|EDITORS: Rowicka, Grażyna J.; Carlin, Eithne B.
TITLE: What's in a verb?
SUBTITLE: Studies in the verbal morphology of the languages of the Americas
SERIES: LOT Occasional Series
PUBLISHER: Netherlands Graduate School of Linguistics / Landelijke
Michael Maxwell, Center for Advanced Study of Language, University of Maryland
The morphology of verbs is typically more complex than that of nouns, and this
is nowhere more true than in the Americas. This volume brings together a dozen
studies, each around 20 pages, of the verbal morphology of a diverse set of
languages of North, Central and South America. It goes without saying that most
of these languages are endangered, and some are moribund (or even extinct as I
write this). Each article is written by a linguist from the Netherlands, or a
linguist who has been affiliated with an institution in the Netherlands. Themes
range from a sketch of inflectional affixation to the semantics of a particular
Together, the papers give a reasonably broad view of the typology of morphology
in the Americas. As one might expect, the quality of the papers varies. Since I
cannot cover all twelve articles in detail, I will concentrate on some high
points. A theme that recurs is that these languages do things that one might not
expect if exposed only to languages of other parts of the world (or indeed, to
just one region in the Americas). Many of the articles can therefore be
profitably read as a way to broaden one's typological horizons.
Peter Bakker's article on Cree morphology manages to introduce three topics of
interest that those who only know languages from other parts of the world may
not have encountered: obviation, stem selection based on animacy, and person
hierarchies. The latter refers to the fact that both subject and object
agreement is marked on Cree transitive verbs, but not by separate subject and
object affixes. Rather, first, second and third persons (and the obviative
person) are marked by affixes; but whether a given affix represents the subject
or the object is dependent on a hierarchy in which, for example, second person
outranks first, in conjunction with a separate affix which determines whether
the higher ranked person is subject (so-called direct mode) or object (inverse
mode, perhaps analogous to a passive in more familiar languages).
Jan van Eijk's article in this book constitutes a typological overview of person
marking systems, and could profitably be used as an introduction to Bakkers'
more detailed exposition of a direct/ inverse system.
The transitivity vs. intransitivity distinction should be familiar to most
readers. But Annette Veerman-Leichsenring brings to the reader's attention a
different categorization of verbs from a Popolocan language of Mexico. This
categorization is based on marking of subject and/or object, but it completely
cross-cuts the transitivity/ intransitivity distinction. I confess to having
approached this article with a sense of disbelief; how can an intransitive verb
mark an object? (The answer is that the intransitive verbs in question have only
implicit subjects, with meanings like ''(it) is known to a person'' or ''(it) gets
lost to a person.'' Those interested in the unergative/ unaccusative distinction
are advised to read this paper carefully.) Other verbs are obligatorily
reflexive, or while transitive mark the subject but not the object. The latter
appears again to be related to person hierarchies, although this point of
contact between analyses is not developed.
Another article developing a typologically unusual theme is that of Mily
Crevels, concerning ''verbal number'' in a Bolivian language. While the concept is
not unknown to typologists, Crevels nicely lays out the semantics of the
construction in this particular language.
Willem Adelaar discusses directional suffixes in a variety of Quechua.
Intriguingly, the analysis draws a distinction between so-called ''final'' and
''non-final'' suffixes. While the non-final suffixes tend to be more derivational
than the final suffixes, this is not always true, and they have apparently
shifted their status in both directions during the development of the Quechuan
languages. Adelaar traces the development of what were clearly directional
affixes (and as such largely derivational) at an earlier stage of the language,
but which often have an aspectual (and therefore more inflectional) usage now.
The result is perhaps analogous to the distinction between the so-called
'conjunct' and 'disjunct' affix domains in Athabaskan languages.
Simon van de Kerke undertakes a study of Leko, a moribund language of Bolivia,
and in particular the nature of object indexing on verbs. A frustration in
reading this paper is the fragmentary nature of the data; the author draws
conclusions, but one can't help the feeling that a different analysis is in
order, or else that further data is needed to justify the conclusions. His
conclusion, for example, is that the use of ''object cross-reference markers...is
not unconstrained and not fully predictable. The constraints are
syntactic/semantic but also pragmatic.'' In other words, we don't know what the
constraints are. The author is aware of the problems, and of course cannot be
blamed; the language was declared dead some years ago, and if it had not been
for van de Kerke's fieldwork (done with elderly speakers who had partially
shifted to Spanish), we would know even less about the language. Salvage
linguistics, at its best and its (unavoidably) worst.
Hank Nater studies alternations in the affixes of Tahltan, an Athabaskan
language. Athabaskan languages are rightly notorious for their complex verbal
morphology, and Tahltan is no exception. This is nowhere more apparent than in
the fusion of affixes resulting from phonological reduction processes operating
across morpheme boundaries. Nater presents a bewildering number of cases of
allomorphy, particularly with stems, and suggests that the synchronic morphology
can only be clarified by reference to the diachronic changes, and therefore
through comparison with related languages. Those who studied generative
phonology in the 1960s or 1970s will recognize this theme, and will be aware of
the potential abuse of this methodology.
Other papers touch on historical reconstruction of morphology (Grażyna J.
Rowicka's article on a Salish language, Pieter Muysken and Katja Hannss's
article on an Uru or Uru-Chipaya language of Bolivia), address the semantics of
affixes (Eithne B. Carlin's description of verbalizers in a Cariban language),
evidentiality (Stella Telles and Leo Wetzels' article on a Nambikwara language
of Brazil), or the language-particular status of parts of speech (Sérgio Meira's
article on stative verbs in a Tupian language; the question is whether and how
they differ from nouns).
What you will not find in this work is theoretical studies of morphology, nor in
most cases much discussion of the theoretical implications of the results. You
will also not find an in-depth description of the verbal morphology of any of
these languages. Both limitations are a result of the article length: twenty or
thirty pages is simply not enough to accomplish either of these goals, given the
complexity of verbal inflection in these languages. What you will find is
bite-sized sketches, suitable for reading at a single sitting, with frequent
comments that further study is needed (a comment that is surely superfluous,
given that it could apply to any topic in linguistics and language description).
Given this perspective, many of the articles in this book will be of interest to
morphologists, to those who specialize in one or another language family of the
Americas, or to typologists. Nevertheless, it seems unlikely that most
researchers would need this book on their shelves, and while the book's price is
quite low, one wonders how many libraries would order it, given that all the
papers are available on-line, at http://lotos.library.uu.nl/index.html. Perhaps
the time has come for our field to consider web-published papers (if they have
been reviewed and accepted for ''publication'') as having equal standing with
papers which appear in printed books or journals.
As is typical with analyses that display considerable complexity in the realms
of morphology, phonology or diachronic changes, one may wonder whether some of
the analyses actually work. In my opinion, it is high time that linguists verify
their analyses computationally. The sheer number of phonological processes and
etymologies in Nater's paper, to take just one example, makes it virtually
impossible to hand-verify the analysis. (I speak from the experience of writing
morphological and phonological analyses which, when implemented on a computer,
turn out not work.) Computational morphology and phonology is still not as
simple as one might like, but it is doable, particularly for the straightforward
phoneme-based rules which are used in most of these papers. (Confirming an
OT-based analysis, or even a rule-based analysis in autosegmental phonology, is
Another drawback to some of the papers in this book is the use of Americanist
phonemic characters in place of IPA characters. While the use of orthographic
transcriptions is reasonable, it is not clear in this day why non-IPA phonemic
transcription should be used.
ABOUT THE REVIEWER
Dr. Maxwell is a researcher in computational morphology and other computational
resources for low density languages, at the Center for Advanced Study of
Language at the University of Maryland. He has also worked on endangered
languages of Ecuador and Colombia, with the Summer Institute of Linguistics.