Review of Deixis and Alignment
AUTHOR: Zúñiga, Fernando
TITLE: Deixis and Alignment
SUBTITLE: Inverse systems in indigenous languages of the Americas
SERIES: Typological Studies in Language 70
PUBLISHER: John Benjamins
Katharina Haude, Department of Linguistics, University of Cologne
This book is a functionally-based study of languages whose morphosyntax is
organized according to the referential properties of the arguments. With its
high degree of technical detail, the book is most appropriate for an audience
thoroughly familiar with alignment typology in general and with hierarchically
based language systems and inverse-marking patterns in particular.
The book is divided into eight chapters. Chapters I, II, and VIII deal with
theoretical aspects of direction and alignment; the remaining five (III-VII)
provide case studies of different Amerindian language families: Algonquian,
Kutenai, Sahaptian, Kiowa-Tanoan, and Mapudungun. There are three appendices,
one of Algonquian paradigms, one of Kiowa personal prefixes, and the third
presenting optimality-theoretic accounts of inverses. Separate indices list the
languages, authors, and subjects cited in the volume. In what follows I will
concentrate on some central aspects of the study.
An indexability hierarchy (cf. Bickel and Nichols in press), also known as
''empathy hierarchy'', ''person hierarchy'', ''nominal hierarchy'' and the like, is a
hierarchy in which, roughly speaking, speech-act-participants (SAPs) are located
over third persons, which in turn are ranked according to their animacy and/or
topicality values. In their concrete shape, indexability hierarchies vary from
language to language and, as the book shows, also with respect to different
construction types within a single language. Typologists either try to
accommodate these languages within the SAO framework (S being the single
argument of the intransitive clause, A the most agent-like argument of the
transitive clause, and O the most patient-like argument of the transitive
clause, cf. Dixon 1994), and look at the ways in which these roles are formally
expressed, or they postulate a separate ''hierarchical alignment'' type (e.g.
Nichols 1992). Zúñiga adopts the latter view. His reasoning is that the basic
organization principle of hierarchical systems is not the mapping of A and O
onto different syntactic functions; rather, the position of the arguments in the
indexability hierarchy determines the way in which they are encoded.
Zúñiga furthermore makes a fundamental distinction between hierarchical
alignment and ''direction''. Hierarchical alignment means that indexability
hierarchies ''constrain the mapping between the grammatical role and the
grammatical relations tiers'' (p. 66). Direction, in contrast, is the
morphological marking of the way in which an action proceeds between two
participants. Morphological direct and inverse markers (like the Plains Cree -a
'direct' and -ikw 'inverse'), where direct marking indicates that the argument
higher on the indexability hierarchy is A and inverse marking indicates that the
higher-ranking argument is O, are instances of this. But also the indication of
the ''grammatical role'' (A or O) of one argument on the verb is a case of
direction marking (e.g. when there is a morpheme marking first person O), even
though this may have nothing to do with hierarchical alignment.
An important point concerning the study of direction is the relationship between
inverse marking and passive voice. Like a passive, an inverse construction can
encode the argument with the O role as the topical argument. Therefore,
functional approaches in the line of Givón (1994) do not separate inverse
constructions from voice operations. Other analyses, in contrast (e.g. Dixon and
Aikhenvald 1997), regard inverse constructions as completely different from
voice operations, since they do not change the valency of the verb and are
directly connected to the referential status of the arguments. Zúñiga adopts a
stance between these two positions. He looks at the ways in which indexability
hierarchies surface in the different languages of study, and at the
morphological measures these languages take to mark the mapping of roles and
relations. This can be direct/inverse marking; but also passives that are
sensitive to an indexability hierarchy (as for example in Tanoan languages) are
seen as cases of direction marking.
Zúñiga introduces two major functional parameters along which direction-related
phenomena in different languages can be described: ''direction domains'' and
''focality''. Direction domains concern the parts of the indexability hierarchy
that affect direction marking. The three domains that can be distinguished are
''mixed'' (SAP interacting with third person), ''non-local'' (third persons
interacting), and ''local'' (SAPs interacting). Depending on the domains that are
covered by direction marking, several direction-marking types and subtypes can
be distinguished. A language may employ direction marking only for the mixed
domain, as is the case in Kiowa. Other languages, such as Kutenai, may display
non-local direction only, i.e., they only mark the interaction between third
persons. The Sahaptian language Nez Perce marks only local direction, i.e. the
interaction between speech-act participants. Most of the languages covered in
the study, however, display ''global direction'', which means that direction
marking involves the entire indexability hierarchy.
''Focality'' concerns the specificity of direction markers. Z postulates a
''focality continuum'' that consists of four stages: ''non-focal (unrestricted)
direction'', where the role of one person only is marked (e.g. a morpheme marking
first person singular O); ''low-focal direction'', where verbal morphemes indicate
the direction between the persons as they are located on the indexability
hierarchy, but which do not specify the persons themselves (e.g. the Algonquian
direct/inverse markers); ''mid-focal direction'', which is more specific about the
persons involved than the former type, but where e.g. the hierarchical distance
between the participants involved is indicated; and finally ''high-focal
(particular) direction'', where both persons involved in the event are made
explicit (e.g. a morpheme indicating that the first person singular acts on the
second person). The concept of focality is seen as an ''analytical tool to track
extensions and reductions'' (p. 247), i.e. it helps to show that a language makes
more explicit distinctions in certain domains than in others.
Formal aspects of ''direction'' include the locus of marking (detached, head,
dependent, or double-marking) and the relationship between direction and
morphosyntactic alignment, i.e., the ways in which direction influences e.g. the
access to particular morphological slots or syntactic functions. As far as the
locus of marking is concerned, direction is generally associated with
head-marking morphology, and this is also the case in most languages
investigated in the book. The issue of grammatical relations is far more
complex. Especially with regard to Algonquian languages, there is some
controversy as to the effect of inverse marking on morphosyntactic alignment. It
seems that in Plains Cree, there is no such effect, whereas in Central Ojibwa,
inverse marking causes a remapping of A and O onto grammatical relations.
Equipped with these formal and functional criteria, Zúñiga analyzes twelve
different languages from five different families. It turns out that languages
that show grammatical reflexes of indexability hierarchies differ widely in the
ways in which they mark direction, and that the indexability hierarchies
themselves are organized differently not only in the various languages but also
depending on the type of construction. As the author states, ''[j]ust as the
morphosyntax of a given language may show different pivots in different realms,
it might be the case that different phenomena are governed by different
underlying hierarchies'' (p. 253). In this way, the study shows how heterogeneous
languages with hierarchical alignment can be and in how many different ways
indexability hierarchies can be reflected.
The strongest part of the book is the description of the alignment systems of
the individual languages (chapters III-VII). The accuracy with which the author
presents and analyzes the data and the accounts given by other linguists is
simply impressive. Zúñiga does not oversimplify anything, and he never jumps to
conclusions. When the data do not allow a clear-cut conclusion, he leaves the
question open for further discussion or for the eventual future availability of
more data (which, as he stresses, is problematic in view of the fact that most
of the languages under study are in danger of extinction). The same holds for
Zúñiga's treatment of the different theoretical approaches to hierarchical or
inverse systems, which he discusses and weighs carefully. In this way, the book
presents an excellent comparative overview of the different ways in which
inverse or hierarchical systems are dealt with.
Another merit of the book is that, while (as the author himself admits), the
language sample is somewhat random, always several members of one family are
discussed. In this way, and in connection with the analytical tools developed in
the first two chapters, it can be observed how easily hierarchical effects can
arise or be blurred in the course of time and due to language contact.
The presentation of the theoretical framework might have been clearer, however.
Sometimes the use of the terminology seems a bit inaccurate, despite the fact
that the author is highly aware of terminological issues. Several terms are used
inconsistently. For example, ''grammatical roles'' are defined as being S, A, O,
and E (the third argument in an extended transitive clause, cf. Dixon and
Aikhenvald 2000), but the author often refers to these categories as
''functions''. This is confusing since he also uses the term ''syntactic functions''
as a synonym of ''grammatical relations'', which refers to language-dependent
categories such as subject and object. On other occasions, A, O, and E are even
called ''macroroles'' (p. 199, 209f.), a term that is associated with the notions
''actor'' and ''undergoer'' of Role and Reference Grammar. I assume that these
terminological problems arise from the not very clear-cut status of the
categories S, A, and O, which have both syntactic and semantic characteristics
and were, accordingly, initially called ''semantic-syntactic primitives'' by Dixon
(1994; cf. also Payne 1997).
It is sometimes hard to find straightforward definitions, for example of the
term ''direction''. In principle, Zúñiga uses this term to refer to all kinds of
formal indications of the way in which an action proceeds between two
participants. However, a definition given on p. 28 suggests that direction is
not independent of hierarchical alignment: ''direction reflects the alignment
between the indexability hierarchy and a relational hierarchy where A's outrank
O's. When the higher referent is an A, a predicate or a whole clause is marked
as direct. Inverse is the label used for constructions where the higher referent
is an O''. On p. 31, Zúñiga characterizes ''direction'' as ''a general term that may
comprise spatial, temporal, and/or personal/actional direction'', but then he
continues saying that ''[c]onsequently, I will henceforth use the term
DIRECTION-MARKING SYSTEM as the preferred label and INVERSE SYSTEM as a
shorthand for the former'' (p. 31; emphasis in the original). Note that in
general, only the term ''direction'' is used. I am convinced that the author has
clear concepts in mind, but as a reader, I felt slightly confused.
The problems of terminology and definition, in part caused by the complicated
topic, might have been helped out by a good index. However, the index is not as
useful as it could be. In those cases where two terms are used for the same
category, only one of them is cross-referenced. Some central terms are omitted
altogether: ''direction'', ''grammatical role'', ''primary/secondary argument'',
''voice''. I would also have liked to have been able to find, by means of the
index, the discussions of controversial issues such as of the notions ''subject''
and ''object'', which are discussed on several occasions, e.g. in the chapters on
Algonquian and Mapudungun. Since for me, this book will be an important source
of information in the future, I found myself elaborating my own index and
glossary - a task that should not normally be that of the reader.
Despite these drawbacks, it has to be stressed that with its useful analytical
framework, the thorough description of hierarchical alignment systems in
individual languages and its comparison of different theoretical approaches,
this book is a very important contribution to the study of an extremely complex
linguistic phenomenon. It will be indispensable for anyone interested in
indexability hierarchies and inverse systems.
Bickel, Balthasar and Johanna Nichols. In press. ''Inflectional Morphology.'' In:
Shopen, Timothy (ed.), _Language typology and syntactic description_. Cambridge:
Cambridge University Press. Second edition.
Dixon, R. M. W. 1994. _Ergativity_. Cambridge: Cambridge University Press.
Dixon, R. M. W. and Alexandra Y. Aikhenvald. 1997. ''A Typology of
Argument-Determined Constructions.'' In: Bybee, Joan et al. (eds.), _Essays on
Language Function and Language Type: Dedicated to T. Givón_. Amsterdam:
Benjamins. pp. 71-113.
Dixon, R. M. W. and Alexandra Y. Aikhenvald 2000. ''Introduction.'' In: Dixon, R.
M. W. and Alexandra Aikhenvald (eds.), _Changing Valency: Case studies in
transitivity_. Cambridge: Cambridge University Press. pp. 1-29.
Givón, T. 1994. ''The pragmatics of de-transitive voice: functional and
typological aspects of inversion.'' In: Givón, T. (ed.), _Voice and Inversion_.
Amsterdam/Philadelphia: John Benjamins. pp. 3-44.
Nichols, Johanna. 1992. _Language Diversity in Space and Time_. Chicago: The
University of Chicago Press.
Payne, Thomas. 1997. _Describing Morphosyntax: A guide for field linguists_.
Cambridge: Cambridge University Press.
ABOUT THE REVIEWER
Katharina Haude studied General Linguistics in Cologne and Comparative
Linguistics of Amerindian languages in Leiden. For her doctoral thesis at the
Radboud University Nijmegen, she wrote a grammatical description of Movima, a
native language of lowland Bolivia. Currently she holds a post-doc position at
the University of Cologne, where she works on a documentation project for
Movima, financed by the Volkswagenstiftung. Her research interests include South
American Indian languages, alignment typology (in particular, inverse systems
and ergativity), and word-class distinctions.