
go.SG.PST
b) Peo-su?
Peter-and
And Peter? (Did Peter go?)
c) e’e,
(ju)na chea kaa
NEG,
DEM
really NEG
‘No, he did not go.’
(Speaker 1)
(Speaker 2)
siika.
go.SG.PST
(Speaker 1)
48
Therefore, the input adopted here is just reduced to what a speaker produces. For that
reason, we have to explain how fragments are structured. The structure of the fragment
for (79b) is represented as follows:
(80)
Input: {-su, Peo}, output:
NPcoord
NP
-su
Peo
‘and Peter’
Complete sentences containing a coordinator are not necessarily syntactically
coordinated to another sentence. The interchange in (81) has a sentence in (81b) which
contains a coordinator in it:
(81)
a) itepo
tuuka
Maria-ta
1PL
yesterday
Maria-NNOM.SG
‘Yesterday, we visited Maria’
b) ¿jitá
into eme’e bwa-ka?
what
and
2PL
eat-PST?
‘And what did you eat?’
pasiyaloa-k
visit-PST
(Speaker 1)
(Speaker 2)
c) bwakabak-ta
(Speaker 1)
bwakabaki-NNOM
‘Bwakabaki.’ (Yaqui food made with beans, meat and other ingredients).
The input for (81b) adopted in this work is represented in (82. It consists just of the
lexical items used by the Speaker (null pronouns, functional projections can be
introduced by Gen):
(82)
Input: {jita, into, eme’e, bwaka}
The structure for the sentence (81b) is shown in (83):
49
(83)
CP[coord]
Jitai
CP[coord]
into
CP
eme’e ti bwaka
The proposal can be easily extended to sentence coordination where two conjoined
sentences like (84) have the structure in (85):
(84)
Maria nojim
ya’a-su-k
Peo into
make-TERM-PST
Peter and
Maria tamal.PL
‘Maria finished doing the tamales and Peter sold them.
am=nenka-k.
3NNOM.PL=sell-PST
The host CP has the feature [COORD] which licenses the addition of another CP.
Because of topicalization, which will be seen in chapter 3, the host CP has a fronted NP
but such topicalization does not block the adjunction of new material. The representation
implies that we do not need to differentiate between specifiers and adjunction sites.
(85)
CP[coord]
CP
CP[coord]
Maria nojim ya’asuk Peo
CP[coord]
Into
CP
am=nenkak
1.8
Summary
In this chapter we have seen the introduction to the coordinated structures of Yaqui.
We have set the empirical and theoretical goals of this work. There is a background on
the type of categories that Yaqui can coordinate and in some aspects of word order. There
50
is information about the logical coordinators and the interaction between them. In the
second part of this chapter I presented the model of OT and some basic assumptions
about coordination. In short, this chapter serves as background for the research developed
in the rest of this dissertation.
The next chapter contains a literature review and a reflection related to coordination
as a relation adjunct-host.
51
2
2.1
LITERATURE REVIEW
Disagreement in the Literature
The main purpose of this chapter is to show that in spite of the fact that coordination
has been formally analyzed through the years, few scholars agree about it. Initial ideas
such as that of Conjunction Reduction proposed by Chomsky (1957) opened the field for
research and pioneering researches like that of Ross (1967) established questions that still
are at the center of the debate: Among others, Is the coordination structure symmetric or
asymmetric?, Does the coordinator form a unit with a conjunct or not? Is the coordinator
a head? In this chapter I summarize different approaches to coordination that belong to
distinct frameworks. Therefore, we expect to have distinct answers for a single question. I
have selected six works; the first one is located in the HPSG framework, the second one
in the Minimalist framework, the next is located in OT framework, the following in LFG
framework, another is within the Autolexical framework and the last one is a revision of
why coordinate structures can not be Conjunction Phrases.
It is obvious that I left out other equally important approaches; however, the main
purpose of this chapter is to motivate a reflection on what is happening in the
coordination phenomenon nowadays. However, many proposals of these works that are
not touched here will be called upon when necessary in the description and/or in the
analysis of Yaqui data.
52
An additional purpose of this chapter is to evaluate in a global manner the various
proposals in order to adopt what can be considered most appropriate for describing and
explaining the behavior of Yaqui coordination.
2.1.1
An HPSG approach (Abeillé 2003)
“It is striking that no agreement has been reached
on the structure of basic coordinate constructions”
(Abeillé 2003:1).
Abeillé, working within a HPSG framework, shows the validity of her (previous)
claim by revising what some researchers says about this issue and drawing her own
conclusions.
Her proposal holds that coordinated structures are asymmetric: the conjunction
makes a subconstituent with one of the conjuncts. For her, this Conj X constituent has
several functions, including adjunct. Abeillé’s paper explore two important questions: is
the structure hierarchical or flat? And do the daughters have the same function or not?
After reviewing linguistic and theoretical facts she concludes that a) it is necessary to
distinguish Conjunction as a type of word and Coordination as a type of construction, b)
Conjunctions are weak syntactic heads that yields a Conjunct phrase, and c) incidental
conjuncts and some asymmetric conjuncts are adjuncts. From her point of view Conjunct
phrases can enter into several constructions: head-only-phrases, head-adjunct-phrases and
coord-phrases (Abeillé 2003:19).
This researcher rejects approaches where the coordinator is a head and where the
coordinate structures are reduced to X-bar schemata, such as those of Kayne (1994) and
53
Johannessen (1998). For her, a structure like the following is not viable (Abeillé
(2003:3)):
(1)
Spec-head-complement. Kayne (1994) and Johannessen (1998) cited in Abeillé
(2003:3):
Conj P
spec
head
XP
head
Conj’
cplt
Conj
John
YP
and
Mary
For Abeillé the most viable structures are (3a) and (3b); however, the (3b) structure
needs to be revised. She considers that the structure in (3a) accounts for n-ary
coordinations and for coordinations with multiple conjunctions. Structure (3b) accounts
for asymmetric coordinations such as Russian comitative coordination, where the case of
the NP is that of the first conjunct (McNally 1993, cited in Abeillé 2003:4):
(2)
(3)
a) Anna
s
Anna-NOM with
Petej
Peter-INSTR
pridut
are-coming-PL
b) *Petej
Anna
pridut
s
a) Head-head. Sag et al (1985), Gazdar et al (1985), cited in Abeillé (2003:3):
NP
head
head
NP[CONJ nul]
NP[CONJ and]
marker
cplt
John
Conj
NP
and
Mary
54
b) Head-adjunct. Munn (1992), (2000), cited in Abeillé (2003:3).
NP
head
adjunct
NP
BP
head
Boolean
John
and
cplt
NP
Mary
In order to analyze some conjuncts as adjuncts (as the example (2a)) Abeillé proposes
that the category of the adjunct should vary with its complement (NP, PP…)
After her analysis of the French particle car ‘since’, this researcher concludes that car
introduces an adjunct phrase and that all coordinating conjunctions can introduce adjunct
phrases in French.
The analysis of incidental coordination in French (i.e. coordination with incidental
prosody which forms, according to her, is S Conj XP.) shows that these constructions do
not involve coordination and that such conjuncts can be of various categories: NPs, PP’s,
Ss. The next example contains what is considered an incidental coordination Abeillé
(2003:7):
(4)
John read the book (and) avidly.
The claim that these types of constructions do not involve coordination is supported
by the lack of reversibility between “conjuncts” and because extraction is allowed out of
the first “conjunct” (Abeillé 2003:8):
(5)
a) *John avidly and read the book.
b) The book that John read, and avidly.
55
Additional evidence that these constructions must be analyzed as adjoined phrases is
obtained from the mobility of the construction: they tend to have the same mobility as
incidental adverbs (Abeillé 2003:8):
(6)
a) Jean,
Jean,
et
and
c’est
it is
heureux,
fortunate,
a
has
lu
read
b) Jean
a,
et c’est heureux,
c) Jean
a
d) Jean
a
votre livre.
your book.
lu
votre livre.
lu,
et c’est heureux,
votre livre.
lu
votre livre, et c’est heureux.
And from agreement facts: real coordinate NP’s trigger plural agreement whereas
incidental NPs do not Abeillé (2003:8):
(7)
a) Jean
John
et
and
Marie liront/*lira
Marie will:read:PL/*SG
b) Jean
John
lira/*liront
will.read.SG/*PL
votre livre.
your book.
votre livre, et
your book, and
marie aussi.
Marie too.
The same author rejects an analysis of constructions like (4) in terms of unlike
coordination (as in Progovac 1998). She rejects too an analysis of (6) and (7) as S (or VP)
coordination with the incidental conjunct being a reduced S (or VP) because extraction
can involve only the main clause and not the incidental conjunct. This violation of the
CSC would be odd if we do not have and adjunct. If we consider that these constructions
are adjuncts, then it is predicted that as any adjunct, they will be mobile and an island for
extraction.
The author extends the adjunct conception to Welsh serial coordination. These
constructions have several characteristics (many of them, as we will see, also appear in
Yaqui): Tense is marked only on the first conjunct, the others involve “verbal nouns”, the
56
order of the conjuncts is fixed (and usually indicative of narrative progression), and the
subject is shared between the conjuncts. The construction does not obey the CSC.
Abeillé considers that conjunction is a weak head that shares most of its syntactic
features with its complement. Then conjunctions take (at least) one complement and
inherits most syntactic features from it, except for the lexical feature CONJ which is
specific for each conjunction (Abeillé 2003:12). Conjunctions can head phrases as
indicated (Abeillé 2003:13):
(8)
a)
NP [CONJ et]
head
b)
comp
[CONJ et] [NP CONJ nul]
Et
Paul
AP [CONJ ou]
head
comp
[CONJ ou]
AP [CONJ nul]
ou
célèbre
Incidental conjuncts have a representation as (9). In such structures, the adjunct is
represented with a Boolean head incident feature, as in Bonami and Godard (2003). The
representation shows that incidental conjuncts are treated as V adjuncts, which could
enter into a Head-adjunct-phrases or Head-complements-adjuncts-phrases Abeillé
(2003:17).
(9)
S
NP
VP
head
[1] VP
Paul
viendra
adjunct
NP[CONJ ou]
MOD [1]
INCIDENT +
ou
Marie.
57
Abeillé proposes that there are two subtypes of conjunction words: basic-conj-word
and discourse-conj-word. Basic-conj-words are marked as INCIDENT and share (by
default) the INCIDENT value of their complement. They also inherit the MOD value of
their complement. On the other hand, discourse-conj-words have a specific MOD V
feature, which they do not necessarily share with their complement, and an INCIDENT +
feature, which their complement does not have. These kinds of conjunctions are binary
relations and take the phrase they modify as one of their arguments (Abeillé 2003:16).
Abeillé uses basically the same lexical entries for conjuncts as main clauses or
fragments, such as the following Abeillé (2003:17):
(10)
a) Mais Paul est parti!.
‘But Paul is gone!’
b) Et Paul?.
‘And Paul?’
Because those fragments can denote questions, propositions, or exclamations, Abeillé
takes the notion of “messages” from Ginzburg and Sag (2000) and introduces it in the
lexical representation, so the conjunction takes two semantic arguments: its complement
(interpreted as a proposition) and another clause available in the discourse context.
2.1.2
A Minimalist approach (Camacho 2003)
“The internal structure of coordination was usually
left unanalyzed, or assumed to be ternary branching...”
Camacho (2003:1)
Camacho’s work, in a Minimalist framework, tries to capture two main properties of
coordination: c-command asymmetry and licensing symmetry. The first one refers to the
fact that one of the conjuncts c-commands the other(s), and the second one to the fact that
58
coordination must be symmetric with respect to a licensing head, i.e. each conjunct
should reflect the same structural properties as if it were in a simplex sentence (Camacho
2003:1).
Camacho accepts the underlying idea behind Chomsky’s conjunction reduction and
claims that conjunction always involves a set of sentential functional projections.
According to his view, coordination is propositional in nature. The structure of
coordination is asymmetrical and the conjuncts are the specifiers of or complements to
sentential functional (propositional) projections (Camacho 2003:2).
Camacho mentions that the exceptions to Wasow’s generalization (the requirement
for symmetry (balancedness) among conjuncts) are of two types: a) cases where only one
of the conjuncts satisfies the requirement of the factor (Unbalanced Coordination in
Johannessen’s (1998) terms) and b) cases in which the features of the conjuncts do not
exactly match these of the factor, giving rise to feature resolution (see Corbett 1983) or
feature indeterminacy (see Dalrymple & Kaplan 2000). Feature resolution is exemplified
in (11). There the verb (factor) does not match the features of the individual coordinated
nouns Camacho (2003:11):
(11)
Juan y
yo
comimos
Juan and
1SG ate.1P.PL
‘Juan and I ate omelette.’
tortilla
omelette
Feature indeterminacy is exemplified with a sentence from Polish. The word kogo
‘who’ satisfies the genitive and accusative case required respectively by the verbs
Dalrymple and Kaplan (2000), cited in Camacho (2003:11:
59
(12)
Kogo Janek lubi a
Jerzy nienawidzi?
Who Janek likes and
Jerzy hates
‘Who does Janek like and Jerzy hate?’
The analysis of Spanish shows that “temporal/aspectual adverbs with scope over both
conjuncts requires temporal/aspectual parallelism” (Camacho 2003:13). In other words,
only person, number, gender and case are subject to resolution rules in Spanish.
He follows Munn’s (1992, 1993) proposal for asymmetric c-command between
conjuncts. He argues against Progovac’s (1997) objection to c-command explanations in
coordinate structures. Camacho’s conclusion is that “one of the conjuncts should be
structurally higher than the other” (2003:22).
Looking at the interpretation of coordination, Camacho distinguishes three types of
approaches: those that favor a propositional analysis of it (Gleitman 1965, Goodall 1987,
Schein 1992); those that favor treating coordination as a group forming operator that
behaves like plurals (Link 1983, Munn 1993), and those that favor both the (a and b) type
of proposals (Partee and Rooth 1983, Johannessen 1998, among others).
Camacho’s (2003) analysis favors a propositional approach to coordination. His
arguments are based on the observation that plurals differ from conjunction: they are not
licensed in the same structural position in a sentence, they have different entailment
relations and they behave differently with respect to adverbs: propositional adverbs do
not modify simplex DP’s, but they can modify conjuncts. This last characteristic is
exemplified next. On it, the modal adverb can not scope out of the conjunction; so, the
following reading is impossible: *the set of people possibly formed by Harvard students
and Columbia students Schein (1992), cited in Camacho (2003:27).
60
(13)
The Columbia students and possibly the Harvard students formed an unbroken
chain around the pentagon.
One of the central proposals of Camacho’s work is that “conjunction is a sentential
functional projection head that has propositional content. Its subcategorization
requirements are minimum in the general case of and, but can be more specific for other
conjunctions” (Camacho 2003:38). The representation of and is shown as follows
Camacho (2003:38):
(14)
and
[+ PROP]
…
The general structure for coordination that Camacho proposes is the following:
(15)
XP
X’
Conj1
X
XP
Conj2
X’
X
YP
In the representation the first X represents the conjunction, the second X any sentential
functional projection, such as INFL, Agr, etc. Thus for subject coordination we have the
following representation.
61
(16)
IP
Subj1
I’
and
IP
Subj2
I’
I
VP
As support for treating conjunction as a functional projection linked to sentential
inflection, Camacho analyzes switch reference systems, commutative constructions,
adverbial coordination and clausal coordination. The explanation is given in a minimalist
framework (Chomsky 1995).
Following Kayne (1994), he proposes that “heads (and parts of words) can not be
conjoined” (Camacho 2003:62). Therefore, the conjuncts must be maximal categories.
This conclusion is supported by the behavior of clitics which, being heads, can not be
conjoined Kayne (1994), cited in Camacho (2003:65).
(17)
*Jean te
et
John CL(2p.ACC) and
me
vois
CL (1p.ACC) sees
souvent
often
An important implication of his proposal is that it derives constituency effects without
a coordinate phrase. The structure allows him to explain important facts as why
coordinate DPs, for example, can act as antecedents of anaphors, why they can bind
infinitival PROs, and why they can undergo DP movement.
In relation to DP movement, for coordinate subjects that seem to move, Camacho
suggests that they are coindexed with a category located in the thematic position, instead
of moving as separate constituent to the position where they appear at the surface, as
indicate below:
62
(18)
Johni and Maryj seem proi+j to ti+j have been called ti+j
He formalizes the idea that coordination entails a chain between the conjuncts and a
silent category by proposing local feature insertion to coordination. i.e. part of the
features of the chain are inserted in the lowest position and they move to the two
conjuncts. Lets take the example of two conjoined subjects. The agreement features of
the conjoined DPs will always be generated in the specifier of IP, as illustrated in the
derivation of the following Spanish sentence Camacho (2003:83):
(19)
Lucía y
Yesi corren.
Lucia
and
Yesi run
‘Lucia and Yesi run.’
(20)
yP
DP Lucía
y’
θ
CASE
SG
3P
y
IP
TNS
DPYesi
NOM
SG, SG
3P
I’
-θ
-CASE
SG
3P
I
VP
TNS
NOM
SG, SG
3P
DP
V’
AGENT
NOM
SG, SG
After movements and feature checking, the derivation has the following
representation:
63
(21)
yP
θ
CASE
SG
3P
DP Lucía
y’
y
IP
TNS
NOM
SG, SG
3P
tYesi/x
I’
I
VP
TNS
tx
V’
As we can see in the derivation, for Camacho, a plural is a sum of singulars, contrary
to Dalrymple and Kaplan’s (2000) conception of plural as a primitive feature.
For partial agreement, Camacho distinguishes two types of agreement: PF and LF
agreement. The first one does not have semantic consequences (i.e. the co-reference
possibilities are still those of the whole coordinate structure), while the second one does.
So, for an example of LF partial agreement, Camacho reinterpret ABS analysis of Arabic
coordination. The following sentence has the indicated representation.
(22)
Neem
Kariim
w
Marwaan
Slept(3P.MAS.SG) Kariim
and
Marwaan
‘Kareem and Marwaan slept in the room.’
fəl-l-biit.
in-the-room
64
(23)
FP
F
neemi
XP
VP
Kariimj
X’
V’
X
ei
w
XP
VP
Marwaan
X’
V’
V
PP
ei
fəl-l-biit
After spell-out, the higher subject will move to the spec-FP, checking agreement with
the verb in F0.
(24)
FP
Kariimj
F’
F
neemi
XP
VP
ej
X’
V’
X
ei
w
XP
VP
Marwaan
X’
V’
V
PP
ei
fəl-l-biit
On the other hand, Camacho proposes that separateness of events could be related to
the level of coordination. For separate events, the coordination could be at the level of TP
65
or CP, for a single event with sub events, the coordination has to be lower in the tree. For
that reason, the following sentences would vary in the level they coordinate:
(25)
a) John came and Peter went.
b)
2.1.3
John came and went.
An OT approach (Gáspár 1999)
“OT is well-positioned to tackle issues in the theory of
coordination that have caused problems for researchers
working in hard constraint-based approaches”
Gáspár (1999:1)
Gáspár is one of two researches that I am aware of who treats coordination in the OT
framework (the other research is developed by Hendricks 2005). Gáspár tries to explain
within this framework some of the most salient problems that coordination poses: “how
to fit the coordinate structure into x-bar theory, how to analyze coordination that can not
be treated as sentential coordination on conceptual grounds, and how to account for
differences between languages in unbalance coordination” (1999:157). In OT, constraints
are violable. For that reason, what seems to be a stipulation in the Johannessen (1998)
minimalist approach, i.e. that the specifier and the complement are not required to be
maximal projections, in OT could be seen as a violation of that restriction. The constraint
is defined as:
(26)
SPEC-COMP-PHRASE
*X, if X is in Spec or Comp position and X is not maximal.
Gáspár proposes a constraint that merges segments (rather than ellipsis or deletion),
he follows in this sense the ideas of Johannessen (1993). Some conditions for merging
66
are that they must occur in the same position in their trees and that they must not have
conflicting features. The constraint is defined as follows:
(27)
FUSION
X must be fused with Y, where X and Y are input elements.
In addition to this constraint, Gáspár uses the faithfulness PARSE constraint of
McCarthy and Prince (1993), reinterpreting it in the following way: “as long as one token
of an input element is present, PARSE, is satisfied, no matter how many tokens are in the
input” (1999:161). Other constraints are SAME-THETA, which demands that conjuncts of
a and P bear compatible theta roles; FILL, a faithfulness constraint that forbids the
addition of new elements in addition to those of the input; and FULL INTERPRETATION
(FI), a semantic constraint demanding that output forms be interpretable.
For a coordinate sentence like the following, Gáspár shows the interaction of PARSE
and FUSION. He proposes the input seen in the tableaux. GEN poses several candidates,
but, after the evaluation, only the candidate (b) is optimalGáspár (1999:162) 4
(28)
Table that shows the interaction of PARSE and FUSION.
{Like[1],[2], [1]=John, [2]=mayor, hate[3],[4], [3]=Mary, PARSE
[4]=mayor}
a. John liked the mayor and Mary hated the mayor
b. )John liked and Mary hated the mayor.
c. John and Mary liked the mayor.
*!
d. John and Mary hated the mayor.
*!
4
FUSION
*[36]8!
*[21]10
A reviewer made the observation that the (28a) choice is grammatical, just a violation of Gricean
principles. Gáspár’s approach does not use pragmatic constraints. However those constraints can
potentially be integrated in any OT approach. The aim of such approaches would be to distinguish between
grammaticality and felicitousness. That is not pursued in Gáspár’s paper.
67
With respect to RNR structures, this researcher proposes that a sentence like the
following can have the representation indicated below. As we can see, the winning
candidate has a double mother. For Gáspár this kind of representations could be well
formed as long as they do not cross branches:
(29)
John liked and Mary hated the mayor.
(30)
&P[IP]
IP
&’
VP
NP
John
V’
&
IP
and
VP
V
NP
liked
the mayor
NP
V’
Mary V
NP
hated
The constraint that avoids crossing branches is defined as follows:
(31)
NO-CROSS
Crossing branches are forbidden.
Gáspár (1999) analyzes Unbalanced Coordination (UC), Extraordinary Balanced
Coordination (EBC) and Ordinary Balanced Coordination (OBC). In UC only one
conjunct bears the grammatical features associated with the conjunction phrase, but all
the conjuncts are interpreted as if they had the same features. In EBC both conjuncts have
deviant features; whereas in OBC both conjuncts have the expected features.
Gáspar adopts the structure proposed by Johannessen (1998). So, UC would be
represented as follows:
68
(32)
AgrP
CoP[NP]
NP
han
‘he’
Agr’
Co’
Agr
Co
NP
og
‘and’
meg
‘me’
var
‘were’
And he introduces some additional constraints. The first one is a constraint
responsible for spec-head-agreement, defined as follows Gáspár (1999:171):
(33)
SHA
An element in [Spec, XP] position must agree with the element in [X] position.
Two more constraints are defined as indicated next. DEFAULT would be responsible
for introducing default values, in this case, default case. SAME FEATURE requires both
conjuncts to bear the same features Gáspár (1999:172-173):
(34)
DEFAULT
*If default form is not adhered to.
(35)
SAME-FEATURE
[Spec, CoP] and [Spec, XP]
The different ranking of these constraints allows explaining UC, EBC and OBC.
Finally, a constraint which function is to ensure semantic resolution (i.e. it ensures
that two singular NPs as subjects trigger plural agreement) is defined:
(36)
SEMCA
Determine agreement features of a coordinated construction from both the
specifier and the complement.
69
Because OT is an input-based theory, Gáspár considers that it is in better position to
explain some ambiguities related to coordinated structures. The following ambiguity can
be explained by the existence of two inputs which produce the same sentence Gáspár
(1999:163).
(37)
a) the pictures of John and Mary] were underexposed.
b) x [x = picture (John & Mary) underexposed (x)]
c) x [x = picture (John vs Mary) underexposed (x)]
The inputs are given in what follows:
(38)
a) {Underexposed[1], [1]= pictures[2], [2]=John, [2]=Mary}
b) {Underexposed[1],[1]=pictures[2],[2]=John,underexposed[3],[3]=pictures[4],
[4]=Mary}
2.1.4
A LFG approach (Peterson 2004)
“An adequate and theoretically satisfying account
of coordination has long remained an elusive goal”
Peterson (2004:643)
Peterson’s (2004) work is located in the LFG framework. His main purpose is to
explain some elusive topics in coordination: Distribution of grammatical functions,
ability to coordinate unlike categories and lack of distribution of lexical properties.
The first property of coordination is illustrated with the following sentence. In it, the
subject and object grammatical functions are distributed across all conjuncts: The subject
Kate is interpreted as the subject of both verbs faxed and emailed, whereas the object the
results is interpreted as the object of each verb too (Peterson2004:645).
(39)
Kate faxed and emailed the results to Paul.
70
The second property --coordination of unlikes- is illustrated next. The sentence
contains the coordination of an AdjP and a NP. In short: the conjuncts do not need to
belong to the same grammatical category (Peterson 2004:648):
(40)
Paul is stupid and a liar.
The third property --Non-Distribution of lexical properties- refers to the fact that
features do not percolate up to the coordination node. “This is equivalent to stating that
coordination is not endocentric: it is not a “headed” construction” (Peterson 2004:650).
The next example indicates that the coordinate subject, but not the individual conjuncts,
must have the property [plural]; i.e. ‘number agreement’ does not distribute (Peterson
2004:651).
(41)
a) The dog and the cat are in the garden
b) *The dog are in the garden and the cat are in the garden.
Peterson’s solution is based in the idea that “functional structure of a coordination of
constituents is the set of functional structures of the coordinated elements” (2004:651).
Following Kaplan and Maxwell (1988), Peterson considers that the identity of a
conjunction does not enter into any syntactic or functional generalization. The
conjunction, therefore, is not included in the functional structure at all. Its information is
necessarily encoded only at the semantic level of representation. So, Peterson proposes
the following rule schema for coordination. We can see that no information is carried by
the conjunction(Peterson 2004:652):
(42)
a. X →
X
C
↑ε↓
Y
↑ε↓
71
Some important assumptions hold: a verb carries with it a skeleton form of the fstructures that it can occur in; “the elements of a coordinate structure carry exactly those
grammatical functions that they would have carried if they had appeared alone in place of
coordination.” (Peterson 2004:654).
Peterson explores his proposal in relation to phenomena such as subcategorization,
anaphora and control. For him sentences such as the following have different functional
structures, therefore, conjunction reduction is rejected:
(43)
John cooked and ate a pie.
(44)
John cooked a pie and John ate a pie.
Their respective f-structures are shown below. In the first case, there is only one
instantiation for John and only one for pie. However, in the second case, there are two
instantiations for John and two for pie. That difference is responsible for the contrast
indicated in the previous sentences.
(45)
f1
f2
SUB
TENSE
PRED
OBJ
f5 PRED ‘John’
PAST
‘cook <(f2 SUBJ) (f2 OBJ)>‘
f4
PRED ‘pie’
DEF --
f3
SUB
TENSE
PRED
OBJ
f5
PAST
‘eat <(f3 SUBJ) (f3 OBJ)>‘
f4
72
(46)
f1
f2
SUB
TENSE
PRED
OBJ
f5 PRED ‘John’
PAST
‘cook<(f2 SUBJ) (f2 OBJ)>‘
f4
PRED ‘pie’
DEF --
f3
SUB
TENSE
PRED
OBJ
f6 PRED ‘pie’
PAST
‘eat <(f3 SUBJ) (f3 OBJ)>‘
f7
PRED ‘pie’
DEF --
Coordination of unlikes is explained by proposing that it is the grammatical function
which determines the ability to coordinate. The f-representation of coordinate unlikes is
very close to the ones seen before. Two unlikes coordinate if they share the same
grammatical function. Because a conjunction is not a head, lexical properties will
percolate only as far as the node dominating the individual conjunct. They are not shared
across the coordination as a whole.
More interesting is Peterson’s discussion of non-distribution of lexical properties.
He claims that only grammatical function attributes are distributed, but that all lexical
properties show non-distributivity. His claim is supported by data from several
languages. In the following examples we find two singular NPs functioning as subject,
with the verb also in singular (Johannessen 1996), cited in Peterson (2004: 670):
(47)
[Qafar]
Lubak-kee
yanguli
Lion.M.SG.ABS-and hyena.M.SG.NOM
‘A lion and a hyena were seen.’
yumbulle.
was-seen.M.SG
73
(48)
(49)
Mi
ke
le
I
and
he
‘I and he sit’
ta
sit.SG
[Slovene]
Groza
in
strah
je
Horror.F.NOM.SG
and
fear.M.NOM.SG is
vas.
village.ACC.
‘Horror and fear seized the whole village.’
prevzela
seized.F.SG
vzo
whole.ACC
Peterson affirms that there is grammar underspecification (at least for English) in the
area of agreement with coordinated subjects, so speakers resort to various strategies to
determine verbal number. Therefore, variability is expected. A strategy (in the sense of
Corbett 1991), is a working principle which speakers use for “patching up” gaps left by
the grammar. However, Peterson (in footnote 22, 2004:672) considers that in some
languages certain strategies are grammaticalized and that maybe a core rule has to be
stipulated (with the corresponding extra-cost to the grammar).
With respect to person and gender, he holds that non-distribution applies to them. For
case, he mentions that all combinations of case in any order are tolerated in English NP
coordinations. This observation contradicts Johannessen’s claim that only the second
conjunct could be in a non-canonical case. The following examples show the affirmed
variation (Peterson 2004:673).
(50)
a) % Me and him are coming to your party.
b) % Him and me are coming to your party.
c) % Him and I are coming to your party
d) % Me and John are coming to your party.
74
2.1.5
An Autolexical Approach (Yuasa and Sadock 2002)
“Language is a multi-faceted affair and
what is coordinate in one structure
might be subordinate in a parallel one”
Yuasa and Sadock (2002:88)
Yuasa and Sadock (2002) analyzed what they consider mismatches between
coordination and subordination in the framework of Autolexical Grammar (Sadock 1991,
1993). This theory assumes the autonomy of different components of the grammar.
Therefore, a sentence could be coordinated at the syntactic level but subordinated at the
semantic one (pseudo coordination) and vice versa, subordinated at syntactic level but
coordinated at the semantic one (pseudo subordination). Their work only focuses in this
last type of construction.
For them, coordination and subordination are defined as follows:
(51)
“A coordinate constituent is one of two or more sister nodes whose categorical
information percolates to the mother node” (Yuasa and Sadock (2002:89)).
(52)
“A subordinate constituent is a node whose categorical information does not
percolate to the mother node while that of at least one sister node does” (Yuasa
and Sadock (2002:90)).
The diagrams that represent those definitions are given below (Yuasa and Sadock
2002:90):
(53)
a) Coordination
b) Subordination.
X
X
X1,
X2,
…Xn
X
Y…
Z
75
The representations intend to capture the fact that, for coordination, the daughter Xs
do not necessarily belong to the same category, but the categorial information of all the
conjuncts can contribute to the categorial information of the mother node, whereas for
subordination, the subordinate constituents Y… Z does not percolate to the mother node,
however, that of their sister X does.
An instance of clausal pseudo-subordination is the following. In it, the verb hatarai
‘to work’ which belong to the first conjunct is not inflected for the past tense, whereas the
verb shi ‘to do’ in the final conjunct is inflected for it(Teramura 1991:221) cited in Yuasa
and Sadock (2002:92):
(54)
Ojiisan-ga
yama-de
hatarai-te
Old man-NOM mountain-LOC work-and
obaasan-ga
old woman-NOM
mise-no
ban-o
shi-ta.
store-GEN
sitting-ACC do-PST
‘The old man worked at the mountain, and the old woman tended the store.’
Yuasa and Sadock suggest that in examples like the previous one, only categorial
information of the final clause percolates to the mother node of the entire structure,
therefore all the structure is interpreted as past tense.
They follow Culicover & Jackendoff (1997) who claim that the semantics of a
construction determines whether the construction is subject to the CSC. They applied this
and four additional tests to --te-coordination and concluded that it is semantically
coordinated. The results are the following and are the expected ones if semantic
coordination is happening:
76
(55)
a) The construction is reversible and truth conditions are preserved.
b) The construction obeys the CSC.
c) Backward pronominalization is not allowed.
d) Any number of conjuncts can occur in coordinated constructions.
e) Scope considerations: under semantic coordination both conjuncts are affected
by negation.
The --te-coordination behaves at semantic level as a coordinated construction. Given
the previous facts, a dual structure is assumed for --te-coordination (Yuasa and Sadock
2002:98).
(56)
S[+Fin]
S[-Fin]
S[+Fin]
NP
VP[-Fin]
Taroo-ga
Osaka-e it-te, Hanako-ga
Kyooto-e ik- u
Arg
Pred
Pred
Prop
NP
VP[+Fin]
Tns Arg
O
Prop
Prop
Tns
O
Prop
and
O
Prop
In the representation semantics involves coordination of like semantic structures,
while the syntax involves subordination. We can see that only the final clause allows
percolation of the categorial feature to the mother node of the complete structure. Of two
semantic tenses, only the last is associated with any surface morpheme.
77
In addition to analyzing -te-coordination, Yuasa and Sadock (2002) explore pseudosubordination of NPs in Yiddish and in West Greenlandic. They give the following
examples:
Pseudo-subordination:
(57) a) der
tate
mit
der
mamen.
mother.DAT
The.NOM father with the.DAT
‘Father and mother.’ (Lit. ‘The father with the mother.’)
Simple coordination:
b) der
tate
um
di
The.NOM father and
the.NOM
‘The father and the mother.’
mame.
mother.NOM
Simple subordination:
c) der
rebe
mit-n
hunt.
The.NOM rabbi.NOM
with-the.DAT dog
‘The rabbi with the dog.’
Although (57a) and (57c) have the same syntactic representation, the structure in
(57a) is coordinated at the semantic level for the following reasons: a) the terms are
reversible without change in reference (that does not happens with (c) where the first
conjunct refers to a particular entity), b) the verb agreement with pseudo-subordinate
subjects is plural (in subordination it is singular): it occurs with predicates whose
meanings demand plural subjects, c) more than two NPs can be connected by pseudosubordinate NPs (all conjuncts “are understood as parallel, a property we would expect if
we are dealing with semantic coordination” (Yuasa and Sadock 2002:102).
For Greenlandic the conclusions are similar. The basic difference with Yiddish is that
“the marker of the construction in Yiddish is a preposition which is otherwise a
subordinator, whereas in West Greenlandic, it is a clitic which is otherwise a coordinator”
(Yuasa and Sadock 2002:107).
78
More important for the purpose of this work is the use of the Greenlandic --lu ‘and’
coordinator. In the following two coordinate clauses, one of them occurs in a
subordinated mood called the Contemporative, while the mood of the other determines
the mood of the entire constituent. The construction is pseudo-coordinated. The
coordinator --lu is a clitic and “attaches as a suffix to the first word of the conjunct that
follows it in much the same way as Latin --que ‘and’ does” Yuasa and Sadock
(2002:fn14). The coordinated sentence containing the coordinator --lu in second position
is given and represented in what follows(Yuasa and Sadock 2002: fn14):
(58)
Atuarfik-Ø
angi-voq
600-nil-lu
atuartoqar-luni.
600-INSTR.PL-LU have.students-CONT.RSG
School-ABS.SG be.big-IND.3SG
‘The school is big and has 600 students.’
(59)
S[ind]
S[ind]
NP
S[cont]
VP
VP
Atuarfik
Angivoq
NP
V
600-nil-lu
atuartoqar-luni
However, as we can see under this approach the position of the coordinator seems to
be irrelevant at the syntactic level. Because of the independence of syntactic and semantic
levels, the coordinator is treated as an operator at the semantic level. They talk about
percolation at the syntactic level and, as they recognized, percolation is the main feature
of headship; therefore, they define coordinate constituents as co-heads, but avoid explicit
use of the concept of headship because it implies notions such as functor, subcategorized,
79
morphological locus, government, and concord, which may be independent of percolation
(Yuasa and Sadock 2002:fn3). In that sense, it appears that a coordinator is a marker and
not a head in their conception.
2.1.6
A P&P approach (Borsley (2005))
“I hope that I have contributed to progress
by showing that the ConjP analysis of
coordinate structures is a dead end”
Borsley (2005:481)
Borsley (2005) focuses on the exploration of the reasons for which the Conjunction
Phrase is rejected in frameworks outside of Principles and Parameters (P&P). Borsley’s
first observation is that in frameworks outside of P&P scholars are reluctant to accept
ConjP’s. For example, in Head Phrase Structure Grammar (HPSG) (Pollard and Sag,
1994), Lexical Functional Grammar (LFG) (Dalrymple and Kaplan, 2000) and Categorial
Grammar (CG) (Bayer 1996, Steedman 2000). Borsley (2005) argues that Conjunction
Phrases are unacceptable because they face problems that fall into the following types: a)
the distribution of coordinate structures; b) the coordinate structures with more than two
conjuncts; and c) the coordination of non-maximal projections; d) languages which
appear to have as many conjunctions as conjuncts; and e) agreement facts about
unbalanced coordination. These are summarized in what follows.
The distribution of coordinate structures is problematic because of the fact that there
is a link between its distribution and the nature of the conjuncts: The contrast in the (60)(64) indicates that “what conjuncts a coordinate structure can contain depends on where it
appears and where it can appear depends on what conjuncts it contains” (Borsley 2005:
80
463). The example in (60) indicates that the coordination is licensed for the equality of
conjuncts (Borsley 2005:463):
(60)
Hobbs bought [a book] and [a newspaper]. (DP & DP)
(61)
*Hobbs bought [a book] and [have a drink]. (DP & VP)
(62)
*Hobbs [go home] and [a newspaper]
(63)
Hobbs may [go home] and [have a drink]. (VP & VP)
(64)
*Hobbs may [go home] and [newspaper]
(VP & DP)
(VP & DP)
The skepticism of Borsley emerges from data as in (60)-(64) because “it will be
necessary for ConjP to have different sets of feature specification in different contexts
and for its specifier and the complement to have the same features in the case of non-NP
coordination and related features in the case of NP coordination” (Borsley 2005: 466)
The coordinate structures with more than two conjuncts but just a single conjunction
present a problem for a ConjP structure because it is a common assumption that a phrase
has a finite number of specifiers or a finite number of complements. Therefore, it is not
possible to generate sentence (65) without the stipulation of an empty head between the
noun Hobbs and the noun Rhodes. In addition, Borsley shows that example (65) is not a
coordinate structure with two conjuncts.
(65)
Hobbs, Rhodes, Barnes and Gunn.
The coordination of non-maximal projections undermines the analysis of ConjPs
because it is standardly assumed that specifiers and complements must be maximal
projections. Therefore, the conjuncts must be maximal projections. But there are many
illustrations in the contrary. Consider for example (66):
81
(66)
Hobbs criticized and insulted his boss.
If we assume the idea that conjuncts are maximal projections, then (66) would arise
from the deletion process of the next sentence:
(67)
Hobbs criticized his boss and insulted his boss.
But as the meaning indicates, the sentences cannot be considered to be derived one
from the other because they have different meanings. In (66) the sentence has a joint
reading, whereas (67) has a disjoint reading. Therefore, a deletion approach is not
appropriate for sentence (66).
The case of languages which appear to have as many conjunctions as conjuncts
requires an analysis where the conjunctions have quite different combinatorial properties.
The first has no specifier and takes a ConjP as complement. The second takes a specifier
and a complement. That makes the analysis undesirable. The sentence is shown in (68)
and the representation in (69) (Borsley 2005:473-474):
(68)
Et
Paul et
Michel
And Paul and
Michel
‘both Paul and Michael’
(69)
ConjP
Conj
ConjP
NP
Conj’
Conj
et
Paul
et
DP
Michel
The unbalanced coordination is split in several particular cases. Let’s take the case
where and external head agrees with just one conjunct like in (70) (Borsley (2005:475):
82
(70)
Pujdu
tam ja
Will.go-1SG there I
‘I and you will go there.’
a
and
ty.
you
(Czech)
Borsley rejects Johannessen’s idea that ConjP acquires φ and Case features from its
specifier through Spec-head agreement and agreement between a phrase and its head. His
argument is based on the observation that the agreement mechanism as conceived in
Spec-head agreement elsewhere does not do the necessary work in coordination. There
are cases where a phrase must not share either φ features or case with its specifier, like in
(71). In addition, a DP like that in (71) as a subject must be nominative but its specifier
is genitive.
(71)
[DP The children’s room] is/ *are untidy.
Borsley adds the observation that in CP, specifier and phrase can differ in number;
they can differ in case too. In short, he concludes that “there is not evidence that
independently motivated mechanisms will ensure that ConjP acquires φ and Case features
from its specifier” (2005: 477).
2.1.7
Summary
This literature review shows that even a single (but not easy) question, such as what is
coordination? would have different answers according to the framework that we adopt.
So it is not strange that two related and central questions that could elucidate the
coordination phenomenon are still under debate: Is the structure of coordination
hierarchical or flat? Are the conjunctions syntactic heads or not? The answer to the first
question has adherents on both sides. The conception that coordinate constructions are
structurally asymmetric began with Ross (1967) and continues to the present with
83
researchers such as Abeillé (2003), Camacho (2003), Johannessen (1998), Sag et al
(1985), Kayne (1994), and Munn (2000), among others, while others conceive that
coordinate constructions are flat: Peterson (2004), Yuasa and Sadock (2002), Dalrymple
and Kaplan 2000, Sag and Wasow (1999), among others. The second question is relevant
as well, and some specialists hold that it is a head or a weak head: e.g. Johannessen
(1998), Abeillé (2003), Camacho (2003), and Gáspár (1999); while some others deny this
claim: e.g. Borsley (2005), Peterson (2004), Yuasa and Sadock (2002), Cormack and
Smith (2005), Dalrymple and Kaplan (2000), and Bresnan (2000).
Some other issues that emerge from this literature review are established as the
following questions: Is coordination propositional in nature? Do some conjuncts function
as adjuncts? What would be a better way to approach the so called UBC or Pseudo
subordination? Do languages conjoin only maximal projections? What are the properties
of coordination that any theory should explain? What could be a promising framework
for approaching to the coordination phenomenon?
Given the intricate nature of the field and the multiple sides of coordination, as the
previous ideas indicate, my work is centered upon the following question: What
properties does Yaqui have that can contribute to answer some of those persistent and
important questions? There are three main aspects of Yaqui coordination that I consider
important to describe and analyze.
1. Sentence coordination poses several challenges because of their patterns. The
coordinator into ‘and’ can occur in several positions. These not so common patterns ask
for a clarification about what the structure of Yaqui coordination actually is.
84
2. Yaqui shows, in general, Ordinary Balance Coordinated constructions; however, it
has some examples of nominal UBC and verbal UBC. Since Johannessen’s (1998)
research these structures enter completely into a theoretical discussion that still does not
end. The description of the Yaqui structures will enrich the field. Moreover, an
explanation of them in OT will give us the opportunity to test this theory on these issues.
3. The language has ‘unexpected’ patterns of coordinate noun-verb agreement on
number which, for their account, seem to require the splitting of number features into two
types: CONCORD and INDEX features, as suggested by Halloway King & Dalrymple
(2004).
The next three chapters treat these three general topics of Yaqui. A description
and a theoretical account in the OT framework are presented in each chapter.
85
3
THE STRUCTURE OF COORDINATION
“It thus appears that the constituent structure
of coordinating constructions is much more
problematic than has been generally thought”
(Haspelmath 2004:9).
3.1
Sentence Coordination
The aim of this chapter is to describe, analyze and explain the sentence coordination
patterns found in Yaqui. The discussion is centered in the coordinator into(ko) ‘and’. It
shows unusual patterns which present a challenge to any theory of language which treats
coordination as containing a tripartite structure.
3.1.1
Distribution of into ‘and’
The coordinator into ‘and’ can occur basically in three different positions when
conjoining two sentences: at the beginning of the second conjoined sentence, after the
first element on the second conjoined sentence, and in final position of the second
sentence. Let’s begin with the pattern of into ‘and’ in second position, given that this can
be considered the unmarked pattern of Yaqui coordination.
3.1.1.1 Into ‘and’ in second position
The basic patterns of sentence coordination where into ‘and’ occurs in second
position are shown in this section. Given a question like (1), a possible answer is given in
86
(2). As we can see, it is a coordinate sentence where the coordinator appears after the first
element of the second conjoined sentence5:
(1)
jitá
yeu
siika?
What out
go.SG
‘What’s going on/ what happened?’
(2)
[ju’u chu’u misi-ta
ke’e-ka],
[DET dog cat-NNOM.SG bite-PST]
‘The dog bit the cat and Diana hit it.’
[Diana
[Diana
into a= beba-k].
and 3NNOM.SG=hit-PST]
Because the answer contains only new information (Choi 2001), this can be
considered the unmarked pattern for the coordinator position. As can be seen from the
examples too, the unmarked word order is SOV. Other possible answers to the question
in (1) show the coordinator in second position too. In the following examples, the
coordinator appears after the subject of the second conjoined sentence:
(3)
[Pablo ji’osiam
jinu-k],
[María into yokia-ta].
[Pablo book
buy-PST],
[Mary and pen-NNOM.SG]
‘Pablo bought a book and Mary a pen.’
(4)
[inepo Diana-ta
bicha-k,], [apoik
achai into
[1SG Diana-NNOM.SG
see-PST], [3SG.POSS father and
‘I saw Diana and her father (saw her) too.’
(5)
[empo yeewe-k],
[inepo into
[2SG play-PST]
[2SG and
‘You played and I slept.’
ketchia].
too]
kocho-k].
sleep-PST]
5
This pattern emerges when the conjoined sentences contain different subjects. If the subject is the
same, the coordinator occurs between the conjoined sentences. The following example shows that a
correferential pronoun occurs after the coordinator (see section 3.1.1.2. for more evidence).
(i)
[Aapo
kuchureo]
[into
aapo
[3SG
fisherman]
[and
3SG
‘Hei is a fisherman and hei is a shoemaker.’
bochareo].
shoemaker].
87
In the above examples the position of into ‘and’ is obligatory. So, the following
sentences where the coordinator appears between both conjuncts are ungrammatical:
(6)
[*ju’u chu’u misí-ta
ke’e-ka],
into
[ DET dog cat-NNOM.SG bite-PST]
and
(‘The dog bit the cat and Diana hit him.’)
[Diana a=beba-k].
[Diana 3NNOM.SG=hit-PST]
(7)
*[Pablo
ji’osiam
jinu-k],
[Pablo
book
buy-PST]
(‘Pablo bought a book and María a pen.’)
[Maria yokia-ta].
[Mary pen-NNOM.SG]
(8)
[*inepo
Diana-ta
bicha-k],into [apoik
[1SG
Diana-NNOM.SG.
see-PST], and [3SG.POSS
(‘I saw Diana and her father as well.’)
(9)
[*empo
yeewe-k],
[2SG
play-PST]
‘You played and I slept.’)
into
and
into
and
achai ketchia].
father too]
[inepo kocho-k].
[1SG sleep-PST]
All the previous examples contain the subject before the particle into ‘and’, i.e. they
are NPs. However, that is not the only category that can go before into ‘and’. In what
follows it is shown what kind of elements can go before the coordinator. Most examples
are taken from a glossed story narrated in Crumrine (1961). I decided to use this kind of
material in order to get the coordination meaning from a broader context other than that
in isolated sentences. The data were checked with a Yaqui speaker from Casas Blancas,
Sonora, and the spelling was modified according to the one used in this work. Where the
speaker disagreed in any aspect of the Yaqui sentences found in these stories, it is shown
in a footnote. The next two examples show that into(k)6 ‘and’ can occur after adverbials
such as ian ‘now’ and kaa ‘not’. In these examples the subject was introduced in the first
6
There are three allomorphs of this coordinator: into, intok, and intoko. These are treated in a separate
section, the meaning of these allomorphs in the examples given here is ‘and’.
88
coordinated sentence (which is not presented here in order to focus on the position of the
coordinator).
(10)
(nii
juya)... [ian intok ujúyoisi
sawa-k].
this tree [now and
beautifully
leaves-POSS]
‘This tree … and now it’s beautiful with leaves.’ (Crumrine 1961:13)
(11)
(ilí
chu’u)..[kaa intok a’a=jajáse-ka]
[intok a’a=ta’áru-k].
(little dog)… [not and
3NNOM.SG=follow-GER][and 3NNOM.SG=lost-PST]
‘(The little dog) is not following it and lost it.’ (Crumrine 1961:19)
Few examples show into ‘and’ after the determiner of a nominal phrase, the only two
attested in Crumrine’s (1961) Yaqui stories are the following7:
(12)
[ií
into o’óu im mesa-ta
bepa juka kuj
kutá-ta]
[this and man here table-NNOM.SG upon this cross wood-NNOM.SG]
[toó-siká].
[leave-go.PL]
‘And this man has laid the rosary wood on top of the table (and) left.’ (Crumrine
1961:24)
(13)
[ií
into yoéme jak-su-ma
yeu siika],
[jaibu
[this and
man
where-INT-there
out go.SG.PST] [already
aman aánne-ka
jum kuj-ta
bepa a’a=
yonder be-SUB
there cross-NNOM.SG
on
3NNOM.SG=
‘And this man, coming out from somewhere again, is again there by
(Crumrine 1961:35)
juchi
again
kate-k].
sit-PST]
the cross.
Until now, we have several observations that must be incorporated into any analysis
of Yaqui coordinated sentences: a) the unmarked order for the coordinator into ‘and’ is
second position, b) the Yaqui unmarked word order is SOV, c) into ‘and’ can occur in
second position if the coordinate sentences contain different subjects. Before into ‘and’
7
My consultants consider these two sentences ungrammatical. For them the coordinator must be after
the full NP. I put them here in order to have a more complete register of coordination in Yaqui. The
examples show that at least historically that position could be occupied by a coordinator. This position is
not considered in my further analysis.
89
several types of elements, can appear, the exemplified ones are: nouns, adverbials, and
determiners.
3.1.1.2 Into ‘and’ in first position
The coordinator can appear in first position in several cases: First, when the subject of
the second sentence doesn’t appear in overt syntax, as in the following example, where
the subject of both sentences is the same. I used a Ø symbol to indicate that the subject is
not present in overt syntax. The sentences (15) and (16) can be an answer to the question
in (1), repeated here for convenience as (14).
(14)
jitá
yeu
siika?
what out
go.SG
‘What’s going on/ what happened?’
(15)
[Joan chu’u-ta
beéba-k]
[John dog-NNOM.SG hit-PST]
‘John hit the dog and hit the cat.’
[into Ø8
[and Ø
miísi-ta
beéba-k].
cat-NNOM.SG hit-PST]
Second, the coordinator must appear too in first position when we have a coordinated
XP (a coordinated subject in this example) in the second sentence9:
(16)
[Yoeme
bwiíka]
[into [Peo into
[Man
sing.PRS]
[and Peter and
‘The man sings and Peter and Diana dance.’
Diana]
Diana
(17)
*[Yoeme
bwiíka]
[[Peo into Diana] into
[Man
sing.PRS]
[[Peter and
Diana] and
(‘The man sings and Peter and Diana dance.’)
ye’e-mme].
dance.PRS-3PL]
ye’e].
dance.PRS]
8
I suggest that a null pronoun occurs after the coordinator. The evidence is seen in sentence (19) in
this section. In it, a correferential overt pronoun with the previous subject appears after the particle into
‘and’.
9
A reviewer made the suggestion that this effect might be the result of a processing constraint to avoid
garden path. This observation seems to be on the right track. The constraint responsible of this effect would
be undominated in Yaqui because the order of the constituents in (16) obligatory.
90
There are semantic effects related to the position occupied by into ‘and’ in the
sentence. These effects can be seen when the coordinated sentences contain similar
subject pronouns. The into ‘and’ particle must be used in first position in order to indicate
that the subject in the second conjoined sentence is the same as the one in the first
sentence. Look at the following contrast between (19) and (21), where two copulative
sentences are conjoined. The first one could be an answer to a question where we ask
something about a determined person, whereas the second one could be an answer to a
question asking the occupations of several persons:
(18)
Jita-po
aapo tekipanoa?
3SG work.PRS
What-LOC
‘What does he work on?’
(19)
[Aapo kuchureo]
[into aapo bochareo].
[3SG fisherman]
[and 3SG shoemaker]
‘Hei is a fisherman and hei is a shoemaker.’
The above sentence (19) contrasts with the next in (21), which could be an answer to
the question in (20). In the answer, the coordinator is in second position and the preferred
reading is disjoint. If we consider, following Dedrick and Casad (1999), that the
coordinator into ‘and’ is a pivot for topicalization together with the proposal of Lee
(2001) that topics have the features /+PROMINENT, -NEW/, then this semantic effect is
predicted because the pronoun in the second conjunct in (19) will be interpreted as /PROMINENT, -NEW/
and does not have to be fronted. Whereas the features of the second
pronoun in (21) would be /+PROMINENT, -NEW/ and therefore the pronoun must be
fronted, appearing before the coordinator into ‘and’. In that sense, the sentence (21)
patterns with the sentences (2-5) which contain different subjects.
91
(20)
Jita-po
bempo tekipanoa?
What-LOC
3PL
work.PRS?
‘What do they work on?’
(21)
[Aapo kuchureo]
[aapo into bochareo].
[3SG fisherman]
[3SG and
shoemaker]
‘Hei is a fisherman and hej (another guy) is a shoemaker.’
The sentence (19) with into ‘and’ in first position is similar in correferential meaning
to the next one (22) where the subject is not in overt syntax10:
(22)
[Aapo kuchureo]
[into Ø
bochareo].
[3SG fisherman]
[and Ø
shoemaker]
‘He is a fisherman and a shoemaker.’
This contrast is attested in coordinate sentences with same subjects (SS) vs. different
subjects (DS). The following sentences contain intransitive verbs:
(23)
[Aapo bwiíka ]
[into aapo ye’e].
[3SG sing.PRS]
[and 3SG dance.PRS]
‘Hei is singing and hei (the same guy) is dancing.’
(24)
[Aapo bwiíka]
[aapo into ye’e].
[3SG sing.PRS]
[3SG and
dance.PRS]
‘Hei is singing and hej (another guy) is dancing.’
(25)
[Aapo bwiíka]
[into Ø
[and Ø
[3SG sing.PRS
‘He is singing and dancing.’
ye’e].
dance.PRS]
Example in (26), taken from Crumrine (1961:19), reinforces the observation that into
occurs in first position when in the discourse, the subject is understood as the same as the
previous coordinated sentence:
10
An alternative analysis is to take example (22) as constituent coordination with the structure He is
[X and Y], i.e. as [VP & VP] coordination. However, I consider that in this kind of examples there is a null
pronoun in subject position. This conception is based in the constraint proposed by Blutner and Zeevat
(2004:4), who defined it as follows: DROP-TOPIC “Arguments coreferents with the topic are structurally
unrealized”. So, for these researchers the subject tends to be dropped if previously mentioned in the
discourse. See the analysis of verbal chains in Chapter 4, example (108).
92
(26)
[…júébena wakásim
áe
áwi-ne]
[intok áe
ji’ibwa-ne],
[…much
cattle
with.it fatten-FUT] [and with.it eat-FUT],
[into jipi’ikim
júébená-ne]…
[and milk
much-FUT]…
‘Much cattle will fatten with it, and will eat it, and will be plenty of milk…’
(Crumrine 1961:19)
As we have seen above, when the subject is not present, the coordinator, in general,
can occur in first position; however, if there is not a subject but there is a topicalized or
focused element, the coordinator must be in second position. We can see this fact in the
following examples where the sentence contains a postpositional phrase (Post-P). The
coordinator can appear in second position (after the Post-P), or in first position (before
the Post-P). This is illustrated with an example adapted from Dedrick and Casad (1999):
(27)
[juchi ‘ae=koni-la
sik-aa]
[intok jo’o-t
‘a’a=siise-k].
again 3SG=circle-ADV go-PPL
and
back-LOC
3NNOM.SG=urinate-PST
‘And having going around him, it urinated on his back.’
(28)
[juchi ‘ae=koni-la
sik-aa]
[jo’o-t
intok ‘a’a=siise-k].
again 3SG=circle-ADV go-PPL
back-LOC and
3NNOM.SG=urinate-PST
‘And having going around him, it urinated on his back.’
Finally, the particle into tends to occur before some phrasal adverbs such as jumák
‘maybe’, junén ‘thus’, junuén ‘that way’, and clitics like ne(e) ‘I’.
(29)
[intok júmak ne
kaa
am =teakaate-k(o) ]
[inepo intok
3NNOM.PL=find-COND
1SG and
and maybe 1SG not
ino
tá’aruka’ate-k(o)]…
1REFL lost-COND…
‘And perhaps if I don’t find them, and if I get lost...’ (Crumrine 1961:16)
(30)
[…into junen au=jia
kaa
ama yoeka’ate-k juni’i kia]
…and thus 3SG.OBL=say.PRS
not
there escape-PST
even just
…’And thus say to him, I did not escape…’ (Crumrine 1961:23)
(31)
[…intok
junuen jum joara-po
waiwa kora-po an-si-sime]
…and
that-way there home-LOC
inside fence-LOC walk-RED-go.PRS
…‘And he is acting that way in this house inside the fence.’ Crumrine (1961:31)
93
(32)
[into=ne
kaa
jabe-m
neu
yajak
junii…]
and=1SG
not
someone-PL 1SG.OBL
come.PL.PST even
‘And even if someone does not come by’… (Crumrine 1961:17)
Summarizing this section, we observed that into ‘and’ occurs in first position under
the following conditions: a) the grammatical subject is the same in both coordinated
clauses, b) the subject of both coordinated sentences is a pronoun and there is
correferentiality between them, c) there is not a topicalized element in the second
conjunct sentence. In addition to these observations, we noted that into ‘and’ can occur in
first position before adverbials or function as a host for clitic pronouns.
3.1.1.3 Into in last position
In the next examples we can see that the coordinator can appear in sentence final
position. These data are taken too from the stories found in Crumrine (1961). The
evidence that intok ‘and’ is in final position of the bracketed sentence in (33) is supported
by the pause after intok and by the occurrence of the particle --su ‘and’ in the second
conjunct. The particle --su ‘and’ functions as a coordinator in the example given here (for
more information of the particle --su ‘and’ see section (3.1.3) of this chapter. Additional
evidence that intok ‘and’ is attached to the end of the first conjunt (as indicated by the
bracketing) comes from the fact that we have two conjoined sentences with disjoint
reference in (33). Consequently, normal conjoining intok would have to occur in second
position if it were really just between the conjuncts. So intok ‘and’ links a previous
sentence in the discourse and appears in the final position of the first conjunct.
94
(33)
...[ito am
bit-bae-o
intok] [bempo-su
bina11 botana
...[1PL 3PL.OBJ see-INTT-TEMP
and] [3SG-and
this
side
itom ane’e beas yajak...].
1PL are
of
arrive.PL.PST…]
‘And (since) we wanted to see them, and they came over to this side where we
are.’ (Crumrine 1961:21)
The following two sentences have intoko in final position, however, in the original
text from (Crumrine 1961), it had intok kía ‘and trully’ in final position12:
(34)
[ju’u
o’ou kia
au=
‘omtemta
benasi],
[DET
man just
3SG.OBL=
angry
like]
[amau a’a=
to’o simlataka], [káa au= bitchu
intoko].
[back 3NNOM.SG= leave went]
[not 3SG.OBL=look
and just]
‘The man looks as though he is angry with her, so he is leaving her behind and
does not even look at her.’ (Crumrine 1961:22)
11
For the consultant, the word is biná, however, the original text have bimá, the same happens with
botana wich in the original was bétana. With respect to the verb, the consultant used yajak instead of itóm
áaneka wich was in the original text, but it didn’t make sense to the consulted Yaqui speaker. Therefore,
the sentence presented here is a different sentence to that in the original text.
12
These sentences contained into kia ‘and really’ instead of intoko ‘and (just)’. However, they were
ungrammatical for the Yaqui speaker which helped me in the verification of this data. For him, the sentence
is perfectly acceptable if we have intoko in final position instead of into kía wich was in the original text.
He considered that it is possible to use intok kia between the coordinated sentences, as shown next:
(i) [ju’u
ó’óu kía
au=‘ómtemta
benási], [amáu a’a=tó’o
[DET
man
just
3SG.OBL=angry like]
[back 3NNOM.SG= leave
símlataka],
[intok kía
káa áu
bitchu]
look.PRS]
went]
[and
just
not 3SG.OBL
‘The man looks as though he is angry with her, so he is leaving her behind and does not even look
at her.’
(ii) ju’u
DET
a=téaka-me
3NNOM.SG-owns-REL
[tuisi
[well
a’a=
súa-e]
3NNOM.SG=care-IMP]
[into
[and
náke-ka
love-GER
[á’a= bitchá] [intok kía
káa-beta
áma
kíkimútúa ].
just
no-one
there
go.in.allow]
[3NNOM.SG=see] [and
‘The owner (said): take well care of it and loving it, see it and just don’t let enter anyone.’
95
(35)
ju’u
a=téaka-me
[tuisi a’a=súa-e]
[into náke-ka
DET
3NNOM.SG=owns-who well 3NNOM.SG=care-IMP and love-GER
á’a=bitchá]
[káabeta
áma kíkimútúa
intoko].
3NNOM.SG=see.PRS [no.one
there go.in.allows and.just]
‘The owner (said): take well care of it and loving it, see it, and just don’t let enter
anyone.’ (Crumrine 1961:23)
The next example shows how the coordinator can occur in final position; both
sentences have the coordinated particle in final position. It is important to observe that
the coordination in the last sentence is intoko,13 whereas it can be into ‘and’ after the first
sentence:
(36)
[…ápo a’a=
tú’ute
into] [a’a=bá’atúa
and] [3NNOM.SG=waters
[…3SG 3NNOM.SG= clean
‘…And he cleans it up and he waters it.’ (Crumrine 1961:23)
intoko].
and]
An important question is why and when intoko occurs in final position. The answer is
related to the several functions that into(ko) ‘and’ can take: the above examples show that
into(ko) in adition to express coordination, introduces an additional meaning: ‘and just’.
It functions as adverb too.
Another important fact is that the coordinator can appear in final position of a
coordinated sentence, and the particle boetuk ‘because’ introduces the sentence:
(37)
13
[inime kábuim
ne
am=tá’áya]
[boetuk ne júébenasi
[because 1SG many.times
[these mountains
1SG 3NNOM.PL=know]
‘ám= bit-la
intoko].
3NNOM.PL=seen-PFV and]
‘This mountains, I know them and because I have been through here so many
times.’ (Crumrine 1961:23)
The consultant used the particle intoko in these examples; however, the original text only has intok.
This could be due to variation in the use of Yaqui language. The text from Crumrine was published in
1961, and the speaker was giving the information in 2000-2002. There are several generations of difference
between consultants.
96
(38)
...[bóetuk bá’a
jú’ebenáe-ka-n
intok] [ité
intok
aet
…[because water much.was
and] [1SG moreover on.it
kia
jiba paséalo
restémcha
réjtem]…
just
always joy.wandering
traveling.as traveled]
…’And because much water was there. And moreover we traveled about it just as
though always joy-wandering…’ (Crumrine 1961:38)
Finally, into can be in final position accompanied by elements which usually go in
initial position too: into juchi ‘and again’, and ian into ‘and now’. Look at the contrast
inside of the example (39). It shows the occurrence of juchi ‘again’ and int-uchi ‘and
again’ in final and initial position. The example (40) shows ian intoko ‘and now’ in final
position.
(39)
[pá’akun yéu nóité-ka]
[kúj-taú
sejtul nóité-ka int-uchi]
[outside out go-GER]
[cross-DIR
once go-GER and-again ]
[juchi nótte-ka]
[ínt-uchi
bemélasi
júchi
nótte-ka]
[again return-GER] [and-again
anew
again
return-GER]
[hum puétau
í’an kikte-k
int-uchi].
now door
now stands-PST
and-again]
‘Again returning, and again one more time returning, and again now he stands in
front of the door.’ (Crumrine 1961:24)
(40)
[ápo a’a
éa-po
a’a= páttáika]
[áma
3SG 3SG.POSS
will-LOC
3NNOM.SG=closed
there
[káá yéu
wéye ían
intoko].
not
out
walk now and
‘He determined to close it, and now he doesn’t want to come out.’ (Crumrine
1961:32)
Summarizing: into(ko)14 can be in final position under certain circumstances: a) when
it functions more like an adverbial than as a simple coordinator; in this case, it means
14
It is important to mention here that intoko can be considered the full form of the coordinator. In
Yaqui it is common for full forms emerge in final position or in isolation. For example, the demonstrative
inii’i ‘this’ in (i) must appear in its full form because it is at the end, after jitasa ‘what’; however, if it is not
at the end it can be in its short or in its full form as indicated in (ii).
(i)
¿jitása inii'i?
(ii)
¿inii(‘i) jitása?
What is this?
What is this?
97
‘and just’, b) when another particle such as boeytuk ‘because’ introduces the sentence, it
has to be in final position. It is common to find the coordinator in final position forming a
kind of compound such as int-uchi ‘and again’15.
3.1.2
Other uses of the particle into
In general, we can say that into(ko) functions as a coordinator, however, as we
perceive from the data of into(ko) in final position, the Yaqui particle is more than a
single coordinator. The next examples show that into has the meaning of ‘and more’, ‘and
more(over)’; i.e. in addition to their coordination feature, it is like an adverbial particle.
This meaning was attested in four cases: a) when intoko is final position, b) when into(ko)
is in second position, c) when into(ko) is after the negation kaa, d) and when into(ko) is
after the particle --su ‘and’. The distribution of into(ko) with this meaning is not clear
from the obtained data. I leave this matter open for now. When intoko has the mentioned
meaning and a coordinator with the ‘and’ meaning is required, it must be used with the
particle --su ‘and’, as shown in (44).
(41)
[báj-ta
juébenaku yuku-mak], [ju’u báso yu’in tobóktila intoko]16.
grass-NNOM.SG where.much rain-COM DET grass plenty risen
and.more
‘And the grass, with the rain, has come up high.’ (Crumrine 1961:20)
15
A reviewer made the comment that these cases seem to be more instances of subordination than
coordination. Although subordinators in Yaqui are in sentence final position, the examples seen here do not
have the case marking usually associated with subordination. For example, a singular subject in a
subordinated sentence is marked with –ta ‘NNOM.SG’ or, if it is a pronoun, it has to be in genitive case. But
if we look at example (37), we can see that both subject pronouns are in nominative case as expected in
coordination. However, from the data is clear that into(ko) is more than the single logical coordinator ‘and’.
16
Again, the original text only has intok, but the consultant used intoko in this construction.
98
(42)
[áma jumak ápo wáiwa bó’oka]. [Júmak ramáa-po],
[there maybe 3SG inside lying].
Maybe ramada-LOC
[káá into bó’o-bae-kai].
[not more lie-INTT-SUB]
‘He is inside lying down; not wanting to lie down in the ramada anymore.’
(Crumrine 1961:29)
(43)
[bóetuk
bá’a jú’ebenáe-ka-n
intok].
[Ité
intok
[because
water much-PST-CONT
and].
We
moreover
áet
kia
jiba paséalo
restémcha
réjtem]
on.it just
always joy.wandering
traveling.as
traveled
…’And because much water was there. And moreover we traveled about it just
as though always joy-wandering…’ (Crumrine 1961:38)
(44)
í’án-su intok empo káa im yúm
jó’e-báe-te-k (o)
now-and more 2SG not here tiredness rest-INTT-COUNT-COND
‘And now if you don’t want to rest here even.’ (Crumrine 1961:16)
3.1.3
Other particles that indicate ‘and’ coordination
júni’i
even
There is another particle used to indicate a continuation in the discourse which has a
similar meaning that the particle into has. It is the particle --su ‘and’, however its use is
restricted to the following contexts:
It is a suffix and it is affixed to nominals and pronominals. It is used on interrogative
and declarative constructions:
(45)
inepo-su
empo-su
aapo-su
itepo-su
eme’e-su
bempo-su
‘and I’
‘and you’
‘and (s)he’
‘and we’
‘and you’
‘and they’
(46)
inepo kaa
ye-yena,
¿empo-su?
1SG not
RED-smoke, you-and
‘I don’t smoke, and (do) you?’
99
(47)
[ito
am=
bit-bae-o
intok] [bempo-su biná17 botana
[1PL 3NNOM.PL
see-INT-when and]
2PL-and
this
side
itom ane’e beas yajak...].
1PL
are
of
arrived.PL.PST
‘And since we wanted to see them, they came over to this side were we are…’
(Crumrine 1961:21)
It was found affixed to adverbials, such as ian ‘now’ and che’awa ‘much’. The
meaning can be that of ‘more(over)’ attested with the particle into ‘and’.
(48)
i’an-su intok
empo kaa
im
yum
jo’e-bae-te-k(o)
now-and more
2SG not
here tiredness rest-INTT-COND
‘And now if you don’t want to rest here even.’ (Crumrine 1961:16)
juni’i.
even
(49)
in
kaba’i che’awa-su awi lobolai intok pappeya...
1SG.POSS
horse much.more fat
round and
active…
‘My horse is much fatter and round and is very active…’ (Crumrine 1961:20)
3.1.4
Setting the problem
There are several interesting aspects about Yaqui coordination. In this section I am
going to focus on three central aspects: a) into ‘and’ breaks the unity of the second
coordinated sentence, this aspect is a problem for theories which suggests that the
coordination has a flat structure, b) into ‘and’ has adverbial characteristics. I will show
that it shares several properties that other Yaqui adverbials have. Therefore, we can
consider that into ‘and’ is adjoined to the sentence where it appears. This aspect is
important if we want to explain appropriately the distribution of the into coordinator in an
OT framework, c) into can co-occur with other coordinators. This fact suggests that this
17
For the consultant, the word is biná, however, the original text have bimá, the same happens with
botana wich in the original was bétana. With respect to the verb, the consultant used yajak instead of itóm
áaneka wich was in the original text, but it didn’t make sense to the consulted Yaqui speaker. See the
footnote (2) of this chapter.
100
particle has characteristics of a subordinator and/or compound. This aspect shows that the
limits between coordination and subordination in Yaqui are not always clear cut.
3.1.4.1 Into ‘and’ breaks the unity of the coordinated sentences
As we have seen before, the Yaqui logical coordinator into ‘and’ occurs in unusual
patterns which present a challenge to theories that suggest that coordination has a flat
structure (among others: Naijt 1979, Peterson 2004, Yuasa and Sadock 2002), such as
shown in (50).
(50)
C
X
C
X
Where C stands for coordinator and X for maximal projections.
We have seen that a very common way to coordinate two sentences in Yaqui is that
indicated in (51), where the coordinator appears within the second sentence; more
precisely, after the subject or after a topicalized element. In other words, the coordinator
breaks the unity of the second sentence. As examples (51) and (52) indicate, the
coordinator cannot appear between the two sentences. Therefore, a flat structure like (50)
does not seem to be appropriate for (51):
(51)
[Maria
[Joan bwika-k]
[Mary
[John sing-PST]
‘John sang and Mary danced.’
(52)
*[Joan bwika-k]
into [Maria ye’e-ka].
*[John sing-PST]
and
[María dance-PST]
‘John sang and María danced.’
into
and
ye’e-ka].
dance-PST]
In section 3.2.3.6 of this chapter I am going to argue that into does not have clitic
properties that would account for its second position placement in a sort of surfacy,
101
morphophonological way. That is, the second-position placement of into is a syntactic
fact, not a morphophonological one. The second position is clearly a position that follows
a topic slot of some kind. Clitics don’t care whether they attach to topics or some other
kind of XP.
3.1.4.2 into ‘and’ is like other adverbials
Another important characteristic of the particle into(ko) ‘and’ is that it has adverbial
properties. Remember that it has sometimes the meaning of ‘and more(over)’. When we
look at the adverbial particles, we realize that many of these particles have the same
distribution as into, in what follows I give evidence of this distribution.
Sentential adverbials tend to occur in second position, for example, the adverb jumak
‘maybe’:
(53)
intok júmak
And maybe
ne
1SG
káa
not
am=
3NNOM.SG=
téakaate-k(o)
find-COND
inepo
intok ino
tá’aruka’ate-k(o).
1SG
and
REF
lost-COND
‘And perhaps if I don’t find them, and if I get lost...’ (Crumrine 1961:16)
(54)
ini
jumak
chú’u
This maybe
dog
‘Maybe this is a dog’ (Crumrine 1961:18)
Other particles tend to occur in sentence final position, including the following: jajáni
‘perhaps’, ja’ani ‘somehow’ (expresses doubt), juni(‘i) ‘even’, and o’oben ‘nevertheless’,
jumaku’u ‘probably’:
(55)
ilí
pánim
o
jitasa
puáto-ta
jajáni.
little bread
or
whatever
plate-3NNOM.SG
perhaps
‘There is perhaps a little bread, or something on the plate.’ (Crumrine 1961:18)
102
(56)
nií
wíkit juma techóe
ja’ani.
this
bird might do-bad-omen somehow
‘This bird might be of bad omen somehow.’ (Crumrine 1961:35)
(57)
...(uusi-m)… yum
jo’oe-bae-te-k
juni…
…(boy-PL)... tiredness
rest-INTT-PST
even
…’(boys)… if you want to take a rest...’ (Crumrine 1961:14)
(58)
ite
jumak [wói-ka]
[[káa báe-ka]
juni’i] nábuhtia kátne.
1PL
maybe two-GER
[[not want-GER]
even beyond go-FUT…
‘Maybe we, there being two of us, even if we don’t want to, will go beyond…’
(Crumrine 1961:14)
(59)
[puéta-ta
ala
etapóka
o’oben]
[ta
bea].
Door-NNOM.SG
is
opening
nevertheless but
already
[into kaá
eu
weáma]…
[jaisa jumák jume
auláta-kai].
and
not
out
walk
what might the
have be-SUB
‘Nevertheless he opened the door but he doesn’t want to walk out…’ (Crumrine
1961:32)
(60)
[két
né
hunen hiáu-su]
[=hú sénu yoéme
i’im
Yet
1SG thus saying-and] [=DET one
man
here
táawa-báe
jumakú’u].
remain-POT
probably
‘And now, as I was saying, one man probably wants to remain here.’ (Crumrine
1961:14)
As shown above, adverbials such as junii ‘even’ or juni’i kía ‘even just’ goes in final
position. Actually, in the following sentences into ‘and’ and those particles cannot cooccur in final position (if we try to use the full form intoko with júnii or juni’i kía in final
position, the sentences (61) and (62) becomes ungrammatical). In other words, into has to
occur in first or second position, but not in last position.
(61)
[into=né
káá
jabe-m
néu
yaják
not
someone-PL 1SG.OBL
come.PL.PST
and=1SG
‘And even if someone does not comes by…’ (Crumrine 1961:17)
(62)
[[into júnén aú = jía]
káá
áma yóeka’ate-k
And
thus 3SGOBL=say not
there escape-PST
‘And thus say to him, I did not escape.’ (Crumrine 1961:23)
júnii]…
even
juni’i kía].
even just
103
3.1.4.3
Into ‘and’ can co-occur with another coordinators
With respect to the occurrence of into with other coordinators, we can see that it can
appear with (bwe)ta ‘but’ which is a logical coordinator and which position is restricted
to occurring only between two full sentential clauses. This fact tells us that the function
covered by into in these cases is not that of a real coordinator. According to Oirsow
(1987) two real coordinators cannot co-occur. These facts show that subordination and
coordination are not clear cut in the Yaqui language, or that into ‘and’ leads a double life
as both coordinator and subordinator.
(63)
[iní’i má
ó’ou ían
kábái-ta
áma yéu
tójak
there out
bring-PST
This so
man now horse-NNOM.SG
jum kóra-po]
[intok á-et
ja-já’amu]
there fence-LOC
and on.it
RED-mounting
[ta
intok ket
juni únna wákíla]
but
and
yet
even very skinny
má
chea káá
‘a=túa
yúumaka.
so
more not
3SG.OBL= truly
unable.to.carry
‘This man now brought the horse inside the fence and is attempting to mount it
but it (the horse) is very skinny so the horse is unable to carry him.’ (Crumrine
1961:37)
(64)
into inwain mesa-po
juka
jí’ik
And hither table-LOC
DET.NNOM.SG
needle
wáata
máne-ká-m-ta
jáiwa
ta
basket
stands-PST-NMLZ-NNOM.SG searching
but
ju’u áma wáate-wa-me
into ka=jita
áma
DET
there want-PASS-NMLZ
and no-something there
‘She came up here to the table searching for the basket where it
she wants in the basket is not there.’ (Crumrine 1961:24)
(65)
intok
and
áu-k.
exist-PST
stands, but what
náiya’a-báe-m-ta
benásia
tá
into jumak
Burn-INTT-NMLZ-NNOM.SG like
but
and
might
jume báji ibáktim
káá
juébena.
those three armfuls
not
enough
‘As though he might want to build a fire, but maybe the three armfuls are not
enough.’ (Crumrine 1961:33)
104
3.2
Proposal about the structure of coordination
In this section I propose that Yaqui coordination patterns can be explained if we adopt
a set of alignment constraints, faithfulness constraints, and markedness constraints,
together with the idea that into ‘and’ is an adjunct that attaches to an XP category.
3.2.1
Background
3.2.1.1 The syntactic structure is not flat
Let us begin with a single definition of coordination taken from Dik (1997:89).
Within a functional framework, Dik defines coordination as “a construction consisting of
two or more members which are functionally equivalent, bound together at the same level
of structure by means of a linking device”. As a general symbolization he proposes the
following schema (Dik 1997:89-90):
(66)
CO
M1
&
M2
&…& Mn
Where “CO” is the coordination as a whole, the “M”s are the members (n>1), and
“&” symbolizes the “linking device” by means of which members are combined.
As we can see, those members are combined at the same structural level, i.e., none of
the members M is in any way subordinate to, or dependent on any of the others. They are
all on a par, and equal members of the coordination CO.
Dik considers that the coordination patterns in languages are adjusted to the following
possibilities, depending on the prepositive or postpositive nature of coordinators (Dik
1997:191):
105
Prepositive nature:
(67) a.
M1 CO M2
b.
M1 CO M2 CO M3
Pospositive nature:
c.
M1 M2 CO
d.
M1 M2 CO M3 CO
In order to see that Yaqui sentence coordination does not fit to any of those predicted
patterns, I repeat here a sentence with the coordinator into ‘and’ after the subject of the
second clause. As we saw before, clausal Yaqui coordination (in the unmarked case) is
as follows:
(68)
[Joan bwiika-k]
[María into
[María and
[John sing-PST]
‘John sang and María danced.’
ye’e-ka].
dance-PST]
Therefore, flat structures as those given in (66) seem to be inappropriate for Yaqui
sentence coordination.
3.2.1.2 Typological description of Yaqui coordination
In a more recent work, Haspelmath (2004), from a typological view, establishes that
languages of the world show asymmetric coordinated structures. He postulates four
logical types for monosyndetic coordination. They are listed below in descendent order of
linguistic frequency. According to him, the fourth type does not seem to occur in any
language of the world and the third type is very rare. Interestingly, Haspelmath (2004)
does not even consider the existence of a symmetric coordinated structure. (Haspelmath
2004: 6):
(69)
a. [A] [co B]
e.g. Hausa
Abdù dà
‘Abdu and
Feemì
Femi’
b. [A co] [B]
e.g. Lai
vòmpii=lee
‘A bear and
phèŋtee
a rabbit’
106
c. [A] [B co]
e.g. Latin
senatus populus-que romanus
‘The senate and the roman people’
d. [co A] [B]
From this perspective, Yaqui uses structures like that in (69a) and (69c). This claim is
supported in what follows:
Haspelmath (2004:7) mentions the following criteria for determining the constituency
of coordinating constructions: clisis, intonational phrasing and extraposition. Using these
criteria we get the following results for Yaqui:
Clisis This criterion requires that the coordinator be “clearly phonologically attached
to one of the coordinants, either as a proclitic or as an enclitic” (Haspelmath 2004: 7).
The next example indicates that into ‘and’ is the host for the negative particle e’e ‘not’ in
Yaqui. The subject of the second sentence is before the coordinator. The example clearly
indicates that the coordinator goes with the second conjunct. The pattern is that of (69a)
or (69c).
(70)
[María ejkuela-u
siika],
[Peo int-e’e].
go.SG.PRS], [Peter and-not]
[Mary school-DIR
‘Mary went to school and Peter did not.’
Intonational Phrasing This criterion establishes that “when the coordinators are
short, a coordinating construction A co B is pronounced as a single intonation phrase, but
when they are longer (e.g. two full clauses), there is usually an intonation break between
them, and the coordinator is then either attached at the beginning of the second phrase or
at the end of the first phrase. The intonation break is indicated by a comma” (Haspelmath
107
2004:7). The next example, of two coordinated sentences, indicates that after the break
the coordinator and the second conjunct form a unit:
(71)
Jabé
biba-m
jinu-k, [into jabé
vino-ta
Someone
cigar-PL
buy-PST [and someone wine-NNOM.SG
‘Someone bought cigars and someone bought wine.’
jinu-k].
buy-PST]
Extraposition This criterion requires checking whether the language allows
“extraposition to the end of the clause, so that, the construction is no longer continuous”
(Haspelmath 2004: 7). The following contrast indicates that in Yaqui the coordination
can be continuous or discontinuous. If it is discontinuous, the coordinator always goes
with the second conjunct. Therefore, the pattern of (73) is that shown in (69a):
(72)
inepo [kowí-ta
into misí-ta]
bwuise-k.
cat-NNOM.SG] grasp-PST
1SG [pig-NNOM.SG and
‘I caught the pig and the cat.’
(73)
inepo [kowí-ta]
bwuise-k
1SG [pig-NNOM.SG]
grasp-PST
‘I caught the pig and the cat.’
[into
[and
misí-ta].
cat-NNOM.SG]
The survey of Yaqui coordinated structures indicates that it is asymmetric. The
coordinator always goes with the second conjunct. These examples indicate that Yaqui
uses both types of structures (4a) and (4c). Any account of Yaqui must reflect this
property of coordination. In other words, we have to consider the internal configuration
of coordinated constituents.
3.2.2
Alternatives for the structure of coordination
One thing that has to be clarified is the structure of coordination. Across time
researchers have proposed several alternatives. In this section I confront those approaches
108
in the light of Yaqui data. As we have seen before, Yaqui clausal coordination is the
following.
(74)
[Joan bwiika-k]
[María
[John sing-PST]
[María
‘John sang and María danced’
into
and
ye’e-ka].
dance-PST]
To repeat, the coordinator appears after the subject of the second clause. The first
type of proposal is a flat tripartite structure (see Chapter 2) which we have seen is unable
to explain sentences such as the one above.
In the second place we have proposals where the coordinator is a head. For example,
for Johannessen (1998:107) the structure of the conjunction phrase is as in (75). It is
headed by a conjunction generated from the lexicon; moreover, the conjunction needs
two arguments in order to have a saturated phrase (i.e. it needs a conjunct as a
complement and another as a specifier). So, the conjunction phrase follows the principles
of X-bar theory.
(75)
a.
CoP[X]
X
first
conjunct
b.
CoP[X]
Co’
Co
conjunction
Co’
Y
second
conjunct
Co
X
second
conjunct
first
conjunction
conjunct
Johannessen (1998:175-76) suggest that conjuncts are attached to CoP by a
transformation: coordinate-alpha. This operation is general in the sense that it can
coordinate any category with any other category at any stage in the syntactic derivation.
From her point of view, a clausal coordination is the union of two CP’s.
109
An important assumption is that the input structures on which coordinate-alpha
operates are fully projected CPs. She gives the example (76) of clausal coordination and
shows its derivation in (77) (Johannessen 1998:177). Because she follows the minimalist
approach, whether the CPs were attached to the CoP before or after their separate
derivations is impossible to tell; either is possible. In other words, the two CPs might
have been underived or derived at the point where CoP attached to them.18. The CoP,
strictly speaking, is now a CoP[CP]. As we can see, the derivation follows the principles
of X-bar theory and has implicit principles of deletion.
(76)
Mary saw a mouse and Martha heard an elephant.
(77)
CoP[CP]
Co’
CP
DP
Co
C’
Mary-i C
and
DP
VP
Martha-i C
saw-j DP
t-i
CP
V’
V
t-j
C’
VP
heard DP
DP
a mouse
t-i
V’
V
t-j
DP
an elephant
According to this theory, a prediction for Yaqui coordination is that the coordinator
will appear between the two sentences. But, as seen before, such structure is not correct:
18
As a reviewer appropriately notes, in the Minimalist Program it would be important that the CPs be
fully-derived structures before coordination applied to them, because to insert unarticulated CP nodes into
the structure and then insert material inside the CP would violate a constraint called ‘Extended Target’
which states that material may be added to a tree at its root. Because OT is not a derivational model and
Gen can generate an infinite set of candidates, this constraint can be dispensed with.
110
(78)
*Joan bwiika
into María ye’eka.
John sing.PST
and
María danced.PST
‘John sang and María danced.’
Another alternative is found in Munn (1987, 1993) who suggests that coordination
relates to a Boolean phrase. The conjuncts form a strong unit together. In the
representation, the conjunction B (for Boolean) takes the second conjunct as a
complement, and projects to a BP which is in turn a complement of the first conjunct or
adjoined to it:
(79)
NP
NP
John
BP
B
NP
and
Mary
Again, the prediction from this structure is that Yaqui would have the pattern in (78)
which does not hold in the language.
A more recent approach is that of Camacho (2003), who considers that the structure
of coordination is the following one (Camacho 2003: 52). On it, the first X represents the
conjunction, the second X any sentential functional projection, such as INFL, Agr, etc.
(80)
XP
Conj1
X’
X
XP
Conj2
X’
X
YP
111
For him, clausal coordination implies the coordination of two events. Thus a sentence
like (81) would be derived as in (82) (Camacho 2003: 56-57). In the derivation, Camacho
establishes that “the lower event head will not rises to the specifier of the higher head,
unlike in the case of adverbs. This yields independent temporal readings for conjoined
clauses but co-dependent temporal readings for adverbial coordination” (Camacho 2003:
56).
(81)
John arrived home and Mary will leave today.
(82)
EvP
Ev’
TP1
Ev
EvP2
TP2
TP2
John arrived home
and
Mary will leave
Ev’
AdvP
Ev
tTP
today
This proposal predicts again that the events will be tied in overt syntax by a
coordinator between both sentences. That is not the pattern of Yaqui coordination.
However, an interesting constraint proposed by Camacho is that events must share
speech time, a constraint that will be used in my analysis of the Yaqui coordination.
On the other hand, Agbayani & Golston (2002) explore coordination constructions
where the coordinator is in second position. They follow the idea that the coordinator is a
head (Munn 1993, Johannessen 1998, Progovac 1998, Zoerner 1999) and they agree with
the idea that the basic coordinate structure is universally the same. In their work they
112
explore three types of coordination structures that cover, according to them, all types of
coordination structures: full form, clitic form and asyndetic form. For them the enclitic
pattern is derived form asyndetic movement of the first word of the second conjunct to
the position commonly occupied by the clitic coordinator. For example, the nominal
Greek coordination in (83) is derived as shown in (84) (Agbayani & Golston 2002: 4):
(83)
Egoòn Akhileús=te
I
Achilles=and
‘I and Achilles’
(84)
CoP[X]
X
Egoòn
Co’
Co
Y
Akhileús-i=te
t-i
This kind of movement in Greek is obligatory in order to avoid an ungrammatical
construction. So, Agbayani & Golston rule out the following type of structure because the
clitic coordinator stands alone:
(85)
*Egoòn
=te
Akhileús
I
=and Achilles
‘I and Achilles’
This explanation is extended to clausal coordination. Then the next example
(Agbayani & Golston 2002: 4) can be represented as indicated below. In the
representation the verb in the second conjunct moves up and attaches to the clitic =te
‘and’, satisfying the clitic requirements of the coordinator. They do not mention if there is
a topic requirement on the element that adjoins to the clitic in the Greek construction, but
113
examples like that in (83) are probably evidence that the coordinator is prosodically
deficient and the movement is just for clitic reasons.
(86)
Epanésteesan peithontó=te
poiméni
laóon
After arose persuaded=and
shepherd
of.army
‘They arose after him and persuaded the leader of the army.’
(87)
CoP[X]
X
Co’
Epanésteesan Co
Y
peithontó-i=te
t-i poiméni laóon
Thus, from this point of view, the coordinator into ‘and’ in sentence (74) (represented
in (88)), should be considered a phonological clitic prosodically dependent in a way that
first position coordinators in the language are not:
(88)
CoP[X]
X
Co’
Joan bwiika Co
Mariai=into
Y
ti ye’eka
However, there are many reasons to reject the idea that into ‘and’ is an enclitic. If it
were an enclitic there would be two initial predictions: the first one is that two
coordinated nouns will have this coordinator in final position, as shown in (89), and,
second, that a structure with into ‘and’ between the nouns would be ungrammatical (90).
114
But contrary to predictions, (89) is ungrammatical and (90) is grammatical (the result is
opposed to examples (83) and (85) of Greek):
(89)
*inepo Joan into.
1SG John and
(‘I and John’)
(90)
inepo into
1SG and
‘I and John’
Joan.
John
There is more evidence that shows that it is inadequate to take into ‘and’ as a clitic: a)
into ‘and’ can appear in first position (i.e. between clauses. See data in section 3.1 of
Chapter 3), an unexpected behavior if were a clitic; b) the coordinator into ‘and’ has
stress by itself and consists of a minimal word in Yaqui (bimoraic trocaic foot). i.e. it is
not prosodically deficient. This is contrary to Yaqui clitics which are monosyllabic and
unstressed (Escalante 1990); c) The coordinator into ‘and’ can be a host for other clitic
particles: ex. /into e’e/ ‘and not’ > inte’e ‘and not’ (see sentence (70) in this chapter); d)
Coordinated noun phrases can be discontinuous, and when that happens, the coordinator
always appears with the second conjunct, crucially, preceding the second conjunct. That
fact indicates that it does not form a unit with the first conjunct (in other words, it is not
attached to the first conjunct and can not be a clitic); e) it is not the closest element in the
second conjunct which “moves” to first position in Yaqui. The position can be occupied
by any topicalized element; finally f) the coordinator into has several allomorphs: into,
into-k, into-k-o ‘and’. The last one is used in clausal coordination and tends to appear
more frequently in final position (i.e. after the second sentence). These claims are
supported in the next section.
115
If intoko ‘and (moreover)’ tends to occur in final position, then, from Agbayani and
Golston’s (2002) proposal it is the entire CP which has to move in order to satisfy the
clitic requirements of intoko:
(91)
Joan bwiika-k,
María ye’e-ka
intoko.
Mary dance-PST
and
John sing-PST
‘John sang and (moreover) Mary danced.’
(92)
CoP[CP]
CP
Co’
Joan bwiika, Co
Maria ye’eka-i=intoko
CP
t-i
This movement seems to be inappropriate for Yaqui: the particle into ‘and’ is not a
clitic, therefore, if movement happens, it has to be for other reasons. Dedrick and Casad
(1999) hold that coordination in Yaqui is a pivot for topicalized items. My analysis of the
language indicates that this conception is correct. Therefore, if the coordinator into ‘and’
is an adjunct and not a head, the kind of structures present in coordinated clauses must be
different.
3.2.3 The coordinator into ‘and’ is not a head
3.2.3.1 Evidence from affixation that into is an adjunct
The Yaqui coordinators come from several sources and they present characteristics
similar to adjuncts. In the next section I show how some coordinators take suffixes. This
fact strongly suggests that the coordinator into ‘and’ must be grouped together with
adverbials (at least for the adjunction process). Although they are lexicalized, it is
116
possible to recover evidence that they were inflected. In the next section, I show the type
of affixes that into ‘and’ and other coordinators and adverbs can take.
3.2.3.2 Coordinators which take suffixes
In this work I restrict my study to analyze constructions where into(k-o) is involved.
When compared with other logical coordinators such as bweta ‘but’, o ‘or’ --ko
‘if…then’, we realized that into seems to be a lexicalized form where the verbal affixes -k and --o are attached to a coordination base. Moreover, into ‘and’ is not alone in relation
with this characteristic: other sequential coordinators present this special property.
If we review the sentences in this chapter, we can find that there are three allomorphs
for into ‘and’: into, intok, and intoko. The distribution of them is not clear and it seems
that the differences are blurred. However, there is a preferred position for the allomorph
intoko ‘and’ in final position. In this final position, the allomorph usually has the
additional meaning ‘and just’, ‘and moreover’. Examples with the allomorph into, intok
and intoko19:
(93)
in
uusi tajkaim
bwaka,
in
1SG.POSS
son
tortillas
eats
1SG.POSS
into ‘a=
bitchu.
and 3NNOM.SG= see.PRS
‘My son is eating tortillas and my brother is looking at him.’
(94)
jume bemela
jamuchim
emo chike-k
intok
DET.PL young
women
3REFL comb-PST
and
ejpejopo
emo bicha-k.
mirror
3REFL see-PST
‘The young women combed themselves and saw themselves in the mirror.’
19
saila
brother
The allomorph intoko can appear between the conjuncts. See example (107) of this chapter.
117
(95)
aapo juka
bweu teta-ta
3SG DET.NNOM.SG big
stone-NNOM.SG
‘And did she charge the big stone?’
puate-k
charge-PST
intoko?
and
3.2.3.3 Origins of the Yaqui coordinators
The coordinators and subordinators in Yaqui show that they come from several
sources, the most common being demonstratives and locatives. They show inflection with
the suffixes --k, -o, -n, etc.. This is true for several adverbs which introduce sentences.
Many of these forms are in free variation, as we can see in the groups formed below:
(96)
juna’a
juna-k
juna-k-o
junak-sa-n
‘that’
‘then’
‘then’
‘then, and then’
(97)
junu’u
junu-e-n
junu-e-n-i
‘that’
‘thus’
‘really’
(98)
jun-i
jun-tu-k
jun-tu-k-o
jun-tu-k-sa-n
jun-e-n
jun-e-n-su
‘so, thus’
‘for that reason’
‘well...’
‘that is why’
‘thus’
‘that’s why’
(99)
ju’u
ju-le-n
ju-le-n-sa-n
ju-le-n-tu-k-o
ju-ma-k
ju-ma-k-sa-n
‘the’
‘that’s why’
‘that’s why’
‘for that reason’
‘it would better if’
‘may be so’
(100) jeewi
jeewi-ma
‘yes’
‘yes, then’
118
(101) chuuba
chuba-la
chuba-la-tu-k-o
chuba-tu-k-o
‘for a while’
‘momentarily’
‘in a while’
‘in a little while’
(102) ini’i
ini-a-n
ini-le-n
‘this’
‘in this way’
‘in this way’
(103) iyi-le-n20
iyi-min-su
‘in this way
‘over there’
It’s easy to see that the coordinator into ‘and’, is affixed with --k, and --o. Those are
the only suffixes which into ‘and’ can host.
(104) into
into-k
into-k-o
‘and’
‘and’
‘and (just)’/’(moreover)’
3.2.3.4 Into and the suffix -k
The exploration of the possible meaning of intoko brings us to the field of verbal
inflection. However, there is no evidence that the coordination introduces the meanings
that --k-o ‘COUNT(ERFACTUAL)-if’ introduces when affixed to a verbal root.
When the suffix --k is attached to verbal roots, it expresses perfective aspect as the
primary meaning. However, it is used for expressing counterfactual and conditional
meaning too. In such cases it is accompanied by the suffix --o ‘when’.
(105) Joan-ta
yepsa-ko,
arrive-COND
John-NNOM.SG
‘If John arrives, Mary will go to Vicam.’
20
Synonym of ini-an.
Maria Vicam-me-u
María Vicam-me-DIR
sim-bae.
go-INTT
119
This use of the affixes is not unusual, Comrie (1993:19) analyzing English establishes
that:
Although most uses of the English past tense do serve to locate situations
prior to the present moment, there are several uses that do not. One is
counterfactuals, e.g. if you did this I would be very happy, where did clearly
does not have past time reference, but refers rather to a potential action in the
present or future. For some speakers of English, there is a distinction between
the form of the verb be used in such constructions and the form of the verb
used with past time reference --cf. John was here (past time reference), but if
John were here (counterfactual present)- so that one might argue that here we
are simply dealing with two distinct but homophonous (for most verbs, or, for
some speakers, for all verbs) forms. (Comrie 1993: 19)
The analysis of intoko reveals that we cannot say that when --k is attached to into
‘and’ it adds a counterfactual or conditional meaning to the sentence, but it is clear from
the paradigm that the now lexicalized particle is composed from several morphemes.
3.2.3.5 Into and the suffix --o
Again, if we examine the primary meaning introduced by --o ‘when’ in verbal roots,
we realize that this meaning is not present when the coordinator into ‘and’ has it. The
meaning is clear when attached to verbal roots, as in the following example:
(106) Joan-ta
yepsa-o,
Maria Vicam-me-u
arrive-TEMP María Vicam-me-DIR
John- NNOM.SG
‘When John arrives, Mary will go to Vicam.’
sim-bae.
go-INTT
120
However, Dedrick & Casad (1999) establish that the conjunction “serves as the base
for attaching the conditional suffix --o it conjoins two clauses that are discourse closer as
a whole. In any event, this sentence illustrates a formulaic use of the conditional and
provides another case in which the dividing line between subordination and coordination
gets blurred” (p. 408):
(107) jiba kaita
into-k-o
junum chupu-k.
there finish-PST
Only nothing
and-COUNT-TEMP
‘There is nothing else and it ends there.’ Dedrick & Casad (1996:408)
My data do not reveal a special meaning for intoko, except that in final position of the
sentence it can mean (in addition to ‘and’) ‘and just’.
Although the coordinator has become lexicalized and currently it is difficult for a
Yaqui speaker to distinguish between the use of into, intok and intoko, it is clear that --k
and --o are affixes. These verbal affixes are attached to other elements, shown above and
grouped together in what follows:
(108) juna-k-o
jun-tu-k-o
ju-le-n-tu-k-o
chuba-la-tu-k-o
chuba-tu-k-o
into-k-o
‘then’
‘well...’
‘for that reason’
‘in a while’
‘in a little while’
‘and (just/when/moreover)’
3.2.3.6 Evidence from cliticization
Agbayani and Golston (2002) suggest that if a coordinator occurs in second or final
position, it must be treated as a clitic. However, as we can perceive next, this claim is not
supported by Yaqui. There are several reasons for this: clitics in this language tend to be
monosyllabic, as exemplified by the following pronouns and their respective clitic forms.
121
(109) Full Pronoun
inepo
itepo
aapoik
Clitic Form
=ne(e)
=te
=a
Gloss
‘I’
‘we’
‘him/it’
Into ‘and’ is itself a base for cliticization. The next examples show that into ‘and’
functions as a host for several types of particles. This process, although not obligatory, is
very common. The next examples show that into ‘and’ merges with particles such as ju’u
‘that’ and juchi ‘again’, jitasa ‘what’, juka ‘DET.NNOM.SG’, i’an ‘now’, aapo ‘3SG’, im
‘here’, um ‘there’, i’i ‘this, au ‘to him’, among others.
(110)
a) into-ju’u
into-juchi
into-jitasa
into-juka
int-u’u
int-uchi
int-itasa
int-uka
‘and that’
‘and again’
‘and what’
‘and the (NNOM.SG)’
b)
int-i’an
int-apo
int-im
int-um
ínt-i’i
Int-au
‘and now’
‘and (s)he’
‘and here’
‘and there’
‘and this’
‘and to him’
into-i’an
into-aapo
into-im
into-um
into-i’i
into-au
The examples are the following:
(111) i’an int-u’u
ilí
chu’u buásiata
yoa-ka.
Now and-that
little dog tail-NNOM.SG wag-GER
‘And now that little dog wagging his tail…’ (Crumrine 1961:18)
(112) i’an int-uchi
jumee bakoch-im
a’abo itóm=
jariu
Now and-again
those snake-PL
here 3PL.OBL=
search:PRS
‘And now the snakes come on this side to look for us.’ (Crumrine 1961:21)
(113) ...kaa nooka
…not talking
21
int-uchi,
and-again
int-itasa21
and-what
jumak ama joa sisime.
maybe there do going.HAB
In the original text the word was intasa ‘and what’, but my consultant rejected it as ungrammatical
and only accepted intitasa ‘and what’.
122
‘…and, he doesn’t say anything, and what he is doing there (we don’t know).’
(Crumrine 1961:28)
(114) ian
int-uka
Now and-this.NNOM.SG
a’a=
pattai-ka
3NNOM.SG= close-PST
‘And now, though having
1961:33)
pueta-ta
apo
mijmo
gate-NNOM.SG
3SG
himself
into kaa=‘a =
tetea-ka.
and
not=3NNOM.SG=
find-PST
shut the gate himself, he cannot find it.’ (Crumrine
(115) inti’an
into jumee naba’aso-m
jume kuusim ae
And-now
and
those blade-NNOM.SG
that rosaray with.it
a’a =
kutukta-ne-’u
kaa
te-teaka.
3NNOM.SG = carve-FUT-REL
not
finding
‘And now he cannot find the knife with which he carves the beads.’ (Crumrine
1961:24)
(116) int-aapo
intok kaa
ju’unea
jitasa jumak a’a= waata-’u
And-3SG
and
NEG
know.PRS
what maybe 3NNOM.SG=want-REL
o
a’a= ju’uneyea.
or
3NNOM.SG=know
‘And maybe he does not know what does he want or it knows.’ (Crumrine
1961:27)
(117) i’an int-im
ramaata
puntta-po
weeka yen-taite-k.
Now and-here
ramada
end-LOC
standing smoke-INCEP-PST
‘And now, standing at the end of the ramada, he has start to smoking.’ (Crumrine
1961:30)
(118)
int-um
kari beju’uku
kate-ka.
And-there house leaves-LOC sit-GER
‘And he sits under the leaves.’ (Crumrine 1961:27)
(119) ian
int-i’i
sami-t
jikau tajtajti
weam-su-ka...
Now and-this
adobe-LOC
up
through
walk-TERM-SUB…
‘And now, after having walked up and down on the adobe …’ (Crumrine 1961:36)
(120)
int-a-u
bo’oka
a’a=bitchu
ili
chu’u.
And-3NNOM.SG-DIR lay.down.PST
3NNOM.SG= look.PRS little dog
‘And the little dog laying down besides him and is looking at him.’ (Crumrine
1961:37)
123
It is important to realize that into ‘and’ may fuse with those particles independently if
it is in first, second or final position. In other words, it does not occupy those positions
for clitic reasons. We have to remember that the process is optional and that into can
occur in those positions without the union of any particle.
The next example shows into-(j)uchi ‘and again’ three times in initial position:
(121) [int-uchi
juka yoem-ta
sim-su-k]
[int-uchi
jaku’ubo
somewhere
and-again
that man-NNOM.SG go-CONT-PST and-again
suma
yeu
sika].
maybe
out
come.PST
[int-uchi
ko’om yepsaka
jum ramaata
betuk]
and-again
down arrived
DET
ramada
under
‘And when the man was gone again, it (the bird) again came from somewhere.
And again arrived going under the ramada.’ (Crumrine 1961:36)
The final example illustrates into-im ‘and here’ in second position.
(122) ‘amak int-im káwi
ááka-m-ta
bétukún
Sometimes and-here mountain pithaya-NMLZ-NNOM.SG under
bicha
matánsa-ú
bichá
saká’a-ne.
toward
Matanza-DIR toward
go-FUT
‘And sometimes we would go toward the mountain which has pitaya on it to the
slaughter house.’
3.2.4
The structure of coordination: A proposal
At the beginning of this chapter we supported the view that a coordinated sentence
groups the coordinator with the second conjunct, as in (123). Clausal coordination now is
represented as in (124). On it, the subject of the second conjunct has been fronted
because of topicalization and is adjoined to CP. An additional adjunction process
introduces a full CP (first sentence). This adjunction process is licensed by the presence
of the feature [coord] in the CP.
124
S1
S2
(123) [Joan bwiika] [Maria into ye’eka]
(124)
CP[coord]
CP
CP[coord]
Joan bwiika
NP
CP[coord]
María-i
into
CP
t-i ye’eka
This explanation for Yaqui coordination contrasts with proposals such as that of
Camacho (2003). He considers that coordination is a functional head (whereas here it is
considered as an adjunct). For example, subject coordination in Camacho’s proposal
holds that if a subject in a simplex sentence is licensed as a specifier of INFL, each
conjunct in a conjunction of subjects will be licensed as a specifier of an INFL-like
propositional projection. The coordinated subject from a Spanish sentence (125) is
represented in (126) (Camacho 2003:39):
(125) Paulina
y
yo
Paulina
and
1SG
‘Paulina and I arrived.’
(126)
llegamos.
arrive.PST.1PL
IP
DP
Paulina
I’
I
y
{tns, φ,…}
IP
DP
yo
I’
I
VP
llegamos
{tns, φ,…}
125
There are several assumptions that are important to keep in mind. He assumes that a
conjunction will copy all the features from another category present in the numeration.
Depending on the position of the conjunction, a different licensing head with different
feature specifications will give the conjunction content. In this sense, the distribution of
conjoined elements will depend on their licensing position in the tree. Because my
approach is framed within OT, many of these assumptions can be avoided and will have
to be derived from the interaction of constraints.
3.2.4.1
Coordination as a process of adjunction
In this section I show that Yaqui coordination can be treated as an adjunction process.
The idea that coordination is a head has been rejected by many authors, e.g. Peterson
(2004), Munn (1993), among others.
In this work I adopt the position that coordination particles are adjuncts. As such they
are sisters of phrasal nodes, as pointed out by Adger (2003: 11): adjunction ensures that
there is a parallelism between adjuncts and specifiers and complements. Complements
are sisters of lexical items; specifiers are sisters of x’-nodes and adjuncts are sisters of XP
nodes. A common view about adjunction is that adjoined elements are incorporated into a
sentence but not via the checking of selectional features. The adjunction is represented as
follows:
126
(127)
XP
XP
Specifier
adjunct
X’
X
complement
Taking coordination as adjunction has the advantage that the phrasal level of XP does
not change, since there is no satisfaction of a selectional feature. This conception of
coordination explains its ability to conjoin different categories. The proposal conforms to
the principles of X’-theory. In addition, the assumption that adjunction is not linearized
explains why a coordinator like into ‘and’ in Yaqui can appear in the positions seen
before: first, second, and last in a sentence containing it.
Adger (2003: 113) gives evidence that in English an adjoined item like the adverb
quickly can appear on either side of the phrase. The next sentences are represented as
follows:
(128) a. Quickly kiss Anson.
b. Kiss Anson quickly.
(129)
a.
VP
quickly
b.
VP
e
VP
V’
kiss
VP
e
Anson
quickly
V’
kiss
Anson
If a coordinator is an adjunct, then the prediction is that it would have this property.
Yaqui data confirm that this happens. The next example contains a coordinator in initial
127
position (sentence (19) is repeated here as (130)). I am assuming here that the coordinator
is a member of the second conjunct:
(130) [aapo kuchureo]
[into aapo bochareo].
[3SG fisherman]
[and 3SG shoemaker]
‘Hei is a fisherman and hei is a shoemaker.’
But a coordinator can also appear in final position. The next example is a question
with a coordinator in final position (of course, it is not a simple coordination, as in (130)):
(131) Joan jitá
bwa-ka
John what eat-PST
‘And what did John eat?’
intoko?
and
In addition, the coordinator can appear in second position (sentence (21) is repeated
here as (132)):
(132) [aapo kuchureo]
[aapo into bochareo].
[3SG fisherman]
[3SG and
shoemaker]
‘Hei is a fisherman and hej (another guy) is a shoemaker.’
If the coordinator is not a head but an adjunct, then these patterns are easily accounted
for because it would be able to occupy different positions in a sentence. The coordinator
can be represented as follows (I assume that the wh-question in (133b) is in situ). In the
structure, the coordinator introduces a feature [coord] which enables the CP to acquire
another adjoined category: the first conjunct.
(133)
a.
b.
CP[coord]
into
CP
aapo
CP[coord]
CP
IP
kikreo
Joan
intoko
IP
jitá bwaka
128
The feature [COORD] is different from that of [+PROP] proposed by Camacho
(2003:38). Camacho suggests that a coordinator introduces a propositional feature
[+prop]. The conception here is that [coord] is a feature that allows the adjunction of
another element which would be taken as the first conjunct. In that sense, the feature does
not imply necessarily that we have a propositional feature in each conjunction process.
For Camacho (2003), the conjunction is a sentential functional head that has
propositional content. In the present approach, it is not a head. There are cases where
coordination of two nouns does not necessarily imply (at least directly) a feature [+prop].
Sentence (134) contains two coordinated nouns, but they are not the main argument,
although, they are coreferent with it. Being optional, we can think that they are adjoined
to a projection inside the sentence.
yeu-sajak,
[joan into
(134) bempoi
3PL
out-go.PL.PST [John and
‘They left, John and Peter.’
peo]i.
Peter]
The facts about into ‘and’ in second position are related to topicalization (Dedrick and
Casad 1999). The topicalized NP is fronted. Then the subject in the example has an extra
movement up, adjoining to the extended CP22:
22
Heidi Harley (p.c.) suggests that this example shows that adjuncts and specifiers are not distinct, or
need not be distinct. This is a big issue in many approaches to X-bar theory and related phrase structure
theories (e.g. Kayne’s LCA-type approach, Chomsky’s Bare Phrase Structure) which deserves more
research.
129
(135)
CP
aapoi
CP[coord]
into
CP
ti’
IP
ti kikreo
There is additional evidence for this: into ‘and’ occurs in positions where sentential
adverbs tend to occur in the language: initial second and final position (already
exemplified at the beginning of this chapter)
Under this approach, a coordinator is not a head, it is an adjunct and as such it is not
assigned θ-roles (nor does it assign any theta-roles itself). Therefore, it is adjoined to
some projection. This approach predicts that coordinators are optional (at least in cases
like (136), although in cases like (130) they are obligatory), because (in a minimalist
approach) the operation Adjoin is not an operation which is triggered by feature-checking
requirements. Look at the contrast between (131) and (136). The sentence (131) has the
coordinator into ‘and’ adjoined to it, but the sentence (136) does not, there the
coordinator is optional.
(136) Joan jitá
bwa-ka?
John what eat-PST?
‘What did John eat?’
It also predicts that coordination may be recursive, since the output of the adjunction
operation is still the same category to which the coordination adjoins:
(137) Peo into Maria into Joan emo ta’a.
Peter and
Mary and
John 3REFL know.PRS
‘Peter and Mary and John know each other.’
130
It has been observed that coordination is asymetrical in the sense that the coordinator
seems to be more tied to the second conjunct than to the first one (Ross 1967).
Researchers like Camacho (2003:60) keep this observation in order to explain conjoined
verbal projections. At the beginning of the derivation Camacho takes the coordinator to
be part of the second conjunct, although at the end of the representation he holds that “the
conjunction does not form a constituent with the second conjunct, contrary to standard
assumptions…there is no single constituent that groups all conjuncts and the conjunction
leaving all other nodes out” (2003: 69). The Yaqui data indicate that we have to evaluate
whether the coordinator really groups with the second conjunct. At first glance, this
seems to be true. A coordinator in the second and in the final position suggest that idea.
Therefore, the proposed structure must reflect that fact. On the other hand, we have
discontinuous coordination in the language. The coordinator always goes with the second
conjunct but not with the first one:
(138) [inepo Sandra-ta
[1SG Sandra-NNOM.SG
‘I like Sandra and Eva.’
tu’ule]
like.PRS]
[into
[and
Eva-ta]
Eva-NNOM.SG]
In addition, it is possible to have a coordinator in both sentential conjuncts. The
request in the next sentence could be answered by the following sentence. It is shown that
each sentence has its coordinator after each subject (although the first one is not a pure
coordinator).
(139) [bweta nee=tejwa],
[jitasa into
[what and.more
[but 1SG.OBL=tell.PRS]
‘But tell me, and what else did they do?’
yaa-ka-mme?
do-PST-3PL
131
(140) [María into
bwika-k]
[Peo into
[Mary and.more
sing-PST]
[Peter and
‘And moreover Mary sang and Peter danced.’
ye’eka].
dance-PST]
Therefore, the structure for that coordinated sentence is shown in (141).23
(141)
CP[coord]
CP
CP[coord]
Maria into bwiika NP
Peoi
CP[coord]
into
CP
ti ye’eka
In the structure, there is only one slot for coordination. This conception contrasts with
that of Agbayani and Golston (2002), who allow the introduction of empty positions in
order to allow the introduction of another conjunct. The following sentence presents a
contrast with the next one. As we can see, the presence of a coordinator between both
clauses is possible. The sentence is not so bad, (the informant’s intuition is that it
introduces a kind of emphasis or redundancy), as shown below:
(142) [Joan bwiika-k]
[Maria into
[Mary and
[John sing-PST]
‘John sang and Mary danced.’
ye’e-ka].
dance-PST]
(143) ?[Joan bwiika-k]
into [Maria into
[John sing-PST]
and
[Mary and
‘John sang and Mary danced.’
23
ye’eka].
dance-PST]
Agbayani & Golston (2002) assume that a coordinator is a head, therefore, in their representation,
when the specifier of the coordinator is full, an abstract coordinator is needed for licensing the addition of
another conjunct. However, for Yaqui, it would be hard to maintain the well-formedness of that kind of
structure for sentences like (142).
132
The movement of a topicalized element to initial position (leaving the coordinator in
second position) is obligatory. Therefore, sentences like the following are ungrammatical
due to the lack of movement of the topicalized noun Maria to first position in the second
conjunct.
(144) *[Joan bwiika-k]
[into Maria ye’eka].
[and Mary dance-PST]
[John sing-PST]
‘John sang and María danced.’
However, when we introduce a sequential coordinator, the sentence is totally
grammatical. The two coordinator meanings do not enter into conflict and the sentence is
grammatical. For that reason, we need to specify that the input contains information
about the type of coordinator that adjoins to the syntactic structure. For now let’s assume
that there are features [COORD
&]
and [COORD
then].
The coocurrence of two logical
coordinators can be ruled out then by a constraint that forbids such a situation.
(145) Joan bwiika-k
[junakbea
Maria into
John sing-PST
then
Mary and
‘John sang and then María danced.’
(146)
ye’e-ka].
dance-PST
CP[coord then]
CP
CP[coord then]
Joan bwiikak Junakbea
CP[coord&]
NP
María-i into
CP[coord&]
CP
t-i ye’eka
On the other hand, the coordination of subjects (Joan into Maria ‘John and Mary’)
and objects (joan-ta into maria-ta ‘John-NNOM and Mary-NNOM’) is indicated next. The
representations stand for AgrsP or AgroP respectively:
133
(147)
YP
Y
AgrP[coord]
DP1
Joan
Joan-ta
AgrP[coord]
into
AgrP
DP2
Maria
Maria-ta
Agr’
Agr
XP
Another pattern of coordination in the language is the following. A coordinated
subject in the second conjunct forces the occurrence of the coordinator in first position.
(148) [Joan bwiika-k]
into [[Maria
John sing-PST]
and
[[Mary
‘John sang and Mary and Peter danced.’
into
and
(149) *[Joan bwiika-k]
[[Maria into Peo] into
[John sing-PST]
[[Mary and
Peter] and
(‘John sang and Mary and Peter danced.’)
Peo] ye’e-ka].
Peter] dance-PST]
ye’e-ka].
dance-PST]
This can be explained using the Coordinate Structure Constraint (CSC) (Ross 1967),
as follows. There is no way in which both DP1 and DP2 may be fronted. They occupy two
different slots. The only option would be to move the first DP1, but that would produce
the ungrammatical structure with repetition of two coordinators: *[DP1 into into DP2]
violating the well known coordinate structure constraint (CSC) which forbids movement
of a single conjunct from the conjoined structure.
134
(150)
CP
DP1i
into
CP[coord]
Co
CP
AgrsP
C
DP1
Maria
AgrsP
into
AgrsP
DP2
Peo
Agrs’
Agrs
XP
Finally, the pattern of coordination with serial verbs in the following, which shows
unbalanced coordination, can be explained by the multiple adjunction of clauses as
indicated in the representation.
(151) u
yoi
a=karo-wa
tucha-kai, uka
liacho-ta
DET
(white).man 3SG.POSS=car-POSS stop-SUB DET.NNOM.SG bag-NNOM.SG
tobokta-kai a=kari-wa
bicha wee-taite-kai uka
take-SUB
3SG.POSS=house-POSS
toward go.SG-begin-SUB DET.NNOM.SG
pueta-ta
etapo-kai,
(into) a=jubia-wa
tebotua-k.
door-NNOM.SG open-SUB
(and) 3SG.POSS=wife-POSS greed-PST
‘The man stopped his car, took the bag, went to his house, opened the door and
greeted his wife.’
135
(152)
CP[coord&]
Sn
V-kai
CP[coord&]
S5
V-kai
CP[coord&]
S4
CP[coord&]
V-kai S3
CP[coord&]
V-kai S2
CP[coord&]
V-kai (into)
CP(S1)
V-k
This view of the coordination process is very close to Munn’s approach (1993), who
treats iterative conjunction by simply adjoining to a Boolean Phrase (BP). Thus conjoined
NPs of the sort Tom, Dick, Harry and Fred are represented as indicated in (153) (Munn
1993:24):
(153)
NP
NP
Tom
BP
NP
Dick
BP
NP
Harry
BP
B
and
NP
Fred
Munn considers that coordinate structures are adjunction structures containing a
Boolean Phrase. However, by using a set of constrains we are able to derive the effects
seen in coordinate constructions and establish some relations with subordination. It is a
common claim that it is hard to distinguish coordination from subordination. What this
136
means in an OT framework is that several constraints interact in such a way that we seem
to have a continuum between coordination and subordination.
3.3
Analysis in OT
In this section I propose several constraints in order to explain the variation in the
position of into ‘and’ in the Yaqui language. I begin with the explanation of the
unmarked pattern of coordination: into ‘and’ in second position.
3.3.1
Into in second position
The basic patterns of sentence coordination with into ‘and’ in second position are
repeated here for convenience. Some of the examples are taken from Dedrick and Casad
(1999). The coordinator is obligatorily placed in second position and can not appear in
these cases in first position24.
Maria into
(154) a) Joan
bwika-k
John
sing-PST
Mary and
John sang and Mary danced.’
ye’e-ka.
dance-PST
b) *Joan
bwika-k
into Maria ye’e-ka.
John
sing-PST
and
Mary dance-PST
‘John sang and Mary danced.’
24
Recall that the coordinator into 'and' can be affixed with -k and -o. These suffixes, when attached to
verbs, mark tense: -k indicates perfective aspect whereas -o has temporal adverbial characteristics and can
be glossed as 'when'. However it is not clear if these suffixes add the same temporal distinctions when
attached to into (see Dedrick and Casad (1999)). For example, the sentence (154a) have the following two
variants, without apparent change in meaning:
(i) joan bwika-k
maria into-k ye'e-ka
joan bwika-k
maria into-ko ye'e-ka
'John sang and Mary danced'
137
In the next example, the sentence contains a postpositional phrase (Post-P). The
coordinator can appear in second position (after the Post-P), or in first position (before
the Post-P), as illustrated below:
(155) a) juchi ‘ae=koni-la
sik-aa jo’o-t
intok ‘a’a=siise-k.
Again 3NNOM.SG-circle-ADV go-PPL back-LOC and 3SGPOSS=urinate-PST
‘And having going around him, it urinated on his back.’
b) juchi
‘ae-konila
sik-aa intok ho’o-t ‘a’a= siise-k.
again
3NNOM.SG-circle-ADV go-PPL and back-LO 3SG.POSS=urinate-PST
‘And having going around him, it urinated on his back.’
Looking at the data in (154) and (155a), we can be tempted to say that the coordinator
into ‘and’ behaves as a clitic and occupies the second position for clitic reasons, as
suggested by Behloul and Harbert (2002) for other languages where the coordinator is in
second position. But we have seen that it is difficult to hold for Yaqui that into ‘and’
occupies the second position for clitic reasons (see section 3.2.2). Yaqui, as most Utoaztecan languages, is a “second position language"; but, as Steele (1979) pointed out,
second position is derived from the importance of the first position. She made the
following observation (Steele 1979: 244):
[T]opic, negation, quotatives, modals and tense tend to occur in sentence
initial position, if sentence initial position is defined to include second
position. Obviously, not all these elements can occur initially within a
single sentence. One factor in the relative position of these elements is the
potential that the scope relationships between these elements will be
manifested in their surface relationship to one another.
138
In OT terms, we can say that these elements tend to be aligned to the leftmost edge of
a clause, but there being only one left edge, only one of them can be the winner. That
amounts to saying that we have a set of constraints responsible for the allocation of
lexical elements to positions.
The question to be addressed, then, is what forces the allocation of the Yaqui
coordinator into to second position? The answer seems to be related to topicalization
processes. Steele (1979: 245) suggests that the second position of modals, tense,
quotatives, and possibly negatives is that topic tends to win the battle for first position.
A similar observation was made by West (1986) in her analysis of the Tucano
adversative coordinator (which appear in second position too): purica ‘but’ occurs
following simple noun phrases (noun, pronouns, or locative words) when these acts as the
topic of the sentence" (West 1986: 202). In Tucano, sentence topic is marked by a clause
constituent’s being moved to the first position in the sentence.25
Dedrick and Casad (1999) suggest that Yaqui coordination is a “topic pivot” for
topicalized nouns and temporal adverbs. Although this observation is basically correct, it
is not entirely accurate because we predict that given a non-topicalized (subject) noun, a
sentence like (154b) (repeated here as (156)) would be grammatical. However that does
not happen.
(156)
25
into
*Joan
bwika-k
John
sing-PST
and
(‘John sang and Maria danced.’)
Maria ye’eka.
Mary dance-PST
The normal word order of Tucano and Yaqui is SOV. Other languages that place coordinators in
second position such as Gaviao and Guaraní seem to be SOV too. I didn't investigate the correlation
between word order and this phenomenon but it seems to be worth pursuing it. I leave this issue for further
research.
139
Steele (1979) mentions the hypothesis that topics tend to solidify in sentence initial
position. Given that the Yaqui language is SOV, I propose that, in absence of another
topicalized element, subjects are interpreted as the topic of the sentence and must be
fronted. This fact will force “the movement” of the subject from the Spec of IP to a
higher position in the clause. In OT terms, the candidate with the subject in first position
and the coordinator in second position will be chosen by the ranked constraints. The
candidate with coordinator in first position and the subject in second position will be
ruled out as non optimal.
That this approach is right is supported by the fact that in the presence of another
topicalized element, the subject doesn’t rise to initial position whereas the topicalized
element does. The sentence in (157a) contains a topicalized (therefore, fronted) direct
object. It contrasts with the sentence (157b) where the subject is raised to topic position
(adjunction to CP, in this case). Note that the order in (157) is the unmarked SOV:
(157) a) [Joan
bocham jinu-k] [panim into
and
[John
shoe.PL buy-PST] [bread
‘John bought shoes and bread Mary sold it.’
Maria am=nenka-k]
Mary 3NNOM.PL=sell-PST]
b) [Joan
bocham
jinu-k ]
[Maria into
[John
shoe.PL
buy-PST]
[Mary and
‘John bought shoes and Mary sold bread.’
panim nenka-k].
bread sell-PST]
Following Grimshaw (1995) and Choi (2001), among others, I assume that topic and
focus are marked in the input, such as illustrated below:
(158) <sing (x), x=topic, x=John>
In what follows, I propose the constraints responsible for the patterns of Yaqui
coordination. The first one is related to the topicalization process which places
140
topicalized elements at the beginning of the sentence. It is defined as follows and it
belongs to the family of Information structuring constraints. It is a general constraint
which I adopt from Lee (2001):
(159) Top-L
(Lee, 2001:81)
Topic aligns left in the clause.
In order to be more concrete, let us define some members of the family of topics
mentioned in (159) as follows:
(160) Top-Subj
A subject bearing a topic feature must align left in the clause26.
(161) Top-Post-P
A Postpositional Phrase (Post-P) bearing a topic feature must align left in the
clause.
The constraint responsible for coordinators’ allocation is defined as follows. It is
derived from the function covered by a coordinator: it is the glue between two units
(NP’s, VP’s, S’s) and, therefore, must be at the leftmost edge of the following conjoined
element27. The next constraint stands too for a family of constraints (languages usually
have several coordinators).
26
Although these constraints resemble alignment constraints in phonology (McCarthy and Prince
1993, Prince and Smolensky 1993), it seems that they can not be defined in such terms. This is because
the satisfaction of them doesn't relate to the measure of how far the elements in the candidates are from
the edge. Suppose that in the hypothetical example in (i) the constraint A requieres XPy to be fronted. A
candidate like (ia) will be optimal, but the candidate (ib) which violates the constraint once, is not
necessarily better than the candidate (ic) which violates the constraint four times.
(i)
a. )
b.
c.
27
Candidates
XPy + XPz + XPu + XPv + XPw
XPz + XPy + XPu + XPv + XPw
XPz + Xpu + XPv + XPw + XPy
Constraint A
!*
!****
This constraint is defined in a broad sense (family of constraints). Beyond addmitting coordination
of sentences, it allows coordination of two VP's (eat and drink), or two single NP's (John and Mary). In the
141
(162) Coord-L
A coordinator must occupy the leftmost edge of XP.
The constraint that will place the coordinator into ‘and’ at the beginning of a
coordinated sentence is defined as follows. I am using the word into for mnemonic
reasons. The constraint is universal and not particular for Yaqui.
(163) Into-L
A logical coordinator must be allocated to the leftmost edge of XP (S in this case).
From the data with into ‘and’ in second position, we can conclude that Top-Subj
dominates Into-L. In other words, the family of Topicalization constraints must dominate
Into-L. This is illustrated in the following table (164). The winning candidate contains the
subject in Spec of JP, therefore, it doesn’t violate the highest ranked Top-Subj and, in
spite of violating the lower ranked constraint Into-L, it emerges as optimal.
(164) Tableau indicating the ranking Top-Subj >> Into-L.
Input: {Maria<Top>, into...}
a. )
b.
Top-Subj
CP[DP Mariai CP[Coor into IP[Spec ti...]]]
CP[
Ø CP [Coor into IP[Spec Maria...]]]
Into-L
*
!*
The data in (155), where into ‘and’ may be in first or second position, depending on
topicalization facts, are explained too by ranking Top-PostP >> Into-L. In the following
table, the input contains a Post-P (ho’o-t ‘on his back)’ marked for topic. The winning
candidate (165a) doesn’t violate Top-PostP because the Post-P, being marked for topic,
face of more complicated data the definition could be refined, however, for the present purposes, it is
sufficient.
142
appears in the highest position in the projection of IP. The candidate (165b) does not
violate the lower ranked constraint Into-L, but violates the higher ranked Top-Post-P,
therefore, it is rule out as non-optimal.
(165) Tableau indicating the ranking Top-Subj >> Into-L.
The postpositional phrase is marked for topic in the input.
Input: {ho’o-t<Top>, into...}
a. )
CP[Post-P
b.
CP[Ø CP[Coord
3.3.2
Top-PostP
[ho’o-t]i CP[Coor into...Post-P[ti] ]]
into... Post-P[ ho’o-t ]]]
Into-L
*
!*
Into in first position
This section shows that into ‘and’ occurs in first position, when there is not an XP
marked for topic in the sentence. Interestingly enough, when postpositional phrases are
not marked for topic, Top-Post-P is inert and the coordinator must appear at the
beginning of the sentence. This is illustrated by the following table where the winning
candidate does not violate any constraint.
(166) Table illustrating the ranking Top-Post-P >> Into-L. The postpositional phrase is
not marked for topic in the input.
Input: {ho’o-t, into,...}
a. )
CP[Coord
into... Post-P[ ho’o-t ]]
b.
CP[Post-P
[ho’o-t]i CP[Coor into...Post-P[ti] ]]
Top-PostP
Into-L
!*
The explanation given for the possibilities for placing into ‘and’ relative to
postpositional phrases can be extended to the following pattern of coordination. In the
143
example (167) the coordinator conjoins two clauses without an overt subject and, as the
example indicates, it does not have to appear in second position:
(167) tu’isi kaa aa
ye’e
intok kaa aa
eteho,
good not
able dance.PRS
and
not able
converse.PRS,
kia
tene-ka
kik-ne.
only mouth-SUB stand-FUT
‘He was not able to dance well, and he was not able to tell stories very well, he
would just stand there with his tongue in his mouth.’
The example (167) is explained by the absence of a topicalized element to be fronted
in the conjoined sentence. The lack of an overt subject -which has been solidified as
topic- implies that if there is not a topicalized element in the clause, the coordinator does
not have to appear in second position in S.
3.3.3
Analysis of two coordinators
More challenging is the pattern where two lexical elements which usually function as
coordinators are found in the same clause. The following example contains the sequential
coordinator junak ‘then’ and the logical coordinator into ‘and’. As exemplified, junak
must always be in first position, into in second and the subject in third position.
(168) a) junak
into joan kaa
then
and
John not
‘And then John did not arrive’
yepsa-k.
arrive-PST
b) *into
junak joan
kaa
yepsak
c) *Joan
junak into
kaa
yepsak
The constraint responsible for the allocation of junak ‘then’ belongs to the family
defined in (162). The constraint is defined as follows.
(169) Junak-L
A sequential coordinator must be allocated to the leftmost edge of S.
144
The data in (168) suggest that Junak-L is ranked over Into-L. The following table
shows that in the presence of an input containing both junak and into, the former always
occupies the first position in S.28 In the next table, the winning candidate (170a) does not
violate the undominated constraint Junak-L, and, in spite of violating Into-L, it emerges
as optimal. The candidate (170b) violates the undominated constraint Junak-L and is
ruled out as non-optimal. If we reverse the ranking, we have the output of languages such
as English and Spanish (and then... ; y entonces...).
(170) Tableau indicating the ranking Junak-L >> Into-L.
Input: {junak, into...}
Junak-L
a. )
CP[Coord
junak CP[Coord into ...]]
b.
CP[Coord
into
CP[Coord
junak...]]
Into-L
*
!*
In order to explain all the data in (168), we need to appeal to another constraint which
is defined below. It is derived from the observation that subjects tend to occur in the Spec
of IP. In the minimalist framework this fact is attributed to Nominative Case marking (I
assume a Split IP, where the higher projection is AGRs; for simplicity reasons, I only
refer to the Specifier of IP):
28
The constraint Junak-L is dominated in languages such as Gavião and Guaraní. In the following
example from Gavião (Stute 1986: 10) the demostrative appears first and the coordinator second:
(i) è
bó
tá-máh
mató-á
that
connector (then) 3PL-DECL+PAST 3SG+show+FINAL
‘Then they showed it.’
145
(171) Subj-SpecIP29
Subjects must check case overtly in the Spec of IP.
The pattern introduced in (168) is explained by the following table. We can see that
the winner candidate is (172a). It does not violate the constraint which requires that
subjects check Nominative Case in the Spec of IP. Its closer competitor (172b) has the
subject in a higher position in the sentence and, therefore violates the constraint SubjSpecIP and is ruled out as non optimal. It is important to realize that the evaluation over
constraints requiring fronting of lexical elements does not count how far the lexical
elements are from the edge. For example, in the optimal candidate (172a) the subject is
twice as far away from the left edge compared with the candidate (172b), which has only
one lexical element between it and the left edge. If the evaluation of those constraints
were counting lexical elements, the candidate (172b) would be the winner. However, the
evaluation process only looks at the satisfaction or non-satisfaction of constraints.
Because neither candidate (172a) nor (172b) satisfied it, both are starred. The final
decision in favor of (172a) is done by the lower ranked constraint Subj-SpecIP. With
respect to the candidates (172c) and (172d) which contain the subject in first position,
they are ruled out because Junak-L is ranked over Top-Subj and can not emerge as
optimal. The candidates (172e) and (172f) which contain the coordinator into ‘and’ in
first position, are out too because the constraint Junak-L dominates Into-L. The
constraints Into-L and Subj-SpecIP are unranked with respect to each other.
29
Grimshaw defines the constraint responsible for the subject allocation in the Spec-of-AGRs as
follows: SUBJECT: The highest A-specifier in an extended projection must be filled. The difference in the
definition of this constraint and that adopted here is clear. Grimshaw's definition will allow the filling of
Spec-of-AGRs by another element, whereas the one adopted here does not. I leave this matter unsolved
until more evidence in favor of one or another may be found.
146
Into-L
*
*
*
*
*
!*
*
*
!*
*
*
Junak-L
Top-Subject
Input: {junak, into, hoan<Top>...}
a. )
CP[ Coor
junak CP[Coor into IP[Spec hoani...[ti]]]]
b.
CP [Coor
junak CP[DP joani
c.
CP[DP
joani
d.
CP[DP
joani CP [Coor into CP[Coor junak IP[ Spec ti...]]]]
e.
CP[Coor
into CP[Coor junak IP[Spec joani ...[ti] ]]]]
!*
*
f.
CP[Coor
into CP[DP joani [Coor junak IP[ Spec [ti]...]]]]
!*
*
3.3.4
Into in last position.
CP[Coor
CP[Coor into IP[Spec ti
...]]]]
junak CP[Coor into [ IP[Spec ti ...]]]]]
Subj-SpecIP
(172) Tableau indicating the ranking Junak-L >> Top-Subj >> Into-L; Subj-SpecIP.
*
We have seen that coordination can be viewed as the adjunction of a coordinator
which introduces the feature [coor] in the top node of the projection to which it is
adjoined. This process of adjunction of a coordinator allows for the possibility of another
adjunction to the same category. This fact explains into ‘and’ in second position as the
result of raising and adjunction to the maximal node of a topicalized subject NP, as
illustrated in what follows:
(173)
CP
NP
Maríai
CP[coord]
into
IP
NP
VP
ti
ye’eka
147
This idea of adjunction allows us to also explain into(ko) ‘and (just)’ in final position.
We can see that into(ko) ‘and’ in final position dominates the entire clause which is
coordinated. It only is aligned to the right, that is the reason why we don’t attest a mirror
image of second position phenomena in coordinators (i.e. there is not into ‘and’ before
the final word). The following illustration indicates the way in which the coordinator is
aligned to the right and appears in final position:
(174)
CP[coord]
CP
NP
IP
jitái
NP
María
VP
intoko
ti jinuk
The constraint responsible for the adjunction of intoko ‘and just’ to the right of IP
conflicts with the constraint into-L which requires fronting of the coordinator. There are
two conditions under which intoko go in final postion: a) when intoko ‘and (just)’ has
additional to its conjunctive meaning an adverbial meaning. It is like an adverb. b) when
it is forced by other particles which introduce a sentence such as the particle bweytuk
‘because’.
The constraint responsible for intoko in final position is an informational-structural
one, in Yaqui completive information or background information which has the features
[-prominent] is aligned to the right edge in a sentence, I suggest that this constraint is
responsible for this pattern. Let’s take the following interaction:
148
(175) tuuka
jita
empo ya’a-k?
yesterday
what 2SG do-PST
‘What did you do yesterday?’
A possible answer could have the following two variations, where the comitative
noun may appear in a preverbal or postverbal position. Let us analyze first the answer in
(176). Here the speaker is highlighting the fact that he was accompanied by Maria:
(176) inepo Mariata-mak sentro-u
1SG Mary-COM
center-DIR
‘I with Mary went to the center’
noite-k.
go-PST
In terms of features, the subject, being [-new, +prominent], is the topic and the rest is
the focus of the sentence. It is the focus because the lexical items have the features
[+new, +prominent].
(177) inepo
-new
+prom
Mariatamak
+new
+prom
centro-u
+new
+prom
noitek.
+new
+prom
Now, let us analyze the answer (178). Here the speaker mentions Mary’s company at
the end of the sentence. This fact has the effect of removing importance to Mary’s
company. The act of going downtown seems to be more important. The topic is still the
subject referred to by the pronoun inepo ‘I’, the focus is centrou siika ‘went to the
center’, but the commitative noun is completive information only.
(178) inepo centro-u
siika
go.PST.SG
1SG downtown-DIR
‘I went (to) downtown with Mary.’
Mariata-mak.
Mary-COM
Because the speaker reduces importance to the company of the person referred to by
the conmitative noun, it has the features [+new, -prom]. It only functions as completive
information and, therefore, is aligned to the right:
149
(179) inepo
-new
+prom
centro-u
+new
+prom
siika
+new
+prom
Mariatamak
+new
+prom
If the first speaker continues with the conversation and asks the following sentence,
the lexical items will have the features indicated.
(180) ¿jitá Maria jinu-k
What Mary buy-PST
‘And what did Mary buy?’
intoko?.
and
The subject is the topic and the object and the verb are the focused elements. The
coordinator, functioning more like an adverb is [+new, -prom] and, therefore, has to be
aligned to the right:
(181) jitá
+new
+prom
Maria
-new
+prom
jinuk
+new
+prom
intoko?
+new
-prom
The answer could emphasize the thing bought for the person who asked the above
question, as illustrated next. The features are indicated below.
(182) aapo e
betchi’ibo
wepul supem
3SG 2SG.OBL
for
one
shirt
into wepul ko’arim
and
one
skirt
‘Lit: She for you a shirt bought and a skirt’
jinu-k
buy-PST
As we can see, the subject aapo ‘3SG’ is the topic, whereas the NP object senu supem
‘one shirt’ and the benefactive NP enchi betchi’ibo ‘for you’ are the focus of the
sentence. All they have the features [+new, +prom]. The coordinator into ‘and’, being a
logical coordinator, does not have adverbial meaning; it is not in final position; its
features are [+new, -prom], the same than when it is an adverbial. The NP senu ko’arim
is completive information [+new, -prom] and must be aligned to the right of the sentence.
150
A question arises here. Why is into ‘and’ not in final position? I suggest that the answer is
related to the fact that a coordinator must dominate the element which it coordinates. If
we allocate it in final position, it would be understood as modifying or coordinating the
entire sentence and not only the NP senu ko’arim. In addition, the sentence would be
ungrammatical because the extraposed NP would not have a coordinator which
established the union with the previous NP.
(183) aapo e
betchi’ibo
-new +new +new
+prom +prom +prom
wepul supem
+new +new
+prom +prom
(184) *aapo e
wepul supem jinuk wepul ko’arim
betchi’ibo
jinuk into wepul ko’arim
-new +new +new +new
+prom -prom -prom -prom
into
Another interchange where we can attest the informational status of the lexical items
which go in final position is the following. In the answer, the NP juka mariata ‘the
Maria’ is optional and correferential with the object clitic pronoun a= ‘her’:
(185) empo kaa
Maria-ta
2SG not
Mary-NNOM.SG
‘Don’t you like Mary?’
tu’ule?
like.PRS?
(186) naaka, inepo a=waata
juka
Maria-ta.
Yes, 1SG 3NNOM.SG=love.PRS DET.NNOM.SG Mary-NNOM.SG
‘Lit: Yes, I love her, the Mary.’
The features of the answer are the following: As we can see, juka Mariata ‘the María’
has the features [-new, -prom]; in other words, it is background information. It is
correferential with the clitic pronoun a= and because it is background information that is
not prominent, it must go in final position.
(187) naaka,
+new
+prom
inepo
-new
+prom
a=
-new
+prom
waata juka Maria-ta
+new -new -new
+prom -prom -prom
151
The constraint is defined as follows Lee (2001:81):
(188) Non-Prom(inent)-R(ight):
Information not prominent must be aligned right in the clause.
The next table indicates that given the input with intoko having the features <+new, prom>, the optimal candidate is (189a) because it does not violate the constraint NonProm-R which is ranked higher than Into-L. The candidates (189b,c) violate fatally the
constraint Non-Prom-R and are left out as non optimal.
(189) Tableau indicating how emerges intoko in final position.
The ranking is Prom-R >> Into-L.
NonProm-R
a.)
{Jitá, maria, jinuk, intoko
+new -new +new +new
+prom +prom +prom -prom}
CP[NP Jitá [CP[[IP[SpecMaría[VP jinuk]]] [Coord intoko]]
b.
CP[NP Jitá CP[Coor
!*
c.
CP[Coorintoko CP[NP
Input:
Intoko IP[Spec María [VP jinuk ]]]]
jitá IP[Spec María [VP jinuk ]]]]
IntoL
*
*
!*
In closing this section, I present the explanation of bweytuk (or boetuk) ‘because’ in
first position and into(ko) ‘and (just)’ in final position. It does not require additional
machinery. Bweytuk has the features [+new, +prom] and intoko [+new, -prom], the last
one must be to the right and the first one to the left by the constraint which states that
coordinators must be aligned to the sentence left edge. The example (37) is repeated here
as (190). The table (191) shows how the order of these constituents is obtained in OT.
The candidate (191a) does not violate the higher ranked constraint Non-Prominent-Rigth
and emerges as optimal. The candidate (191b) violates it and is ruled out as non optimal.
152
(190) [inime kábui-m
ne
am =
tá’áya]
[boetuk
ne
These mountain-PL 1SG 3NNOM.PL= know]
[because
I
júébenasi
ám=
bit-la
intoko]
many.times 3NNOM.PL= see-PFV
and]
‘This mountains I know them, and because I have been through here so many
times.’ (Crumrine 1961:23)
(191) Tableau which shows the interaction between boetuk ‘because’ and intoko ‘and
(just)’. It has the ranking Prom-R >> Into-L.
Input: {boetuk, …
intoko}
+new
+new
+prom
-prom}
a.) CP[[CP[boetuk…] [Coord intoko]]]
+new
+new
+prom
-prom
b.
CP[[Coord intoko]CP[boetuk…]]
+new
+new
-prom
+prom
Non-Prom-R
Into-L
*
!*
In this last section I have shown how the constraints over topicalization, the
constraints over location of coordinators and constraints over non-prominent information
explain the different positions where the coordinator appears: first, second and last. The
framework of OT allows us to integrate several insights in a unified account in order to
give a better understanding than previous frameworks.
3.4
Summary of Chapter 3.
In this chapter I have given evidence that it is hard to maintain that a coordinator is a
head. We saw that a coordinator occupies different positions in a coordinate sentence.
Those positions require a structural explanation and the adjunct-host relation seem to be
more appropriate for explaining the mobility of the coordinator into(ko) ‘and’ than the
specifier-head-complement relation. The use of constraints that regulate word order in
153
Yaqui allows us to explain the sentence coordination patterns in Yaqui. For example a
constraint over topics requires the fronting of topics, a constraint over non-prominent
right puts to the right of a sentence the non-prominent information. The OT approach has
enough flexibility for explaining the different patterns which we found in the use of the
coordinator into ‘and’ in Yaqui.
154
4
OBC AND UBC IN YAQUI
The purpose of this chapter is to present a description and OT analysis of Ordinary
Balanced Coordination (OBC) and Unbalanced Coordination (UBC)30. In the first part I
define and describe both the OBC and the UBC. After that it is shown that the UBC
should be classified at least as semantic coordination because it fits with several tests
used in Yuasa and Sadock’s (2002) analysis: The construction respects the Coordinate
Structure Constraint (CSC), and the construction is reversible and backward
pronominalization is not allowed, among others. In the next section I analyze some of the
most salient characteristics of --kai constructions within the OT framework. In the last
part
of
this
chapter
I
revise
the
characteristics
of
Pseudosubordination,
Pseudocoordination and Coordination.
4.1
Verbal coordination
This section explores verbal conjuncts (i.e. verbs and verb phrases as well as clausal
conjuncts). We will see that it is difficult to separate verbal coordination from sentence
coordination as the language does not allow the conjunction of individual verbal heads.
Next, the reader will find a description of the main characteristics of these types of
coordinations.
30
The third typological possibility; the Extraordinary Balanced Coordination (EBC), was not attested
in Yaqui.
155
4.1.1
Verbal balanced coordination
In what follows we can see that in general, verbal coordination could be considered
balanced in Yaqui. The concept of verbal balanced coordination as used here refers to a
situation where both coordinated verbs are inflected in the ordinary way by tense, aspect
and mood, and various agreement features such as person and number (i.e. it is the
opposite of the unbalanced coordination of Johannessen 1998).
The next example illustrates ordinary balanced coordination. It shows that both verbs
(there could be more) are inflected the same way. Both are marked for past tense. Then,
for these types of examples, coordination is balanced. The example contains two
intransitive verbs.
(1)
U
ili
usi
small boy
‘The boy jumped and ran.’
DET
[chept-e-k
[jump-INTR-PST
into
and
buit-e-k].
run-INTR-PST]
4.1.1.1 Yaqui coordination tends to be balanced for tense
With respect to tense, Yaqui coordinated verbs can be inflected the same in past (as
above) present and future. The next example contains two bare verbs. A Yaqui verb
without inflection signals a continuous present:
(2)
yoi
[bwika
into ye’e].
(non-Yaqui).man
[sing.PRS
and
dance.PRS]
‘The (non Yaqui) man is singing and dancing.’
The sentence in (3) contains verbs in the future tense. The verbs can contain different
tenses too, as indicated in (4):
(3)
u
uusi
[chept-i-ne
DET
boy
[jump-INTR-FUT
‘The boy will jump and will run.’
into
and
buit-i-ne].
run-INTR-FUT]
156
(4) u
uusi [chept-e-k
DET
boy [jump-INTR-PST
‘The boy jumped and will run.’
into
and
buit-i-bae].
run-INTR-INTT]
However, as we will see in the section about verbal unbalanced coordination, tense is
the feature where it is possible to find unbalancedness.
4.1.1.2 Yaqui verbal coordination is balanced for number
In relation to other features such as person and number, Yaqui verbs, in general, don’t
mark them. They don’t mark gender either. However, there are a set of suppletive verbs
which are conditioned by number and some few verbs that use reduplication for marking
plurality. Those can be tested to discover how they behave under coordination. The next
example indicates that verbal coordination is balanced in this respect: Both conjuncts
require their plural forms:
(5)
bempo torimme-u
[{saja-k/ *siika}
into aman ko-kocho-k].
{go.PL-PST/*go.SG.PST} and there RED.PL-sleep-PST]
3PL torim-DIR
‘They went to Torim and slept there.’
4.1.1.3 Yaqui verbal coordination is balanced for aspect
The following example indicates that verbal coordination is balanced for aspect as
well. Each verb can be inflected by different aspectual suffixes. It is not the case that one
depends on the other for aspectual interpretation. In the following example, the inceptive
suffix --taite ‘began’ does not affect the meaning of the first conjunct, showing that both
verbs have independent aspect.
(6)
a maala-wa
[hoara-u yepsa-k
into aman jichik-taite-k].
his mother-POSS [house-DIR arrive.SG-PST and there sweep-INCEP-PST]
‘His mother arrived at the house and began to sweep there.’
157
4.1.1.4 Yaqui verbal coordination is balanced for mood
Mood is also balanced in verbal coordination. The example in what follows indicates
that the reduplication marks modality over the last conjunct but it does not affect the
meaning of the first conjunct (i.e. the “decide” meaning introduced by the reduplication
does not spread to the first conjunct).
(7)
Aapo pueplo betana yepsa-k
into
ji’i-bwa-ba-bae-k.
He
town from arrive.SG-PST and
something-eat-RED-INTT-PST
‘He arrived from the town and decided to eat something.’
‘*He decided to arrive from the town and decided to eat something.’
4.1.2
OT Constraints for explaining Balancedness
This section explores some constraints useful in explaining balancedness in Yaqui.
Tense, mood and number are the characteristics explained here.
4.1.2.1 Tense, number, and mood balancedness
The constraints used for explaining balancedness in tense, number and mood are
based on economical considerations. The underlying idea here is that it is more
economical to avoid morphological tense, number and mood marking than inserting it.
The constraints are defined as follows:
(8)
*TENSE MARKING: Avoid morphological tense marking.
(9)
*NUMBER MARKING: Avoid Morphological Number marking.
(10)
*MOOD MARKING: Avoid morphological mood marking.
158
These constrains are beat by a constraint requiring feature satisfaction. I assume that
lexical items in the input carry information of the type shown in (12). Those features
must be morphologically (or semantically) satisfied:
(11)
SATISFY FEATURE: lexical feature requirements must be morphologically
satisfied.
So, given an input as in (12), some of the most viable candidates are shown in
(12a,b,c):
buika
<PRS, __, IND>
buika-k
<PST, IND>
buika
<PRS, IND>
into
and
into
and
into
and
!*
*
*MOOD
a) ) Ye’e
<PRS, __, IND,>
b)
Ye’e
<PRS, IND>
c)
Ye’e-ka
<PRS, IND>
*NUMBER
buika]
<TNS:PRS>
<Num: _>
<Mood:IND>
*TENSE
Input: [ye’e,
into,
<TNS:PRS> <and>
<Num: _>
<Mood:IND>
FAITH-I-O
Ranking: SAT-FEAT, FAITH-I-O >> *TENSE, *NUMBER, *MOOD
SAT-FEAT
(12)
**
**
!*
In the previous tableau, we can see that there is no way for Yaqui to satisfy the
demand of the constraint SAT-FEAT because there is not a morphological affix in open
syntax for marking indicative present tense. The single verbal root marks indicative
present tense and does not convey information about number. Therefore, all the most
viable candidates violate the SAT-FEAT constraint. However, the candidate (12a) respects
FAITH-I-O whereas candidates (12b) and (12c) do not. FAITH-I-O is violated in (12b)
because the second conjunct has a different tense marking than the one required in the
159
input. Candidate (12c) does not bear the indeterminacy for number present in candidate
(12a). Therefore, it is not optimal too and (12a) emerges as the optimal one.
Because Yaqui allows the union of CP’s with different tense markings, we have to
allow coordination with different tense values. For a sentence like (13), we must have a
verbal input as that indicated in (14).
(13) ian
buika-k
into yooko
yi’i-ne.
today sing-PST
and
tomorrow
dance-FUT
‘(He) sang today and will dance tomorrow.’
*MOOD
yi’i-ne
<FUT,IND, __>
ye’e
<PRS, IND>
yi’i-ne
<FUT, IND>
into
and
into
and
into
and
*NUMBER
a) ) buika-k
<PST, IND, __>
b)
buika
<PRS, IND>
c)
buika-k
<PST, IND>
*TENSE
yi’i-ne]
<TNS:FUT>
<Num: _>
<Mood:IND>
FAITH-I-O
Input: [buika-k,
into,
<TNS:PST> <and>
<Num: _>
<Mood:IND>
SAT-FEAT
(14) Ranking: SAT-FEAT, FAITH-I-O >> *TENSE, *NUMBER, *MOOD
**
!** **
!** **
The tableau indicates that candidate (14a) satisfies both SAT-FEAT and FAITH-I-O.
Therefore, it wins against candidate (14b) which violates both of them. It wins too against
candidate (14c) which only violates the higher ranked constraint FAITH-I-O.
4.1.3 The Coordinate Structure Constraint (CSC)
Yaqui data indicate that the language does not allow extraction from any conjunct.
Therefore, we can establish that this behavior is produced by the higher ranking of a
constraint regulating extraction. The CSC (Ross 1967) is seen here as a universal,
violable constraint. The definition in OT is as follows:
160
(15)
*EXTRACTION: Extraction from a conjoined structure is not allowed.
This constraint avoids extraction from any conjunct, as indicated in the next
examples. In (16) we have two declarative coordinate full sentences, if we try to extract
from the conjuncts, it is not possible to have a grammatical sentence.
(16)
Joan Paola-ta
atea-k Maria into Peo-ta
John Paola-NNOM.SG meet-PST Maria and Pedro-NNOM.SG
‘John found Paola and Mary greeted Peter.’
tebotua-k.
greet-PST
First, if we try to extract the object from the first conjunct, the result is an
ungrammatical sentence like (17):
(17)
*jabe-ta
Joan Ø atea-k
Maria into Peo-ta
tebotua-k.
who-NNOM.SG John Ø meet-PST Maria and Peter-NNOM.SG greet-PST
(‘Who did John find and Maria greeted Peter.’)
Second, if we try to extract the object from the second conjunct, the result is again an
ungrammatical sentence:
(18)
*jabe-ta
Joan Paola atea-k
Maria into Ø tebotua-k.
who-NNOM.SG John Paola meet-PST
Maria and Ø greet-PST
(‘Who did John find Paola and Maria greeted?’)
Finally, it is not possible either to have Across the Board Extraction, as indicated by
the following ungrammatical sentence:
(19)
*jabe-ta
Joan Ø
atea-k
Maria into Ø tebotua-k.
meet-PST
Maria and Ø greet-PST
who-NNOM.SG John Ø
(‘Who did John find and Maria greeted?’)
The only way to ask for the objects is by having a WH-question inside each conjunct.
Therefore, the constraint *EXTRACTION is not violable in the language31:
31
A reviewer made the suggestion of checking if it is possible that movement of a DP to the left in
Yaqui is topicalization, but WH-words automatically are focused elements, and hence cannot be
161
(20)
jabeta Joan atea-k
into jabeta Maria tebotua-k.
who John meet-PST
and
who Maria greet-PST
‘Who did John find and who did Maria greet.’
The previous example could be explained if we consider that the constraint
*EXTRACTION interacts with a constraint that forbids coordination of non-maximal
projections.
This constraint is supported both theoretically and empirically. On the
theoretical side there is a common view that languages coordinate maximal projections.
Kayne (1994) rejects the coordination of verbal heads in English. He proposes the
coordination of VPs for English; Johannessen (1998) proposes that coordination joins
CPs. Empirically, Yaqui shows that the language only licenses the coordination of VP’s
but not V’s. The constraint is defined as follows:
(21)
*COORDINATION OF NON-MAXIMAL PROJECTIONS (*COORD-NON-MAX):
Coordination of heads is not allowed.
In addition to the previous constraints, we have the presence of a constraint based on
the observation that in coordinate structures we have the distribution of grammatical
functions (Peterson 2004). This constraint forces the reduction of lexical material to the
minimal amount required for covering the functions in a coordinate structure. For
example, in the Spanish sentence in (22) the grammatical functions of subject and object
topicalized. If this were true, the ungrammaticality of (17) and (18) would be explained by this fact. The
following sentence indicates that the WH-word in the second conjunct can be topicalized (it appears in the
slot for topicalized items: before the coordinator into ‘and’). Therefore, the ungrammaticality of these
sentences ((17), (18)) must be attributable to the extraction of WH-words.
(i)
jabeta Joan
atea-k
jabeta into
who
John
meet-PST
who
and
‘Who did John find and who did Maria greet.’
Maria
Maria
tebotua-k.
greet-PST
162
are distributed because they appear just once, but they are interpreted as the subject and
object of each verb.
(22)
El
maestro
abrió
y
cerró la
puerta.
The teacher
opened
and
closed the door
‘The teacher opened and closed the door.’
The constraint is defined as indicated next:
(23)
DISTRIBUTION OF GRAMMATICAL FUNCTIONS (DGF): The attributes of
grammatical functions must be distributed in a coordinate structure.
The example in (24) can be explained the by the interaction of these constraints. As
indicated in table (25), the input is unordered. The most viable candidates are (25a, b, and
c). Among them, candidate (25a) is optimal because it does not violate the higher ranked
constraint *EXTRACTION, whereas candidates (25b) and (25c) do. It is interesting to note
that candidate (25b) shows the distribution of the grammatical functions subject and
object, like example (22) in Spanish and English. However, it does not emerge as optimal
because the constraint *EXTRACTION is higher in the hierarchy.
(24)
Jabeta Joan atea-k
into jabeta aapo
Who John find-PST
and
who 3SG
‘Who did John find and who did he greet?’
(25)
Tableau with the ranking *EXTRACTION >> DGF, *COORD-NON-MAX.
Input: {Jabeta, joan, ateak, into, jabeta, aapo,
tebotuak}
a.) Jabeta joan ateak into jabeta aapo tebotuak
tebotua-k.
greet-PST
* EXTRACTION
DGF
*COORDNON-MAX
**
b.
Jabeta Joan ateak into tebotuak
!*
c.
Jabeta Joan ateak into aapo tebotuak
!*
*
*
The next section will treat some of the most important characteristics of what we call
here “--kai-constructions”.
163
4.1.4
Verbal unbalanced coordination
In this section we are going to see that Yaqui verbal coordination has only one of the
two typological patterns (assigning and receiving types) proposed by Johannessen (1998):
the receiving type. The assigning type is ruled out because the language does not allow
verbal head coordination and therefore it is not possible to find a situation where the
features of the objects enter in conflict. The descriptive concept of verbal unbalanced
coordination that I use is that suggested by Johannessen (1998). It is split into two types
of unbalancedness: The receiving type of UBC and the assigning type of UBC. The first
one happens when “one verb is inflected in the ordinary way; by tense, aspect and mood,
and various agreement features such as person and number. The other conjunct(s)
occur(s) in their base form, or in some or other non finite form” (Johannessen 1998:34).
Amharic:
(26) [yi-rrammε-inna
yi-rət’-al.
3SG.M-walk-and
3SG.M-run-3SG.M.NON-PAST
‘He walks and (then) runs/will run.’
The second type (assigning) happens when “the verbs in each conjunct have different
subcategorization properties; they assign, e.g., different case to their complements”
(Johannessen 1998: 38). Examples (27) and (28) show that the closest verbal conjunct
assigns its case to the object den Mann ‘the man’: the verb half ‘helped’ requires to
assign dative case, whereas the verb begrüste ‘greeted’ requires to assign accusative case.
(Rolf Thieroff, cited in Johannessen 1998:38):
German:
(27) Maria [begrüsste
und half] dem /*den
María greeted
and
helped the.DAT/ *the.ACC
‘María greeted and helped the man.’
Mann.
man
164
(28)
Maria [half und begrüsste]
*dem/
den
María helped and
greeted
*the.DAT/ the.ACC
‘María helped and greeted the man.’
Mann.
man
The German examples indicate that the sentences become ungrammatical if we try to
use the case marking of the first verbal conjunct. So, the coordination is unbalanced in
the assignment of case marking.
4.1.4.1 Yaqui lacks the assigning type of UBC
With respect to the assigning type of UBC, the data indicate that Yaqui does not seem
to presents case conflicts. The language marks nominative with a zero marker and nonnominative singular with --ta (the plural --(i)m never co-occurs with --ta ‘NNOM.SG’).
Two classes of ditransitive verbs (Escalante 1990) which could potentially enter in
conflict were analyzed. Those verbs what requires the object marker with --ta ‘NNOM.SG’
vs. those requiring the object marker with --ta-u ‘NNOM.SG-DIR’. The contrast is shown
below:
(29)
inepo Peo-ta
1SG Peter-NNOM.SG
‘I gave corn to Peter.’
bachi-ta
corn-NNOM.SG
miika-k.
give-PST
(30)
inepo Peo-ta-u
bachi-ta
1SG Peter- NNOM.SG-DIR corn- NNOM.SG
‘I sold corn to Peter.’
nenka-k.
sell-PST
Under coordination each sentence gets its own arguments (i.e. each transitive verb
must have its objects). There is never a case where a single object could be “shared” by
both verbs, suggesting that in Yaqui, more than verbal coordination we have clausal
coordination (or VP coordination at least). Moreover, the conjuncts could be considered
to be balanced:
165
(31)
inepo [Peo-ta-u
bachi-ta
nenka-k]
1SG [Peter- NNOM.SG-DIR corn- NNOM.SG sell-PST]
[a-a=miika-k].
[3NNOM.SG-3NNOM.SG=give-PST]
‘I sold and gave corn to Peter.’
(32)
inepo [peo-ta
bachi-ta
1SG [Peter-NNOM.SG
corn- NNOM.SG
[a-w-a=
nenka-k]
[3NNOM.SG=DIR-3NNOM.SG sell-PST]
‘I gave and sold corn to Peter’
into
and
miika-k]
give-PST]
into
and
In relation to transitive predicates, the following examples indicate that Yaqui
coordinated verbs require both objects. In general, two transitive verbs cannot be
coordinated like two intransitives. Each verb requires its own object in overt syntax,
hence the following contrast. It shows too that coordination is balanced: each verb
requires its own tense marking and its object argument.
(33)
Joan [karo-m
jiinu]
buy.PRS]
John [car-PL
‘John buys and sells cars.’
into
and
[am=nenka].
[3NNOM.SG-PL=sell.PRS]
(34)
*Joan karom jiinu
into
John car- PL buy.PRS
and
(‘John buys and sells cars.’)
nenka.
sell.PRS
4.1.4.2 Yaqui has a Receiving type UBC
Yaqui has a verbal construction that can be classified as receiving type UBC. It
happens in serial verb constructions like the following one. In it, the verbs of the series
are marked with the suffix --kai ‘SUB’ and only the last one is marked for tense (past
tense in this case). The whole construction is understood as marked with the tense of the
final conjunct. Let’s call these kinds of examples “--kai-constructions”.
166
(35)
[u
yoi
a=
karo-wa tucha-kai],
DET
(non-Yaqui) man
3SG.POSS= car-POSS stop-SUB
[u-ka
liacho-ta
tobokta-kai],
DET-NNOM.SG bag- NNOM.SG take-SUB
[a=
kari-wa
bicha wee-taite-kai]
3SG.POSS=
house-POSS toward go.SG-begin-SUB
[u-ka
pueta-ta
etapo-kai],
DET-NNOM.SG door-NNOM.SG open-SUB
[a=
jubia-wa tebotua-k].
3SG.POSS= wife-POSS greet-PST
‘The man stopped his car, took the bag, went to his house, opened the door and
greeted his wife.’
The structure that I propose for this type of -kai chaining structure is the following.
As the representation indicates the structure is the adjunction of CP’s to a tensed CP
which gives the temporal interpretation of the whole sentence:
(36)
CP[PST]
CP
u yoi a karowa tucha-kai,
CP[PST]
CP
uka liachota tobokta-kai,
CP[PST]
CP
a kariwa bicha taite-kai,
CP[PST]
CP
CP[PST]
uka puetata etapo-kai, a jubiawa tebotua-k
4.1.4.3 Verbal chaining structures: --kai-construction
Because of their status as UBC (Johannessen 1998) or as Pseudo-subordination
(Yuasa and Sadock 2002), --kai-constructions deserve being described and analyzed for
their theoretical implications. For that reason, in what follows it is shown first that the -kai suffix can be considered a subordinator and second, that some constructions where it
appears are tied to what can be considered as coordination.
167
4.1.4.3.1 --kai as a subordinating particle
Most researchers of the language (Dedrick and Casad 1999, Escalante 1990,
Lindenfeld 1973, among others) treat --kai as a subordinating particle. My own data tend
to confirm this claim. But there are subtle uses that are important to clarify. The next
example indicates a common use of -kai as subordinating particle; generally it has a
gerundive meaning:
(37)
Maria tajkaim
ya’a-su-kai
Maria tortillas
make-TERM-SUB
‘After finishing making tortillas, Maria ate them.’
am=bwa-ka.
3SG.NNOM.PL=eat-PST
Even semantic cases where we can talk about pseudosubordination must be treated
syntactically as adjoined clauses. Example (38) can be interpreted semantically as
coordinate or subordinate, but the syntactic marking is clearly subordinated (the Yaqui
language does not have a copulative marker).
(38)
u
yoi
tebe-ta-kai
anukichi.
liar
The (non-Yaqui) man
tall-NNOM.SG-SUB
‘That white man is tall and a liar/ that white man, being tall, is a liar.’
The complex sentence (38) is represented in (39). The --kai clause is adjoined to the
host CP.
(39)
CP2
CP1
u yoi tebeta-kai
CP2
anukichi
The --kai clauses have several characteristics that are explored in the following
section.
168
4.1.4.3.2 --kai-constructions are coordinate at the semantic level
In this section I show that --kai-constructions are coordinate at the semantic level.
The criteria used for stating this claim are the following (Yuasa & Sadock 2002): a) The
construction is reversible and truth conditions are preserved, b) The construction obeys
the Coordinate Structure Constraint (CSC), c) Backward pronominalization is not
allowed, d) Any number of conjuncts can occur in coordinated constructions, e) Scope
considerations: under semantic coordination both conjuncts are affected (ex. by
negation).
4.1.4.3.2.1 Analysis
Because --kai-constructions resembles in some aspects the Japanese -te-coordination,
I applied the semantic criteria used by Yuasa & Sadock (2002) in order to see if the
construction can be considered coordinated (at least at the semantic level).
In the first place, I checked if the construction is reversible and truth conditions are
preserved. As the following example indicates, it fulfills this requisite. The coordinate
complex sentence (40) has the order (S1-kai & S2-TNS), whereas in (41) the order is
reversed (S2-kai & S1-TNS):
(40)
32
yoi32
a=
karo-wa
DET
(non-Yaqui) man
3SG.POSS=
car-POSS
[a=
jubia-wa
tebotua-k].
3SG.POSS=
wife-POSS
greed-PST
‘The man stopped the car and greeted his wife.’
[u
tucha-kai],
stop-SUB
The concept yoi or yori is opposed to the concept yoeme. The last one is used for referring to the
Yaqui men, whereas the first is used for all non Yaqui men. These were identified first with the white men
that arrived first to the Yaqui land. Actually, the term is used for all non-Yaqui persons.
169
(41)
[u
yoi
a=
jubia-wa
DET
(non-Yaqui)man
3SG.POSS=
wife-POS
[a=
karo-wa
tucha-k].
3SG.POSS=
car-POSS
stop-PST
‘The man greeted his wife and stopped the car.’
tebotua-kai],
greet-SUB,
Another criterion for deciding if a --kai-construction is coordinate, is to observe if it
obeys the CSC, This principle states that “in a coordinate structure, no conjunct may be
moved, nor may any element contained in a conjunct be moved out of that conjunct”
(Ross 1967:89)33. We can take the following --kai-sentence and check the results:
(42)
Peo Maria-ta
bicha-kai
Peter Maria-NNOM.SG
see-SUB
‘Peter saw Mary and heard Juana.’
Joana-ta
Juana-NNOM.SG
jikaja-k.
hear-PST
As the following ungrammatical sentences indicate, it is not possible to extract any of
the objects. The symbol Ø indicates the site of possible extraction:
(43)
*jabe-ta
Who-NNOM.SG
Peo Ø bicha-kai
Peter Ø see-SUB
Joana-ta
Juana-NNOM.SG
(44)
*jabe-ta
Who-NNOM.SG
Peo Maria-ta
Peter Mary-NNOM.SG
bicha-kai
see-SUB
jikaja-k.
hear-PST
Ø
jikaja-k
hear-PST
Because there are some exceptions to the CSC34, Ross (1967) added the Across the
Board Exception (ATB) which allows some specific extractions. So the CSC holds
…unless each conjunct properly contains a gap paired with the same filler. Therefore, the
33
The CSC has a continuation which masks its violability, the Across the Board Exception (ATB).
This hedge allows extraction if the extracted element is moved out of all conjuncts: ex. what did Mary cook
and John eat? In this section it is shown that even these cases are not allowed in the language.
34
The CSC has been questioned by researchers working in approaches that avoid the use of empty
elements (e.g. Sag et al 2003), because it is usually violated. However, in any OT approach violability of
constraints is expected and the violation of the CSC is predicted in some languages where the constraint is
not highly ranked. The constraint is defined here as a constraint that forbids extraction from any conjunct.
For this reason, in all the cases where extraction is possible, the CSC is violated. On the other hand, Yuasa
and Sadock (2002) use the CSC as a test that we are faced to semantic coordinated constructions.
170
ATB predicts that the following sentence would be grammatical. However, as we can see,
the sentence is ungrammatical.
(45)
*jabe-ta
Who-NNOM.SG
Peo
Peter
Ø
bicha-kai
see-SUB
Ø
jikaja-k.
hear-PST
The only way to ask an object WH-question in Yaqui is to use WH-words in each
conjunct. But in this case, the use of the coordinator into ‘and’ becomes obligatory and it
has to be in second position. The second position of the coordinator suggests that the
WH-word in the conjunct has been fronted. The wh-words are not in situ because they
occupy the first position in the conjucts, which is not a canonical position of the object;
we have to remember that the language is SOV35.
(46)
jabe-ta bicha-kai
jabe-ta
into
Who-NNOM.SG see-SUB
who-NNOM.SG and
‘Who did Peter se and hear?’
(47) *jabe-ta
bicha-kai
into
Who-NNOM.SG see-SUB
and
(‘Who did Peter see and hear?’)
jikaja-k
hear-PST
DET
ju
jabe-ta
jikaja-k
who-NNOM.SG hear-PST
DET
ju
Peo?
Peter
Peo?
Peter
The full tensed coordinate sentences have the same syntactic pattern: It is not possible
to extract the object. Each Wh-question occurs in its conjunct. Therefore, the behavior of
the -kai-construction is not particular to it. Rather, it has to be explained by general
principles of Yaqui syntax.
35
Although WH-in situ is common in Yaqui, the languge allows WH-movement too, as indicated in
example (i):
(i)
jitá
jume
jaamuch-im
What DET.PL woman-PL
‘What will these women eat?
bwa-bae?
eat-INTT
171
(48)
jabe-ta
bicha-k
into jabe-ta
jikaja-k
Who-NNOM.SG see-PST
and
who-NNOM.SG hear-PST
‘Who did Peter see and who did he hear?’
ju
DET
Peo?
Peter
One way to ‘extract’ a non WH-object from a --kai-construction is by postposing or
preposing it. The canonical position for the object is occupied by a resumptive pronoun.
In such cases, we cannot talk about movement.
(49)
jume libro-m,
Joan am= ji’oste-kai
the
book-PL,
John 3NNOM.PL= write-SUB
‘John wrote the book(s) and sold it (them)’
am=
2NNOM.PL=
nenka-k
sell-PST
(50)
aapo am=nenka-kai
juchi aman am=
poa-k
3SG 3NNOM.PL=sell-SUB again there 3NNOM.PL= pick up-PST
jume na’aso-m36
DET.PL orange-PL
‘The oranges, (s)he sold them and again (come back to) pick up more there’ //
‘The oranges, selling them, he come back to pick up more there’.
But even the use of resumptive pronouns does not improve the grammaticality of a -kai-construction (even a coordinate sentence with full tense-marking over the verbs may
be ungrammatical) containing an extracted WH-question:
(51)
*Jabetai
Who
Peo ai=
bicha-kai
(into) ai=
jikaja-k.
Peter 3NNOM.SG=see-SUB (and) 3NNOM.SG=hear-PST
36
This construction contains a potential problem for the backward pronominalization test. In order to
avoid the problem we must propose that the extraposition ‘movement’ left the nominal in a higher position
where it c-commands the pronouns. The next examples support the view that the noun must be in a higher
position than the pronoun. The coordinate sentence (i) contains a full NP (jume librom) which c-comands
the pronoun (am=), therefore the sentence is grammatical. However, sentence (ii) has the reverse order and
the full NP(jume librom) does not c-command the pronoun (am=). Therefore, the sentence is predicted as
ungrammatical.
(i)
Joan
jume
libro-m
ji’oste-kai
John
the.PL book-PL
write-SUB
‘John wrote the book(s) and sold it (them)’
(ii)
*Joan
am=
ji’oste-kai
jume
am=
nenka-k
them= sell-PST
librom nenka-k
172
The analysis of simple sentences of Yaqui indicates that we have a contrast between
interrogative and declarative sentences. A preposed object WH-question never combines
with a pronoun in the canonical position, whereas in a declarative sentence the postposed
or preposed object can be coindexed with such a pronoun. Therefore, the same principles
are playing a central role in the coordinate sentences above.
(52)
jabe-ta
Joan kesum (*a=)
Who-NNOM.SG
John chess (3SG=)
‘Who does John gave chess.’
maaka-k.
give-PST
Postposed object:
(53) Joan keesum
ai=
John chess
3NNOM.SG=
‘John gave chess to Maria.’
maaka-k
give-PST
Preposed object:
(54) [u-ka
maria-ta]i
DET-NNOM.SG Maria-NNOM.SG
‘To María, John gave chess.’
Joan
John
[u-ka
DET-NNOM.SG
keesum
chess
maria-ta]i.
Maria-NNOM.SG
a i=
maaka-k.
3NNOM.SG=give-PST
The next test is called backward pronominalization. As the name suggests, in a
coordinated structure, a nominal referential expression cannot be coindexed with a
pronoun in a previous conjunct. The contrast among the sentences below confirms that
the principle is respected in Yaqui --kai constructions:
(55)
37
ye’e-kai,
bwika-kai
into (aapoi)37
yoemei
Yaqui man drink-SUB
sing-SUB
and
(3SG)
u
pajko-po.
DET
feast-LOC
‘The Yaqui man drank, sang, and (he) slept in the feast.’
kocho-k
sleep-PST
This is a streesed pronoun which seems to add emphasis. Therefore, this sentence could be
considered to be marked. In the unmarked case, it is not possible to have a pronoun.
173
(56)
*aapoi ye’e-kai,
bwika-kai
into Yoemei
kocho-k u pajko-po
3SG drink-SUB
sing-SUB
and
Yaqui.man sleep-PST the feast-LOC
(‘Hei drank, sang, and the Yaqui mani slept in the feast.’)
We can see that in this case, the --kai construction behaves like a real coordination
that disallows backward pronominalization. As expected, that is the case too in full
tensed conjoined sentences:
(57)
Peoi aman siika into Aapoi aman ji’ibwa-k
Peter there went and
3SG here eat-PST
‘Peteri went there and hei ate there’.
(58)
*Aapoi
aman siika Peoi into aman ji’ibwa-k
3SG there went Peter and
there eat-PST
(Hei went there and Peteri ate there).
The subsequent criterion that I apply to discover if --kai constructions are coordinated
requires that the potential number of conjuncts occuring in the construction is unlimited.
This is a central property of coordination. As examples (35) and (59) and indicate --kaiconstructions fulfill this requisite.
(59)
Joan tienda-u
buite-kai,
biba-ta
jinu-kai,
cigar-NNOM.SG
buy-SUB
John store-to
run-SUB,
(into) pesio-u
bicha siika.
and Hermosillo-DIR
toward go.PST
‘John run to the store, bought a cigar, and went to Hermosillo.’
Finally, looking at scope considerations, the requirement is that under semantic
coordination both conjuncts be affected by negation. In the following sentence the
negation affects both conjuncts:
(60)
Ka
tua
ke
kowi-ta
nenka-kai
uka
wakas-ta
Not true that pig-NNOM.SG sell-SUB
DET.NNOM.SG cow-NNOM.SG
jinu-k ju
Peo.
buy-PST
DET
Peter
‘It is not true that Peter sold the pig and bought the cow.’
174
Aditionally, as in the Japanese --te-constructions analyzed by Yuasa and Sadock
(2002), when the subordinator -kai introduce a real subordinated clause in semantics, the
negation does not affect it:
(61)
Kat= nee
kokos-ayu-k, weche-kai.
Not= 1SG hurt-be-PST fall-SUB
‘I did not hurt, when I felt down.’
This section has shown that the --kai construction meets the criteria for being
considered semantically coordinated, in spite of the fact that some of the conjuncts are
marked with the subordinator --kai.
4.1.4.3.3 Characteristics of the --kai-construction
This section describes the characteristics of --kai-constructions. The most salient are
the following: Tense is marked only in the final conjunct, the order of the conjuncts tends
to be fixed, the subject is shared between the conjuncts, the particle --kai is obligatory on
each verb of the series, the subjects are not repeated in the --kai-clause, the particle --kai
only adjoins to predicates, --kai-constructions tend to be interpreted as a single event, the
coordinator can not occur between the --kai-clauses in a series (except some cases of real
subordination that we will describe).
4.1.4.3.3.1 Tense is marked only in the final conjunct
As we can see in the next example, only the final verb is inflected for tense, however,
the whole construction is understood as past tense. For Yuasa and Sadock (2002) this is
an indication that we have a structure where only the tense feature in the last conjunct
percolates up to the mother node.
175
(62)
U
jamut jichi-kai,
sankoa-ta
nau
DET.SG womansweep-SUB
garbage-NNOM.SG
together
mekka goota-k.
away throw- PST
‘The woman swept, pick up the garbage and throw it away.’
toja-kai,
pick.up-kai
Following Yuasa and Sadock (2002) I agree that pseudosubordination of this type
could be explained if we assume that syntactically the -kai clauses are subordinated.
Tense is in the final conjunct in Yaqui because the head parameter is involved here.
Yuasa and Sadock conclude that there are basically two possibilities for structures,
depending on whether or not languages are head-left or head-right. Yaqui is OV,
therefore, being head-right it is predicted that the tensed clause will be in final position,
as it actually is. Languages that are head-left present the tensed clause at the beginning.
The structure of a Yaqui subordinated --kai clause like (63) is represented in (64)
(adapted from Yuasa and Sadock (2002:98)):
(63) Joan
yepsa-kai,
Maria-ta
tebotua-k.
John
arrive-SUB, Maria-NNOM.SG.SG greet-PST
‘John arriving greeted Mary/John arrived and greeted Mary.’
(64)
CP[+Fin]
CP[-Fin]
NP
VP[-Fin]
Joan1
yepsa-kai,
CP[+Fin]
NP
pro1
VP[+Fin]
Maria-ta tebotua-k
For a chained pseudosubordinated clause, where the into ‘and’ particle can optionally
occur between the last --kai clause and the tensed one, the optimal candidate will have the
following structure.
176
(65)
CP[+Fin, Coord]
CP[-Fin]
V-kai,
CP[+Fin, Coord]
CP[-Fin]
CP[+Fin, Coord]
V-kai,
(into)
CP[+Fin]
V-TNS
The optionally of into ‘and’ indicated in (65) is related to the chain. It can never occur
with just one --kai clause, but it can occur before the tensed clause with two or more
sequential --kai-clauses38
4.1.4.3.3.2 Tense in final position: an OT analysis
Given that in Yaqui the tensed clause occurs in final position, we need to allow at
least the following two closely competing candidates in (66) and to rule out the nonoptimal one (66b).
(66)
a) V-kai, V-kai, V-kai V-TNS.
b) *V-TNS, V-kai, V-kai, V-kai.
The constraints responsible for the alternation are head-right and head-left. They are
defined as follows.
(67)
HEAD-LEFT: the head of a construction must be at the left edge of it.
(68)
HEAD-RIGHT: the head of a construction must be at the right edge of it.
38
The structure that I am interested here is the one where the coordinator can optionally occur before
the tensed clause, as in example (65) (V-kai, V-kai (into) V-TNS). Those are the examples of pseudosubordination. The –kai clauses can be themselves joined by a coordinator, but in that case they are clearly
subordinate to the tensed clause ([V-kai into V-kai], V-TNS). In such case, into ‘and’ can never occur
between the coordinate V–kai and V-TNS clause (* [V-kai into V-kai] into V-TNS). The subordinated status
comes from the evidence that in such cases, the coordinate V-kai clauses are understood as gerundive. They
are not interpreted as containing the same tense than the final tensed-clause.
177
As usual in OT, the input is unordered, and the constraints will evaluate a set of
competitors. However, only the two closest competitors are presented in the tableau.
Candidate (69a) wins the battle against candidate (69b) because in Yaqui HEAD-RIGTH is
over HEAD-LEFT. The inverse order of the constraints will produce the pattern found in
HEAD-LEFT languages, i.e. candidate (69b) would be the winner.
The constraints are ranked as indicated; they interact as indicated in (69).
(69)
Ranking: Head Right >> Head-left
Input: [V-kai, V-kai, V-kai, V-TNS]
Head-Right
a) ) V-kai, V-kai, V-kai, V-TNS
b)
V-TNS, V-kai, V-kai, V-kai.
Head-Left
*
!*
4.1.4.3.3.3 Possibility of inserting into ‘and’ in a chain
Remember that one subordinate --kai clause cannot co-occur with into ‘and’, as
illustrated in (70), i.e. into ‘and’ can optionally occur before the tense clause if we have a
chain. The representation in (70) shows the position of into ‘and’.
(70)
a) V-kai (*into) V-TNS.
(subordination)
b) V-kai, V-kai, V-kai, (into) V-TNS.
(pseudosubordination)
We can see that this alternation really is at the border between subordination and
coordination. It could explain too why a --kai chain can be considered semantically
coordinated but syntactically subordinated. I propose that --kai has the features [+SUB, TNS],
so, the non-occurrence of into ‘and’ in (70a) is expected (because of the
inconsistency of the [+SUB] feature and the [+COORD] feature introduced by into
‘and’). However, sentences with the structure (70b) indicate that into ‘and’ can occur
178
before the tensed verb. For that reason, it seems that it can be considered a marker that
the chain is about to finish. Then the constraint responsible of this alternation seems to be
pragmatic, more than syntactic (i.e. it could be a cooperative Gricean constraint).
(71)
COMM(UNICATE): Communicate that you will end your chaining.
The constraint can be fulfilled by the introduction of into ‘and’ or by a simple pause
between the relevant elements. The constraint enters into play when more than two
clauses are in the input. I assume that it is violated if into ‘and’ is present in a candidate
that it is not a chain. As usual in OT, the nature of the input is important for the
generation of viable candidates by Gen. An input can either contain a coordinator or not.
Therefore, if into ‘and’ is in the input, it has to appear in the output (by the Faith-I-O
constraint), if it is not, the candidate with the pause will be the winner.
An additional constraint is required for explaining the alternation: the one that avoids
incompatible features. This constraint avoids the combination of the features [+SUB] and
[+COORD]. It is defined as follows:
(72)
AVOID-CONTR(ADICTORY)-FEAT(URES): Don’t mix [+SUB] and [+COORD].
The next tableau explains the subordinated structures as in (70a). Candidate (73a)
does not violate any of the three proposed constraints, whereas candidate (73b) violates
all of them. For that reason candidate (73b) with the coordinator into ‘and’ is not optimal.
The winner is candidate (73a):
179
(73)
COMM, FAITH-IO>>AVOID-CONTR-FEAT
Input: [V-kai, V-TNS]
COMM
FAITH-IO
AVOID-CONTR-FEAT
!*
*
*
a) ) V-kai V-TNS
b)
V-kai into V-TNS
The next tableau indicates that if we allow into ‘and’ in the input, we still have as
winner the candidate without a coordinator (74a) because we are using a mechanism for
marking a chain where there is not a chain. So the candidate (74b) can never emerge as
optimal.
(74)
COMM, FAITH-IO>>AVOID-CONTR-FEAT
Input: [V-kai, into, V-TNS]
COMM
a) ) V-kai V-TNS
b)
V-kai into V-TNS
FAITH-IO
AVOID-CONTR-FEAT
*
!*
*
The next tableau shows the situation in (70b) when into ‘and’ occurs in the candidate.
We have to keep in mind that when we have a chain, there are two ways to mark that it is
about to be finished: by a pause or by the occurrence of into ‘and’ before the final
(tensed) clause. The next tableau indicates the competition between both candidates. If
into ‘and’ is in the input, it appears in the output. In such a case the candidate (75b) with
the pause loses in the competition because it violates the FAITH-IO constraint.
(75)
COMM, FAITH-IO>>AVOID-CONTR-FEAT
Input: [V-kai, V-kai, V-TNS]
COMM
FAITH-IO
a) ) V-kai, V-kai into V-TNS
b)
V-kai, V-kai, V-TNS
AVOID-CONTR-FEAT
*
!*
180
When into ‘and’ is not in the input, the candidate with the pause will be the winner.
As the tableau (76) indicates, candidate (76a) violates the constraint FAITH-IO and the
constraint AVOID-CONTR-FEAT, whereas candidate (76b) does not violate them.
Therefore, it emerges as the optimal one.
(76)
COMM, FAITH-IO>>AVOID-CONTR-FEAT
Input: [V-kai, V-kai, V-TNS]
a)
COMM
V-kai, V-kai into V-TNS
b) ) V-kai, V-kai, V-TNS
c)
V-kai, V-kai V-TNS
FAITH-IO
AVOID-CONTR-FEAT
*
!*
*
!*
In tableaux (75) and (76) above, both winner candidates satisfy COMM by the
introduction of into ‘and’ or by a pause. The candidate (76c) without a pause can not
emerge as winner because it violates
COMM.
The pause is introduced by Gen in the
candidate (76b). Therefore it violates
FAITH-IO
but it does not violate AVOID-CONTR-
FEAT. For that reason it is optimal.
4.1.4.3.3.4 The order of the conjuncts tends to be sequential
Because chaining structures usually indicate narrative progression, the order of the
conjuncts tends to be fixed; as we can see in the following example, the --kai-clauses
must antecede the tensed one.
(77)
yoeme
ye’e-kai,
bwika-kai,
jita
je’e-kai
sing-SUB
something
drink-SUB
(Yaqui) man dance-SUB
kocho-ka-n
u
pasko-po.
sleep-PST-CONT
DET
feast-LOC
‘The Yaqui man danced, sang, drank something and slept in the feast.’
into
and
181
If we try to reverse the order, extrapossing the -kai-clauses, the sentence becomes
highly degraded39:
(78)
*yoeme
pasko-po
kocho-ka-n,
yi’i-su-kai
(Yaqui) man
feast-LOC
sleep- PST-CONT
dance-TERM-SUB
bwik-su-kai,
into jita
je’e-su-kai.
sing-TERM-SUB,
and
something
drink-TERM-SUB
(‘The Yaqui man slept in the feast, after finishing dancing, singing and drinking
something.’)
Even under a split or discontinuous series the sentence is not totally acceptable.
Therefore, there must be a pragmatic constraint playing a role in avoiding the common
discontinuous coordination (see nominal coordination).
(79)
?Joan pasko-po
yi’i-su-kai
kocho-k
into buik-su-kai,
John feast-LOC
dance-TERM-SUB
sleep- PST
and
sing-TERM-SUB
into jita
ji’i-su-kai.
and
something
drink- TERM-SUB
‘John finished dancing in the feast and slept, and finished singing and finished
drinking something.’
The sequentiality is due to the nature of the narration of the events. The chains
(initiated by the --kai clauses) are conformed by a set of sub-events, where they are
39
If the postposed –kai clauses are intended to be contemporaneous with the main verb. That is, if they
get a ‘while’ interpretation the sentence is more acceptable. See the following contrast. From these two
examples the consultant prefers the –kai clause before the tensed verb. The examples with the ‘while’
interpretation seem to be clear cases of subordination. The cases that we are interested here are the ones
with a sequential interpretation. The sequentiality of events is a characteristic of the verbal chains analyzed
in this work.
(i)
Joan
[buika-kai]
yi’i-bae.
dance-INTT
John
sing-SUB
‘John will dance singing.’
(ii)
?Joan yi’i-bae,
[buika-kai].
sing-SUB
John
dance-INTT
‘John will dance, singing.’
182
presented according to the speaker’s intention. The constraint again could be a pragmatic
one (again in the Gricean sense):
(80)
BE-ORDERED: present eventive information in sequential order.
Sequential order could be interpreted as equivalent to cardinal order, if the input has
predicates with indices that indicate the order of the events. (This order could be altered
by other constraints that forces changes in the order). The idea here is that we have
several sub-events which give rise to an entire eventive set. (Camacho 2003 talks about
the coordination of eventive phrases; I propose to talk about eventive features in the
predicate, that are brought by the predicates and that have a reflex in the order of them).
So an input would contain the features and indices that will produce the order of the
predicates.
In relation to the possibility of reversibility, I assume that it is a logical property of
Gen. Gen can posit any structure; therefore, commutation must be a property of it. I did
not explore the situation under which a conjunct can be commuted without an apparent
change in meaning. There are some sentences where the order of presentation in some
contexts does not seem to be important (John and Mary are tall, Mary and John are tall).
That is a topic which needs to be explored. Right now, I assume that the speaker presents
the information by respecting this constraint:
183
(81)
Tableau with the constraint Be-Ordered choosing the optimal candidate.
Input: [V1, V2, V3]
BE-ORDERED
a) ) V1, V2, V3
b)
V2, V1, V3
!*
c)
V3, V2, V1
!*
4.1.4.3.3.5 The subject is shared between conjuncts and not repeated
The previous examples, like the next one, show that the suffix --kai is only used in
subject missing constructions. The subject is understood as the same in the whole
complex sentence:
(82)
[Maala
yoowe traste-ta
baksia-kai], [teopo-u
Mother
old
dish-NNOM.SG wash-SUB
church-DIR
‘The grandmother washed the dishes, and went to the church.’
siika].
go.SG.PST
So, the following example containing different subjects cannot get the --kai suffix.
Each verb is marked obligatorily for tense in the series. The clauses are connected with
the coordinator into ‘and’.
(83)
[Maala
yoowe traste-ta
baksia-{k/*-kai}],
[ume
Mother
old
dish-NNOM.SG wash-{PST/*-SUB} DET.PL
ili
si-m
into bu-busa-{k/*-kai}],
[into bem
small child-PL
and
RED-get up-{PST/*-SUB},
and
2PL.POSS
mala-wa
into bem=
achai-wa
teopo-u
saja-k]
mother-POSS and
3PLPOSS=
father-POSS church-DIR go.PL-PST
‘The grandmother washed the dishes, the children got up and their mother and
their father went to the church.’
The proposed structure implies then that the subject is missing in each CP. The
subject in a higher position binds the null subject in the following clauses. The constraint
184
DROP TOPIC which requires that topicalized subjects be dropped, explains why the subject
is not repeated.
(84)
CP
CP
CP
[maala yoowe]i trasteta baksia-kai
proi teopou siika
The subject cannot be repeated in the -kai-clause; however, if the person has an
available clitic pronoun, this can occur (usually in the subject position of the tensed
clause).
(85)
née yeepsa-kai, yeste-kai,
nim
juubi
1SG arrive.SG-PST sit.SG-SUB
1POSS.SG
wife
bitchu-su-kai,
(nee) a=
tebotua-bae.
see-TERM-SUB,
(1SG) 3NNOM.SG= greet-INTT
‘I will arrive, I will seat down, I will see my wife and I will greet her.’
4.1.4.3.3.6 --Kai: a same subject construction, and an OT analysis
As we saw before, -kai constructions make reference to the same subject throughout,
and the subject may not be repeated in the construction. In what follows, I suggest that
we need to use the constraints used by Blutner and Zeevat (2004:4): SUBJ(ECT) and
DROP-TOPIC. They are defined as in (86) and (87), respectively. The first one forces the
occurrence of the subject in a preverbal position. The second one demands that
correfential arguments be unrealized.
(86)
SUBJ: The highest A-specifier in an extended projection must be filled.
(87)
DROP-TOPIC: Arguments coreferent with the topic are structurally unrealized.40
40
Such as it stands, the constraint of DROP-TOPIC implies that if an object is a topic, then the
subsequent occurrences of arguments correferent with the object must be unrealized. For example, in
185
For Blutner and Zeevat (2004), the subject, being the topic, tends to be dropped. So
the Yaqui chain (88) can be analyzed as containing several correferential arguments with
the subject (the topic in this construction):
(88)
u
yoi
a=
karo-wa
tucha-kai,
(non-Yaqui).man
3SG.POSS=
car-POSS
stop-SUB
u-ka
liacho-ta
tobokta-kai,
DET-NNOM.SG bag-NNOM.SG take-SUB
a
kari-wa
bicha
wee-taite-kai
3SG.POSS
house-POSS toward
go.SG-begin-SUB
u-ka
pueta-ta
etapo-kai,
a
jubia-wa
tebotua-k.
the-NNOM.SG door-NNOM.SG open-SUB
3SG.POSS wife-POSS greet-PST
‘The man stopped his car, took the bag, went to his house, opened the door and
greeted his wife.’
DET
For analyzing this characteristic, we need to use again the constraint of FAITH-IO,
which requires that elements in the input be preserved in the output. In order to simplify
the representation, I only put the subjects and predicates of the previous chain (88),
(again: the input is not constituted by sentences, but unordered elements). The table does
not give evidence of the ranking of SUBJECT, but it is assumed that all the candidates in
the table respect it. The winner is the candidate (89a) because it respects the constraint
DROP TOPIC whereas the candidate (89b) violates it. This constraint is higher than the
constraint FAITH-IO, so in spite of the fact that candidate (89b) respects FAITH-IO, it can
not emerge as optimal.
English the sentence *Mary, singing, he kissed to mean that Mary, the object, was singing, not he, is
ungrammatical. Because this is not possible, we need to restrict the constraint to just topicalized subjects.
That is the sense that the constraint has in this work.
186
(89)
Tableau that shows the winner in a chain. The ranking is Drop Topic>> Faith-IO
Input: [Joani tucha-kai, Joani tobokta-kai, SUBJECT
Joani wee-taite-kai Joani etapo-kai,
Joani tebotua-k]
a) ) Joani … tucha-kai, proi… tobokta-kai,
proi ... wee-taite-kai, proi … etapo-kai,
proi … tebotua-k
b) Joani … tucha-kai, Joani… tobokta-kai,
Joani .. wee-taite-kai, Joani … etapo-kai,
Joani … tebotua-k
DROP
TOPIC
FAITH-IO
*
!*
One of the advantages of the constraint DROP-TOPIC is that it is not specific for
coordinate strucutures. Then, if we see other sides of the grammar, we expect to find the
effects of DROP-TOPIC. The effects are seen in subordinated clauses with a correferential
subject. Let’s take the next example of Yaqui:
(90)
Joan inien
ea
[kari-ta
John this.way think.PRS
[house-NNOM.SG
‘John thinks that he is going to buy a house.’
jinu-pee-sime].
buy-DESID-go.SG.PRS]
In the previous sentence, the subject of the subordinated clause does not appear. The
account of this could be the same as the one for the coordinated structure. However, a
pronoun can appear in that subordinated clause, suggesting that pronoun licensing has to
do with pragmatic constraints too.
(91)
ea
[a-ui kari-ta
jinu-pee-sime].
Joani inien
John this way
think.Prs [him-to house-NNOM.SG buy-DESID-go.SG.PRS]
‘Johni thinks that hei is going to buy a house// to buy a house.’
Altough Yaqui does not have infinitive forms (and the pronoun can be licensed in
theoretical standard terms). Spanish has constructions with infinitive verbs that can
license nominative pronouns, suggesting that the constraint of
DROP-TOPIC
is violated
under certain conditions, as in the following Spanish example (with the pronoun the
187
sentence has the meaning that he is going to buy it by using his own means or resources,
without any help, without using a third person for treading. It is a stressed pronoun.)41:
(92)
a
comprar
(éli) / una
Juani piensa ir
John thinks to.go to
to.buy
(he) / a
‘John thinks about going to buy (himself) a house.’
casa.
house
4.1.4.3.3.7 The particle --kai is obligatory on each verb of the series.
The next example indicates that the particle -kai is obligatory over each clause. An
uninflected verb is not acceptable.
(93)
Joan Peo-ta
ji’osia-m
jinu-ria-*(kai)
John Peter-NNOM.SG book-PL
buy-BENEF-SUB
a-w-am
bit-tua-k.
3NNOM.SG-DIR-3NNOM.SG.PL see-CAUS-PST
‘John bought a book for Peter and sent it to him.’
I suggest that this is a property of the input. I assume that --kai must be in the input,
as are other affixes in Yaqui. Therefore, the constraint of FAITH-IO rules out candidates
without the suffix --kai. Another required constraint is *SUB(ORDINATE)-MARKING which
shows an aversion to mark subordinate elements. The ranking is the following.
(94)
*SUB-MARKING: do not mark subordination.
41
This behavior of Spanish is not weird, even an interrogative main clause can license a nominative
pronoun, as illustrated (the tensed version is available too):
i.
Ganar-le
tú
a
Michael Jordan? No
To win-him you
to
Michael Jordan? Not
Will you beat Michael Jordan? I do not believe it.
lo
it
creo.
believe
Because this issue is beyond of coordination, I just point out that an infinitive form in Spanish does
not block nominative licensing. Other factors are present here and need to be analyzed (for example, the
verb must be fronted).
188
(95)
FAITH-IO >> *SUB-MARKING.
(96)
Input: {V-kai, V-kai, V-TNS}
FAITH-IO
*SUB-MARKING.
) a) V-kai, V-kai, V-TNS
b) V, V, V-TNS
**
**
Candidate (96a) is optimal because it does not violate the constraint Faith-IO which is
highly ranked in the language. Candidate (96b) violates it twice and is non-optimal.
4.1.4.3.3.8 The particle --kai only adjoins to predicates
The particle --kai can be added to lexical words functioning as predicates. It is
important to note that only verbal constructions give rise to UBC, other -kai-predicates
seem to function like adjuncts. Examples:
(97)
Verbal
Adjetival
Nominal
Numeral
Adverbial
Determiner
Pronominal
bwiika-kai
teebe-kai
chu’u-ta-kai
goi-kai
mekka-kai
*hu’u-kai
*bempo-kai
‘singing’
‘being tall’
‘being a dog’
‘being two’
‘being far away’
The following example contains two adjectival --kai-clauses. As we can see, a
coordinator is possible between those --kai clauses. That is not an allowed pattern in
verbal --kai-chaining. Additionally, the tense interpretation in (98) does not depend of the
tensed verb; it has a gerundive meaning as usual in subordinated --kai-constructions.
(98)
Te-tebe-kai
into
wa-wakila-kai
and
RED.PL-thin-SUB
‘Being tall and thin they believe that they are beautiful’
RED.PL-tall-SUB
emo
tu’ure.
3REFL like.PST
189
(99)
Aman mekka-le-kai
into a=
obiachi-le-kai
kaa
There far away-believe-SUB and
3NNOM.SG= difficult-believe-SUB not
aman wee-bae.
there go.SG-INTT
‘Believing that it is far away and believing that it is difficult, he will not want to
go there.’
I suggest that the coordinator into ‘and’ in (98) joints two --kai clauses as in the
following representation:
(100)
CP
CP
CP
CP
CP
CP
Tetebekai
into
wawakilakai
emo tu’ure
4.1.4.3.3.9 --kai-constructions interpreted as sequential events
In the next example, where a clause is marked with --kai ‘SUB’ (and there is not an
open coordinator) it is one event composed of two single sub-events. They are seen as
occurring at the same time or one immediately after the other. This seems to be a case of
real subordination.
(101) Diana chu’u-ta
ibakta-kai
a=muk-tua-k.
embrace-SUB 3NNOM.SG=die-CAUS-PST
Diana dog-NNOM.SG
‘Diana embracing the dog left it dead.’
However, examples like the following contrast with the previous situation. In the
following example the events do not occur at the same time. They only express
sequentiality of events. Because of that, the sentence can be translated as a coordinate
structure in English and Spanish. In other words, we have a different degree of union
190
between clauses (co-occurring/sequential). The status of subordinated is not easy to
maintain in the next example.
(102) Diana chu’u-ta
bicha-kai
a=ibakta-kai
Diana dog-NNOM.SG see-SUB
3NNOM.SG=embrace-SUB
into a=muk-tua-k.
and
3NNOM.SG=die-CAUS-PST
‘Diana saw the dog, embraced it and left it dead.’
There are examples where formally, the clause marked with --kai is subordinated, but
semantically seems to be coordinated, giving rise to what Yuasa and Sadock (2002) call
pseudosubordination:
(103) Aapo jita
jikkaja-kai
Maria-ta
3SG something
hear-SUB
Maria-NNOM.SG
‘(S)he hears (something) and stares at Mary.’
bitchu.
stare.PRS
4.1.4.3.3.10 The construction makes reference to a single event
In this section, following Progovac (1999) I assume that “single coordinations (with
and) are unspecified with respect to single vs. multiple event readings, rather than being
specified for a single event interpretation” (144, note 3). The assumption seems necessary
given the following contrast:
A balanced coordination could be one event or two events, with SS or DS.
SS, one event/two events
(104) Joan buika-k
into
John sing-PST
and
‘John sang and danced.’
ye’e-ka.
dance-PST
DS, one event/two events
(105) Joan buika-k
Maria into
John sing-PST
María and
‘John sang and Maria dance.’
ye’e-ka.
dance-PST
191
SS, two events (only reading)
(106) Joan tuuka
buika-k
into yooko
John yesterday
sing-PST
and
tomorrow
‘John sang yesterday and will dance tomorrow.’
yi’i-bae.
dance-INTT
DS, two events (only reading)
(107) Joan tuuka
buika-k
Maria into yooko
John yesterday
sing-PST
María and
tomorrow
‘John sang yesterday and Maria will dance tomorrow.’
yi’i-bae.
dance-INTT
With SS subordination and pseudosubordination (-kai construction), the reading may
be either a co-occurring or sequential events:
SS, Subordination, sequentiality of events
(108) María tajkaim
ya’a-su-kai
María tortillas
make-TERM-SUB
‘Doing tortillas, María ate them.’
am
3NNOM.PL
bwa-ka.
eat-PST
SS, Pseudosubordination, sequentiality of events; example (35) is repeated here as
(109).
(109) [u
yoi
a=
karo-wa
tucha-kai],
(non-Yaqui).man
3SG.PL car-POSS
stop-SUB
[u-ka
liacho-ta
tobokta-kai]
DET-NNOM.SG bag- NNOM.SG take-SUB
[a
kari-wa
bicha wee-taite-kai]
3SG.POSS
house-POSS toward go.SG-begin-SUB
[u-ka
pueta-ta
etapo-kai],
DET-NNOM.SG door-NNOM.SG open-SUB
[a
jubia-wa
tebotua-k].
3SG.PL wife-POSS
greet-PST
‘The man stopped his car, took the bag, went to his house, opened the door and
greeted his wife.’
DET
In the previous example, the events form a complex event. It is formed by five
predicates, but they have a sequential reading.
192
Therefore, it seems that Progovac’s (1999) claim about unspecification is correct. In
OT we can consider that the input is unspecified, and that the constraints will give the
available reading(s).
With respect to nominal conjunction, Progovac concludes that “what gives rise to a
multiple-event interpretation is the physical presence of an extra conjunction marker”
(1999:145). For her the “multiplicity of events is encoded syntactically; moreover, it is
actually encoded in an iconic way, by an increased number of conjunction markers”
(145).
In a more recent paper, De Vries assumes “that every coordinate structure has DistP
as its maximal projection, for the simple reason that every coordination is interpreted
either collectively or distributively” (2005: 87).
It seems that actually every NP (maybe every CP) must have a DistP. It is unspecified
in the input and it is specified by semantic (as when adverbials are introduced in sentence
conjuncts, as in (109) or by pragmatic constraints (as when the background specifies if
the conjuncts must be understood as distributed or not). The next example indicates that a
single sentence with a plural subject is unspecified for the distributive feature:
(110) Ju-me
maejto-m
libro-m
jinu-k.
DET-PL
teacher-PL
book-PL
buy-PST
‘The teachers bought a book.’ (collectively/distributively)
But it can get the specification from a quantifier adverbial:
(111) Chikti maejto-m
libro-m
jinu-k.
Each teacher-PL
book-PL
buy-PST
‘Each teacher bought a book.’ (distributively)
193
In order analyze of --kai constructions, I assume that DistrP’s are in the input. The
proposed constraints are the following. The first one will force the appearing, if possible,
of unspecified forms. It is defined as follows:
(112) *EVENT-SPECIFICATION: Avoid event specification.
However, specification will emerge if adverbials, different grammatical tenses or
other factors force the specification of the DistPs. Let’s call this constraint EVENT
INTERPRETATION. It is defined as indicated next:
(113) EVENT-INTERPRETATION: Distributive phrases must be specified.
Another constraint is Adverbial-Interpretation. Its definition is shown in (114):
(114) ADV-INTERP: Adverbials with different tense reference are distributed.
The interaction of these constraints is shown in table (115). In it the candidate (115a)
has a violation of EVENT-INTERPRETATION, but because it is unranked with *EVENTSPECIFICATION, nothing is decided until the constraint Faith-IO, which decides as winner
candidate (115a). Candidates (115b) and (115c) do not emerge as optimal because they
violate FAITH-IO.
(115) Ranking: EVENT-INTERPRETATION,*EVENT-SPECIFICATION>>FAITH-
FAITH-IO
) a) <DISTR:__> unspecified
b) <DISTR: + > (disjoint)
c) <DISTR: - > (joint)
*EVENTSPECIFICATION
Input: [Joan, buikak, into, ye’e-ka
<DISTR: __>]
(See ex. 101)
EVENTINTERPRETATION
IO.
*
*
!*
!*
*
194
The following tableau indicates that the winner is candidate (116b). This is so
because candidate (116b) respects the constraint Adv-Interpretation which is higher in the
hierarchy. Candidates (116a) and (116c) violate it and cannot be optimal.
(116) Ranking:
ADV-INTER(PRETATION)>>EVENT
INTER(PRETATION),
*
!*
FAITH-IO
!*
*EVENT-SPEC
a) <DISTR: __> unspecified
) b) <DISTR: + > disjoint
c) <DISTR: - > joint
EVENT-INTERTATION
Input: [Joan, tuuka, buikak, into, yooko, yi’i-bae
<DISTR: __>]
See ex. (106)
ADV-IINTER(PRETATION)
*EVENT-SPEC(IFICATION) >>FAITH-IO.
*
*
*
*
4.1.4.3.3.11 CSC is respected
We saw before that the CSC is highly respected in Yaqui. Therefore, the constraint
must be ranked high. The constraint was defined as a ban on extraction. If extraction
from a conjunct happens, the constraint is violated. The explanation is summarized here:
(117) *EXTRACTION: Extraction from a conjoined structure is not allowed.
Because Yaqui has a syntactic requirement that heads cannot be conjoined, the
constraints were defined as follows:
(118) *Coordination of non-maximal projections (*Coord-non-max). Coordination of
heads is not allowed.
195
(119) DGF (Distribution of grammatical functions). The attributes of grammatical
functions must be distributed in a coordinate structure.
Candidate (120a) is optimal because it does not violate the higher ranked constraint
*Extraction, whereas candidates (120b) and (120c) do.
a.)
b.
c.
… jabeta Joan ateakai into jabeta a tebotuak
… jabeta Joan ateakai into tebotuak
… jabeta Joan ateakai into a tebotuak
DGF
* EXTRACTION
Input: {… jabeta, Joan, ateakai, into, jabeta, a, tebotuak}
*COORD-NON-MAX
(120) Tableau with the ranking *Extraction >> DGF, *Coord-non-max
**
!*
!*
*
*
4.1.4.3.3.12 Interaction of -kai and the particle into ‘and’
It is very common to have only the juxtaposition of --kai clauses, without the use of
the particle into ‘and’. The next example is a case where the particle into ‘and’ cannot
occur between them. As we saw before, the construction is grammatical only if the
subject is the controller of the series.
(121) u
ili
jamut
yepsa-kai
jichik-taite-k.
small woman
arrive-SUB
browse-INCEP-PST
‘The woman, (after) arriving, began to browse.’
DET
The sentence becomes ungrammatical if into ‘and’ appears between both verbs:
(122) *u
ili
jamut
yepsa-kai
into
DET
small woman
arrive-SUB
and
(‘The woman arrived and began to browse.’)
jichik-taite-k.
browse-INCEP-PST
196
However, in serial constructions, when two or more sentences with --kai ‘SUB’ are
put together the particle into ‘and’ can appear optionally between the final --kai clauses
and the finite verbs.
(123) u
achai jibwa-kai,
joboa-kai,
mam-baksia-kai (into) a =
father eat-SUB,
full-SUB
hand-wash-SUB (and) 3SG.POSS
ili
usi-mme-u
etejo-taite-k.
small child-PL-to
talk-begin-PST
‘The father ate, (became) full, washed his hands and began to talk to his children.’
DET
Even with the same subject the language has the option of marking each verb for
tense, but in that case --kai cannot appear, and into ‘and’ can appear between each
conjunct. The tendency is to have in overt syntax only the last into ‘and’ in the series.
The suffix --kai cannot co-occur with any other tense marker, -k ‘PST’ for example.
(124) u
achai jibwa-k(*-kai) (into) joboa-k
(into) mam-baksia-k
DET father eat-PST(*-SUB) (and) full-SUB
(and) hand-wash-SUB
into a =
ili
usi-mme-u
etejo-taite-k.
and
3SG.POSS
small child-PL-to
talk-begin-PST
‘The father ate, (became) full, washed his hands and began to talk to his children.’
4.1.4.3.4 Some -kai clauses are adjoined in Yaqui
At the syntactic level, -kai constructions are subordinated (Takano 2004: 171 reaches
the same conclusion for similar constructions in Japanese), but as Yuasa and Sadock
(2002) point out, they are coordinated at the semantic level.
Takano shows that English verbal coordination poses a problem for an analysis
where the verbal inflectional morpheme is located in T in syntax and merged with the
adjacent verb in the phonological component, because it predicts that only the adjacent
verb will fuse with the inflectional morpheme. This situation favors the idea that the
inflectional morpheme is part of V (i.e. his explanation follows the Checking Theory of
197
Chomsky 1995). However, Takano considers that both types of verbal inflection happen
in languages. The contrast between the next two sentences shows that in the first case the
construction involves a bare verb and an inflected one. The second case has the first verb
affixed with a gerundive particle whereas the second verb is inflected with -ta. Takano
(2004:171):
(125) a.
b.
John-ga
sono ronbun-o
John-NOM
that paper-ACC
John-ga
sono ronbun-o
John-NOM
that paper-ACC
‘John copied and filed the paper.’
kopiisi
copy
kopiisi-te
copy-ing
fairusita.
filed
fairusita.
filed
So for Japanese, after his analysis, he concludes that only sentence (125a) is an actual
coordination, while sentence (125b) is an example of a subordinated one. He suggests
that bare verbs are conjoined as follows:
(126)
T’
VP
DP
T
V
V1
&
-ta
V2
In the representation V1 and V2 are bare verbs, and & is a phonetically null
conjunction. Since the tense morpheme is located in T, it will be attached to V2 and V1
will remain bare, given as output the sentence (125a).
The analysis of Yaqui indicates that it is not possible to have a coordinated bare verb
plus a verb marked morphologically for tense where the marked one gives the tense
reading for the whole construction:
198
(127) u
ili
uusi buite-k
DET
small boy run-PST
‘The child ran and jumped.’
into
and
(128) *u
chepte-k.
jump-PST
ili
uusi buite into
DET
small boy run
and
(‘The child ran and jumped.’)
chepte-k.
jump-PST
As we saw before, serial verbs are marked with --kai and only the final verb is
marked for tense, giving the tense interpretation for all the verbs. Those cases can be
treated as adjoined clauses, similar to (125b). The use of --kai, or into or both: -kai and
into seem to be related to event codification. They are ways in which Yaqui indicates
separateness of events.
I repeat the following examples in order to show the separateness of the events. In the
first example, the two events are more closely tied than the second one, as the glosses
indicate. The first example can be a clear example of an adjoined subordinated clause,
whereas the second one is an example of a coordinated one. The structure of each
sentence is represented in (129b and 130b) respectively.
(129) a) Diana
chu’u-ta
ibakta-kai
a=muk-tua-k.
Diana
dog-NNOM.SG embrace-SUB 3NNOM.SG=die-CAUS-PST
‘Diana embracing the dog, left it dead.’
b)
CP
CP
Diana chu’uta ibakta-kai
CP
a=muktua-k
(130) a) Diana chu’u-ta
bicha-kai,
a=ibakta-kai
Diana dog-NNOM.SG
see-SUB
3NNOM.SG=embrace-SUB
into
a=muk-tua-k.
and
3NNOM.SG=die-CAUS-PST
‘Diana saw the dog, embraced it and left it dead.’
199
b)
CP
CP
CP
Diana chu’uta bicha-kai
CP
a=ibakta-kai
CP
into
CP
a=muktua-k
4.1.4.3.5 Some -kai clauses are coordinated in Yaqui
There is evidence that some --kai clauses are coordinated. The suffix --kai in this case
marks the jointness (cotemporaneousness) of the event. Let’s take the following example
that indicates that the coordination of two relative subordinated clauses can not be
discontinuous:
(131) *Joan [bocha-reo-ta
bicha-ka-me] o-’omte-k
John [shoe-NMLZ-NNOM.SG see-PST-NMLZ] RED-angry-PST
[into maejto-ta
bicha-ka-me].
[and teacher- NNOM.SG
see-PST-NMLZ]
(‘John who saw the shoemaker and who saw the teacher was angry.’)
However, if we use the --kai ‘SUB’ marker instead of --me ‘NMLZ’ used for
introducing relative clauses, the sentence becomes grammatical. It has in addition the
adverb ketchia ‘too’. In this case, it is hard to maintain that we have the extraposition of a
coordinated relative phrase. Instead of that, we can claim that we have the coordination of
two clauses.
(132) Joan [bocha-reo-ta
bicha-ka-me o-’omte-k]
John [shoe-NMLZ-NNOM.SG see-PST-NMLZ RED-angry-PST]
[into maejto-ta
bicha-kai
ketchia].
[and teacher-NNOM.SG
see-PST-SUB too]
‘John who saw the shoemaker was angry and (he) saw the teacher too.’
200
In this case, the into ‘and’ particle can occupy other positions: the coordinator could
be after a topicalized object in the second conjunct, or it could be even in final position.
Those patterns are attested in actual sentence coordination. The occurrence of the
adverbial kechia ‘too’ supports the idea that we have a coordinated sentence.
(133) Joan [bocha-reo-ta
bicha-ka-me o-’omte-k]
John [shoe-NMLZ-NNOM.SG see-PST-NMLZ RED-angry-PST]
[maejto-ta
into bicha-kai
ketchia].
[teacher-NNOM.SG
and
see-PST-SUB too]
‘John who saw the shoemaker was angry and (who) saw the teacher too.’
(134) Joan [bocha-reo-ta
bicha-ka-me o-’omte-k]
John [shoe-NMLZ-NNOM.SG see-PST-NMLZ RED-angry-PST]
[maejto-ta
bicha-kai
intoko42].
[teacher-NNOM.SG
see-PST-SUB and]
‘John who saw the shoemaker was angry and (who) saw the teacher too.’
Another set of sentences that indicates that --kai constructions can be coordinated is
the following. In it the order of the conjuncts tends to be fixed. It is not possible to switch
places between conjuncts: -kai is always before into ‘and’. The coordinator in these cases
is obligatory:
(135) Malia [mala-wa-ta-kai
into papá-wa-i] Diana-ta
betchi’ibo
Mary [mother-POSS-NNOM.SG-SUB and father-POSS-i] Diana-NNOM.SG for
‘Mary is mother and father for Diana.’
(136) Malia a-u
achai-ta-kai
into
Mary 3NNOM.SG-DIR father-NNOM.SG-SUB and
‘Mary is father and mother for him/her.’
42
a-u
mala-wa-i.
3NNOM.SG-DIR mother-POSS-i
Intoko ‘and’ is taking the place of ketchia ‘too’. This is an indicative of the double life of the
particle into(ko) ‘and’ as a coordinator and as an adverbial.
201
4.1.5
Reflection about pseudo-coordination, -subordination, and coordination
In this section I present a reflection about the field of coordination which in the
literature can be found split into these three areas of research: pseudocoordination,
pseudosubordination and coordination.
Pseudocoordinations are constructions that look like VP coordinations, as in The next
example from LØudrup (2002:121):
(137) Han sitter og
skriver dikt.
he
sits
and
writes poems.
‘He is writing poetry.’
And it is usual that in these constructions the Coordinated Structure Constraint (CSC)
be violated (LØudrup (2002:122):
(138)
Hva
sitter han
What
sits
he
‘What is he writing?
og
and
skriver?
writes?
Therefore, their status as coordinate constructions has been questioned. For LØudrup
(2002) a group of what is called peudocoordination (PCO) must be treated as biclausal
subordination whereas another grup must be analyzed as monoclausal structures. This
position is contrary to De Vos (2004) who claims that PCO is coordination. One
characteristic
of
the
construction
is
that
“truth
conditional
semantics
of
pseudocoordination is the semantics of coordinations. Han sitter og skriver dikt ‘he sits
and write poems’ is true iff he sits and he writes poems” (LØudrup (2002, fn.3: 127)
LØudrup’s (2002) analysis of PCOs states that they are grammatically diverse:
control, raising and monoclausal constructions. A simplified monoclausal functional
202
structure is given in (139). The two verbs together constitute one predicate that takes one
set of syntactic arguments within one clausal domain (LØudrup 2002:125).
(139) En
A
mann sitter og
man sits
and
skiver dikt.
writes poems
SUB “man”
PRED “sit-and-write”
OBJ “poems”
And a constituent structure is shown in (137a-b):
(140) a) (han har) sittet
(he has) sat
b)
og
and
skrovet
written
dikt.
poems
VP
V
sat
IP
I
and
VP
V
NP
written
poems
In his account, LØudrup (2002) assumes that the grammatical marker og ‘and’ heads
the IP and that all pseudocoordination has the same basic constituent structure, which
they share with (ordinary) control and raising constructions.
On the other hand, De Vos (2004) states the following properties for PCO (in
English): a) the first conjunct is restricted to limited number of verbs, b) it allows
systematic violations of CSC, c) it yields aspectual interpretations (notably durativity), as
well as ‘surprise’ and pejorative readings, d) both verbs must have the same
morphological form (De Vos 2004:112).
203
According to De Vos (2004), PCO is not subordination for the following reasons: a)
coordinated verbs do not behave like auxiliaries (Pollock 1994): they can not be modified
by a both and they can not raise across negation to T, b) the subject of the embedded
clause cannot be licensed: it can not be PRO because V is not an infinitive; it can not be
pro because English is not a pro-drop language (to propose pro for English would be
counterintuitive); it is not a trace of raising because the V can assign case to the “moved”
NP and the expletive can not occur with it; it is not a trace of ATB because PCO would
be a garden-variety-coordination.
The proposal of De Vos (2004) states that PCO are complex heads derived in the
syntax itself (i.e. the construction is not a compound). The account unifies the behavior of
PCO and what he calls Reduplicative Coordination (ReCo). And example of ReCo is
given in (141) De Vos (2004:185):
(141) What did John read up and read up on?
The proposed structure is the following (De Vos 2004:189):
(142)
…
VP
Spec
V
V0
V
read
sit
go
XP
&
&
and
verbal complement
V
read
204
This proposal, according to De Vos (2004), has the following advantages: a)
extraction is allowed, therefore, not CSC violation happens, b) only a single subject is
projected by the complex predicate, c) V&V PCO and ReCo pattern alike with lexical
verbs in subject-aux-inversion and V to T raising, d) both is not tolerated in PCO and
ReCo because it contrasts two entire events.
For De Vos the particle and marks a transition between the two stages and it is a
two-place ‘sameness’ operator. In addition, “ReCo/PCO and is identical in its lexical
specifications to the garden-variety coordinator and” (De Vos 2004: 189). Therefore, at
the semantic level and takes ‘same’ categories and at syntactic level “the sole difference
between them is that garden variety and projects an entire XP of its own, ReCo/PCO and
projects only a head label” (De Vos 2004: 189).
Yaqui does not have pseudocoordinate structures. The language only has, in terms of
De Vos (2004), garden-variety-coordination and pseudosubordination. However, it is
interesting to analyze why Yaqui lacks that kind of structures and it is important to
analyze the structures used for coordination.
First, Yaqui does not have PCO because conjoined transitive verbs must have
their objects in overt syntax:
(143) u
ili
jamut lapis-ta
jinuk
small woman pencil
buy-PST
‘The girl bought a pencil and sold it.’
DET
into
and
a=nenka-k.
3NNOM.SG=sell-PST
Second, it is not possible to extract the object from a conjunct; therefore, there is no
CSC violation. The next example of ReCo shows that we must repeat the WH-question
in each conjunct if we want to have a grammatical coordinate sentence:
205
(144) Jitá bwiika
into jitá
bwiika
What sing.PRS
and
what sing.PRS
‘What does (the) John sings and sings?’
(145) *Jitá bwiika
into bwiika
What sing.PRS
and
sing.PRS
‘What does (the) John sings and sings.’
ju
DET
ju
DET
Joan?
John
Joan?
John
There are alternative resources for expressing repetition (i.e. alternatives to ReCo).
The main one is reduplication.
(146) Aapo ji-ji’oste-kai-su
lotte-k.
3SG RED-write.book-SUB-TERM be tired-PST
‘He wrote and wrote (book(s)) until he was tired’// ‘He is tired after being writing
and writing (book(s)).’
But, we have too the conjunction of reduplicated verbs, as indicated in the next
example:
(147) Malia jitá
ji-jinu-ka-n into
Mary what RED-buyand
‘What did Mary buy and buy?’
jitá
what
ji-jinu-ka-n.
RED-buy-PST-CONT
Another resource in the language is the repetition of the object:
(148) Joan dulsem
into dulsem
John candies
and
candies
‘John always eats candies and candies.’
jiba bwa-bwae.
always RED-eat
The previous data indicates that PCO was not attested (nor ReCo) in Yaqui. However,
the literature shows that we have to look at Pseudocoordination, Garden VarietyCoordination and Pseudosubordination if we want to have a better explanation of
coordination phenomenon.
In the first place, we have the observation that this classification lies mainly in two
parameters: a syntactic and a semantic one. Syntactically, pseudocoordination emerges
with two (or more) conjuncts joined by a coordinator, but semantically it violates the
206
CSC (Munn 1993, suggests that the CSC is a semantic constraint), although the truth
conditions
are
those
of
coordinated
structures
(LØudrup
2002,
fn.3:
127).
Pseudocoordination, syntactically is a subordinated clause, but semantically it behaves as
a coordinated one: it respects the CSC (Yuasa and Sadock 2002). A garden-varietycoordination syntactically has two (or more) conjuncts joined by a coordinator and
syntactically tends to respect the CSC.
Table 4.1 shows that the CSC is a violable constraint in some languages like Spanish
but not in languages like Yaqui. Therefore, if it exists, the CSC must be a soft constraint:
Table 4.1. Violability of the CSC
PseudoGardensubordination
Variety
Coordination
Pseudocoordination
Obey the CSC?
(Yaqui)
Obey the CSC?
(English)
Obey the CSC?
(Spanish)
PCO
ReCo
Not attested
Not attested
Yes
Yes
No
No
Yes/no
Not attested
No
No
Yes/no
Not attested
Examples:
In Yaqui: PCO and ReCo were not attested. But garden-variety-coordination respects
the CSC:
(149) Jabe-ta
bicha-k
into jabe-ta jikkaja
Who-NNOM.SG see-PST
and
who-NNOM.SG hear.PST
‘Who did Peter see and who did he hear?’
(150) *Jabe-ta
bicha-k
into
Who-NNOM.SG see-PST
and
(‘Who did Peter see and hear?’)
jikka
hear.PST
ju
DET
Peo.
Peter
ju
DET
Peo.
Peter
207
A pseudosubordinated construction respects too the CSC:
(151) Jabe-ta
bicha-kai
jabe-ta
into
Who-NNOM.SG see-SUB
who-NNOM.SG and
‘Who does Peter saw and who does (he) heard?’
jikaja-k ju Peo?
hear-PST DET Peter?
(152) *Jabetai
Who
Peo bicha-kai
Peter see-SUB
(into) jikaja-k.
(and) hear-PST
(153) *Jabetai
Who
Peo ai=
Peter 3NNOM.SG=
bicha-kai
see-SUB
(into) ai=
(and) 3NNOM.SG=
jikaja-k.
hear-PST
In English: Both PCO and ReCo violate the CSC43, as indicated below:
PCO
(De Vos 2002:112):
(154) What has John sat and done all day?
ReCo
(De Vos 2002:185):
(155) What did John read up and read up on?
A garden-variety-coordination does not violate the CSC:
(156) John saw Maria and Peter heard Juana’
(157) *Whom did John see and Peter hear Juana?
(158) *Whom John saw Mary and did Peter hear?
But ATB extraction is possible, therefore the CSC is violated:
(159) Whom did John see and Peter hear?
Also if the conjuncts make reference to a single object, the CSC is violated:
(160) What does Maria buy and sell?
43
Recall that the CSC is seen here as a violable constraint. It bans extraction from any coordinate
structure. The Across the Board principle is not a way to keep the CSC inviolable. We can dispense with it.
208
If we look at chaning structures, we can say that English does not have
pseudosubordination. The following chaining structure (161) contains conjoined
gerundive verbs that do not allow extraction from them (as indicated in (162)) but allow
extraction from the tensed clause (as seen in example (163)):
(161) ‘Looking at Maria, listening to Juana, the husband knew the secret.’
(162) *Whom looking at, listening to, the husband knew the secret?
(163) What, looking at Maria, listening to Juana does the husband knew?
On the other hand, it is not clear that Spanish has PCO because the attested examples
do not fit to the characteristics of English PCO (De Vos 2004). In Spanish the first
conjunct is not so restricted to a limited number of verbs, it does not yield special
aspectual interpretations nor have it ‘surprise’ and pejorative readings. However the
following kind of Spanish examples share the following properties with the PCO in
English: it violates the CSC, it shows too systematic violations of the CSC and it requires
that both verbs have the same morphological form. The next example could be placed in
the Garden-variety-coordination category.
PCO
(164) ¿Qué pensó
e
hizo Juan
What thought
and
did
John
‘What did John think and do all day?
todo
all
el
the
día?
day
ReCo
(165) ¿Qué
leyó y
leyó Juan?
What read and
read John?
‘What did John read and read?’
A garden-variety-coordination does not violate the CSC. The next examples indicate
that extraction is not possible, if the sentences contain different subjects:
209
(166) Juan vio
a
María y
Pedro escuchó a
John saw to
María and
Peter heard to
‘John saw Maria and heard Juana.’
Juana.
Juana
(167) *¿A quien
vio
Juan y
Pedro escuchó a
To whom saw John and
Peter heard to
(‘Who did John see and Peter heard Juana?’)
Juana.
Juana
(168) *¿A quien vio Juan a María y escuchó Pedro?
(‘Who did John see Mary and Peter heard?’)
But ATB extraction is possible, therefore the CSC is violated:
(169) ¿A
quien vio
Juan y
escuchó
To
whom saw John and
heard
‘Who did John see and Peter hear?
Pedro?
Peter?
But if the conjuncts make reference to a single subject, the CSC is violated:
(170) ¿Qué compra
y
vende Maria?
What buy
and
sell
Maria?
What does Maria buy and sell?
It seems that Spanish does not have pseudosubordination. The following chaining
structure contains conjoined gerundive verbs that do not allow extraction from them but
allow extraction from the tensed clause:
(171) Analizando a
Maria, oyendo
a
Juana,
Looking
to
Maria, hearing
to
Juana
el
marido
supo el
secreto.
the
husband
knew the
secret.
‘Looking Maria, listening Juana, the husband knew the secret.’
Extraction from an adjunct:
(172) *A quien
analizando, oyendo
a
Juana,
To whom
looking,
listening
to
Juana
el
marido
supo el
secreto.
the
husband
knew the
secret.
(‘Whom looking, listening Juana, did the husband know the secret?’)
210
Extraction from the main clause:
(173) Qué, analizando
a
Maria, oyendo
a
Juana,
what looking
to
Maria, listening
to
Juana
supo el
marido.
knew the
husband.
‘What, looking Maria, listening Juana, did the husband know?’
4.1.5.1
Insights
The CSC is used as a test for coordinated constituency. It is central in the above
classification. However, the data shows that CSC is just one of several constraints
interacting in the make up of coordinate constructions. In addition, the data shows that is
not easy to establish the line between coordination and subordination.
The occurrence of PCO is an argument against conjunction reduction because we can
not say that sentence (174a) is the source of sentence (174b):
(174) a) John went and John drank beer.
b) John went and drank beer.
The semantic distinction is a reflex of a syntactic distinction.
Another characteristic that is worth noticing is that pseudocoordination is reduced to
sentences in which the same subject (grammatical or logical) is involved; the same
happens in Yaqui with pseudosubordination. (This fact is different for Japanese because
the te-constructions can contain different subjects (Yuasa and Sadock 2002): they are
control structures.) The same subject is understood in all the clauses of the chain, as
shown in Table 4.2.
211
Table 4.2. The
Pseudosubordination
same
subject
requirement
Pseudocoordination
Same subject
PCO
SS *DS
(English)
(Spanish?)
ReCo
SS *DS
(English)
(Spanish)
of
Pseudocoordination
and
Garden-Variety
Coordination
Pseudosubordination
SS, DS
(Yaqui)
(English)
(Spanish)
SS, *DS
(Yaqui)
Other characteristics of these constructions in English, Spanish and Yaqui are
summarized in Table 4.3. The table summarizes various observations:
(175) a) The three languages have garden-variety-coordination.
b) English and Spanish have PCO whereas Yaqui does not have it.
c) English and Spanish does not have pseudosubordination, whereas Yaqui has
it.
d) There are several aspects that occur cross linguistically: Any number of
conjuncts can occur, reversibility, scope of negation and the sameness
constraint.
212
Table 4.3: Some contrasts between Pseudosubordination, Garden Variety
Coordination and Pseudosubordination
Pseudocoordination
Garden-Variety PseudosubCoordination
ordination
PCO
ReCo
*Backward
pronominalization
Any number of
conjuncts can occur
Reversibility
Syntactic
structure
Yes:
*Did hei read up
and did Johni
read up on?
Yes:
(English)
(Spanish)
Juan rezó, rezó
y rezó hasta que
se cansó.
Yes
(English)
(Spanish)
??
Categorial Main verbs
Sort
(English)
(Spanish)
Yes
(English)
(Spanish)
Es falso que
leyó y leyó
hasta que se
cansó.
Reduced to
some verbs in
English.
Common
coordination in
Spanish
Main verbs
(English)
(Spanish)
coordinate coordinate
coordinate
Scope (both affected
by negation)
Sameness
Constraint
Yes:
* Has hei sat and
has Johni done all
his homework?
No
(English)
Yes
(Spanish)
Que pensó, dijo e
hizo Juan todo el
día?
No
English
Yes: (Spanish)
Qué hizo, dijo y
pensó Juan todo
el dia?
Yes
(English?)
(Spanish)
No es cierto que
compró y vendió
el burro.
Semantic
Sort
Yes:
(English)
(Spanish)
(Yaqui)
Yes:
(English)
(Spanish)
(Yaqui)
Pedro trabajó,
estudió, e hizo
la tarea.
Yes
(English)
(Spanish)
(Yaqui)
Yes:
(Yaqui)
Yes
(English)
(Spanish)
(Yaqui)
Yes
(Yaqui)
Any verb can
be coordinated
in this way
(English)
(Spanish)
Any verb
(English)
(Spanish)
(Yaqui)
Main and
auxiliary verbs
(English)
(Spanish)
subordinated
Any verb
(Yaqui)
Yes
(Yaqui)
Yes
(Yaqui)
Main
verbs.
(Yaqui)
213
In (120), repeted here, I showed that in Yaqui, the CSC is always satisfied because of
the ranking *EXTRACTION >> DGF, *COORD-NON-MAX.
a.)
b.
c.
jabeta Joan ateak into jabeta a tebotuak
jabeta Joan ateak into tebotuak
jabeta Joan ateak into a tebotuak
DGF
* EXTRACTION
Input: {jabeta, Joan, ateak, into, jabeta, a, tebotuak}
*COORD-NON-MAX
(176) Tableau with the ranking *EXTRACTION >> DGF, *COORD-NON-MAX ( = 120)
**
!*
!*
*
*
For languages like English and Spanish, where we have reduction of grammatical
roles and coordination of heads is allowed, the ranking will be reversed. In such cases the
candidate with the structure of (167b) or (167c) will emerge as optimal, depending on the
nature of the input and the interrelation within other ranked constraints.
One aspect of PCO is that it seems to be licensed by the coordination of heads. This
aspect is related too to the possibility of having RNR structures in those languages.
We saw that it would be difficult to say that Yaqui has PCO. In a similar way, we
predict that Yaqui will not have RNR structures. At first sight, that seems to hold in
Yaqui, however, if we compare the kind of data introduced by Cann et al (2005) with
similar constructions in Yaqui, we find that similar problems are recreated.
The next example shows a typical example of RNR. But in Yaqui, an overt pronoun
is required in the canonical position, whereas in English and Spanish it is not required:
214
(177) ume ili
usim ka=ai tu’ure amako,
into
DET.PL small boy
not=it like sometimes, and
ket
ka=ai wantaroa
[ame-u
o’omti-wa-ko]i
too
not=it support
3PL.OBL-DIR angry-PASS-when
‘Children do not like and sometimes they do not support the anger towards them’
The following example shows that there can be more than one right dislocated
expression giving rise to apparent non-constituent coordination.
(178) Joan yew=am
go’ota-k,
Peo into ye=am
John out=3NNOM.PL throw-PST
Peter and
HO=3NNOM.PL
maka-k
ume maestro-ta
ji’oste-im
jabe-ta
make-PST
DET.PL teacher-NNOM.SG
writing-PL
who-NNOM.SG
ama a=
wanta juni’i.
there 3NNOM.PL= like even
‘John copied and Peter gave the teacher’s writings to whoever asks for them.’
The following example shows that the dependency occurs into a strong island:
(179) Joan a=
jinu-pea
Peo into junea
Joan 3NNOM.SG= buy-want
Peter and
know
jaisa teak uka
karro-ta
1980 ne-nenka-me.
how name the.NNOM.SG car-NNOM.SG 1980 RED-sell-SUBJ.REL
‘John wants to buy and Peter knows the name of the person who sells a 1980 car.’
But not every pronoun could give rise to a structure that we can consider RNR. The
following example has translations that indicate that they are grammatical in English and
Spanish and are not RNR.
(180) Jose aman pasiyaloa-pea ta
a
beas
but
3SG really
Jose there visit-DES
kopti-la-wa
a=
samai-wa-ta
jo’aka-po.
forget-PFV-PASS
3SG.POSS=
aunt-POSS-NNOM.SG live-LOC
‘Jose wants to visit there, but he really forgot where his aunt lives’
(181) Jose
a-u
pasiyaloa-pea ta
a
beas
Jose
3NNOM.SG-DIRvisit-DES
but
3SG really
kopti-la-wa
[a=
samai-wa-ta]
jo’aka-po
forget-PFVA-PASS
[3SG.POSS= aunt-POSS-NNOM.SG] live-LOC
‘Jose wants to visit her, but he really forgot where his aunt lives’
215
I don’t cover this topic in the OT framework, but instead point it out as an interesting
area for future research into the Yaqui language. It is clear that it is not by accident that
Yaqui does not allow pseudocoordination. The explanation of the three phenomena by
using the same set of constraints with different ranking seems to be promising.
4.2
Conclusion
In this chapter I presented the main properties of verbal coordination. The focus was
the --kai construction because it has subordinating and coordinating characteristics.
This section presents an analysis of the main properties of Yaqui verbal
coordination. The characteristics that are described and explained are summarized here:
a) related to balanced coordination: Balancedness for Tense, Number and Mood and nonviolation of CSC; b) related to unbalanced coordination (pseudosubordination): The tense
marking occurs in the final conjunct; the order of the conjuncts tends to be sequential, but
reversibility is possible; the construction makes reference to a same subject and it is not
repeated; the suffix -kai appears on each verb; the construction makes reference to a
sequential event; in a semantic coordinated chain, into ‘and’ can not occur between --kai
clauses. This chapter uses the notion of a coordinator like an adjunct which attaches to a
host CP and licenses the addition of a new CP (the first coordinator). We saw too that in
Yaqui the CSC is respected. In the final part of this chapter I have presented an overview
of pseudocoordination, pseudosubordiantion and coordination.
216
5
NOMINAL COORDINATION
This chapter is about Yaqui nominal coordination. However, in order to have a better
understanding of the principles that regulate Yaqui nominal coordination, it is necessary
to describe first the morphology of nouns and verbs. So, the reader will find first a
description of nominal and verbal classes and, after that, an explanation about how the
number morphology of coordinated nouns interacts with verbal number requirements. In
the final part, I analyze in the OT framework some challenging asymmetries in agreement
that can be problematic for an LFG account along the lines of Halloway King &
Dalrymple (2004). Those researchers split number features into two types: CONCORD
features and INDEX features. This partition allows them to explain agreement facts
between coordinated nouns and determiners in English. I apply those ideas to the
agreement between coordinated nouns and verbs and show that the idea of two number
features is useful but that there are some unexpected patterns that can be explained using
the OT framework.
5.1
Background on Nominal and verbal classes
5.1.1
Number in nouns and in verbs
In this section I show first that nouns are subject to morphological requirements and
they form three classes based on their ability to take a singular or plural marker. After
that, I show that verbs form three classes too, according to their requirements of singular
or plural arguments. Finally I analyze the interaction between morphological
217
requirements of nouns and morphological requirements of verbs in connection with
coordination of nouns.
5.1.1.1 Nominal classes
Nouns are subject to different morphological requirements for inflection. They can be
divided into three groups: nouns that can be marked for both singular and plural, nouns
that can only be marked for singular, and nouns that can only be marked with plural. Lets
call them class one (N1), class two (N2) and class three (N3). The classes can be seen
below, where the symbol (-) indicates that the noun cannot take the indicated marker:
plural or singular.
(1)
Class one (N1)
SG
PL
Class two (N2)
SG
(-)
Class three (N3)
(-)
PL
Examples from each class are given in what follows. The singular is indicated by the
zero marker in nominative, whereas the plural is the marker -(i)m. When the noun is nonnominative, the singular is marked with --ta and the plural is again marked with -(i)m.
The markers --ta and --(i)m are mutually exclusive.
(2)
Class one: nouns that take both singular and plural. The examples are in nonnominative form.
Singular
Plural
Gloss
kawis-ta
ousei-ta
koowi-ta
kawis-im
ousei-m
koowi-m
‘fox’
‘lion’
‘pig’
218
(3)
(4)
Class two: nouns that take only singular. The examples are in non-nominative
form.
Singular
Plural
Gloss
naposa-ta
seberia-ta
seé’e-ta
*naposa-m
*seberia-m
*seé’e-m
‘ash’
‘cold’
‘sand’
Class three: nouns that take only the plural marker. As indicated by the asterisk,
they can not be marked singular (nominative or non-nominative), they require to
be marked always with --(i)m ‘PL’. However, in spite of this marking, their
meaning could be singular or plural (in other words, they are unspecified for
number).
Singular
Plural
Gloss
*supe/ *supe-ta
*puusi/ *pusi-ta
*boocha/ *bocha-ta
supe-m
puusi-m
boocha-m
‘shirt(s)’
‘eye(s)’
‘shoe(s)’
The semantics of these nouns (N3) have been explored by Buitimea Valenzuela
(2003), who concluded that they make reference to body parts, instruments, large objects,
some reptiles (“medium size”, with legs), clothing, food, diseases, some collectives, and
nouns that express volume. Some examples are given next Buitimea 2003:16-32):
(5)
Noun
Gloss: SG/PL
mamam
jeemam
tepuam
kuetem
sakkaom
bejo’orim
piisam
bwajim
nojim
keesum
keekam
kapichooram
bwassumiam
‘hand(s)’
‘liver(s)’
‘ax(es)’
‘sky rocket(s)’
‘gila monster(s)’
‘lizard(s) (kind of)’
‘blanket(s)’
‘underpants’
‘tamale(s)’
‘cheese(s)’
‘mange(s)’
‘smallpox(es)’
‘tress(es)’
219
opoam
jaakam
‘tear(s)’
‘phlegm(s)’
The N3 class can never be marked with -ta ‘NNOM.SG’ because this suffix is mutually
exclusive with --(i)m ‘PL’ (Escalante 1990, Dedrick and Casad 1999). The exclusion
relation is illustrated below:
(6)
nem
juubi chu’u-ta
tu’ure.
1SG.POSS
wife dog-NNOM.SG like.PRS
‘My wife likes the dog.’
(7)
nem
juubi chu’u-m
1SG.POSS
wife dog-PL
‘My wife likes the dogs.’
(8)
*nem
juubi chu’u-m-ta
1SG.POSS
wife dog-PL-NNOM.SG
(‘My wife likes the dogs.’)
tu’ure
like.PRS
tu’ure.
like.PRS
Based on the collected data, we can establish that class one is the most abundant and
probably the unmarked case: it can host singular and plural markers. Class two and class
three are the marked ones; they only accept one number marker.
(9)
Class one (N1)
SG
PL
unmarked
Class two (N2)
SG
(-)
marked
Class three (N3)
(-)
PL
marked
The morphological requirements of noun classes seen above interact with the
morphological requirements of verbs. The verbal classes are shown below:
5.1.1.2 Verbal classes
Looking at number, Yaqui verbs can be classified in three classes too: the verbs that
can take a singular or plural noun, those that take only singular nouns, and those that take
220
only plural nouns. They can be intransitives or transitives. Although case marking does
not relate in Yaqui to argument structure, the exploration of transitives and intransitives
will be done further because it is relevant for the coordination patterns. We will see a set
of suppletive verbs that have different patterns of coordination when they agree with the
subject (intransitives) than when they agree with the object (transitives).
(10)
Class one (V1)
SG
PL
Class two (V2)
SG
(-)
Class three (V3)
(-)
PL
Exemplification of each class is given in what follows:
(11)
Class one (V1): verbs that take either a singular or plural noun. They are not
marked for singular or plural agreement.
Singular
a) Wiikit
aa
nen-ne’e.
Bird
can
RED-fly.PRS
‘The bird can fly.’
Plural
b) ju-me
wiikich-im
DET.PL
bird-PL
‘The birds are flying.’
(12)
nen-ne’e.
RED-fly.PRS
Class two (V2): verbs that take only singular nouns (suppletive verbs).
Singular
a) Uusi
Vicamme-u siika.
Boy
Vicam-DIR
go.SG.PRS
‘The boy is going to Vicam.’
Plural
b) *Uusi-m Vicamme-u siika.
Boy-PL Vicam-DIR
go.SG.PRS
(‘The boys go to Vicam.’)
221
(13)
Class three (V3): (suppletive) verbs that take only plural nouns as argument. As
asterisks indicate, the nouns cannot be singular, they are required to be always be
understood as plural.
Singular
a) *Peroon-im
pesio-u
siika.
Soldier-PL
Hermosillo-DIR
go.SG.PRS
(‘The soldiers are going to Hermosillo.’)
Plural
b) Peroon-im
pesio-u
sajak.
Soldier-PL
Hermosillo-DIR
go.PL.PRS
‘The soldiers are going to Hermosillo.’
According to this classification, we can again establish that class one (V1) is the
unmarked situation: it can take both singular and plural nouns as arguments. Class two
(V2) and class three (V3) probably are the marked ones, they only accept singular or
plural nouns.
(14)
Class one (V1)
SG
PL
Unmarked
Class two (V2)
SG
(-)
marked
Class three (V3)
(-)
PL
marked
Classes V2 and V3 are composed of suppletive verbs, they can be intransitives or
transitives. The intransitive ones show agreement with the subject whereas the transitive
ones agree with the object of the sentence. Because they behave differently in relation
with coordinated nouns, they are described separately in this work. Now, I want to show
that nominal and verbal classes interact in interesting ways. Let’s analyze this kind of
interaction:
222
5.1.2
Interactions between nouns and verbs
In what follows, the continuous arrows indicate that the noun combine with the
signaled verb: the N1 combine with all verbs, N2 only combine with V1 and V2, and N3
combine with all verbs too.
There are several things to see in this picture. First, why can N3 (plural) combine
with V2 (singular), contrary to what was seen before? And second, why do N2 (singular)
and V3 (plural) not combine in a similar fashion? Another question that needs
clarification is the underspecification for number found in N3 when combined with V1
(see example (22)).
(15)
N1
SG
PL
N2
SG
(-)
N3
(-)
PL
V1
SG
PL
V2
SG
(-)
V3
(-)
PL
The following examples indicate that N1 combine with all verbal classes:
(16)
N1
buuru
bachi-ta
donkey
corn-NNOM.SG
‘The donkey ate corn.’
(17)
N1
buuru
aman
donkey
there
‘The donkey ran there.’
V1
bwa-ka.
eat-PST
V2
buite.
ran.SG.PRS
223
(18)
N1
buuru-m
aman
donkey-PL
there
‘The donkeys ran there.’
V3
tenne.
ran.PL.PRS
The next examples indicate that N2 combines only with V1 and V2, but not with V3.
(19)
N2
See’e
Sand
‘The sand will finish.’
V1
lu’uti-bae.
finish-INTT.
(20)
N2
V2
See’e
kora-po
bo’ote.
Sand
yard-LOC
lying.SG.PRS
‘The sand is lying in the yard (corral).’
(21)
N2
V3
*See’e
kora-po
to’ote.
Sand
yard-LOC
lying.PL.PRS
(‘The sand is lying in the yard (corral).)’
Finally, N3 (plural) combines with all verbs. When combined with V1, the resultant
sentence is ambiguous (or underspecified) for number, i.e. it can be interpreted as
singular or plural; see example (22). When combined with V2 the verb gives the singular
interpretation of the noun marked morphologically with a plural, and with V3 the
interpretation is plural:
(22)
N3
V1
wuikui-m
inim
jo’a.
live.PRS
alligator-PL here
‘The alligator lives here’/’the alligators live here.’
(23)
N3
V2
wuikui-m
jupa-u
weye.
alligator-PL tree-to
go.SG.PRS
‘The alligator is going to the tree.’
224
(24)
N3
V3
wuikui-m
jupa-u
sajak.
alligator-PL tree-to
go.PL.PRS
‘The alligators are going to the tree.’
The observed data present some apparently simple problems that were solved in a
separate paper (Langendoen & Martínez Fabián 2004), such as why the suffixes --(i)m
‘plural’ and --ta ‘non nominative’ do not combine, as indicated in (7) and why N3
(marked with plural) can combine with all verbs. Langendoen & Martínez Fabián
conclude that the N3 class is subject to a set of constraints such as HAVE-AFF(IX) which
indicates a preference of the language system for having inflected forms (that will explain
why --(i)m ‘PL’ must be present in each form of N3). In addition, we have the interaction
of the constraint *CASE, which requires that forms not be inflected for Case. The
constraint FAITH-FS requires faithfulness to feature specification in the input and the
constraint *NUM(BER) rules out candidates marked for number. The next table indicates
the ranking HAVE-AFF, *CASE>>FAITH-FS>> *NUM. The example shows in the input a
noun of the N3 class. It has the feature specification [Accusative & Singular]. The
winning candidate, (25c), violates the constraint FAITH-FS because it emerges with the
feature [Plural], but it respects the higher ranking HAVE-AFFIX and *CASE.
(25)
Choice of supe-m for expressing supe [Acusative] & [Singular]
supe [Accusative] & [Singular]
HAVE-AFF
a.
supe
*!
b.
supe-ta [Acusative]
c.)
supe-m [Plural]
*CASE
FAITH-FS
*!
*
***
*NUM
*
Although the ranking gives the correct output, it predicts that the noun will be
interpreted exclusively as plural, i.e. it does not explain why those nouns are
underspecified for number when combined with a verb of the N1 class. In the section I
show that we need to state that there are two types of features (CONCORD and INDEX
225
features in the sense of the Halloway King and Dalrymple 2004) which give rise to the
patterns found in Yaqui.
Some of these constraints in table (25) will be used in the final part of this section
where agreement patterns are analyzed. For now, because in this work the interest is
centered on coordination, the next section explores the behavior of nouns and verbs under
coordination.
5.2
Noun coordination and verbal agreement
5.2.1
Noun coordination and intransitive suppletive verbs
This section illustrates the behavior of suppletive intransitive verbs. These function in
a different way than transitive verbs with respect to noun coordination. As we saw earlier,
most verbs in Yaqui don’t give information about number; however, there are some
intransitive Yaqui verbs which are suppletive for singular and plural44. The following
examples in (26a) are the intransitive counterparts of the transitive verbs described in the
next section.
(26)
Singular
a) yejte
kikte
bo’ote
Plural
jo’ote
ja’abwe
to’ote
Gloss
‘to sit, to stand up’
‘to stop’,’to get up’,’to put up’
‘to lie down’
44
There are some suppletive forms for past and stative meaning too. For example, the following
singular verbs (and the plurals too) vary according to this aspect. Because this issue is not relevant for
coordination, I leave this aside here.
Singular/Plural
Past/Stative
yejte/katek
kikte/japtek
‘to sit, to stand up’
‘to stop’,’to get up’,‘to put up’
226
b) buite
siika
weeye
tenne
sajak
kaate
‘to run’
‘to arrive’
‘to go’
Coordination of two (or more) singular nouns requires a plural verb (this is contrary
to transitive verbs agreeing with the object: two coordinated singular nouns require a
singular verb). In the following example, we can see that nominative case is recognized
because of the lack of morphological marking.
(27)
yooko
[Joan into Peo] tenni-bae
tomorrow
John and
Peter run.PL-INTT
‘John and Peter will run tomorrowi’
(28)
*yooko
[Joan into Peo] buiti-bae.
Tomorrow
John and
Peter run.SG-INTT
(‘John and Peter will run tomorrow’)
A singular noun coordinate with a plural noun (class 1) combine with a plural verb.
The next example shows that the order of the conjuncts does not matter for the verbal
requirements:
(29)
jume uúsi-m
into ju’u
maejto
DET.PL child-PL
and
DET.SG
teacher
‘The children and the teacher went over there.’
aman saja-k.
there go.PL-PST
(30)
ju’u
maejto
into jume uúsi-m
DET.SG
teacher
and
DET.PL child-PL
‘The teacher and the children went over there.’
aman sajak.
there go.PL.PST
More interesting are the requirements found with nouns of the class 3 (the ones that
must be always marked with --(i)m ‘PL’). A non-coordinate noun combined with a
singular verb is interpreted as singular; if the verb is plural it is interpreted as plural. If
the verb does not mark number, depending of the context, it can be interpreted as singular
or plural:
227
(31)
bejo’ori-m
nas
bui-buite
lizard-PL
DIR
RED-run.SG.PRS
‘The lizard is running (without a specific direction).’
(32)
bejo’ori-m
nas
tet-tenne.
lizard-PL
DIR
RED-run.PL.PRS
‘The lizards are running. (without a specific direction)’
(33)
bejo’ori-m
yumjoe-(mme).
lizard-PL
rest.PRS-(3PL)
‘The lizard(s) is/are resting.’
When the coordinate nouns combine with a verb that does not indicate number, the
nouns can be interpreted as singular or plural. There is uniformity in the interpretation;
both conjuncts must be interpreted as singular or both as plural.
(34)
bejo’ori-m
into sakkao-m
inim yumjoe.
lizard-PL
and
gila.monster-PL
here rest.PRS
‘The lizards and the gila monsters are resting here.’
‘The lizard and the gila monster are resting here.’
*’The lizard and the gila monster are resting here.’
*’The lizard and the gila monster are resting here.’
The use of numerals allows expressing the number of the conjunct:
(35)
wepul bejo’ori-m
into wepul sakkao-m
one lizard-PL
and
one
gila.monster-PL
‘One lizard and one gila monster are resting here.’
(36)
bejo’orim
into wepul sakkaom
yumjoe.
‘The lizard and one gila monster are resting.’
(37)
Wepul
bejo’orim
into sakkaom
‘One lizard and the gila monsters are resting.’
inim
here
yumjoe.
rest.PRS
yumjoe.
Two plural nouns of the N3 class require a plural verb too. The same happens with
the combination of a singular and a plural noun.
228
5.2.2
Summary
The next table summarizes the patters of nominal coordination combined with an
intransitive verb. We can see that all combinations result in a plural agreement.
Therefore, our explanation must allow the generation of these candidate structures too.
(38)
[N + N]
SG+SG
PL+PL
SG+PL
PL+SG
VINTR
PL
PL
PL
PL
Interpretation
PL
PL
PL
PL
And we need to rule out the following unattested patterns.
(39)
[N + N]
SG+SG
PL+PL
SG+PL
PL+SG
VINTR
SG
SG
SG
SG
Interpretation
*
*
*
*
Discontinuous coordination occurs with intransitives too, as the following example
indicates the verb must be singular if the preverbal subject is singular.
(40)
Yoeme
juya-u
siika into
Man
forest-DIR
go.SG and
‘The man went to the forest and the boys (too).’
uusi-m (ketchia).
boy-PL (too)
But this example might not be a real discontinuous coordination; it might be analyzed
as an example of sentence coordination. If it were a real nominal coordination, the
following sentence would be grammatical because in Yaqui preverbal coordinated nouns
require a plural verb, if it were a split coordinate subject, the occurrence of a plural verb
would be expected:
(41)
*Joan sajak
into Peo (ketchia).
and
Peter (too)
John go.PL.PST
(‘John went and Peter (too’)).
229
5.2.3
Analysis
The following section explores the morphological features present in the intransitive
verbs and in the nominal classes. This section is the background for the explanation of an
asymmetry in agreement between conjoined nominals and verbs: conjoined singular
nominals as subjects of intransitive verbs require a plural verb whereas conjoined
singular nouns functioning as object of a transitive verb requires a singular verb.
5.2.4
Analysis of the interaction between coordinate nouns and verbs
This section focuses on the OT analysis of the agreement patterns emerging between
coordinate nouns and verbs. The first part presents some background about the number
system of features proposed by Halloway King and Dalrymple (2004) and in the final
part I present my own analysis.
5.2.4.1 The system of concord and index features
Within the framework of the Lexical-Functional Grammar (LFG), Halloway King
and Dalrymple (2004) propose that there are two types of features active in coordinate
structures: CONCORD features and INDEX features. They use this distinction for explaining
noun agreement in English. I introduce the concepts by looking at the Yaqui examples.
Determiner noun agreement in Yaqui seem to be straightforwardly describable: a
singular determiner goes with a singular noun and a plural determiner goes with a plural
noun.
(42)
Ju’u
wiikit ne-ne’e.
RED-fly
‘The bird flies.’
DET.SG bird
230
(43)
Jume wikich-im
DET.PL bird-PL
‘The birds fly.’
ne-ne’e.
RED-fly
Holloway King and Dalrymple (2004), following Wechsler & Zlatić (2000) suggest
that there are two types of agreement features associated with nouns, CONCORD features
and INDEX features. The first ones are closely related to the declension class of a noun
and generally control agreement between a noun and its determiners and adjectives. The
second ones are closely related to the noun’s semantics, control agreement between a
noun phrase and a bound pronoun and often control verb agreement (Holloway King &
Dalrymple 2004:71).
These researchers exploit the distinction between nondistributive and distributive
features. CONCORD features are treated as distributive (each conjunct must bear it)
whereas INDEX features are taken as nondistributive (i.e. they are carried by the entire set,
in LFG terms). Distributive features allow an explanation of the agreement inside a
coordinate phrase (as, for example, between a singular determiner and a singular noun).
On the other hand, nondistributive features allow us to explain, for example,
agreement between a coordinate phrase as a whole with a verb. The next example
illustrates the idea about CONCORD and INDEX features. I use a different representation
than that used for Holloway King & Dalrymple (2004) because I am not assuming the
LFG framework nor the functional representation proposed by them. We can see that the
CONC(ORD) singular features in the next example license the agreement inside the
coordinate phrase, while the IND(EX) feature corresponding to the coordinate structure as
a whole agrees with the plural verb. The representation only intends to show the idea of
231
how those features work. It does not reflect the use of these features in the LFG
framework. In order to facilitate the reading, CONCORD features are represented as
subindices whereas INDEX features are represented as superindices.
(44)
jamut CONC:SG into ju
yoemeCONC:SG]IND:PL
DET.SG woman
and
DET.SG man
‘The woman and the man left.’
[Ju
sajak.CONC:Ø IND:PL
left.PL
The system proposed by Holloway King & Dalrymple (2004) predicts that there
would be, at least, four types of verbs:
(45)
a)
V
[INDEX]
b) V
[CONCORD]
c) V
[Ø]
d)
V
CONCORD
INDEX
The analysis of Yaqui indicates that the verbs respond to the following
representations:
For class-1 Verbs (the ones that do not mark number) the representation is that of
(45c), repeated here as (46):
(46)
tekipanoa
‘work’
V
[Ø]
The predictions generated by the representation of this type of verbs are correct. First,
it predicts that the verb can combine with singular or plural nouns (see examples (47) and
(48)). I assume that nouns carry CONCORD and INDEX number features. The CONCORD
feature is represented attached to the noun whereas INDEX feature is attached to the whole
parenthesis. The verb does not carry any number information. I use the symbol Ø for
representing the idea that the verb does not have any CONCORD/INDEX number
232
information. However, in the last part of this section, I suggest that the verb is
underspecified for number.
(47)
[UusiCONC:SG]IND:SG
Boy
‘The boy is working.’
tekipanoaCONC:Ø/IND:Ø
work.PRS
(48)
[Yoeme-mCONC:PL]IND:PL
Man-PL
‘The men are working.’
tekipanoaCONC:Ø/IND:Ø
work.PRS
Second, it predicts that this type of verb can combine with coordinate singulars. Such
a prediction is confirmed in the following example (49):
(49)
yoemeCONC:SG]IND:PL
[UusiCONC:SG into
Boy
and
man
‘The boy and the man are working.’
tekipanoa.CONC:Ø/IND:Ø
work:.PRS
Third, it is predicted their combination with coordinated singular and plural nouns.
(50)
[MaejtoCONC:SG
into jaamuch-imCONC:PL] IND:PL
Teacher
and
woman-PL
‘The teacher and the women are working.’
tekipanoa.CONC:Ø/ IND:Ø
work.PRS
Finally, the prediction is that it can appear too with coordinate plurals:
(51)
[Uusi-mCONC:PL into jaamuch-imCONC:PL]IND:PL
Boy-PL
and
woman-PL
‘The boys and the women are working.’
tekipanoa. CONC:Ø/ IND:Ø
work.PRS
Verbs of class V2 have the following representation:
(52)
siika ‘go.SG’
V
CONCORD: SG
INDEX: SG
For that reason, the predictions are the following ones: they can be used with singular
nouns. As before, in the representation, CONCORD features are attached to the noun,
233
whereas INDEX features are attached to the whole unit (to the parenthesis in the
representation). The verb carries both features. The first one is the CONCORD feature and
the last one the INDEX feature. The next sentence shows that both type of number features
match, therefore, the sentence is predicted to be grammatical.
(53)
[[uusi] CONC:SG] IND:SG
boy
‘The boy left’
siika CONC:SG/IND:SG
leave:SG.PST
Because the INDEX feature in the verb is singular, it cannot combine with conjoined
singular nouns (which have a plural INDEX feature). So the following sentence is
correctly ruled out as ungrammatical:
(54)
into
*[uusi] CONC:SG
boy
and
(‘The boy and the woman left.’)
[jamut] CONC:SG] IND:PL
woman
siika CONC:SG/ IND:SG
leave.SG.PST
It is predicted too that the verb cannot be used with mixed conjoined singular and
plural nouns (the order of the conjuncts does not matter). The verbal CONCORD singular
feature does not combine with the CONCORD plural feature of the plural conjunct.
(55)
into uusi-m CONC:PL] IND:PL
*[jamut CONC:SG
woman
and
boy-PL
(‘The woman and the children left.’)
siika CONC:SG/ IND:SG
leave.SG.PST
(56)
*[uusi-m CONC:PL
into jamut CONC:SG] IND:PL
boy-PL
and
woman
(‘The children and the woman left.’)
siika CONC:SG/ IND:SG
leave.SG.PST
The most restrictive situation is when both features (CONCORD and INDEX) are
imposed in the system. So, if we have a coordinate structure with a CONCORD singular
feature, it is predicted that the coordination will be singular. That is, it must refer to a
single individual. This prediction holds in Yaqui. The following example could be
234
considered a coordinate structure in spite of the fact that it does not bear in open syntax a
coordinator. The coordinate nouns make reference to a single individual:
(57)
nim
[compai CONC:SG
jalai CONC:SG] IND:SG
1SG.POSS
fellow parent
friend
‘My fellow parent and friend left.’
siika CONC:SG/ IND:SG.
leave.SG.PST
Intransitive verbs of the V3 class have the following representation. In it, the
CONCORD feature is irrelevant. The INDEX feature must agree with a plural subject.
(58)
sajak
‘go.PL’
V
CONCORD: Ø
INDEX: PL
Therefore, these verbs can combine with a single plural noun but they never combine
with a single singular noun. In (59) the verbal INDEX feature match the nominal INDEX
feature, for that reason the sentence is grammatical, whereas in (60) the verbal plural
INDEX feature does not match the nominal singular INDEX feature. Therefore, the sentence
is ungrammatical.
(59)
[samireo-m CONC:PL] IND:PL
adobe.maker-PL
‘The adobe makers left.’
sajak CONC:Ø/ IND:PL
leave.PL.PST
(60)
*[samireo CONC:SG] IND:SG
adobe.maker
(‘The adobe maker left.’)
sajak CONC:Ø/ IND:PL
leave.PL.PST
It is predicted that the verb must be compatible with conjoined singular nouns if and
only if the resulting phrase refers to more than an individual. The example (60) shows
that the prediction holds in Yaqui. On it, both INDEX features match. If the structure
refers to just one individual, as in (61), the INDEX features do not match and the sentence
is ungrammatical.
235
nim
juubi CONC:SG] IND:PL sajak CONC:Ø/IND:PL
1SG.POSS wife
leave.PL.PST
(61)
[nim
uusi CONC:SG
into
1SG.POSS
boy
and
‘My son and my wife left.’
(62)
*nim
[compai CONC:SG
jalai CONC:SG] IND:SG
1SG.POSS
fellow parent
friend
(‘My fellow parent and friend left.’)
sajak CONC:Ø/ IND:PL
leave.PL.PST
Because only the INDEX feature is relevant for this type of verb, they can appear with
coordinated plural nouns. The next example shows that the INDEX feature of the
conjoined nominal matches the plural INDEX feature of the verb. Therefore, the sentence
is grammatical:
(63)
jaamuchi-m CONC:PL] IND:PL
sajak CONC:Ø/ IND:PL
[uusi-m CONC:PL into ili
boy-PL and
small woman-PL
leave.PL.PST
‘The boys and the girls left’
These types of verbs can be used too with coordinate singular and plural nouns. As
we see in example (63), the INDEX feature of the nouns as a whole match the INDEX
feature of the verb and grammaticality is predicted:
(64)
ili
jaamuchi-m CONC:PL]IND:PL
[uusi CONC:SG into
boy
and
small woman-PL
‘The boy and the girls left.’
sajak CONC:Ø/IND:PL
leave.PL.PST
We have seen that the behavior of intransitive verbs is explained under the
assumption that there are two types of features involved in noun-verb agreement.
However, the picture seen until now is not so clear when we analyze the relation held
between the N-3 class and suppletive verbs. Before analyzing this relation, let’s look at
the suppletive transitive verbs.
236
5.2.5
Noun coordination and transitive suppletive verbs
Consider the following set of verbs which agree with the sentence’s object. The verbs
are suppletive for number: the paradigm is shown in (65):
(65)
Singular object
Plural object Gloss
yecha
kecha
teeka
me’a
joá
ja’abwa
to’a
sua
‘to sit, to put upon’
‘to get up, to put up’
‘to put down’
‘to kill’
The following examples illustrate the verbal requirements for a singular and a plural
NP respectively:
(66)
inepo yoem-ta
1SG man-NNOM.SG
‘I got the man up.’
(67)
inepo yoeme-m
1SG man-PL
‘I got the men up.’
kecha-k.
get.up.NNOM.SG-PST
ja’abwa-k.
get.up.NNOM.PL-PST
When we have the coordination of two singular NP’s, functioning as an object, the
verb must be singular. This is an unexpected behavior if we consider that the union of
two singular nouns should be interpreted as plural:
(68)
Maria yoem-ta
into usi-ta
child-NNOM.SG
Maria man-NNOM.SG and
‘Maria got up the man and the child.’
kecha-k.
get.up.NNOM.SG-PST
(69)
*Mariayoem-ta
into
usi-ta
Maria man-NNOM.SG and
child-NNOM.SG
(‘Maria got up the man and the child.’)
ja’abwa-k
get.up.NNOM.PL-PST
Recall that the coordination of the object can be discontinuous, as the following
example indicates, and the singular verb still requires two singular coordinated NPs.
237
(70)
inepo yoem-ta
kecha-k
into
1SG man-NNOM.SG get.up.NNOM.SG-PST and
‘I get up the man and the child.’
usi-ta.
child-NNOM.SG
We can ask if the marked pattern above could be derived from the coordination of
two sentences each containing a singular verb and a singular object, as in the following
example:
(71)
inepo yoem-ta
kecha-k
into
1SG man-NNOM.SG get.up.NNOM.SG-PST and
‘I get up the man and get up the child.’
usi-ta
kecha-k
child-NNOM.SG get.up.SG-PST
Even so, we have to decide if the following example is a kind of discontinuous
coordination or instead sentence coordination:
(72)
inepo yoem-ta
kecha-k
1SG man-NNOM.SG get.up.SG.OBJ-PST
‘I get up the man and (to) the child too.’
into usi-ta
and child-NNOM.SG
kechia.
too
The next examples show that if the verb is plural, the objects can not be marked with
-ta ‘NNOM.SG’. It doesn’t matter what the position of the object is. In other words, this
kind of verb requires a plural noun as complement:
(73)
*inepo ja’abwa-k
yoem-ta
into
man-NNOM.SG and
1SG get.up.PL.OBJ-PST
(‘I get up the man and the child.’)
usi-ta.
child-NNOM.SG
(74)
*inepo yoem-ta
ja’abwa-k
1SG man-NNOM.SG get.up.PL.OBJ-PST
(‘I get up the man and the child too.’)
usi-ta
(ketchia)
child-NNOM.SG (too)
into
and
Another pattern shows the interaction of verbal requirements and morphological
requirements of the Yaqui NPs. The following nouns belong to N3 class (Buitimea
2003:16-17). We have to remember that these nouns are always required to be marked for
238
plural; it doesn’t matter if they are understood in a singular sense. I illustrate the
paradigm with the following examples. I will refer to these as “morphological plurals”.
(75)
puúsi-m
naka-m
tono-m
boócha-m
reépa-m
‘eye/eyes’
‘ear/ears’
‘knee/knees’
‘shoe/shoes’
‘earring/earrings’
When these kinds of nouns are the object of verbs like those seen above, the
interpretation of plural or singular is indicated by the verb, as shown in the following
examples:
(76)
inepo maeche’eta-m
1SG machete-PL
‘I put up the machetes’
ja’abwa-k
put.up.PL.OBJ-PST
(77)
inepo maeche’eta-m
1SG machete-PL
‘I put up the machete’
kecha-k
put up.SG.OBJ-PST
The same happens with the coordination of two morphological plural nouns
functioning as object. The verb indicates the interpretation as singular or plural.
(78)
inepo mache’eta-m into kuchi’i-m
1SG machete-PL and
knife-PL
‘I put up the machetes and the knifes.’
ja’abwa-k.
put.up.PL.OBJ-PST
(79)
inepo mache’eta-m into kuchi’i-m
1SG machete-PL and
knife-PL
‘I put up the machete and the knife.’
kecha-k.
put.up.SG.OBJ-PST
Under discontinuous coordination the interpretation and the requirements are the
same as above:
(80)
inepo mache’eta-m kecha-k
1SG machete-PL put.up.SG.OBJ-PST
‘I put up the machete and the knife’
into
and
kuchi’i-m
knife-PL
239
Interestingly, a conflict arises when we have the coordination of a plural and a
singular noun: Which verb has to be used in this case? The conflict is resolved by using a
plural verb, it doesn’t matter what the order of the coordinated constituents is. If the verb
is singular, we have an ungrammatical sentence:
(81)
inepo usi-ta
into ilí
1SG child-NNOM.SG
and
little
‘I put down the child and the girls.’
jamuch-im
woman-PL
(82)
inepo ilí
jamuch-im
into usi-ta
1SG little woman-PL
and
child-NNOM.SG
‘I put down the girls and the child.’
to’a-k.
put down.PL.OBJ-PST
(83)
*inepo usi-ta
into ilí
1SG child-NNOM.SG
and
little
(‘I put down the child and the girls.’)
teeka.
put.down.SG.OBJ.PST
(84)
*inepo ilí jamuch-im
into usi-ta
1SG little woman-PL
and
child-NNOM.SG
(‘I put down the girls and the child.’)
5.2.6
Interaction between pronouns and coordination
jamuch-im
woman-PL
to’a-k.
put.down.PL.OBJ-PST
teeka.
put.down.SG.OBJ.PST
Co-referential coordinated nouns agree in number with a plural pronoun. If we use a
plural object pronoun, the sentence obligatorily requires a plural verb in order to be
grammatical. Look at the following contrast. The coordinated nouns are singular and are
co-referential with the plural pronoun.
(85)
inepo usí
o’ou-ta
into usí
jamut-ta
banko-t
1SG child male-NNOM.SG
and
child female-NNOM.SG chair-LOC
am=joá-k.
3OBJ.PL=sit.down.PL-PST
‘I sit them down the boy and the girl in the chair.’
240
(86)
*inepo banko-t
am=yecha-k
juka
usi
1SG chair-LOC
3OBJ.PL=sit.down.SG.OBJ-PST
DET.NNOM.SG child
o’ou-ta
into juka
usí
jamut-ta.
male-NNOM.SG
and
DET.NNOM.sG child female-NNOM.SG
(‘I sit them down in the chair, the boy and the girl.’)
The pronoun allows us to extrapose the coordinated singular noun:
(87)
inepo banko-t
am=joá-k
juka
usí
1SG chair-LOC
3OBJ.PL=sit.down:PL.OBJ-PST
DET.NNOM.SG child
o’ou-ta
into juka
usí
jamut-ta.
male-NNOM.SG
and
DET.NNOM.SG child female-NNOM.SG
‘I sit them down in the chair, the boy and the girl.’
A singular accusative pronoun can be attached to the singular verb. In that case, the
co-referential noun must be singular too. However, it is not possible to have two
coordinated nouns if the pronoun is singular. This is illustrated with the following
contrast:
(88)
inepo banko-t
a=yecha-k
3OBJ.SG=sit down.SG. OBJ-PST
1SG chair-LOC
usí
o’ou-ta.
child male-NNOM.SG
‘I sit him down in the chair, the boy.’
juka
DET.NNOM.SG
(89)
*inepo banko-t
a=yecha-k
juka
3OBJ.SG=sit.down.SG.OBJ-PST
DET.NNOM.SG
1SG chair-LOC
uusí o’ou-ta
into juka
uusí jamut-ta.
child male-NNOM.SG and DET.NNOM.SG child female-NNOM.SG
(‘I sit him/her down in the chair, the boy and the girl.’)
5.2.7 Summary
The following representations illustrate the facts seen in this section:
There are some suppletive Yaqui verbs which agree with the direct object of the
sentence.
241
(90)
Object
NSG
NPL
Verb
VSG OBJ
VPL OBJ
Two (or more) coordinated singular nouns in object position take a singular verb.
Two (or more) coordinated plural nouns take a plural verb:
(91)
Object
NSG + NSG
NPL + NPL
Verb
VSG OBJ
VPL OBJ
In case of conflict arising from the coordination of a singular noun and a plural noun
the verb must be plural:
(92)
Object
NSG + NPL
NPL + NSG
Verb
VPL OBJ
For morphological plural nouns (N3 class), the verb could be singular or plural, the
verb indicates how to interpret the coordinated object:
(93)
Object
NPL + NPL
NPL + NPL
Verb
VSG OBJ
VPL OBJ
gives a singular reading of the coordinate nouns.
gives a plural reading of the coordinated nouns.
The presence of a plural clitic pronoun in co-reference with two coordinate singular
nouns obligatorily requires a plural verb:
(94)
Object
NSG + NSG
Verb
P3PL OBJ=VPL OBJ
(where P represents a clitic pronoun)
However, the presence of a singular clitic pronoun only can be co-referential with a
single singular noun. It can not be co-referential with two coordinated singular nouns:
(95)
Object
NSG
*NSG + NSG
Verb
P3SG OBJ=VSG OBJ
P3SG OBJ=VSG OBJ
(where P represents a clitic pronoun)
(where P represents a clitic pronoun)
242
In short, we need to license candidates in Yaqui that conform to the following
patterns.
(96)
[N + N]OBJ
+
VTR
Interpretation
SG+SG
PL+PL
SG+PL
PL+SG
+
SG
PL
+
PL
PL
And rule out candidates with the following structure.
(97)
[N + N] OBJ
+
VTR
Interpretation
SG+SG
PL+PL
SG+PL
PL+SG
+
PL
*
+
SG
*
As we saw in the previous section, there are some differences between intransitive
and transitive verbs. In the next section I present an analysis of transitive verbs. We will
see that the feature system proposed by Halloway King and Dalrymple (2004) makes
wrong predictions about the Yaqui coordination patterns.
5.3
Analysis of transitive verbs
I propose that the verbs which agree with the object have the following features. The
singular verb has active the feature CONCORD singular. The
any role. The predictions are checked in what follows:
(98)
me’a ‘to kill.SG.OBJ’
V
CONCORD: SG
INDEX: Ø
INDEX
feature does not play
243
It is predicted that the verb combines with single singular nouns in object position and
that it can never combine with a plural noun in object position. The verb requires
matching in CONCORD singular. The candidate in (99) satisfies it but the one in (100)
violates it. Therefore, one is grammatical and the other ungrammatical:
(99)
Yoeme[masó-ta CONC:SG]IND:SG
man deer-NNOM
‘The man killed the deer (sg).’
me’ak CONC:SG/Ø
kill:SG.OBJ.PST
(100) *yoeme
[masó-m CONC:PL] IND:PL me’a-k CONC:SG/ IND:Ø
man
deer-PL
kill.SG.OBJ.PST
(‘The man killed the deer (pl).’)
Because of the feature CONCORD singular must be distributed (i.e. matched or applied
to each nominal covering the role of object), it is predicted that conjoined singular nouns
will produce grammatical sentences. The INDEX feature of this type of verb does not play
any role in the agreement system. For that reason the INDEX plural in the whole nominal
phrase does not affect the grammaticality of the sentence.
into masó-ta CONC:SG]IND:PL me’ak CONC:SG/ IND:Ø
(101) Joan [parós-ta CONC:SG
John hare-NNOM
and
deer-NNOM
kill.SG.OBJ.PST
‘John killed the hare and the deer (sg).’
The verb will never co-occur with conjoined singular and plural nouns. The reason is
that the CONCORD plural feature in one of the conjuncts does not match (i.e. it is not
distributed) with the CONCORD singular demanded by the verb. The ungrammaticality of
(102) is expected:
into masó-m CONC:PL] IND:PL me’a-k CONC:SG/ IND: Ø
(102) *empo [paró’os-ta CONC:SG
2sg
hare- NNOM
and
deer- NNOM
kill.PL.OBJ-PST
(‘You killed the hare and the deer (pl).’)
244
It will never combine with conjoined plural nouns, either. The CONCORD singular
feature in the verb does not match the CONCORD plural feature of each conjoined noun.
Then, the sentence (103) is ruled out as ungrammatical:
into masó-m CONC:PL] IND:PL sua-k CONC:SG/ IND:Ø
(103) *inepo [paró’os-im CONC:PL
1SG
hare-PL
and
deer-PL
kill.SG.OBJ-PST
(‘I killed the hares and the deer (pl).’)
5.3.1
The problems
There are two problems that I want to analyze here: the false predictions of the
system proposed by Holloway King and Dalrymple (2004) and the challenge posed by
the patterns of coordinated nouns.
5.3.1.1 False predictions
Given the four possibilities established by Halloway King and Dalrymple (2004), it is
not possible to generate the behavior of agreement in plural verbs which agree with the
object. The four possibilities are repeated here. I show how they fail in each case:
(104)
a)
V
[INDEX]
b)
V
[CONCORD]
c)
V
[Ø]
d)
V
CONCORD
INDEX
If we assign the plural value to the INDEX feature of plural verbs which agree with the
object, we arrive at the following representation. On it the relevant feature is the INDEX
plural.
(105)
sua
‘kill.PL.OBJ’
V
CONCORD: Ø
INDEX: PL
245
The representation predicts as grammatical two conjoined singular nouns, but the next
example is not a grammatical Yaqui sentence. Observe that the INDEX plural features
match and it would not be the reason for the ungrammaticality.
into masó-ta CONC:SG] IND:PL sua-k CONC:Ø/ IND:PL
(106) *Joan [parós-ta CONC:SG
John hare- NNOM.SG
and
deer- NNOM.SG
kill: PL.OBJ-PST
(‘John killed the hare and the deer.’)
If we analyze the CONCORD feature as the relevant one (see the representation (107)),
the problem is that it wrongly rules out a grammatical sentence like (108). The sentence
is predicted to be ungrammatical because the verbal CONCORD plural does not match the
singular CONCORD in the nominal conjunct, i.e., the CONCORD plural feature is not
distributed.
(107)
sua
V
‘kill:PL.OBJ’
CONCORD: PL
INDEX: Ø
into masó-ta CONC:SG] IND:PL sua-k CONC:PL/ IND:Ø
(108) aapo [paró’os-im CONC:PL
3SG hare-PL
and
deer-NNOM.SG
kill.PL.OBJ-PST
‘(S)he killed the hares and the deer (sg).’
If we consider that the verb does not bear any number feature, as in (109), the
prediction is that it will accept any combination of number values in a coordinate
structure. This conclusion is unacceptable because we lose the plural characteristic of this
type of verb.
(109)
sua
V
[Ø]
‘kill.PL.OBJ’
Finally, if we check the last possibility seen in (110), we still have the problem of
predicting as ungrammatical two conjoined singular and plural nouns. Observe that the
246
CONCORD feature is not distributed on each member of the coordinate structure.
Therefore, sentence (111) is wrongly predicted to be ungrammatical.
(110)
sua
‘ kill:PL.OBJ’
V
CONCORD:PL
INDEX:PL
(111) aapo
IND:PL
[paró’os-ta CONC:SG
into
masó-m CONC:PL] IND:PL
3SG hare-NNOM.SG
and
deer-PL
‘(S)he killed the hare and the deer (pl).’
sua-k
CONC:PL/
kill.PL.OBJ-PST
However, this representation has the advantage that it predicts all the others patterns
attested in the language. It predicts that the verb will never combine with a singular noun
but just with plural nouns. The sentence in (112) is ungrammatical because the singular
nominal INDEX feature does not match the plural verbal INDEX feature. The sentence in
(113) is grammatical because both CONCORD and INDEX features match:
(112) *inepo
[usi-ta CONC:SG] IND:SG
child-SG
1SG
(‘I laid down the child’)
(113) inepo [usi-m CONC:PL] IND:PL
1SG
child-PL
‘I laid down the children.’
to’a CONC:PL/ IND:PL
lay.down.PL.OBJ.PST
to’a CONC:PL/ IND:PL
lay.down.PL.OBJ.PST
It predicts too that the verb will never combine with conjoined singular nouns. That
prediction holds in the language. The ungrammaticality is due to the fact that the verbal
CONCORD plural feature is not distributed, as indicated in the following sentence:
into masó-ta CONC:SG] IND:PL sua-k CONC:PL/ IND:PL
(114) *Joan [parós-ta CONC:SG
John hare- NNOM.SG
and
deer- NNOM.SG
kill. PL.OBJ-PST
(‘John killed the hare and the deer.’)
247
The grammaticality of conjoined plural nouns combined with a verb containing
CONCORD plural and INDEX plural is expected. In such cases, the CONCORD features are
distributed and the INDEX features match. Therefore, there is not a conflictive situation:
(115) Amureo
[paró’os-im CONC:PL
into masó-m CONC:PL] IND:PL sua-kCONC:PL/ IND:PL
Hunter
hare-PL
and
deer-PL
kill.PL.OBJ-PST
‘The hunter killed the hares and the deer (pl).’
5.3.2 Solving the problem in OT terms
If we analyze the conditions under which conjoined singular and plural nouns fail to
be generated by the system of CONCORD and INDEX features, we arrive at the following
situation. Let’s recall first the representation of singular verbs:
(116)
me’a ‘kill.SG.OBJ’
V
CONCORD: SG
INDEX: Ø
As we said before, the kind of singular verbs which agree with the object are unable
to generate as grammatical the coordinate structure [singular + plural (it does matter the
order)] for the following reason: the concord singular feature of the verb is not distributed
over each conjunct (or, in a checking conception, it does not match with the plural feature
of the conjoined nominal).
into masó-m CONC:PL]INDEX:PL me’ak CONC:SG/ INDEX: Ø
(117) *empo [paró’os-ta CONC:SG
2SG hare-NNOM.SG
and deer-PL
kill.SG.OBJ-PST
(‘You killed the hare and the deer (pl).’)
Now, let’s revise the condition under which the plurals verbs do not license the
coordinate structure [singular + plural]. The verbal representation is given in (118):
248
(118)
sua
V
‘kill.PL.OBJ’
CONCORD: PL
INDEX: PL
The sentence containing [singular + plural] is predicted to be ungrammatical because
the CONCORD plural does not distribute to the singular conjoined nominal.
into masó-m CONC:PL] IND:PL sua-k CONC:PL/ IND:PL
(119) *Aapo [paró’os-ta CONC:SG
3SG hare-NNOM.SG
and
deer-PL
kill.PL.OBJ-PST
‘(S)he killed the hare and the deer (pl).’
As we can see, in both cases (117) and (119), the CONCORD feature is not distributed.
The candidates are tied in this aspect. However the sentence in (119) is grammatical in
the language. The question is then: Why does the language select the plural verb for
expressing conjoined singular and plural nouns? Let us depart from the idea that this
meaning has to be expressed, if possible, by using the resources of the language.
Therefore, two viable candidates are the one with the singular verb and the other with the
plural verb. The candidate containing the singular verb lost the battle against the plural
verb. Why is that the case? I suggest that the singular verb has an additional failure than
the one expressed before (lack of distribution of CONCORD singular). If we assume that
the INDEX feature, instead of being empty, is unspecified (as indicated in (120), where the
line indicates underspecification), and that unspecified features must be filled with the
features of the nominal for which the verb is subcategorizing for, then the candidate with
the singular verb must have the representation in (121):
(120)
me’a ‘kill.SG.OBJ’
V
CONCORD : SG
INDEX : ___
249
(121) *empo [paró’os-ta CONC:SG
into masó-m CONC:PL] IND:PL me’a-k CONC:SG/ IND: PL
2SG hare-NNOM.SG
and
deer-NNOM.SG
kill.SG.OBJ-PST
(‘You killed the hare and the deer (pl).’)
As we can see, the candidate with the verb me’ak ‘to kill.SG.OBJ’ has CONCORD
singular and INDEX plural. This is an undesirable specification of features as pointed out
by Halloway King & Dalrymple (2004). They rule out determiners which require singular
CONCORD and plural INDEX; such determiners could be used in cases where coordinate
structures refer to more than one individual in which each conjunct is singular. They rule
these determiners out “by requiring determiners to impose uniform numbers
specifications: there are not determiners that impose a different value for CONCORD and
INDEX.” (84). If that is true, then we can think that the verb me’ak ‘to kill.SG.OBJ’ violates
such a constraint. Let’s call it NUMBER UNIFORMITY:
(122) NUMBER UNIFORMITY: Verbs must bear number uniformity.
This constraint will force that both CONCORD and INDEX features have the same value
over the verb. Therefore, we can say that candidate (121) violates two constraints: the
constraint that requires distribution of the singular feature and the constraint that requires
NUMBER UNIFORMITY.
On the other hand, the candidate with the plural verb just violates the CONCORD
feature, but it does not violate NUMBER UNIFORMITY: the INDEX plural on the verb
matches the plural INDEX in the noun phrase. Therefore, to express conjoined singular
and plural nouns is less costly using the plural verb, than using the singular verb. The
next tableaux show that it does not matter if the input is the verb me’a ‘to kill.SG.OBJ’ or
sua ‘to kill.PL.OBJ’, the winner in both cases is the verb with plural meaning.
250
Two additional constraints are required one that forces the checking of the plural
INDEX in the candidates, and one that requires the distribution of the CONCORD feature of
the verb. They are defined as follows.
(123) CHECK INDEX. Index features must match in each candidate.
(124) CONCORD DISTR. Concord features of the verb must be distributed to the nominal
arguments.
I suggest the ranking in (125). The candidates show the CONCORD features as
subscripts and the INDEX features as superscripts.
a. )[paró’os-taSG
into
masó-mPL]PL
sua-kPL/PL
b.
[paró’os-taSG
into
masó-mPL]PL
me’akSG/PL
c.
[paró’os-imPL
into
masó-mPL]PL
sua-kPL/Ø
!*
***
d.
[paró’os-taSG
into
masó-mPL]PL
me’akSG/ SG
!*
*
e.
[paró’os-imPL
into
masó-mPL]PL
sua-kPL/PL
f.
[paró’os-taSG
into
masó-mPL]PL
me’akSG/ Ø
We can see that NUMBER
UNIFORMITY
!*
CONCORD
DISTR
FAITH-I-O
CHECK
INDEX
input:
{paros-taCONC:SG, into, maso-mCONC:PL, me’akCONC:SG/ IND:Ø}
NUMBER
UNIFORMITY
(125) Tableau with the ranking CHECK INDEX >> NUMBER UNIFORMITY >> FAITH-IO>> DIST. CONCORD:
***
*
*
*
*
!****
!*
*
is violated by candidate (125b) because the
features on the verb are not the same, i.e. they are not uniform. Therefore, it is rule out of
the competition. The CHECK
INDEX
constraint rules out candidates (125c), (125d), and
(125f), because the INDEX feature in the verb does not match the INDEX feature of the
whole conjoined NP. The FAITH-I-O constraint requires that the features in the input be
251
present in the output. However, we can see that they change the nominal and verbal
number feature specifications. Further, there are changes in the phonological information
of the verbal root. So, for example, candidate (125e) has four violations of FAITH-I-O.
The tableau that shows the results with a plural verb is the following.
a. )[paró’os-taSG
into
masó-mPL]PL
sua-kPL/PL
b.
[paró’os-taSG
into
masó-mPL]PL
me’akSG/PL
c.
[paró’os-imPL
into
masó-mPL]PL
sua-kPL/Ø
!*
**
d.
[paró’os-taSG
into
masó-mPL]PL
me’akSG/ SG
!*
***
e.
[paró’os-imPL
into
masó-mPL]PL
sua-kPL/PL
f.
[paró’os-taSG
into
masó-mPL]PL
me’akSG/ Ø
CONCORD
DISTR.
FAITH-I-O
NUMBER
UNIFORMITY
CHECK INDEX
(126) Input: {paros-taSG, into, maso-mPL, suakPL/PL}
*
!*
**
*
*
!*
!*
***
*
5.3.2.1 Conjoined nominal class 3 and the verbs
As it was mentioned above, for conjoined nouns of the class three, it was suggested in
a separate paper (Langendoen & Martínez Fabián 2004) that those nouns are subject to a
highly ranked constraint demanding affixation over the noun. It is called HaveAF. The
interaction of this constraint with some other constraints result in the fact that those nouns
are always marked with the --(i)m ‘PL’ affix. The next example shows part of the
analysis. The input contains the nominal root, it has the feature specification of
[Accusative, Singular] (in such case, if the noun were of class-1, it would be expect to
have the --ta marking). We can see that the winner candidate is the one marked with -(i)m ‘PL’.
252
(127) HAVEAF: lexical forms must have affix.
(128) FAITH-FS: Features in the input must be preserved in the output
(129) *CASE: Avoid case marking.
(130) *NUMBER: Avoid number marking.
(131) Selection of supe-m for expressing supe [Accusative] & [Singular]
supe [Accusative] & [Singular]
supe
HAVEAF
FAITH-FS
*!
*
*NUMBER
*!
supe-ta [Accusative]
)
*CASE
supe-m [Plural]
***
*
Remember that the verb gives the interpretation of the noun: singular or plural, as
repeated in next examples. As we can see in the representation (132) and (133), the
CONCORD feature is responsible of the singular/plural interpretation:
(132) inepo mache’eta-m CONC:SG IND: Ø
1SG machete-PL
‘I lay down the machete.’
teeka CONC:SG IND: Ø
lay.SG.OBJ
(133) inepo mache’eta-m CONC:PL IND: PL
1SG machete-PL
‘I laydown the machetes.’
to’a CONC:PL IND:PL
lay.PL.OBJ
The coordinate nouns are interpreted in the same way. Each noun is interpreted as
singular if the verb is singular and each noun is interpreted as plural if the verb is plural.
The whole conjoined construction has a plural INDEX feature.
into mache’eta-m CONC:SG] IND:PL teekaCONC:SG IND:PL
(134) inepo [kuchi’i-m CONC:SG
and
machete-PL
lay.SG.OBJ
1SG knife-PL
‘I lay down the knife and the machete.’
253
(135) inepo [kuchi’i-m CONC:PL
into mache’eta-m CONC:PL]IND:PL
1SG knife- PL
and
machete- PL
‘I lay down the knife and the machetes.’
to’a CONC:PL IND:PL
lay.PL.OBJ
With verbs of class one, which do not mark number, nouns of class three, in spite of
being marked with --(i)m ‘PL’, are unspecified for number too. They can be interpreted
either as both singular or both plural, as shown by the translations:
(136) Peo [kuchi’i-m CONC:Ø into mache’eta-m CONC:Ø]IND:PL
Peter knife-PL
and machete- PL
‘Peter bought the/a knife/knives and the/a machete(s).’
jinu-k CONC:Ø/ IND:Ø
buy-PST
In the analysis given here I assume that nouns of class three are unspecified for
number in the input. The constraint HAVE-AF demands that nouns appear affixed. Among
the affixes of the language, the set of constraints selects the candidate with the affix -(i)m ‘PL’, as seen in the previous table. My analysis shows that it is appropriate to
consider that the nouns of this class are underspecified for number and that the suffix -(i)m has the following representation. On it the CONCORD feature is underspecified but
the INDEX feature is plural:
(137)
kuchi’i
[CONCORD: __]
[INDEX: __ ]
-(i)m
[CONCORD: __]
[INDEX: PL ]
Because nouns of class three always emerge in overt syntax with the suffix --(i)m, I
consider them to have the following representation. In it, the affix gives the Plural INDEX
feature to the class three nouns. It is represented as follows in the inputs:
(138)
kuchi’i-m
[CONCORD: __]
[INDEX: PL ]
254
This representation implies that the CONCORD feature can be left unspecified (in cases
where it does combine with verbs which do not carry number information and when there
are no numerals indicating number) or specified with the value SG or PL. That makes
several predictions (most of them are fulfilled). Let’s look at them.
With the verbs of class three the coordinated nouns are left undefined in their
CONCORD feature that can be interpreted as singular or plural (by pragmatic principles),
but it have a plural
INDEX
(that means that the Yaqui speakers look at those nouns as if
they were composed of several parts (as suggested by Buitimea Valenzuela p.c.). The line
after the CONCORD and INDEX features means underspecification: There is no CONCORD
value that the verb can distribute:
(139) Joan [macheta-m CONC:_] IND:PL
John [machete-PL ]
‘John bought a machete(s).’
jinu-kCONC:_/IND:_
buy-PST
For the patterns of Yaqui coordination with these nouns I suggest the following
analysis. I just put into the input the coordinated nouns and the verb with which they
agree. The following table shows that the candidate (a) is the winner because it does not
violate the higher ranked constraint, whereas candidates (b) and (c) do.
255
)
a kuchi’i-m
into
[Case: NNOM]
[Concord: _]
[Index: PL]
kuchi’i-ta
into
[Case: NNOM]
[Concord:SG]
[Index: PL ]
c kuchi’i
into
[Case: NOM]
[Concord: SG]
[Index: PL]
dkuchi’i-m
into
[Case: NNOM]
[Concord: PL]
[Index: PL]
macheta-m
jinuk
[Case: NNOM] [Case: NNOM]
[Concord: _ ] [concord:_]
[Index: PL ] [Index: _ ]
macheta-ta
jinuk
[Case: NNOM] [Case: NNOM]
[Concord: SG] [concord: SG]
[Index: PL ] [Index: PL ]
macheta
jinuk
[Case: NOM] [Case: NNOM]
[Concord:SG] [concord: SG]
[Index: PL ] [Index: _ ]
macheta-m jinuk
[Case: NNOM] [Case: NNOM]
[Concord:PL] [concord: PL]
[Index: PL ] [Index: PL ]
*NUMBER
DISTR. CONCORD
FAITHH-SF
jinuk
[Case: NNOM]
[concord: _ ]
[Index: _ ]}
CHECK INDEX
into, machetam,
[Case: NNOM]
[Concord: __]
[Index: PL ]
*CASE
Input:
{kuchi’i-m,
[Case: NNOM]
[Concord:__ ]
[Index: PL]
HAVEAF
(140) Tableau that shows the noun class 3 interacting with the verb class 1
**
!*
!*
**
**
**
**
!*
**
*
With the verb siika ‘to leave.SG’, the predictions are fulfilled. We can see that in this
case the winning candidate would have the distribution of the singular CONCORD feature
and it is interpreted as a singular entity which is composed of several parts (koari-m
‘skirt’ or which is part of a plurality (ex. chobe-m ‘buttock(s)’). The features of (141) are
shown in the winning candidate in the table (143):
(141)
[bejo’ori-m]
lizard-PL
‘The lizard left.’
siika.
left
The impossibility of the following sentence is attributed to the fact that the singular
INDEX of the verb is incompatible with the INDEX of the whole nominal phrase.
256
(142) *[bejo’ori-m CONC:SG
[Case: Nom]
[Concord: SG]
[Index: PL]
into porowi-m CONC:SG] IND:PL
[Case: Nom]
[Concord: SG]
[Index: PL]
siika CONC:SG/IND:SG
[Case: Nom]
[Concord: SG]
[Index: SG]
The winning candidate is shown in the following tableau. It has the features indicated.
siika
[case: NOM]
[concord:SG]
[index: SG ]
siika
[case: NOM]
[concord:SG]
[index: SG ]
siika
[case: NOM]
[concord:SG]
[index: SG ]
siika
[case: NOM]
[concord:SG]
[index: PL ]
b.
c.
d. )
!*
!*
*NUMBER
bejo’ori-m
[case: NOM]
[concord:SG]
[index: PL ]
bejo’ori-ta
[case: NNOM]
[concord:SG]
[index: PL ]
bejo’ori
[case: NOM]
[concord:SG ]
[index: PL ]
bejo’ori-m
[case: NOM]
[concord:SG]
[index: PL ]
DISTR.
CONCORD
a.
FAITH-SF
siika
[case: NOM]
[concord:SG]
[index: SG ]}
CHECK INDEX
{bejo’ori-m
[case: NOM]
[concord:__ ]
[index: PL ]
*CASE
Input:
HAVE AF
(143) Tableau with noun class 3 and verb class 2.
!*
*
*
*
*
*
*
*
**
*
The analysis predicts the behavior of the sajak ‘to leave.PL’ type verbs interacting
with conjoined nouns of the N3 class:
The coordinate structure contains unspecified nouns for CONCORD number.
Therefore, it can be interpreted as CONCORD singular or plural, but the INDEX feature is
plural. For that reason, the conjoined nominals can not be interpreted as referring to a
257
single unit. The INDEX features of the verb and the nouns match. The features of example
(144) can be seen in the winning candidate (145a):
(144) bejo’ori-m
into porowi-m
Lizard-PL
and
porowi45-PL
‘The lizard and the porowi left’
saja-k
leave.PL-PST
a.)
b.
c.
d.
bejo’ori-m into porowi-m
saja-k
[case: NOM]
[case: NOM] [case: NOM]
[concord:__]
[concord:__] [concord:__]
[index: PL ] [index: PL ] [index: PL ]
bejo’ori-ta
into porowi-ta
saja-k
[case: NOM]
[case: NOM] [case: NOM]
[concord:SG]
[concord:SG] [concord:SG]
[index: PL ] [index: pl ]
[index: PL ]
bejo’ori
into porowi
saja-k
[case: NOM]
[case: NOM] [case: NOM]
[concord:__]
[concord:__] [concord:__]
[index: PL ]
[index: PL ] [index: PL ]
bejo’ori-m
into porowi-m
saja-k
[case: NOM] [case: NOM]
[case: NOM]
[concord:PL]
[concord:pl] [concord:PL]
[index: PL ]
[index: pl ] [index: PL ]
*
!*
!*
*
**
*
*
*
**
*
!**
*
*
*
The ranking predicts the behavior of the verbs which agree with the object as well
(e.g. the verbs of the me’a ‘to kill.SG.OBJ’ class). The nouns are interpreted just as
singular entities, which as a whole have a plural INDEX. The reason for the singular
interpretation is that the singular CONCORD feature of the verb distributes to the
45
*NUMBER
DISTR.
CONCORD
FAITH-SF
saja-k
[case:NOM]
[concord:__]
[index:PL ], into}
CHECK
INDEX
porowi-m
[ase: NOM]
[oncord:__]
[ndex: PL ],
*CASE
Input
{bejo’ori-m
[case: NOM]
[concord:__]
[index: PL ],
HAVE AF
(145) Table with noun class 3 and verb class 3.
The porowi is a kind of lizard (Dinosaurus dorsalis). It seems that these kinds of small animals
belong to the same plural category.
258
coordinate nominals, whereas the INDEX features of the nouns and the verb match giving
the plural reading to the coordinate nouns.
into porowi-m CONC:SG] IND:PL me’a-k CONC:SG/IND:PL
(146) [bejo’ori-m CONC:SG
lizard-PL
and
porowi-PL
kill.SG.OBJ-PST
‘((S)he) killed the lizard and the porowi.’
a.)
b.
c.
d.
bejo’ori-m into
[case: NNOM]
[concord:SG]
[index: PL]
bejo’ori-ta
into
[case: NNOM]
[concord:SG]
[index: PL]
bejo’ori
into
[case: NOM]
[concord:__]
[index: PL ]
bejo’ori-m
into
[case: NOM]
[concord:PL]
[index: PL ]
porowi-m
[case: NNOM]
[concord:SG]
[index: PL]
porowi-ta
[case: NNOM]
[concord:SG]
[index: PL]
porowi
[case: NOM]
[concord:__]
[index: PL ]
porowi-m
[case: NOM]
[concord:PL]
[index: PL ]
me’ak
[case: NNOM]
[concord:SG]
[index: PL]
me’ak
[case: NNOM]
[concord:SG]
[index: PL]
me’ak
!*
[case: NOM]
[concord:SG]
[index: PL ]
me’ak
[case: NOM]
[concord:PL]
[index: PL ]
!*
*NUMBER
DISTR.
CONCORD
FAITHH-SF
CHECK INDEX
*CASE
Input
{bejo’ori-m
porowi-m
me’ak
[case: NNOM] [case:NNOM]
[case: NNOM]
[concord:__]
[concord:__] [concord:SG]
[index: PL ] [index: PL ] [index:__], into}
HAVEAF
(147) Table with a verb class 2 which agree with the object, and nouns class 3.
*
*
*
*
*
*
*
*
*
*
*
*
!*
*
*
*
*
*
*
*
For the verbs of the sua ‘to kill.PL.OBJ’ class, the predictions hold too. Because the
verb has the specification Concord plural and Index plural, the distribution of concord to
the noun gives the correct result. The coordinate structure can only be interpreted as
conjoining two pluralities.
259
(148) [bejo’ori-m CONC:PL
into porowi-m CONC:PL]
[Case: NNOM]
[Case: NNOM]
[Concord: PL]
[Concord: PL]
[Index: PL]
[Index: PL]
‘He killed the lizards and the porowis.’
sua-k CONC:PL
[Case: NNOM]
[Concord:PL]
[Index: PL]
In order to explain it in an OT framework, it is assumed, as before, that the nouns
have the specification given in the input. The competing candidates are shown in next
table. In order to get the correct output, we need to use the previous constraint of
Uniformity. This constraint is ranked before the FAITH-I-O constraint:
suak,
[case:NNOM]
[concord: Pl]
[index: PL],
a.
)
bejo’ori-m into porowi-m
[case:NNOM]
[case:NNOM]
[concord:PL]
[concord:PL]
[index: PL]
[index: PL]
bejo’ori-ta
into porowi-ta
[case:NNOM]
[case:NNOM]
[concord:SG]
[concord:SG]
[index: PL]
[index: PL]
bejo’ori
into porowi
[case: NOM]
[case: NOM]
[concord:__]
[concord:__]
[index: PL]
[index: PL]
bejo’ori-m
into porowi-m
[case:NNOM]
[case:NNOM]
[concord:__]
[concord:__]
[index: PL]
[index: PL]
suak
[case:nnom]
[concord:PL]
[index: PL]
suak
[case:NNOM]
[concord:SG]
[index: PL]
suak
[case: NOM]
[concord:__]
[index: pl]
suak
[case:NNOM]
[concord:PL]
[index: PL]
b.
c.
d.
!
*
!
*
FAITHH-FS
DISTR. CONCORD
*NUMBER
porowi-m,
[case:NNOM]
[concord:__]
[index: PL ]
Input
{bejo’ori-m,
[case:NNOM]
[concord:__]
[index: PL ]
into}
HAVEAF
*CASE
CHECK INDEX
UNIFORMITY
(149) Table with a verb class 3 which agrees with the object and noun class 3.
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
!
*
*
*
* *
* *
260
5.4
NP conjunction and separateness of the events
Givón (2001) mentions that NP conjunction is not merely a syntactic device for
rendering two propositions about two separate events into the more economical surface
structure, rather it is a device for coding a single event. (Givón 2001:16). The following
observations are made: a) Separate events will tend to be coded as separate clauses, b)
Clauses with conjoined subject or object NPs tend to code single multi-participant events;
c) The order of an event could interact with the order of the conjoined items. In addition
to these observations, we can add the following: d) a single participant might develop a
series of multi-events.
For those reasons, Givón considers that it is necessary to look at some pragmatic
principles responsible for the order of the conjuncts. The next section explores such
pragmatics principles.
5.4.1
Observations about the Relative order of conjoined NPs
It has been noted that “human language, unlike propositional logic, does not seem to
be quite as neutral to serial order” (Givón 2001: 17). The order of participants in an event
is reflects their relative importance or relevance. In other words, there are pragmatic
principles involved in that ordering. Cooper and Ross (1975), cited in Givón (2001:17),
note the following hierarchies in frozen expressions with conjoined NPs.
(150)
a.
b.
c.
d.
e.
f.
near > far
adult > young
male > female
singular > plural
animate > inanimate
agent > patient
(now and then, ?then and now)
(father and son, ? son and father)
(man and wife, ? wife and man)
(one and all, ? all and one)
(life and death, ? death and life)
(cat and mouse, ? mouse and cat)
261
g.
h.
large > small
positive > negative
(large and small, ? small and large)
(more or less, ?less or more)
The exploration of this type of contrast in Yaqui shows that ordering restrictions in
conjunctions is reduced, as in English and apparently in Spanish too, to some frozen
idiomatic expressions. It does not seem to be a phenomenon that pervades living
grammar. Some of the frozen expressions found in Yaqui are given in the pairs below.
However, as we can see, the contrast is not totally unacceptable.
Near > far (now and then, ?then and now)
(151) a.
Joan imi’i into aman ket
John here and
there too
‘John was here and there too.’
b.
(152) a.
b.
?Joan aman into imi’i ket
John there and
here too
(‘John was there and here.’)
weama-n.
walk-PST.CONT
weama-n.
to be-PST.CONT
jiba bena iani junak bena-sia.
alreadyseem today then seem-sia
‘It is the same now and then.’
?jiba bena junak into ian
bena-sia.
alreadyseem then and
today seem-sia
(‘It is the same then and now.’)
Agent > patient (cat and mouse, * mouse and cat)
(153) a.
Em
chu’u into em
miisi nau=nassua.
Your dog and
your cat
together=fight.PRS
‘Your dog and your cat are fighting.’
b.
?Em miisi into em
chu’u nau=nassua
Your cat
and
your dog together=fight.PRS
(‘Your cat and your dog are fighting ‘)
But most of these pragmatic principles do not apply in the language. The next
examples illustrate this fact:
262
Adult > young
(154) a.
Ju
(father and son, ?son and father)
achai into a=
usi-wa
nau
DET
father and
3POSS=son-POSS
together
‘The father and his son are singing together.’
b.
Ju
uusi into a=
achai-wa
nau
boy and
3POSS= father-POSS together
‘The boy and his father are singing together.’
DET
bwiika.
sing.PRS
bwiika.
sing.PRS
Large > small
(large and small, ?small and large)
(155) a.
bwere-m
into ilitchi sotoi-m
tu’ule.
Big-PL
and
small pot-PL
like.PRS
‘(S)he likes big and small pots.’
b.
ilitchi-m
into bwere
small-PL
and
big
‘(S)he likes small and big pots.’
sotoi-m
pot-PL
tu’ule.
like.PRS
Male > female
(man and wife, * wife and man)
(156) a.
uka
o’ou-ta
into wepul jamut-ta
DET.NNOM.SG man-NNOM.SG and
one
woman-NNOM.SG
bicha-k-an.
see-PST-CONT
‘I was looking a man and one woman.’
b.
jamut-ta
into o’ou.-ta
nee
and
man-NNOM.SG 1SG
woman-NNOM.SG
‘I saw a woman and a man over there.’
bicha-k
see-PST
nee
1SG
amani.
there
Additional evidence that the distinction male/female is not playing a largte role in the
language is given by the following pair of sentences. On them the verbalized nouns used
for husband and wife are used in both cases for expressing marriage:
(157) a.
b.
bempo emo ku-kuna-k.
They REFL PL-husband-POSS
‘They are married.’
bempo emo ju-jube-k.
They REFL PL-wife-POSS
‘They are married.’
263
5.4.2
Summary
The previous section explores the use of pragmatic constraints that could alter the
order of the conjuncts. However, the Yaqui language does not present the kind of
contrasts found in some other languages. An OT treatment of these facts suggests that
those constraints in Yaqui are unranked with respect to each other, or that they occupy a
lower position in the hierarchy of constraints. I suggest an introductory analysis in the
following section.
5.4.3
OT analysis of pragmatic constraints
I suggest that many of the pragmatic constraints codify the importance or relevance of
the participants mentioned in the conjuncts. They can be seen as statements about the
expected sociolinguistic behavior of the speaker. So, we can define the following
constraints related to what has to be mention first:
(158) MALE FIRST: Mention first male than female.
(159) ADULT FIRST: Mention first adult persons than young persons
(160) SINGULAR FIRST: Mention first singular than plural.
(161) POSITIVE FIRST: Mention first positive things than negatives.
These constraints interact with Gricean constraint demanding order. We can just
assume that lexical items in the input carry indexical information which is codified in
each candidate and that is indicative of order of presentation. The order can be taken as
cardinal order. The constraint is defined as follows:
(162) BE ORDERED: Present the information in cardinal order.
264
So an input would contain the features and indices that will produce the order of the
predicates.
As a way of exemplification, let’s take the first three constraints in (158-160)
together with the Yaqui sentence in (163). The order jamut into uusim ‘the woman and
the boys’ is seen in the table (164). It shows how the order of the conjoined nouns is
generated in OT terms.
(163) [Jamut
into uusi-m]
teopo-po
bwuik-bae.
Woman
and
boy-PL
church-LOC sing-INTT
‘The woman and the boys will sing in the church.’
The conjunct jamut ‘woman’ in (163) has the features [female, adult, singular],
whereas the conjunct uusim ‘boys’ has the features [male, young, plural]. For that reason,
given an input as that in the table (164), the winner is the candidate (164a) which does
not violate the higher ranked constraint BE ORDERED, whereas the candidate (164b)
violates it. The nouns in the input carry a subindex which can be considered the number
that indicates the position that a speaker wants it to occupy in the conjoined structure.
(164) Tableau with the order jamut into uusim ‘the woman and the boys’.
The ranking is BE ORDERED >> MALE FIRST, ADULT FIRST, SINGULAR FIRST.
Input:
a. )
b.
{jamut1, into, uusim2…}
[Jamut1 into uusim2…]
[Uusim2 into jamut1…]
BE
MALE
ADULT SINGULAR
ORDERED
FIRST
FIRST
FIRST
*
*
*
!*
The inverse order of the conjuncts in (163) is represented in sentence (165). If the
input has the information given in (166), then the winner will be the candidate (166b)
because it does not violate the constraint BE ORDERED. Candidate (166a) violates this
constraint and is rule out as non-optimal.
265
(165) [uusi-m
into jamut]
teopo-po
bwuik-bae.
Boy-PL
and
woman
church-LOC sing-INTT
‘The boys and the woman will sing in the church.’
(166) Tableau with the order uusim into jamut ‘the boys and the woman’. The ranking is
BE ORDERED >> MALE FIRST, ADULT FIRST, SINGULAR FIRST.
Input: {uusi-m1, into jamut2, …}
a. [Jamut2 into uusim1…]
b. ) [Uusim1 into jamut2…]
BE
ORDERED
MALE
FIRST
!*
*
ADULT
FIRST
SINGULAR
*
*
FIRST
We can say that Yaqui does not have the kind of restrictions in the order of the
conjuncts seen previously because the pragmatic constraints are ranked lower in the
hierarchy of constraints. This approach is tentative but makes predictions that can be
tested typologically. For example, it predicts that the inverse order between BE ORDERED
and the rest of the constraints will produce a language where these constraints will favor
the loosing candidate in table (166). It predicts, too, that there would be rankings between
the constraints, so, if we have MALE FIRST >> ADULT FIRST >> SINGULAR FIRST and the
input contains two nouns, one with the features N[male, adult, singular] and another with the
features N[male, adult, plural] the order will be: [N[male, adult, singular] & N[male, adult, plural].
5.5
Noun coordination and case marking
In Yaqui all conjuncts must be case-marked (except the plural nouns, marked with --
(i)m ‘Pl’, which are mutually exclusive with the suffix --ta ‘NNOM.SG’). In other words,
each conjunct bears information about case-marking. It is never the case that a single case
marking is applied to a coordinate construction. The contrast is shown in sentences (166)(167) where the conjuncts cover the grammatical function of direct object:
266
(167) u
cu’u [[buru-ta]
into
DET
dog donkey-NNOM.SG
and
‘The dog bites the donkey and the horse’
[kaba’i-ta]]
horse-NNOM.SG
(168) *u
cu’u [[buru
into kaba’i]-ta]
DET
dog donkey
and
horse-NNOM.SG
(‘The dog bites the donkey and the horse.’)
ke’e-ka.
bite-PST
ke’e-ka.
bite-PST
The next examples contain coordinate constructions covering several grammatical
functions. As we can see, each nominal gets its own case-marking. (The nominative is
recognized by the absence of morphological marking).
Subject:
(169) [Bochareo
into kuchureo]
jo’ara-po
nau
etejo.
Shoemaker and
fisherman
house-LOC
together talk:PRS
‘The shoemaker and the fisherman are talking in the house.’
Indirect object (with --ta ‘NNOM.SG’):
(170) [Sandra-ta
into Joel-ta]
=ne yokia-m
Sandra-NNOM.SG and
Joel- NNOM.SG =1SG marker-PL
‘I gave the markers to Sandra and to Joel.’
maka-k.
give-PST
Indirect object (with --ta-u ‘NNOM.SG-DIR’):
(171) [Rosa-ta-u
into Patricia-ta-u]
=ne na’aso-m
Rosa- NNOM.SG-DIR and
Patricia- NNOM.SG-DIR=1SG orange-PL
toja-k.
bring-
PST
‘I brought oranges to Rosa and to Patricia.’
Comitative:
(172) Aapo [Lupe-ta-mak
into Lolis-ta-mak]
tekipanoa.
3SG Lupe-NNOM.SG-COM and Lolis- NNOM.SG-COM work.PRS
‘(S)he works with Lupe and with Lolis.’
Genitive:
(173) Joan into
[a= ako-wa
into a=
sai-wa]
uka
John and
3POSS= sister-POSS and
3POSS=brother-POSS DET.NNOM.SG
kari-ta
su’utoja-ka-me.
house-NNOM.SG
left-PST-3PL
‘John and his sister and his brother left the house.’
267
I suggest that each noun is case-marked as a consequence of a constraint over what
can be coordinated. Yaqui data indicate that we only have the coordination of maximal
projections. So, we have a constraint forbidding the coordination of non-maximal
projections. It is defined as follows:
(174)
*COORD OF NMAX-PROJ: Avoid coordination of non-maximal projections.
This constraint is ranked above *CASE and is well founded on empirical and
theoretical grounds. Researchers such as Johannessen (1998) have suggested that
coordination conjoins two (or more) CP’s46. From this point of view, a coordinate
sentence like the following has the indicated structure. Two maximal projections are
coordinated:
(175) Joan e’echa
into
and
John sow.PRT
‘John sows and works’
(176)
tekipanoa
work.PRT
CoP[CP]
CP
Co’[CP]
C’
Co
IP
C
Spec
Joani
CP
into
I’
VP
ti,tj
IP
Io
e’echaj
C
Spec
proi
ti,tl
46
C’
I’
VP
Io
tekipanoal
The coordinator for Johannessen is a functional head, contrary to my proposal, which considers the
coordinator to be an adjunct.
268
On the other hand, Kayne (1994) suggests that languages only coordinate maximal
projections, but not necessarily CP’s. Kayne proposes that Universal Grammar does not
allow coordination of heads. For him, English RNR (Right Node Rising) structures
involves FV coordination always. In the following example, [e] in the first conjunct is an
elided object. It is not the coordination of two finite verbs. However, as pointed out by
Takano (2004), not all cases of V and V in English can be analyzed in this way.
(177) John read [e]i and reviewed [the article]i
I suggest that the interaction of *COORD
OF
NMAX-PROJ with *CASE (avoid case
marking) and Faith-FS (Features in the input must be preserved in the output) gives rise
to the pattern seen in Yaqui. The next table shows the interaction of those constraints.
Candidate (179a) does not violate the constraint *COORD
OF
NMAX-PROJ whereas
candidates (179b) and (179c) do violate it. Because this constraint is ranked above *CASE
and FAITH-FS candidate (179a) emerges as optimal and wins over candidates (179b) and
(179c), which violate that undominated constraint. The winning candidate mentioned in
(166), repeated here as (177), would win the battle against its closest competitor,
candidate (167), repeated here as (178), when they are evaluated by the ranked
constraints:
(178) u
cu’u [[buru-ta]
into
dog donkey-NNOM.SG
and
‘The dog bites the donkey and the horse.’
DET
(179) *u
[kaba’i-ta]]
horse-NNOM.SG
cu’u [[buru
into kaba’i]-ta]
DET
dog donkey
and
horse-NNOM.SG
(‘The dog bites the donkey and the horse.’)
ke’e-ka.
bite-PST
ke’e-ka.
bite-PST
269
(180) Tableau with the ranking *COORD OF NMAX-PROJ>>*CASE>>FAITH-FS, showing
how each noun gets case marking.
Input: {buru,
kaba’i, into…}
[NNOM:SG], [NNOM:SG]
a.)
…[[buru]-ta] into [kaba’i]-ta]…
*COORD OF
NMAX-PROJ
b.
…[[buru] into [kaba’i]]-ta…
!*
c.
…[[buru] into [kaba’i]]…
!*
*CASE
FAITH-FS
**
*
*
**
We have seen in this chapter the interaction of the nominal and verbal classes of
Yaqui. It was shown that the Index and Concord features are useful but not enough in the
explanation of Yaqui agreement between nominal arguments and nouns. It was necessary
to introduce a set of constraints that explains the alternations found in the coordination of
the Yaqui language.
The final ranking proposed in this chapter is shown next. The exploration of nouns cl3 and the different type of verbs indicates that Yaqui has the following final ranking. It
allows the explanation of all patterns described here and that holds between the four types
of verbs and the nominal class three.
(181) HAVE-AF >> *CASE >> CHECK INDEX >> PRESERVE LEXICAL FEATURES >>
UNIFORMITY>>FAITH-I-O>>DISTR.CONCORD>>*NUMBER.
5.6
Summary of chapter 5
This chapter has focused on the description of the morphological number features of
nouns and verbs of Yaqui. We have seen that there are three classes of nouns. The regular
nouns which take singular and plural marking (N1), the nouns which requires singular
marking (N2) and the nouns which requires just plural marking (N3). The (N3) class has
270
a subset of nouns which must be always marked with plural but which are underspecified
for singular or plural. My analysis shows that there are five types of verbs: the verbs that
does not mark number (the jinuk ‘to buy’ class), the singular suppletive intransitive verbs
(the siika ‘leave.SG’ class), the plural suppletive intransitive verbs (the sajak ‘leave.PL’
class), the singular suppletive transitive verbs (the me’a ‘to kill.SG.OBJ’ class), and the
plural suppletive transitive verbs (the sua ‘to kill.PL.OBJ’ class). I analyzed some
interactions related to agreement between nouns and verbs, the order of the conjuncts and
case marking. We saw that the feature system proposed by Halloway King and
Dalrymple (2004) was unable to explain the Yaqui patterns of coordination and we
applied an OT analysis that shows that it is able to predict the alternations described in
this chapter.
271
6
6.1
CONCLUSIONS AND TOPICS FOR FUTURE RESEARCH
Conclusions
This dissertation has described and explained three main topics regarding Yaqui
coordination: the nature of sentential coordinate structures which are treated like adjuncthost relations, the --kai construction that in some cases emerges like a subordinate
structure and in other cases like a coordinate structure, and finally, in Chapter Five, I
provide an analysis of the agreement between nominals and verbs.
This work gave evidence that the ConjP hypothesis is not appropriate for explaining
the coordinate structures of Yaqui sentences. A ConjP approach does not predict and is
unable to account for into ‘and’ in second and last position. This failure is evident
because the ConjP approach predicts that the specifier of the projection must be occupied
by the first conjunct whereas the complement position will be occupied by the second
conjunct. The coordinator, being the head, will be between both conjuncts. Therefore, the
ConjP approach predicts as grammatical sentence (1), and predicts as ungrammatical
sentence (2). However, the opposite situation happens in Yaqui: (1) is ungrammatical and
(2) is grammatical:
(1)
*[Joan bwika-k]
into [maria
and
Mary
John sing-PST
(‘John sang and Mary danced’)
(2)
[Joan bwika-k]
[maría into
John sing-PST
Mary and
‘John sang and Mary danced’
ye’e-ka]
dance-PST
ye’e-ka]
dance-PST
272
We saw that Agbayani and Golston (2002) attempt to rescue the ConjP hypothesis by
claiming that coordinators which appear after one constituent like in (2) are clitics.
Therefore, the moved element attaches to the coordinator, as illustrated in (3):
(3)
ConjP [CP
[Joan buikak] Conj’[Conjo Mariai=into][CP[ti ye’eka]]]
But in Chapter Three we saw that into ‘and’ in Yaqui is not a clitic. It is not
prosodically deficient: it is a minimal word in the language, it is a disyllabic trochaic
foot, it has stress, it is a host for clitizicing other particles and it can occur in first
position. In Yaqui, topicalized elements are located in front of CP. For that reason, it was
suggested that sentence (2) contains a fronted subject and that the landing site of a
topicalized item is not the head of a projection. I have suggested that the coordinator is an
adjunct which attaches to a Maximal projection (CP in this case) and that the topicalized
element is fronted as indicated in (4):
(4)
CP[coord]
Mariai
CP[coord]
into
CP
ti maria ye’eka
Because (4) is now marked with the feature [coord], it licenses the addition of another
CP (the first conjunct) as indicated in (5)47:
47
Recall that into ‘and’ is conceived in this work to be a marker (the sense in which the word “marker”
is used here is similar to the sense used when taking about agreement markers) and an operator (by its
logical properties (truth values)).
273
(5)
CP[coord]
CP
CP[coord]
Mariai
CP[coord]
into
CP
Joan bwikak
ti maria ye’eka
In the final part of Chapter Three, I show that some OT constraints are able to handle
the Yaqui patterns of sentence coordination.
On the other hand, the idea that coordination is the result of an adjunct-host relation is
extended to the analysis of the --kai constructions in Chapter Four. A set of tests indicates
that the --kai verbal construction is marked as subordinated but that it can emerge
syntactically as coordinated (if is a chaining structure) or subordinated (if it is not a
chain). So we saw the following basic representations:
The tree in (6) stands for a --kai coordinate series (see example (35) in Chapter Four).
The into ‘and’ is optional. In this case into ‘and’ cannot occur between the --kai clauses
(*v-kai into v-kai, (into) V-tns).
(6)
CP[(coord)/tns]
CP
CP[(coord)/tns]
V-kai CP
V-kai
CP[(coord)/tns]
CP
(into)
CP
V-tns
A subordinate clause emerges without the occurrence of into ‘and’, as indicated in the
contrast of (7a) and (7b) (see example (38) in chapter four):
274
(7)
a)
CP
V-kai
CP
b)
CP
*CP[coord/tns]
CP
V-tns
CP[coord/tns]
V-kai into
CP
V-tns
The following structure was proposed for cases where into occurs between --kai
clauses. In such a case the (gerundive) tense of the --kai clauses does not depend on the
tense of the final verb in the series. The occurrence of into between the last --kai clause
and the tensed verb is not possible (*V-kai into V-kai into V-tns). The --kai clauses are
subordinated (see examples (98) and (99) in Chapter 4).
(8)
CP
CP[coord/ger]
CP
V-kai
CP
CP[coord/ger]
CP
CP[ger]
into
V-kai
V-tns
The analysis of the --kai clauses indicates too that the construction respects the CSC
and that it is an inviolable constraint in Yaqui. Typologically the --kai construction is a
pseudo-subordinated structure (i.e. it is marked as subordinated but it behaves as
coordinated). There was no attestation of any case of pseudo-coordination in the language
(cases that syntactically are coordinated but that are really subordinated). The use of
constraints helped us to explain the main characteristics of the --kai construction.
Finally, Chapter Five presented a description of nominal and verbal
morphosyntactic features of number. We detected an asymmetry in nominal-verb
275
agreement: two (or more) coordinate nouns combine with a plural verb if the verb is
intransitive, whereas they combine with a singular verb if it is both transitive and agree
with the object. The asymmetry is shown in (9) and (10).
(9)
Joan into Peo sajak/
John and
Peter go.PL.PST/
‘John and Peter left’
*siika
go.SG.PST
(10)
Andrea Joan-ta
into peo-ta
kecha-k/
*ja’abwa-k
Andrea John-NNOM.SG and Peter-NNOM.SG get up.SG.OBJ-PST get up.PL.OBJ-PST
‘Andrea got up John and Peter’
My analysis of the nominal and verbal features of number indicates that there are
several classes of nouns and verbs. These classes interact in such a way that a system of
features like that proposed by Holloway King & Dalrymple (2004) does not explain all of
the resulting patterns of the language. The system appeals to a distinction in number
features: CONCORD features and INDEX features. Chapter Five shows that we need to
recast the observations made by Holloway King and Dalrymple into OT terms. The use
of additional constraints helped us to explain the Yaqui data.
6.2
Topics for future research
I want to close this work by pointing out two interesting areas for future research into
the Yaqui language. One is the possibility of exploring the nature of the input. The other
is the possibility that the coordinator into ‘and’ can be a complementizer in the language.
The nature of the input is worth exploration because in Yaqui it seems that it is
possible to derive the set of sentences in (11)-(14) from a common source. We can
suppose that there is a set of constraints regulating the pronunciation of lexical items. If
this is so, the pronunciation of lexical items will be tied to constraints regulating the
276
intended meaning. The glosses indicate that the meaning tends to be different for each
case. For example, sentence (11) will be used just for a disjoint subject reading. Such a
reading is not available in the other cases. What makes this proposal interesting is the
idea that we have a single input which is able to produce four patterns of coordination
attested in Yaqui: sentence coordination (11), VP coordination (12), NP coordination
(13), and NP discontinuous coordination (14). The question is, then: Could we say that
the following sentences are derived from a common source48?
(11)
(12)
Sentence coordination:
[jabe bibam jinu-k]
[jabe into bino-ta
Who cigars buy-PST
who and
wine-NNOM:SG
‘Who bought cigars and who bought wine?’
(Subject: disjoint reading; two disjoint events)
VP coordination:
[jabe
bibam jinu-k]
[jabe into bino-ta
Who cigars buy-PST
who and
wine-NNOM:SG
‘Who bought cigars and bought wine?’
(Subject joint reading, two disjoint/joint events.)
(Focuses the act of buying, it repeats twice the verb.)
NP coordination:
(13) [jabe bibam jinu-k]
[jabe into
Who cigars buy-PST
who and
‘Who bought cigars and wine?’
(Subject joint reading, one joint event.)
Discontinuous NP coordination:
(14) [jabe bibam jinu-k]
[jabe into
Who cigars buy-PST
who and
‘Who bought cigars and wine?’
(Subject joint reading, one joint event.)
48
jinu-k]
buy-PST
jinu-k]
buy-PST
bino-ta
wine-NNOM:SG
jinu-k]
buy-PST
bino-ta
wine- NNOM:SG
jinu-k]
buy-PST
Oirsow (1987) suggested that coordination is an optional rule that applies over well-formed
sentences of a language. However, as pointed out by Johannessen (1998), that approach is unable to explain
unbalanced coordination (e.g. you and me will go to the party) where one of the conjuncts can not be a
grammatical sentence of the language (*me will go to the party).
277
In addition to the previous examples, we can have the non-pronunciation of the
coordinator, as indicated in (15):
(15)
[jabe bibam jinu-k]
[jabe into bino-ta
Who cigars buy-PST
who and
wine- NNOM:SG
‘Who did buy cigars, who did buy wine?’
(Disjoint / joint subject reading, disjoint/joint event)
jinu-k]
buy-PST
Such an approach will require blocking the following types of candidates due to their
ungrammaticality:
(16)
a. [jabe
b. [jabe
c. [jabe
d. [jabe
e. [jabe
f. [jabe
g. [jabe
bibam
bibam
bibam
bibam
bibam
bibam
bibam
jinu-k]
jinu-k]
jinu-k]
jinu-k]
jinu-k]
jinu-k]
jinu-k]
[jabe
[jabe
[jabe
[jabe
[jabe
[jabe
[jabe
into
into
into
into
into
into
into
bino-ta jinu-k]
bino-ta jinu-k]
bino-ta jinu-k]
bino-ta jinu-k]
bino-ta jinu-k]
bino-ta jinu-k]
bino-ta jinu-k]
Finally, I want to point out an observation made by Sheila Dooley, who has suggested
that the into ‘and’ particle could be considered to be a complementizer (p.c.). If that is
true, then the coordinator will be a head. In such a case, the moved subject will be in the
specifier position of CP, as indicated in (17). We saw in Chapter Three that subjectfronting is obligatory for reasons of topicality.
(17)
CP[coord]
Spec
C’[coord]
Mariai C0
into
IP
Spec
ti’
I’
I0
VP
ye’ekaj
ti, tj
278
Because the CP is marked with a [coord] feature, there can be an adjunction of
another CP, as indicated in (18). So, the first conjunct joan bwikak ‘John sang’ is still an
adjoined CP. Therefore, from this point of view, the ConjP will be just the second
conjunct which serves like a host for the first conjunct:
(18)
CP[coord]
CP
CP[coord]
joan bwikak Spec
mariai
C’[coord]
C0
into
IP
Spec
ti’
I’
I0
VP
ye’ekaj
ti, tj
This suggestion is important from a theoretical point of view. It combines two
previously competing sides of the debate on coordinate structures: coordinate structures
as headed constructions and coordinate structures as adjoined structures. Under this
analysis, the coordinator will be a head in the second conjunct, but the first conjunct will
be the product of an adjoin operation.
279
REFERENCES
Abeillé, A. 2003. A lexicalist and construction based approach to coordination. In
Proceedings of HPSG Conference, East Lansing, Michigan.
Adger, David (2003): Core Syntax: A Minimalist Approach. Oxford: Oxford University
Press.
Agbayani, B. & C. Goldston. 2002. A Typology of Coordination Structure. Ms., CSU
Fresno.
Archangeli, Diana. 1997. Optimality Theory: An introduction to linguistics in the 1990’s.
In Optimality Theory, an overview, eds. D. Archangelli & D.T. Langendoen, 1-32.
Oxford: Blackwell Publishers.
Asher and Vieu. 2005. Subordinating and coordinating discourse relations. Lingua,
115(4):591--610.
Aoun, J., E. Benmamoun & D. Sportiche. 1994. Agreement and Conjunction in some
varieties of Arabic.Linguistic Inquiry 25:195-220.
Bahloul, Maher, and W. Harbert. 1993. Agreement asymmetries in Arabic. In
Proceedings of the Eleventh West Coast Conference on Formal Linguistics, 15-31. Stanford, Calif.: CSLI Publications.
Bahloul, M, and W. Harbert. 2002. Postverbal Subjects in Arabic and the Theory of
Agreement. in U. Shlonsky and J. Ouhalla, eds., Themes in Arabic and Hebrew
Syntax. Kluwer.
Bayer, 1996. The coordination of unlike categories. Language 72. 579-616.
Beaver, D. I. 2004. The optimization of discourse anaphora. Linguistics and Philosophy.
Netherlands: Kuwer Academic Publishers. 27: 3-56.
Beaver and Lee. 2004. Input-Output Mismatches in OT. Reinhard Blutner and Henk
Zeevat (eds.), Optimality Theory and Pragmatics, Palgrave/Macmillan, pp. 112-153.
Blutner, R. & H. Zeevat. 2004. “Editors Preface: Pragmatics and Optimality Theory”. In
R. Blutner & H. Zeevat (eds.), Pragmatics in Optimality Theory, Palgrave
McMillan 2004.
Borsley, R. D. 2005. Against ConjP. Lingua 115, 461-482.
280
Bresnan J. 2000. Lexical-functional syntax. Oxford: Blackwell Publishers.
Buitimea V. C. 2003. Los nominales de la lengua yaqui. ms.
Camacho, J. 2003. The Structure of Coordination: Conjunction and Agreement
Phenomena in Spanish and Other Languages (Studies in Natural Language and
Linguistic Theory 57). Dordrecht: Kluwer Academic Publishers.
Choi, H. W. 1999. Optimizing Structure in Context. Scrambling and Information
Structure. Stanford, California: CSLI Publications, Stanford.
Choi, H. W. 2001. Phrase Structure, Information Structure, and Resolution of Mismatch.
In Sells, P., editor, Formal and Empirical Issues in Optimality Theoretic Syntax,
pages 17--62. Stanford: CSLI Publications,.
Chomsky, Noam. 1957. Syntactic structures. The Hague: Mouton.
________. 1995. The minimalits program. Cambridge, MA: MIT Press.
Cooper, William & Ross, John R. 1975. World order. Grossman, L. J. San & T. Vance
(Eds.) Papers from the Parasession on Functionalism. Chicago: Chicago
Linguistic Society, 63-111.
Corbet, G. G. 1983. Hierarchies, targets, and controllers: agreement patterns in Slavic.
London: Croom Helm.
Cormack, A. & Smith, N. 1994. Serial verbs. Working Papers in Linguistics 6:63-88.
Department of Phonetic and Linguistics, University College London.
Cormack, A and N. Smith (2005). What is coordination. Lingua 115, 395--418.
Crumrine, L. S. 1961. The Phonology of Arizona Yaqui, with texts (University of Arizona
Anthropological Papers, 5). Tucson: University of Arizona Press.
Culicover, P. W. & R.Jackendoff 1997. Semantic subordination despite syntactic
coordination. Linguistic Inquiry 28:195-218.
Dalrymple, M. & R. M. Kaplan 2000. Feature indeterminacy and feature resolution.
Language 76: 759--798.
De Vries, M. 2005. Coordination and Syntactic Hierarchy. Studia Linguistica 59, 83-105.
281
Dedrick, J. M. & E. H. Casad 1999. Sonora Yaqui Language Structures. Tucson:
University of Arizona Press.
De Vos, M. 2004. Pseudo co-ordination is not subordination. L.Cornips & J.Doetjes
(Eds). Linguistics in the Netherlands 2004, Amsterdam: John Benjamins, 181192.
Dik, Simon. 1968. Coordination: Its Implications for a Theory of General Linguistics,
Amsterdam: North-Holland.
Dik, Simon, 1997. Vraagzinnen in een Funktionele Grammatika. Spektator 6, 407-412.
Escalante, F. 1990. Voice and Argument Structure in Yaqui. PhD dissertation, University
of Arizona.
Frazier, L. & C. Clifton. 2001. Parsing coordinates and ellipsis: copy. Syntax 4:1, 1-22.
Frank, A. 2002. A (discourse) functional analysis of asymmetric coordination. Butt M. &
T. Holloway King (Eds.) Proceedings of the LFG02 Conference. National
Technical University of Athens, Greece.
Gáspár, M. (1999). Coordination in Optimality Theory. Nordic Journal of Linguistics,
22:157-182.
Ginzburg, J. & I. Sag.2000. Interrogative Investigations: The Form, Meaning and Use of
English Interrogatives Constructions.
http://lingo.stanford.edu/sag/publications.html
Givon, T. 2001. Syntax: An Introduction. Amsterdam: John Benjamins Publishing
Company.
Gleitman, L. 1965. Coordinating conjunctions in English. Language 41:260-293.
Goodall, G. 1987. Parallel Structures in Syntax. Cambridge: Cambridge University
Press.
Grimshaw, Jane. 1995. Projection, Heads, and Optimality. Linguistic Inquiry 28: 73-422.
Grosz, Barbara Jean, Aravind Joshi, & Scott Weinstein.1983. Providing a united account
of the definite noun phrases in discourse. Proceedings of the 21st Annual Meeting
of the Association for Computational Linguistics, 44-50.
282
Halloway King, T. & M. Dalrymple. 2004. Determiner agreement and noun conjunction.
Journal of Linguistics 40:69--104.
Haspelmath, M. 2004. Coordination. Shopen, T. (Ed.), Language Typology and
Linguistic Description. Cambridge, MA: Cambridge University Press.
Heycock, C. & R. Zamparelli. 2003. Coordinated bare definites. Linguistic Inquiry 34,
443-469.
Hoeks, J. & P. Hendriks. 2005. Optimality Theory and human sentence processing:
Towards a cross-modular analysis of coordination. Submitted to Cognition.
Johannessen, J. B. 1993. Coordination: A minimalist approach. Unpublished doctoral
dissertation, University of Oslo, Norway.
________. 1996. Partial agreement and coordination. Linguistic Inquiry 27: 661-675.
________. 1998. Coordination. Oxford: Oxford University Press.
Kaplan R. M. and J. T. Maxwell. 1988. Constituent Coordination in Lexical-Functional
Grammar. Xerox Palo Alto Research Center.
Kayne, R. 1994. The antisymmetry of syntax. Cambridge, MA: MIT Press.
Kibrik, A. A. 2004. Coordination in Upper Kuskokwim Athabaskan. Haspelmath, Martin
(ed.) Coordinating Constructions. John Benjamin Publishing Company. 537--554
Lambrecht, K. (1994). Information structure and sentence form: Topic, focus, and the
mental representation of discourse referents. Cambridge, MA: University Press.
Langendoen, D. T. 2003. “Merge”. Carnie, A., H. Harley & M. Willie (eds.) Formal
Approaches to Function in Grammar. John Benjamin Publishing Company, 307-318.
Langendoen D. T. & C. Martínez Fabián. 2004. Paradigmas nominales de yaqui y la
Teoría de estructura paradigmática. To appear in Memorias del VIII Encuentro de
Lingüística del Noroeste. México: Universidad de Sonora.
Lee, H. 2001 Markedness and Word Order Freezing. Formal and Empirical Issues in
Optimality Theoretic Syntax. Stanford: CSLI Publications.
Lindenfeld, J. 1973. Yaqui Syntax (University of California Publications in Linguistics
76). Berkeley and Los Angeles: University of California Press.
283
Link, G.1983. The logical analysis of plural and mass terms: A lattice-theoretical
approach. In Bauerle, R., C. Schwarze, & A. Von Stechow (eds.) Meaning, Use
and Interpretation of language, 302-323. Berlin: Walter de Gruyer.
Løudrup, H. 2002. The syntactic structures of Norwegian pseudocoordinations. Studia
Linguistica 56(2), 121-142.
McNally, L. (1993). Commitative coordination: a case study in group formation. Natural
Language and Linguistic Theory 11:347-379.
McCarthy, J. & A. Prince. 1993. Prosodic Morphology: Constraint interaction and
Satisfaction. Technical Report #3 Rutgers University Center for Cognitive
Science, New Brunswick, NJ.
McCarthy, John. 2002. Comparative markedness. http://ruccs.rutgers.edu/roa.html.
Munn, Alan. 1987. Coordinate structures, X-bar theory and parasitic gaps. Honours
Research Paper, McGill University, Montreal, Canada.
________. 1992. A null operator analysis of ATB gaps. Linguistic Review, 9, 1-26.
________. 1993. Topics in the Syntax and Semantics of Coordinate Structures. Ph.D/
dissertation, University of Maryland, College Park.
________.2000. Three types of coordination asymmetries. Schwabe, K. & N.Zhang
(Eds.), Ellipsis in conjunction. Tübingen: Max Niemeyer Verlag, 1-22.
Neijt, A. 1979. Gapping: A Contribution to Sentence Grammar. Ph.D. dissertation,
University of Utrecht.
Oirsow, Van R.R. 1987. The syntax of coordination.London: Croom Helm.
Partee, B. & Rooth, M. (1983). Generalized conjunction and type ambiguity. In Meaning,
Use, and Interpretation of Language, eds. R.Bäuerle, C. Schwarze, & A. Aon
Stechow, 361-383. New York, NY: Walter de Gruyter.
Pesetsky, David. 1997. Optimality theory and syntax: movement and pronunciation.
Archangeli, D. & D. T. Langendoen (Eds.) Optimality Theory, An Overview.
Oxford: Blackwell Publishers, 134-170.
________. 1998. Some Optimality principles of sentence pronuntiation. Barbosa, P., D.
Fox, P. Hagstrom, M. McGinnis & D. Pesetsky (Eds.). Is the Best Good Enough?
Cambridge, MA: MIT Press, 337-383.
284
Peterson, P. 2004.Coordination: Consequences of a Lexical-Functional account. Natural
Language & Linguistic Theory 22:643-679.
Phillips, Colin. 2003. Order constituency. Linguistic Inquiry 34:37-90.
Pollard, C. & I. A.Sag (1994). Head-driven phrase structure grammar. Chicago: The
University of Chicago Press.
Prince, A., & P Smolensky. (2004). Optimality Theory. Constraint Interaction in
Generative Grammar. Malden, MA: Blackwell. (Also appeared as Technical
Report TR-2, Rutgers Center for Cognitive Science, New Brunswick, NJ, April
1993).
Progovac, L. 1997. Slavic and the structure for coordination. In Formal approaches to
Slavic Linguistics, The Indiana Meeting 1996, ed. M. Lindseth & S. Franks, 207224. Ann Arbor, Mich.: Michigan Slavic Publications.
________. 1998. Structure for Coordination. Glot International 3.7 and 3.8.
________. 1999. Events and economy of coordination. Syntax 2:2, pp. 141-159.
Pullum, G. and A. Zwicky. 1986. “Phonological Resolution of syntactic Feature
conflict”. Language, 62: 751-773.
Rebuschi, (2005) Generalizing the antisymmetric analysis of coordination to nominal
modification. Lingua, 115(4):445-459.
Roodenburg, J. 2004. French bare arguments are not extinct: The case of coordinated
bare nouns. Linguistic Inquiry 35 (2), 301--313.
Ross, J. R. 1967. Constraints on Variables in Syntax. Doctoral dissertation, MIT,
Cambridge, MA.
Sadler, L. 2003. Asymmetric coordination and agreement in LFG. In Butt, M. & King, T.
H. (Eds.), Nominals: Inside and Out. Stanford: CSLI Publications. 85-117.
Sadock, J. M. 1991. Autolexical Syntax. Chicago: University of Chicago Press.
Sadock, J. M. 1993. An Autolexical sketch of English. Ms., University of Chicago.
Sag, I., G. Gazdar, T. Wasow, & S. Weisler.1985. Coordination and how to distinguish
categories. Natural Language and Linguistic Theory 3:117-172.
285
Sag, I. & T. Wasow. 1999. Syntactic Theory. A formal introduction. Stanford, California:
CSLI Publications.
Sag, I. et al. 2003. Syntactic Theory. Stanford, California: CSLI Publications.
Schachter, P. 1977. Constraints on coordination. Language. 53:86-102.
Schein, E. H.: 1992, Organizational Culture and Leadership. San Francisco: Jossey-Bass.
Sells, P. 2001 (Ed.) Formal and Empirical Issues in Optimality Theoretic Syntax.
Stanford: CSLI Publications.
Speas, M. 1997. Optimality Theory and syntax: Null pronouns and control. D. Archangeli
& D. T. Langendoen (Eds.), Optimality Theory. An Overview. Oxford: Blackwell
Publishing, 171-199.
Steele, S. 1979. Uto-Aztecan: An assessment for historical and comparative linguistics.
Campbell, L. and M. Mithun (Eds.) The Languages of Native America: Historical
and Comparative Assessment. Austin: University of Texas, 444-544
Steedman, Mark. 2000. The Syntactic Process. Cambridge, Mass.: MIT Press.
Stute, Horst 1985. Os auxiliares dinâmicos da língua gavião. Porto Velho Workpapers: 141. Brasília: SIL.
Takano, Yuji. 2004. Coordination of verbs and two types of verbal inflection. Linguistic
Inquiry 35:168-178.
Teramura, H. 1991. Nihongo-no shintakusu-to imi. Tokyo: Kuroshio.
Van Rooy, R. (2004) Relevance and Bidirectional Optimality Theory. Blutner, Reinhard
and Henk Zeevat (Eds.) Optimality Theory and Pragmatics. Palgrave/Macmillan,
pp. 173-210.
Vallduví, E. 1992. The Informational Component. New York: Garland.
Verstraete, J-C. 2005. Two types of coordination in clause combining. Lingua,
115(4):611--626.
West, B. 1986. "The adversative particle in Tucano discourse." In Joseph E. Grimes
(ed.), Sentence initial devices, 201-14. Summer Institute of Linguistics
Publications in Linguistics, 75. Dallas: Summer Institute of Linguistics and the
University of Texas at Arlington.
286
Wechsler, S. & Zlatić, L. 2000. A theory of agreement and its application to SerboCroatian. Language 76. 759--798.
Winter, Y. (2001). Flexibility Principles in Boolean Semantics: coordination, plurality
and scope in natural language. MIT Press, Cambridge, Massachusetts.
Yuasa E. & J. Sadock. 2002. Pseudo-subordination: A mismatch between syntax and
semantics. Journal of Linguistics, 38:87-111.
Zoerner, 1999. One coordinator for all. Linguistic Analysis 29:322-341.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement