Date: Fri, 15 Jul 2005 17:29:33 -0300
From: Aroldo Andrade <aroldo.andrade@gmail.com>
Subject: Language Contact and Grammatical Change
AUTHOR: Heine, Bernd; Kuteva, Tania
TITLE: Language Contact and Grammatical Change
SERIES: Cambridge Approaches to Language Contact
PUBLISHER: Cambridge University Press
YEAR: 2005
Aroldo L. Andrade, unaffiliated scholar
INTRODUCTION
Heine and Kuteva's book aims at demonstrating that transfer of grammatical
meanings and structures across languages is a regular phenomenon, shaped
by universal processes of grammatical change. It also provides an overview
of typological studies on language contact and a good deal of data
relevant to the subject from all over the world.
DESCRIPTION
The book is divided into seven chapters. Its intended audience includes
advanced students on language change, sociolinguistics and linguistic
anthropology.
Chapter 1 show the limits of the research. Among the sources of
similarities between languages, language contact was chosen and, among the
kinds of linguistic transfer available, the transfer of meanings was
focused. Languages are more studied in the book than dialects for lack of
data. Some basic concepts are presented, as those of use patterns and
grammatical (or functional) categories: while the former refers to
recurrent pieces of discourse associated with the same grammatical
meaning, the latter label concerns stable, conventionalized form-meaning
units serving the expression of grammatical functions. Alternative
approaches and terms are compared to those used in the book. Besides, the
authors explain the thesis, the theoretical and methodological assumptions
and offer an overview of the remaining chapters.
Chapter 2 draws on the role of use patterns in the emergence of new
structures as a result of language contact. The framework used is based on
the notion of replication, the transfer pattern from model to replica
language. These notions are similar to Weinreich's (1964 [1953]) source
and recipient language, with the difference that they do not involve form-
meaning units, but only meanings, as in the following example, where
Portuguese (1a) is the model for the young Tariana speakers (1b), who
recognised that relative pronouns are also used as relative clause markers
in their L2 Portuguese (Tariana is an endangered North Arawak language
spoken in Northwestern Brazil; examples are from Aikhenwald 2002).
(1) a.
quem sabia, falava assim.
who knew spoke like.this
b.
kwana ka-yeka-kani kayu-na na-sape.
who REL-know-PAST.REL.PL DEM:ANIM thus-REM.P.VIS 3.PL-speak
'Those who knew, spoke like this'
When the grammaticalization process begins, Heine & Kuteva say, there is
already some rarely used collocation, which they call a minor use pattern.
The rise of a major use pattern relates to (i) increased frequency of use,
(i) use in new contexts (=extension), and (iii) its association to new
grammatical functions. This last item can be related either to the
emergence of new meanings or to narrowing, it means, to the restriction of
a pattern to a particular use, equivalent to one existent in the model
language. Four case studies are provided in order to illustrate the
mechanism. Lastly, the discussion on the transition from a major use
pattern into a full-fledged category paves the way to the next chapter.
Chapter 3 concerns the development of grammatical categories, the phase
involving genuine grammaticalization. The authors distinguish two types of
contact-induced grammaticalization, as they call the process that creates
functional categories due to language contact. Ordinary grammaticalization
is a process by which the model language only provides the category to be
replicated, as shown in (2a). It is stressed that the process is a
creative act, although constrained by (i) universal principles of
grammaticalization; (ii) the nature of the model category and (iii) the
structural outfit of the languages involved. On the other hand, replica
grammaticalization is a process by which the model language provides not
only the category, but also the way it is replicated, as in (2b).
(2) a. Ordinary grammaticalization: Mx = [Ry > Rx]
b. Replica grammaticalization: [My > Mx] >> [Ry > Rx]
where M = model language; R = replica language; x, y = use patterns or
functional categories; > = "develops into"; >> = "is replicated on".
After the distinction, there is an attempt to interpret cases of polysemy
copying -- also referred to as calquing or loan translation -- as also
involving grammaticalization. An exemplification appears in a cross-
linguistic survey of future tense categories. The following consequences
of the analysis are discussed: (i) contact-induced grammaticalization is
unidirectional, with rare exceptions; (ii) the conceptual sources used in
grammaticalization are not different from the ones found elsewhere, but
relate to universal cognitive processes; (iii) there are limits to the
kinds of grammatical structures that can be replicated; most of such
processes will be identified as restructuring -- meaning the
reorganization of the system as a result of change -- or spontaneous
replication -- referring to an under- or overgeneralization of the
principles of replication due to insufficient knowledge of the L2.
Finally, there seems to be a correlation among space, time and degree of
grammaticalization, but the identification of the time when it happened
depends on the availability of records.
Chapter 4 inquires upon the impact of the change on the grammatical
structure of the languages concerned, as some grammatical changes may
incur in a typological change in the language profile. The authors
classify the possible structural effects of grammaticalization in six
types, not mutually exclusive among themselves. The following
classification would depend on the category under investigation. (i) Gap
filling refers to the inclusion of a category in the system or of a
meaning to a previous category. (ii) Coexistence describes a situation in
which the new and the old structures encoding a category coexist, either
by double marking or by variation. (iii) Differentiation means that the
new and the old categories coexist but the structure of the old one is
redefined; it is dealt with as a special instance of coexistence. The
example in (1) suggests that Tariana speakers have retained their relative
construction but also added an interrogative pronoun to it. (iv)
Equivalence (or isomorphism) is the situation found when some category of
the replica is restructured to be equivalent to a corresponding category
of the model language, because they are conceived or described as the
same.
Again in Tariana, innovative speakers have given up the locative case
distinctions of the fellow Arawak languages by replicating a generalized
locative case, as happens in the East Tucanoan languages. (v) Category
extension occurs when a new use pattern is assigned to some old category,
when the grammatical categorization structure of the language is not
affected, only the internal structure of categories. (vi) Category
replacement happens when a new category replaces the old one, this being
an ultimate process, usually triggered by category extension. Besides the
change in a specific category, it is also discussed changes that involve
entire domains of grammar by the introduction of a new conceptual domain,
as tense marking in Nilotic languages as a result of contact with their
Bantu neighbors. Another stage of change involves the overall typological
profile of a language, which can affect its semantics, morphology or
syntax (in this order, following the authors). As regards syntax, it is
not immune to replication, as frequently assumed. Apart from the general
grammaticalization parameters (extension, desemanticization,
decategorialization and erosion), the authors present specific evidence
for a morphological cycle leading from free syntactic structures and
lexical forms to clitics and further to affixes. This cycle may not be
carried through all its stages and renewal may not occur.
Chapter 5 investigates the nature of linguistic areas -- classes of
languages that share a number of features as a result of contact. Three
main types are distinguished: (i) 'sprachbund', defined by Heine and
Kuteva on the basis of a set of linguistic features without reference to
the historical forces that gave rise to it (differently from most of the
work dedicated to this subject); (ii) metatypy, an ideal linguistic area
where the languages concerned exhibit a high degree of
intertranslatability, as a result of wholesale restructuring due to
contact; and (iii) grammaticalization area, considered as a group of
geographically contiguous languages that have undergone the same
grammaticalization process as a result of language contact. A
grammaticalization area prototypically consists of at least three
languages, affected by two instances of the same grammaticalization
process, and is considered the basis for the investigation of the other
two types of areas. Therefore, a sprachbund is normally characterized by a
bundle of grammaticalization areas; properties of the major sprachbunds
identified in the literature, as the Balkans, Meso-America and South Asia,
are reviewed by the authors. By its turn, a metatypy is a result of
grammaticalization areas and also of processes leading to fixed
collocations such as proverbial and idiomatic expressions. Finally, a
comparative of African languages and languages from other continents is
shown to argue that Africa constitutes a large grammaticalization area.
Chapter 6 discusses the limits of replication, i.e., additional factors
generally characterizing situations of language contact. The fairly loose
definition of equivalence adopted can raise problems if looked into in
more detail. There are two main ways in which equivalence has been
defined: structural isomorphism (the cross-linguistic compatibility of the
linguist's theoretical construct of categories) and translational
equivalence (what speakers in situations of contact conceive or treat as
equivalent use patterns or categories). While the first is not consistent
with the properties of corresponding categories, especially during the
early stages of replication, the second can be identified by the study of
translational work. The primary goal of speakers is the semantic
equivalence between constructions of both languages. However, sometimes
this is achieved through rather unexpected ways. For instance, the outcome
of an equivalence process can be a morphologically distinct category from
that identified in the model language. The constraints to replication may
involve: (i) the particular structure of the languages involved; (ii)
genetically motivated forces (=drift); (iii) sociolinguistic factors; and
(iv) the length of contact, which is not considered a decisive aspect for
structural change. The authors admit that besides contact, there are other
variables that may delimit the behavior of use patterns, as phonological
similarity, borrowing and the influence of written discourse. A discussion
on the term 'attrition' shows that it is not incompatible with the idea of
contact-induced grammaticalization. Finally, the distinction
between 'natural' (language internal) and 'unnatural' (language external)
change is considered in the light of the aforementioned generalizations.
Chapter 7 summarizes the most relevant findings discussed in the book. It
is stressed, among other points, that sociolinguistic factors are not
decisive for the occurrence of grammatical replication and that a number
of notions proposed in contact linguistics are of limited relevance for
the study.
CRITICAL EVALUATION
Heine and Kuteva's book presents an organized review of data on language
contact, and represents indeed a useful guide for those interested in
language change. The authors are very clear regarding the thesis proposed,
and also explicit clearly the differences and similarities to previous
studies. Maybe the concern in representing the existing terminology and
theories, although positive, would better be concentrated in the
introduction and conclusion of the book for the sake of clarity. An
example can be observed in chapter 6, where some of the subsections
present limits to replication, while others deny the importance of some
limits proposed in the specialized literature.
The functional-typological orientation displayed by the book has some
advantages, as the explicit methodology and the clear analysis, supported
by many examples. This permits the authors to evaluate the range of their
proposed generalizations without the fear of a failed analysis.
Notwithstanding this, sometimes there is a sort of readiness to admit
exceptions, as in the discussion on replication from an aspect marker to a
topic (or focus) marker in Solomons Pijin, having Kwaio as a model
language, in page 108f. It is not explained why this would be a
counterexample to the principle of unidirectionality or not conceptually
plausible. I believe the comment was unnecessary, once it is sound with a
theory of grammar in which topic/focus are the leftmost categories of the
sentence. Specific examples of grammaticalization of aspect/modal
particles to topic categories are provided by Roberts & Roussou (2003).
In fact, assumptions generally taken in functionalist works were not
explained. Although this would be expected -- as most of
the "grammaticalization community" follows this theoretical background --
some of those assumptions have direct import to the issues dealt with in
the book. I would like to mention specifically the gradualist view of
language change, which has been opposed to the catastrophic view based in
the individual (cf., among others, Lightfoot 1999). The focus on continua
restrains a clear rendering of the distinction between functional category
as opposed to use pattern. A corollary of this relates to the definition
of "markedness" of a grammaticalization process as a result of its low
frequency, which seems insatisfatory when it comes to identify universal
principles of grammaticalization. Moreover, the frequency necessary for
one to consider an element a trigger of changes in the grammatical
structure of a language is another unclear point.
Many problems with the adopted framework were foreseen by the authors.
Firstly, when a replica grammaticalization happens, it is not plausible to
conceive that the speakers of the replica language would have access to
the reconstructed form of the model language. The device of recurring to a
universal principle of grammaticalization seems not to be satisfactory,
once there are rare processes of grammaticalization that cannot be
explained without the recognition of the role of a specific language
contact. One of these examples is the Irish English "hot news" perfect,
e.g. "She's after selling the boat" (meaning 'She has just sold the
boat'), in page 94. The second aspect is frequently mentioned in the
functional literature and relates to the view of grammaticalization as
involving primarily a semantic process. It seems that the instances of
case syncretism indeed offer evidence for change in many domains of
language happening in a simultaneous fashion, as it is normally observed
with other phenomena of language change.
Of course, with these observations I do not wish to invalidate this piece
of scholarship. Overall, the aims proposed in the book were fairly well
achieved. The book demonstrates that language contact is a regular
process, in total connection with the parameters of grammaticalization.
This solely is an invaluable contribution to the field of historical
linguistics. Formal researchers on grammaticalization will definitely
profit from this work as well.
REFERENCES
Aikhenwald, Alexandra Y. (2002) Language contact in Amazonia. New York:
Oxford University Press.
Lightfoot, David. (1999) The Development of Language: acquisition, change
and evolution. Malden, MA: Blackwell.
Roberts, Ian & Anna Roussou. (2003) Syntactic change: a minimalist
approach to grammaticalization. Cambridge: Cambridge University Press.
Weinreich, Uriel. (1964) [1953] Languages in contact. London: The Hague;
Paris: Mouton.
|
