EDITORS: Siemund, Peter; Kintana, Noemi
TITLE: Language Contact and Contact Languages
SERIES: Hamburg Studies on Multilingualism 7
PUBLISHER: John Benjamins
YEAR: 2008
Alessio S. Frenda, Centre for Language and Communication Studies, Trinity
College Dublin
SUMMARY
This book looks at language contact from a broad range of perspectives,
organized in four parts dealing with, respectively, Typology (Part II),
Diachrony (Part III), Acquisition (Part IV) and Translation (Part V). Part I is
taken up by Peter Siemund's Introduction (''Language contact: constraints and
common paths of contact induced language change''), where the reader is reminded
of the importance of continuing to investigate language change on the grounds
that ''robust models'' of contact-induced change ''are still at a premium'' despite
the vast amount of literature already published on the subject, while
''predicting the outcome of a language contact situation remains an immensely
challenging task'' (p.3).
One of the issues discussed in the Introduction is how accessible different
grammatical domains are to the influence of a contact language, and Comrie's
paper (''Inflectional morphology and language contact, with special reference to
mixed languages'') which opens Part II (Typology), is concerned with the
borrowability of a core grammatical domain, inflectional morphology (IM). Comrie
upholds the view that IM is one of the least borrowable features of a language
and argues that within IM itself there exists a borrowability hierarchy, with
agglutinative morphology (AM) more borrowable than fusional morphology (FM). He
also argues that even when FM is borrowed it tends to be reinterpreted as AM in
the borrowing language, based on the analysis of three case studies: (a) cases
of clear-cut borrowing not involving language shift or ''mixing''; (b) cases of
morphological borrowing in ''mixed'' languages; and (c) the intermediate case of
Romani, which is not a mixed language but has undergone a considerable amount of
IM borrowing in the course of its history.
Another central feature of a language, word order preferences, is investigated
in Bernd Heine's ''Contact-induced word order change without word order change''.
Heine's aim is to demonstrate (through the analysis of a good number of
typologically diverse language contact situations) that when a language changes
its word order preferences in a contact situation, it never actually substitutes
a new and hitherto unknown word order for the old one, but merely replaces it by
an already possible, alternative word order which was previously reserved for
pragmatically marked expressions.
Almost as if taking up the gauntlet thrown down by Siemund in the Introduction,
Lars Johanson (''Remodeling grammar: copying, conventionalization,
grammaticalization'') presents a descriptive model for contact-induced processes
of language change, called the Code Copying Framework, and uses it to provide a
systematic and clear discussion of a number of examples already known from the
literature. In particular, Johanson draws a distinction between Global and
Selective Copying, i.e., respectively, between cases in which both the
linguistic material and its content are copied from one code (variety) into
another, and cases in which only certain properties, e.g. the content, are
copied, i.e. realized in the target variety using native structures and
materials. Among the properties that may be copied, Johanson emphasizes
frequency of use, observing how in contact situations speakers may start using a
certain structure of their native variety more frequently until they end up
matching the frequency of use of a corresponding structure in the other variety
(''Frequential copying''). Frequency of use is central to most contributions to
this volume, e.g. as a means by which pragmatic markedness is removed from a
construction (cf. Heine above).
Another point made by Siemund in the Introduction is the importance of the
sociolinguistic conditions in which contact happens, as an important variable
that one should ideally be able to control in order to ensure mutual
comparability between different contact scenarios. Michael Noonan
(''Contact-induced change: the case of the Tamangic languages'') is concerned
precisely with the sociolinguistic factor and its influence in determining the
outcome of contact. He presents the case of a family of languages spoken in
Nepal, the Tamangic group, whose members have been in contact with a number of
non-affiliated languages under different sociolinguistic circumstances, and
focuses on the different outcomes that such different circumstances have
produced in different members of the group.
Part II is concluded by Thomas Stolz's contribution on ''Total reduplication vs.
echo-word formation in language contact situations'', which tries to reconcile
the potential universality of a linguistic feature with its areal diffusion
through contact. Stolz argues that the diffusion of universal structures is not
necessarily only a matter of language-internal development and may be promoted
or inhibited by the increased or decreased frequency of use in areal contact.
This is, I believe, a somewhat controversial hypothesis and I would like to
reserve further comments on it for the Evaluation section of my review.
Opening Part III (Diachrony) is Martin Elsig's paper (''Variability within the
French interrogative system: a diachronic perspective''). Elsig presents a formal
account of variation and change, couched in terms of Chomsky's (1995) Minimalist
program. Using a corpus of literary texts stretching from Old to Modern French,
Elsig investigates changing preferences in the way interrogatives are formed,
especially as regards the use of subject-verb inversion, an expressive strategy
which used to be pervasively employed in earlier varieties and is virtually
absent from modern spoken French. The variety of forms observed is explained as
the coexistence of ''several diastratic, diaphasic and diatopic varieties of
French [...] each having a particular grammatical system'' (p.135), and language
change as the result of reinterpreting the structural aspects of syntactic
inversion. Elsig's paper is the only one featuring a formal approach, and
possibly the least concerned with contact, which is mentioned at the very
beginning of the paper as that occurring between different varieties of French,
and only mentioned again in the final paragraph, where it is stated that ''the
variation in the interrogative system of contemporary varieties of French may be
related to a parametric change which occurred in the late Middle French period
and which did not depend on a language contact scenario'' (p.158), while the
later development of these varieties depended on the interplay of
language-internal factors with ''the influence of prescriptive and normative
pressure'', by which they are nevertheless not being affected despite the fact
that speakers of the vernacular have access to the prescribed Standard
(pp.158f.). Most of the paper is devoted to the formal analysis of the syntactic
properties of subject-verb inversion in Old French interrogative clauses and of
their parametric reinterpretation in the course of history.
How contact may take place not only at the community level at large but as the
result of the work of an intellectual elite is a theme to which this book
dedicates some attention, especially in the Part dedicated to translation.
However, this issue is also dealt with, from a historical point of view, by
Steffen Höder and Ludger Zeevart (''Verb-late word order in Old Swedish
subordinate clauses: loan, _Ausbau_ phenomenon, or both?''), in which the authors
offer an account of how Old Swedish acquired a verb-late pattern in subordinate
clauses. They dismiss the traditional account, whereby the word order in
question arose as a syntactic loan from either Latin or Middle Low German, as
''quite simplistic and therefore unsatisfactory'' (p.164), and having reconsidered
the problem in the light of the available documentary evidence, they conclude by
suggesting that the change in question originated in Old Swedish translations
from Latin originals as an alternative word order available in the language was
grammaticalized as a formal marker of subordination.
Almost mirroring Höder and Zeevart's study of how the complexity of a system may
increase through contact, Conxita Lleó, Susana Cortés and Ariadna Benet
(''Contact-induced phonological changes in the Catalan spoken in Barcelona'')
investigate change in the phonological system of Catalan speakers in two
districts of Barcelona in a cross-generational study aimed at providing an
account of how the relatively complex vowel system of Catalan is undergoing a
simplification process due to contact with the simpler vowel system of Spanish.
They conclude that while internal factors (e.g. the tendency to move towards a
more unmarked system) also play a role, they are not enough to trigger the
observed process in speakers whose environment is mainly Catalan-speaking.
Concluding Part III is Lukas Pietsch's ''Prepositional aspect constructions in
Hiberno-English'', which describes how a linguistic structure which is weakly
represented in terms of frequency can be reinforced in a situation of contact
with another linguistic variety in which a corresponding structure is more
strongly entrenched. Pietsch exemplifies this with the case of English and
Irish, arguing that periphrastic aspectual verbal constructions of the type {be
+ preposition + gerund}, always marginally and transiently attested in other
varieties of English, received a boost in Hiberno-English thanks to the
productivity of the corresponding structure with the verbal noun in Irish.
The section on Acquisition (Part IV) is anticipated in the Introduction by
Siemund's remark that the ''locus of language change'' is the speaker himself or
herself, in whose mind two or more grammars come into contact (p.4). The two
papers in this Part are concerned with two rather distinct subjects: on the one
hand, the acquisition of a second language in a situation of contact; on the
other, the parallels between L2-acquisition processes and language shift. In the
first paper (''Acquisition of Basque in successive bilingualism: data from oral
storytelling'') Margareta Almgren, Leire Beloki, Itziar Idiazabal and Ibon
Manterola compare the acquisition of Basque L2 in successive bilinguals to the
acquisition of Basque L1 by studying two populations of pre-school children in
the Basque Country, finding that L2 learners who acquire Basque in immersion
programs follow the same acquisition patterns as their L1 counterparts as far as
reproduction of narrative structures and verb tense distribution are concerned.
The other paper, Michaela Hilbert's ''Interrogative inversion in non-standard
varieties of English'', addresses a number of issues deriving from the
observation that many local varieties of ''world Englishes'' (or ''New Englishes'')
differ from Standard English in exhibiting, paradoxically at first sight,
main-clause interrogatives without verb-subject inversion and verb-subject
inversion in subordinate interrogatives. The author criticizes the two
alternative explanations that can be found in the literature: the substratal
account, which cannot explain why different varieties of English end up
differing in similar ways from the Standard despite their different substratal
varieties; and the functional-typological hypothesis, which explains this by
appealing to underlying linguistic universals but often fails to consider
language-specific details which are important determinants behind apparently
similar surface configurations. Based on a corpus-based analysis of two New
English varieties (Indian and Singapore English), compared to generalizations
from previous studies on individuals' acquisition of English as L2, Hilbert
concludes that the phenomena under investigation can only be accounted for as
the product of the same rules that govern L2 acquisition (especially the
salience and frequency of ''prefabricated'' formulas), thus undermining the
privileged status of the substratal hypothesis and at the same time suggesting
second language acquisition as a ''domain in which potential 'universal status'
of certain non-standard phenomena of English varieties can be located'' (p.285).
Part V deals with translation as a locus of language contact. Nicole Baumgarten
and Demet Özçetin (''Linguistic variation through language contact in
translation'') discuss the influence of the English models on German translations
in terms of the linguistic variation that is introduced into the target
language. The study is conducted through a corpus-based analysis of the
distribution and frequency of personal deictic pronouns in German originals,
English originals and German translations. The study is presented in the wider
context of the hypothesis that linguistic variation introduced in the target
language by the translation process ''can, over time (through sheer frequency),
marginalize other linguistic means used for the particular communicative
function in the target language'', thus causing a ''change in communicative
preferences'', and the establishment of a new norm (pp.294f.).
A more complex experimental set-up is discussed by the last contribution, Erich
Steiner's ''Empirical studies of translations as a mode of language contact'',
focusing on the explicitness of lexical and grammatical encoding in multilingual
discourse (a term used by the author to refer to the type of written discourse
produced by translation). Steiner reports the findings of an empirical,
multi-corpus and multi-level analysis of German-to-English and English-to-German
translations, as well as of English and German originals, concluding that
bi-directional and in some cases unidirectional influence can be detected at the
discourse level and in relation to different parameters, such as explicitness of
encoding, lexical density and directness, and emphasizing once again the
importance of frequency of use.
EVALUATION
The volume contains a collection of quite diverse contributions, making a global
evaluation of this book quite difficult to formulate. Therefore, in what
follows, I will mostly point out some of the issues I encountered in relation to
individual contributions.
Given the emphasis explicitly placed on the importance of the empirical data by
some contributors (e.g. Heine), I shall first discuss a few problems connected
with the presentation and interpretation of the linguistic material. While
discussing instances in which a grammaticalization process is replicated in a
contact scenario, Heine mentions the case of Breton (pp.51-54), observing that
what was originally a pragmatically marked construction of the type '(it is) X
that Vs' was reinterpreted in Breton as monoclausal and grammaticalized into the
unmarked way of saying 'X Vs' (e.g. _An amzer a zo braw_ {the weather REL is
nice} 'The weather is nice', p.52). This process gave Breton, which had
inherited the VSO unmarked word order from P-Celtic, a SVO word order and
brought it in line with its Romance neighbours. Heine argues that this was
possible in the first place because Breton speakers replicated an identical
grammaticalization process which took place in two local Romance varieties,
local vernacular French and Gascon: cf. Colloquial French _Ton nez qui coule_
{your nose *REL runs} 'Your nose runs' (the asterisk indicates that after the
reanalysis _qui_ was no longer felt to be a relative pronoun and the entire
expression was interpreted as monoclausal). The problem is that the reanalysis
of this construction must have been, in historical terms, as innovative in local
vernacular French and Gascon as it was in Breton, and we do not know whether it
originated in the Celtic or the Romance varieties. Heine cursorily mentions
(fn.9, p.54) the hypothesis that things might have gone the other way round
(i.e. that local vernacular French and Gascon replicated a grammaticalization
process originating in Breton), but he dismisses it with the comment that ''the
evidence available suggests that the Breton structure cannot be traced back to
earlier Celtic'' (ibid.). This statement is somewhat surprising: the structure in
question is common in Gaelic (cf. (1) below) and Welsh (2), where according to
Manning (2004) it had been quite common for some time before being expunged by
puristic efforts to restore the ''native'' construction.
Irish Gaelic
(1) Máire a rinne é
Mary REL did it
'It is Mary who did it'
Welsh
(2) fy mrawd a ddarllenod y bennod echdoe
my brother REL read.PST ART chapter the_day_before_yesterday
'My brother read the chapter the day before yesterday'
(Manning 2004: example (2))
Thus, while it is still reasonable to hypothesize areal convergence in broad
terms (after all, both Welsh and Breton are in contact with SVO languages; but
then, so are Irish and Scottish Gaelic, whose speakers have not abandoned the
VSO order), Heine's hypothesis that Breton speakers copied a grammaticalization
process from the Romance varieties spoken locally does not seem to be supported
by the data presented here. Furthermore, Heine's final comment that ''[t]he
grammatical changes described in this paper concern processes of
grammaticalization that could in principle have happened as well _internally_,
that is, without language contact'' (p.57), is somewhat at odds with the
suggestion of almost intentional mimetic efforts previously made (cf. such
passages as ''In an attempt to replicate the postpositions of Waskia, Takia
speakers developed postpositions'', p.46, or ''a strategy to match the
postpositional structure of the model language'', p.47).
Elsewhere, it seems problematic to reconcile the data with the interpretation
that is given of them. Noonan, for instance, evaluates two different stances on
borrowability against the empirical data he has just presented (p.101):
Moravcsik's (1978) prediction that non-lexical properties of a language cannot
be borrowed unless lexical properties have been borrowed first and Ross's (2001)
hypothesis that they can, lexical borrowing being independent of ''syntactic
borrowing''. Noonan observes at first that the data from this region seem to be
more consistent with Ross's position than with Moravcsik's, only to admit later
on that while there are ''cases of [Tamangic] languages which have borrowed
vocabulary from Nepali without borrowing any syntactic constructions'', we do not
have ''instances of Tamangic languages which have borrowed only syntactic
constructions without borrowing vocabulary'' (p.102). Yet instead of concluding
that his data are fully consistent with Moravcsik's prediction and at variance
with Ross's, Noonan goes on to say that ''[t]he regional Nepali in the Dhaulagiri
Zone has almost no lexical borrowing from Tibeto-Burman languages, but [...] we
find a certain amount of syntactic influence [...] So, regionally, Ross's
hypothesis is confirmed'' (ibid.). This sounds somewhat odd, for ''almost no
lexical borrowing'' undoubtedly means ''at least a few lexical borrowings'', which
again means that Noonan's data confirm Moravcsik's prediction and seem to be at
variance with Ross's.
Some methodological issues are also to be found. In Stolz's contribution, for
instance, there seem to be an issue of ambiguity and one of definition. Having
observed that the evidence in favor of the areal diffusion of the potentially
universal Total Reduplication (TR) as inconclusive and not such as to rule out
independent, language-internal development wherever TR is found, Stolz examines
a variant of TR, which he calls ''total-reduplication-cum-variation'' (TRCV), and
finds that there are sufficient grounds to claim that it can spread through
contact. Having achieved this, he adds that ''[i]f marked variants of TR can be
shown to diffuse via contact then nothing prevents us from assuming the same for
proper TR'' (p.108; note that TRCV is considered a marked variant of ''proper TR''
precisely because it is not as widely and independently attested as TR and
therefore is not a potential universal, pp.128f.). To be sure, what Stolz means
by areal diffusion of TR is slightly ambiguous: towards the beginning of his
paper (e.g. p.112) it is frequency of use which may increase or decrease given
the appropriate contact-induced stimulus, not the possibility of finding TR
itself; towards the end, it is ''the presence of TR in language X'' as the result
of language contact that needs to be proved (e.g. p.128). A linguistic universal
ought by definition to refer to a feature that is attested in so many
genetically unrelated and spatially distant language groups that we can safely
assume it will be attested in any newly-discovered language regardless of its
contact history. Such is the distribution of universal features like vowels (all
known languages have vowels; we don't expect to discover a language without
vowels). Therefore, the fact that a marked variant of a potential universal is
not itself a universal and might have spread through areal contact would not
seem to bear on the discussion of universals. For instance, we could say that
nasal vowels are a marked variant of proper vowels and manage to demonstrate
that they diffuse through contact; but if we then claimed that ''nothing prevents
us from assuming the same for proper vowels'', we would ipso facto imply that
there once existed, and hence might still exist, languages without proper
vowels, hence denying the universality of the claim that all languages have
vowels. In other words, it would appear to me that the distribution of TRCV
cannot tell anything about that of TR, short of suggesting that the latter is
not a universal either.
Another type of methodological issue has to do with the unclear or undeclared
motivations of certain statements. Hilbert's classification of several English
varieties depending on the type of contact scenario in which they have developed
is one example. For instance, it is not explained why varieties such as American
English, Australian English, New Zealand English and South African English are
considered the results of language maintenance, and thus labeled ''L1 varieties'',
while the likes of Irish English, Indian English, Singapore English, Malaysian
English, Nigerian English are considered examples of ''L2 varieties'' deriving
from language shift; why, for instance, would Irish English be more of a L2
variety than American English, given that there undoubtedly was language shift
in the history of both?
Also methodologically problematic seems to be the way in which Steiner lumps
together prepositions, nouns and adjectives as ''nominal word classes'', and
conjunctions, verbs and adverbs as ''verbal classes'', ''based on their preferred
complementation patterns'' and ''rely[ing] on graded notions of
'nominality/verbality' [...] as is usually the case in syntax nowadays'' (p.328
and fn.4). However, it seems to me problematic to set up two so broadly-defined
word classes, especially given that the preponderance of ''verbal words'' or
''nominal words'' in Steiner's various corpora is represented by the combined
incidence rate of all sub-classes (adverbs, for instance, are not distinguished
according to whether they modify a NP or a VP, and all adverbial occurrences
would seem to boost the ''verbal class'' figures). From a statistical point of
view, as the reader is forewarned by Steiner himself, statistical significance
has not yet been assessed for any of the contrasts calculated here (p.327),
which leaves the reader wondering at a few statements which appear to be based
on very small differences (e.g. ''German translations have an increased verbal
quality relative to register parallel originals'', p.332, a statement based on a
difference of 0.18 percentage points). Furthermore, the readability of Steiner's
paper is in many places hampered by such stylistic obscurity as to make an
overall assessment of the paper quite difficult. The indirectness of sentences
like ''avenues of language contact will be modularized by metafunction'' (p.322)
or ''_Explicitness_ [...] is measured through operationalizations of the type we
have indicated in Section 2.1'' (p.335) is remarkable (and there is, by the way,
no ''Section 2.1''); paradoxical definitions such as ''more abstract (and at the
same time, more empirical) types of contrast than have often been at the centre
of theorizing'' (p.318, and again, ''more abstract (and at the same time, more
empirical) properties than have often been at the centre of theorizing'', p.321)
are presented to the reader without explaining in what sense such properties and
such contrasts are said to be at the same time more abstract and more empirical
''than have often been at the centre of theorizing''. Furthermore, in the
description of the complex corpus architecture devised for the study, it is not
clear exactly in what the two corpora of original texts differ from the two
''Reference Corpora'', and likewise left without a definition is the notion of
''translation unit'' introduced on p.324.
On the whole, the volume would have benefited from some more careful
proofreading, which would have avoided the occasional slip of the pen (as when
the English/Breton pair is listed among the examples of language contact to be
discussed in the book, p.6) and some minor oversights (e.g. when the sections of
a paper are referred to as ''chapters'' -- perhaps those of a previous manuscript
or dissertation whose introduction was then ''copied and pasted'').
I would like to conclude by saying that if so much space was devoted to
criticism, it is only because of the diverse composition of this volume, which
required issues of individual contributions to be individually addressed. This
is a book which undoubtedly offers a good number of original and insightful
studies of how contact may promote and drive language change, while generally
treating the effect of contact in promoting and driving language change as a
hypothesis to be verified empirically rather than an assumption to be made a
priori. It contains examples and case studies from a typologically broad variety
of languages and aims to reach a wide audience, both in terms of the issues
covered by its contributors and (generally speaking) in terms of the style of
presentation they adopt, which makes the content accessible to non-specialists.
REFERENCES
Chomsky, N. 1995. _The Minimalist program_. Cambridge, MA: The MIT Press
Manning, H. P. 2004. The geology of railway embankments: Celticity, Liberalism,
the Oxford Welsh reforms, and the word order(s) of Welsh. _Language &
Communication_ 24 (2), 135-163
Moravcsik, E. 1978. Language contact. In: _Universals of human language_, Vol.
1. J. Greenberg, C. Ferguson and E. Moravcsik (eds.). Stanford, CA: University Press
Ross, M. D. 2001. Contact-induced change in Oceanic languages in North-West
Melanesia. In: _Areal diffusion and genetic inheritance: Problems in comparative
linguistics_. A. Y. Aikhenvald and R. M. W. Dixon (eds.). Oxford: University Press
ABOUT THE REVIEWER
Alessio S. Frenda is a research student at the Centre for Language and
Communication Studies of Trinity College, Dublin. He is interested in historical
linguistics and semantics. His current research project concerns the evolution
of grammatical gender in Irish and Welsh from a cognitive and functional point
of view and is supported by the Irish Research Council for the Humanities and
Social Sciences.
|
