Review of Linguistic Diversity and Language Theories
Date: Sun, 13 Nov 2005 23:32:24 -0800 (PST)
From: Solveiga Armoskaite
Subject: Linguistic Diversity and Language theories
EDITORS: Frajzyngier, Zygmunt, Hodges, Adam, Rood, Davis S.
TITLE: Linguistic Diversity and Language theories
SERIES: Language Companion Series 72
PUBLISHER: John Benjamins Publishing Company
, Graduate Student, Department of Linguistics, University of British
Columbia, Vancouver, Canada
Chapter 1 What are we typologists doing? Gilbert Lazard reaffirms
the importance of the typology in linguistics for only thus can we prove
the scientific value of the linguistic endeavour. Lazard argues that the
nature of cross-linguistic invariants is at heart of the typological
research. For him, invariants are neither forms nor meanings
(Saussurean 'signifiants' and 'signifiés'). It must be located at a more
abstract level, namely: these are the relationships appearing in the
correlation between forms and meanings. One of the major problems
of the typological comparison is lack of reliable methodology. Lazard
rejects both the use of linguistic categories (language specific) and
semantic substance (too amorphous). Instead, he argues that intuition
is the way out and advocates ACF, 'arbitrary conceptual frameworks'
as the point of departure for typological research. Lastly, Lazard
critically evaluates the relationship between the typological linguistics
and cognitive science and concludes that it would not '...be a scandal
to say that in the present state of things, the potential contribution of
linguistics to cognitive sciences is more important than the reverse.'
Chapter 2 The canonical approach in typology. Here, Greville Corbett
discusses the canonical approach to typology. Following Chomsky,
the aim is to define the notion of possible human language. Under this
view, the author takes the linguistic definitions (e.g. agreement) to
their logical endpoints and builds theoretical spaces of possibilities.
Only then the actual linguistic data is checked to verify how this
theoretical space is actually populated. The canonical data
('best', 'clearest' prototypes of particular linguistic elements) is not
necessarily the most pervasive and may not be found in natural
languages altogether. The major part of the chapter is devoted to the
specifics of the implementation of the canonical approach, e.g.
construction of the typological database.
Chapter 3 What is an empirical theory of linguistic meaning a theory
of? Pierre-Yves Raccah addresses the broad issue of how scientific
theories work. Contrary to the commonsensical views, science does
not directly account for the 'laws of nature' but rather indirectly
explains how a particular theory describes and predicts the
phenomenon it is supposed to account for. Our only access to the
world is indirect, limited by the capacity of our cognitive apparatus.
These limitations affect the study of linguistics. Thus, the meaning in
natural languages can be the object of an empirical science only
indirectly. We have no access to truth conditional semantic meaning,
but we can study the set of constraints that words and structures in a
particular language impose on the construction of meaning of the
utterances, concludes Raccah.
Chapter 4 Language processes, theory and description of language
change, and building on the past. This chapter by Robert Nicolai
stands out due to the particularly hard questions and entrenched
viewpoints it calls to re-evaluate. Faced with a controversial genetic
classification of Songhay language as Nilo-Saharan, Nicolai questions
the theoretical and practical tools available for language classification.
Must we accept that linguistic processes can be correctly
apprehended only on the a priori hypothesis of one homogenous
structural system? Is monolingualism to be taken as normal state of
affairs? Nicolai invites us to consider the use of real or virtual
multilingual/multidialectal factors as normal theoretical parameters in
building models of language evolution.
Chapter 5 On the part played by human conscious choice in language
structure and language evolution. Due to the impressive range of the
sources and issues covered, it is impossible to give justice to the
chapter contributed by Claude Hagege. In a nutshell, the author
argues that ordinary language users, grammarians and even
politicians consciously shape the evolution of the language. Thus,
concludes Hagege, one should not take for granted the common view
that linguistic classifications never rises into consciousness.
Chapter 6 The challenge of polygrammaticalization for linguistic
theory: fractal grammar and transcategorial functioning. The chapter
contributed by Stephane Robert reads easily, as any linguist is bound
to experience the sense of déjà vu given the topic: the behavior of
transcategorial morphemes, i.e. linguistic units used synchronically in
different syntactic categories. Time and again we come to grips with
the topic (cf. Jespersen 1924, Arad 2003 among many, many others),
yet the definite account proves to be elusive. Robert proposes a
dynamic analysis that relates the polysemy of these morphemes to the
syntactic flexibility. First, the syntactic category of the linguistic
element emerges as a result of its immediate context (or co-text,
utterance environment). Next, the syntactic variation triggers variation
in its semantic scope and activation of contextual properties. In order
analyze the fine details of a transcategorial morpheme, Robert works
out a certain scale (or: the common schematic form) that sums up its
uses and senses and shows how the morpheme functions as a
landmark within the utterance. This schematic form is not to be
confused with a semantic feature; it serves as a matrix for the
construal of new meanings when mapped onto new domains. The
author concludes questioning the status of the grammatical categories
in the light of the transcategorial morphemes.
Chapter 7 On discourse frequency, grammar and grammaticalization.
Regina Pustet seeks to revive the interest in the work of George
Kingsley Zipf (1902-1950), who is remembered for discovering for the
presumably universal statistical correlation between the discourse
frequency of linguistic items and their structural complexity or length.
Pustet reminds that Zipf also established statistical correlation
between discourse frequency and semantic complexity. However the
said correlations were used to study lexical items. It comes as a
natural next step to apply Zipf' s method to the study of
grammaticalization, argues Pustet.
Chapter 8 On the assumption of the sentence as the basic unit of
syntactic structure. Marianne Mithun elegantly argues that the status
of the sentence as the basic privileged unit maybe overrated in the
light of cross-linguistic data. Linguistic structure goes beyond the
sentence into the realm of discourse. Mithun discusses switch
reference markers in Hualapai (Yuman family): their use at the
sentence level for marking syntactic relations among clauses and
exploited to signal cohesion in a larger discourse unit. The author also
reviews the rationale behind the choice between the two markers and
their probable origins. Although the article is generously supplied with
interesting data, a more detailed morpheme break up in the longer
narratives would be appreciated by a curious reader.
Chapter 9 Adpositions as a non-universal category. On the basis of
crosslinguistic evidence from several languages and Klamath in
particular, Scott DeLancey reasons that there is no useful sense in
which the category of adposition can be considered a linguistic
universal. Klamath proves to be an interesting case for the debate. It
offers a singleton category candidate for adposition while other
adpositional functions can be integrated into e.g., certain kind of
compound verbs. In sum, the function of adpositions is always
present in any language, but it's expression cross-linguistically is in
Chapter 10 Understanding antigemination. Juliette Blevins advocates
for Evolutionary Phonology approach in the attempt to explain the
phenomenon of antigemination (instances of blocking gemination).
Alternative approaches by McCarthy (1986) and Odden (1988) are
weakened by the operation of the tier conflation and suggestion that
OCP is not a component of Universal Grammar respectively. Blevins'
Evolutionary Phonology working hypothesis is that common sound
patterns result from common phonetically motivated sound change.
According to Blevins, it provides a simpler solution for antigemination.
Sequences of identical consonants will more readily resist re-
interpretation as geminates in languages with pre-existing
consonantal length contrast, and it will belong to a class of natural
histories. In languages where pre-existing length contrasts are absent,
the antigemination processes will fall into the class of unnatural
Chapter 11 What it means to be rare: the variability of person
marking. Michael Cysouw calls to bring up to date the relationship
between linguistic variability and the explanatory power of linguistic
theory. His research is motivated by the question: how much of
linguistic variation should be explained by a theory for it to be
considered sufficient or good? Cysouw does not offer solutions but
he sure raises good questions. For example, methodological
concerns are a challenge in itself. Cysouw states that his collection of
person marking cases is rather ad hoc; but the general tendency in
typology is to sample the most common cases and leave out the
uncommon ones. It is hard to develop a procedure for sampling
possible, yet rare, cases.
Chapter 12 The principle of Functional Transparency in language
structure and in language evolution. Zygmunt Frajzyngier's
Functional Transparency refers to functional domains coded in the
language. Frajzyngier first discusses the three components of the
principle: i) every utterance has a transparent role in discourse; ii)
every constituent has a transparent role within the utterance; iii) the
choice of the role of the constituent is dictated by the functional
domains coded in the language rather than by addressee's needs to
understand the role of the constituent in some real world. Next, the
author identifies and exemplifies the coding means available across
languages (not necessarily in every language). These are: lexicon
(especially where lexical items encode inherent constituent information
like 'verb'), linear order, adpositional coding, conjunctions and
other 'function' words, inflectional coding, phonological means. The
systems proposed by Frajzyngier appears to be elegant and
straightforward, but it is not immediately obvious how would the author
address the issues such as the role and impact of context, ambiguity,
multifunctional morphemes and the like.
Chapter 13 The importance of discourse analysis for linguistic theory.
Liang Tao promotes a usage based discourse analysis in the study of
language and cognition based on her exhaustive study of Mandarin
Chinese classifier system. The paper is based on a solid corpus of
data from both spoken and written discourse, something which the
previous analysis neglected to consider. Based on the advances in
cognitive processes, Tao builds on the assumption that grammatical
patterns reflect human observations of the physical world around
them. However, while cognitive representations may be constant, the
linguistic coding systems change over time. Phono-syntactic
conspiracy describes the case in point: the dynamic, ever changing
phenomenon of Mandarin Chinese numeral classifier system.
Chapter 14 Compounding theories and linguistic diversity. Anders
Søgard provides an overview of existing theories on compounding
only to conclude that most of them are Euro-centric and fail to predict
productive patterns of compounding attested in non-European
languages. Drawing on typological data and fusing together findings
highlighted by diverse linguistic theories, Søgard puts forward his
theory: the construction hierarchy of compounding. The hierarchy is
comprised of distinct but interrelated levels of more or less schematic
constructions, where every level is associated with conventional
specific formal or semantic features (e.g., phonology, argument
structure, event structure.. and the like). Søgard's original contribution
is that the hierarchy allows for the gradience between literal,
metonymical and metaphorical relatedness as well as finds a place for
Chapter 15 Inalienability and possessum individuation. To Frantisek
Lichtenberk, one of the challenges of linguistics is to investigate the
interplay between functional/cognitive factors and language
structures. In this light, he provides an engaging analysis of (in)
alienable possessum constructions in Toqabaqita (Oceanic)
language. The general pattern is that constructions in which an affix
indexing the possessor appears on the possessum noun encode
inalienable possession (synthetic means), while constructions with
possessive classifiers express alienable possession (periphrastic
means). Lichtenberk focuses on the cases where the same noun can
occur in both constructions. The main factor here, argues Lichtenberk,
is that of possessum individuation: the distinctness of an entity from its
own background, specifically the distinctness of the possessum from
the possessor. When the inalienable possessum is not individuated
('my eye'), the synthetic possessive construction is used; if the
possessum is individuated ('my left eye', 'my this eye'), the periphrastic
construction is used.
Chapter 16 Resultativeness in English. Marina Gorlach reflects on
resultative constructions in English as manifested by V NP Prt
(marked: resultative) versus V Prt NP (unmarked: ambiguous between
resultative/non-resultative interpretations). Her goal is to establish the
category of resultativness as a linguistic universal. The most
interesting part of the discussion is her data where the two English
phrasal constructions -- the marked and the unmarked one -- are
compared in the light of text translations into Russian. Gorlach shows
that there is correlation between the English V NP Prt resultative
interpretation and the use of perfective forms in translation to Russian.
Chapter 17 Encoding speaker perspective: evidentials. Based on
cross-linguistic data, Ferdinand de Haan argues that evidentiality is a
deictic rather than a modal category. De Haan painstakingly goes
through a variety of evidential expressions -- visual, inferential,
auditory, quotative -- to establish that evidentiality asserts the
evidence while epistemic modality evaluates it. Based on the
discussion, he proposes to add evidentiality to category deixis as an
example of propositional deixis. The author does not deny the relation
between the evidentiality and epistemic modality, but such a
relationship is considered to be secondary.
Chapter 18 Distinguishing between referential and grammatical
function in morphological typology. Edward Vajda tackles the question
of the fallacies in the traditional morphological typology (division of
languages into analytic, synthetic and polysynthetic). He first points
out the gaps and inconsistencies of the system and then suggests a
model of 'Holistic Grammar'. Vajda's model aims to replace the fuzzy
notions of derivation and inflection with clear cut dichotomies between
referential, discourse and phrasal functions subdivided into their
possible head vs. modifier roles. The resultant grid is when applied to
the case study of Ket (endangered isolate in Central Siberia).
The book offers a smorgasbord of current theoretical approaches to a
variety of linguistic topics. It succeeds in identifying the "... unstated or
understated fundamental issues in linguistic theory..." which the
editors of the volume named as the main purpose of the International
Symposium on Linguistic Diversity and Language Theories, the basis
for the volume. Due to the number of articles and the wealth of issues
covered it is impossible to single out a few for a close evaluation.
Instead, I would rather single out a few reoccurring topics. In terms of
theory, functionally and cognitively driven approaches stand out, with
a few digressions into the emergent or dynamic view of grammar.
Functionalists present clearly defined claims that easily lend
themselves to further debate, whether one happens to agree or
disagree. Efforts to anchor linguistics cognitively come across more as
assumptions taken from the realm of psychology and are therefore
difficult to evaluate within linguistic context. The issue of exceptional
and/or rare linguistic patterns and how linguistic theories aimed at
regularity deal with it is another particularly interesting topic that
several authors address. The conclusions reached vary as much as
the data they are based upon. The concern about typological
methodology is another prominent question that keeps re-surfacing
throughout the articles. The authors ponder such issues as lack of
methodological uniformity or some standard basis for cross-linguistic
comparison, inadequacy of terminology used, and a priori theoretical
assumptions that interfere with interpretation of data. Overall, it is a
worthwhile book where everyone can find something to enjoy.
Especially if you hit rock bottom and need a challenging linguistic
question or two, look no further.
Arad, M. 2003. Locality constraints on the interpretation of roots: the
case of Hebrew denominal verbs. Natural Language and Linguistic
Theory 21: 737-778.
Jespersen, O. 1924. The philosophy of grammar. London: George
Allen & Unwin. Odden.
| ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Solveiga Armoskaite is a graduate student at University of British
Columbia. Syntax-semantics interface of aspect, typology, and
language documentation/preservation are her main interests. She
works on Lithuanian, Cree and Blackfoot.