LINGUIST List 8.1265

Sun Sep 7 1997

Review: Spanish in Contact, Review #1

Editor for this issue: Andrew Carnie <carnielinguistlist.org>

Directory

1. Teresa Satterfield, Review of Spanish in Contact

[****Editor's note: This is the first of two reviews of this volume. The 
second review will be posted in the next few weeks.-AC]




Spanish in Contact: Issues in Bilingualism. Ana Roca and John B. Jensen,
 editors. Somerville, MA: Cascadilla Press, 1996. Pp v + 226.
 (Paperback)

 Reviewed by Teresa Satterfield, <tsatterumich.edu>,
		 University of Michigan


 This volume grew out of papers presented at the XII Symposium of
Spanish and Portuguese Bilingualism at University of Massachusetts
(Amherst) in 1991. Since that time, the contributions have been revised
and updated for the current publication. (On that note, it is somewhat
ironic that the uninspiring title "Spanish in Contact" does not come close
to reflecting the rather novel premise of the book.) Fortunately, the
goal to advance a more "visionary" perspective on bilingual studies does
come through in the editor's introductory comments as well as in the
inclusion of the large body of (sociolinguistic) work which moves beyond
 the commonplace theme of Spanish-English contact. Unlike a recent
collection of works examining Romance-language bilingualism and diglossia
in Europe (Posner and Green, eds. (1993), ROCA and JENSEN maintain even
tighter thematic cohesion by choosing to limit coverage exclusively to
instances of contact between Spanish-plus-one or more different world
languages, thereby representing an ambitious array of non-European
languages and communities in conjunction with Spanish. Some of the
languages surveyed such as Guarani, and Mexicano are quite on the
periphery of "usual" linguistic inquiry. 
 While the essays visit the issues of languages in contact under many
labels (historical linguistics, dialectology, anthropology, etc.), readers
with sociolinguistic interests will be especially pleased that most of the
case studies in the book place great significance on speaker attitudes and
values, resulting in a clearer understanding of the strong link forged
between language choice and social identification/group membership.
 As a starting point in 'Introduction' (v-xi), the editors provide a
succinct overview of Spanish in its role as a language in contact. The 
background perspective on the study of general bilingualism is equally
compact and serves more readily as a forum for raising the concerns of
bilingual studies. Here R & J lay out blanket questions such as, "What
happens when two or more languages come into contact?" However, they also
make a brief effort to expose the inherent complexities of bilingualism,
acknowledging the validity of a well-established research agenda which has
promoted treatments in this area which are multidisciplinary in scope. 
Nevertheless, R & J clearly advocate a newer and stronger reliance on
linguistics for real answers to the "contact conundrum." Specialists in
bilingual studies will find themselves pondering the points raised and
assessing the development of these new orientations in the field. 
Nonspecialists who are not familiar with these issues may want to first
consult general introductions such as Haugen (1953), Appel and Muysken
(1987), Klee and Ramos-Garcia (1991) in order to fully appreciate the
range of foci discussed in the volume. The introductory section concludes
with a useful outline of the major themes covered by each contributor. The
remaining chapters are organized into three parts, not necessarily by
subject matter, but rather according to the geographical instances of
language contact. 
 Part I presents three bilingual/ multilingual case studies carried
out in Spain, opening with the paper 'Helebiduntasuna Euskadan' (1-12) in
which ROBERT HAMMOND points out some of the typological phenomena of the
Basque lexicon and language, including major phonological processes and
syntactic structures. The data are later compared and contrasted with
monolingual Spanish patterns, as is now a prominent use of Basque in a
range of monolingual comparative studies. From this data, Hammond argues
that certain non-lexical universals may underlie and constrain the
similarities found in Basque and Spanish, rather than interaction stemming
from the two languages' 1200 years of contact. It would have been 
interesting if Hammond had made some references to Hualde et al. (1993) or
other recent work in Basque linguistics, especially since the current
approach offers no firm evidence on aspects such as borrowing, but
instead is left to conclude that more research is required in the Basque
language contexts. JASONE CENOZ rigorously analyzes the role of existing
Basque-Spanish bilingualism in the acquisition of English in 'Learning a
Third Language' (13-27). Cenoz correlates bilingualism with positive
effects on English language achievement, as compared to monolinguals
receiving the same exposure to English. The prediction is based on the
Interdependence Hypothesis, which basically states that given adequate
exposure and motivation, proficiency in one language can be successfully
carried over to proficiency in another language. The empirical results
constitute an unstable link in Cenoz's otherwise solid evidence for the
predicted outcome: motivation, and notbilingualism, tended to be the best
overall predictor of achievement in English acquisition, although
Basque-Spanish bilinguals did fare better when all variables (motivation,
intelligence, exposure, etc.) were equal for monolinguals and bilinguals.
The final paper in this section explores the linguistic behavior of 
adolescent bilingual speakers of Catalan and Spanish in Barcelona. HOPE
DOYLE (29-43) meticulously documents the language attitudes of these
youths, a subgroup of which are claimed to have led to a "revitalization"
of Catalan. To the extent that it has broken through the status of
historically "minority" language, the author asserts that Catalan is
competing with Spanish in all language contexts. Doyle's work is
ambitious, given the seminal effort to analyze attitudes towards language
choice in a systematic manner. For all the attention to detail, the
empirical results falling out from this study remain cloudy, giving the
impression that the importance and value of both languages still hangs
greatly in the balance among the teenage population. In any case, the
fieldwork methodology is excellent and the work makes for an informative
view of young adult subjects as products of a bilingual policy which
integrates formal instruction both of Catalan and of Spanish into the
educational curriculum. 
 Section II contains four papers which focus on various contact
situations in Latin America. MARGARITA HIDALGO, in 'A Profile of
Language Issues in Contemporary Mexico' (45-72), provides an
interdisciplinary account of the post-colonial language situation of
Mexico. Sociolinguistic and dialectological concerns form the core of this
diachronic "profile." While it would not be accurate to say that this work
is data-driven per se, there is certainly no lack of chronological lists
and tables included for reference. Hidalgo traces a linguistic history,
covering 'typology' as it relates both to geographical and linguistic
diversity in Mexico. H then addresses the social issues of minority
language maintenance/ bilingual communities in the face of educational
policies and mass migration to urban areas. A detailed presentation of
dialectal variation among Spanish speakers concludes the paper, with the
explicit point that a continued progression toward a single, official
"Standard Mexican Spanish," variety is all but guaranteed. In an attempt
to shed further light on questions of linguistic interference, CAROL
KLEE'S study of '...The Influence of Quechua on Spanish Language Structure'
(73-91), comes in reaction to Thomason and Kaufman's (1988)
sociolinguistic construct of language permeability. Klee examines two
typologically distinct contact languages, illustrating the effects of
Quechua on clitic-pronoun morphology, word order, and past tense aspectual
features of Andean Spanish. Klee's small samples must be interpreted
tentatively, but findings suggest that a primary determinant in the outcome
of language contact can indeed be attributed to the structure of those
languages, as claimed by Silva-Corvalan (1990). Content aside, most
problematic in the Klee paper is form. First, the alignment of tables
(e.g. Table 8, p. 82) makes for difficult processing of information; 
secondly, knowledge of standard written Spanish is essential, as there are
no English glosses to represent the considerable number of speech examples;
 and lastly, there are some (trivial) typo-s in English, which reflect the
"interference" of Spanish orthography. YOLANDA RUSSINOVICH SOLE reshapes
the discussion of Paraguay as "paradigm of national bilingualism." 
Re-evaluating assumptions rooted in Rubin's (1968) now- standard reference
work on Paraguay's stable condition of bilingualism, 'Language, Affect,
and Nationalism in Paraguay' (93-111) probes the true nature underlying
Guarani-Spanish dynamics among 'Asuncenos (inhabitants of Asuncion)'. In
terms of empirical and conceptual breadth, this inquiry is certainly one of
the most competent of the entire volume. R-Sole brings to the fore an
arsenal of over fifty years of census records, concrete historical
information, and personal interviews with 652 "bilingual" informants,
interpreting the accumulated data according to a multi-dimensional schema. 
(The meaning of 'dimension' is not discussed, but the author applies the
term to her use of interrelated continuums of differentials-- much like
syntactic parameters of variation-- although the former contain social,
cognitive, or situational elements.) To this end, R-S soundly
substantiates a three-dimensional model whose axes consist of age, rural
vs. urban demographics, and linguistic proficiency. In culmination, the
study supplies neat counter-examples to claims that 90% of the population
speaks Guarani (G.Corvalan (1988)), painting instead a linguistic
portrayal where less than half of Paraguay, past and present, can be
considered to be functionally bilingual. An examination of the relatively
strict diglossic patterns manifested in the two languages, plus the future
implications for bilingualism round out this fascinating contribution. 
'... Stable Bilingualism Possible in an Immigrational Setting?' (113-122)
is a self-proclaimed "first," taking the linguistic experiences of 
Argentines of British descent as a "new" data source within bilingual
research circles. By standard accounts (Fishman 1966, among others), the
sociolinguistic conditions fostered in "Anglotinian" enclaves of Buenos
Aires are predicted to not only favor maintenance of the minority language,
English, but to foment it. FLORENCIA CORTES-CONDE applies this
generalization to questionnaire responses elicited from a "representative
sample (p.120)" of present-day British-Argentinians, noting, in fact, the
emergence of quite a different pattern. The author's findings point to
the steady decline of English as the language of preference among
Anglo-Argentinians; although the subsequent explanation of the trend as a
reflex of "peer group identity" requires much more development. In the
final analysis, this study does take a large step in the direction of
theory refinement, showing that on both micro- and macro-levels, the
process of language shift can be much more intricate than previously
implied.

 Seven investigations constitute the final portion of the volume. The
editors state that the works are grouped together not only due to their
relevance to issues of Spanish in the United States, but also because the
internal linguistic-cognitive (in a word, theoretical) aspects of
bilingualism are given more attention in these final papers. As it turns
out, one study in this section has nothing whatsoever to do with Spanish in
the United States, and several other works are at best case studies, rather
than theoretically-grounded contributions bilingual questions. 
Fortunately, 'English Calques in Chicano Spanish' (123-130) fits the
editors' description. While it is definitely not intended as an
introductory reading, no doubt those familiar with the lexical borrowing
literature will find it an interesting piece of data. Co-authors SMEAD
and CLEGG lead with a discussion on borrowing terminologies, invoking
Haugen's (1972) S/I model of lexical innovation and Otheguy and Garcia's
(1988) subsequent 3-dimensional revision. Against this backdrop, S &C
apply a rather opaque working example, which unfortunately for this reader,
did little to explain what the model in question actually defined. The O&G
(1988) lexicon classifying operation is argued for on the grounds that it
allows for the least ambiguous interpretation of lexical borrowing forms; 
unfortunately, the claim is of small consequence, since the authors fail to
discuss rival taxonomies. The implementation of this model on a corpus of
Chicano Anglicisms results in the successful extrapolation of a subgroup of
lexicon items identified as general calques. A second taxonomy credited to
O&G (1989) is then applied, providing a more formalized categorization,
specifying the exact nature of each calque. While these taxonomic models
are demonstrated to be largely effective for the specific corpus selected,
it would be good to see the authors seriously question the O&G models'
inner workings, as well as test to what degree significant generalizations
can be made about other calquing environments. To their credit, S &C
recognize certain overall limitations : a) the assignment of calque-status
to particular lexical items may always prove elusive, since the bilingual
speaker's semantic perception of the underlying meaning in English or
Spanish cannot always be ascertained; and b) findings are preliminary at
best on the question of regional diffusion of this class of lexical
borrowing. In sum, the premises are intriguing, yet this ambitious study
seems fragmentary at best: this being due, it is suspected, to the attempt
to cover so much ground in such a short space. The next three articles
involve aspects of phonology. In 'Difference in Voice Onset Time in Early
and Later Spanish-English Bilinguals' (131-141), MEHMET YAVAS analyzes the
VOT production of two groups of adult bilinguals with English as a second
language, comparing those who acquired the L2 at ages 5-6 with speakers who
began L2 acquisition in later childhood (11-12 years of age). Like several
L2 studies, Y's findings support the notion of a "critical period," such
that, when all other factors are held constant, the potential to attain
native pronunciation was found to gradually diminish with age. In this
case, adults who were earlier child bilinguals demonstrated identical VOT
production to prototypical monolinguals in English, while the later
bilingual patterning was shown to be within that of native speakers, yet
not an exact match. This study is solid as a descriptive work, but shows
little theoretical insight. For instance, Y is not clear in the account of
 individual variation effects of VOT. If Y's observations are accurate,
then age-of-acquisition factors cannot explain the variation among the
informants in terms of "foreign accents" in the L2, since even a small
sample of those speakers within native VOT limits produced noticeable
"non-native"accents. This problem is not a major shortcoming in the work. 
What causes more reservation is the fact that the study has obviously been
influenced by L2 acquisition research. In this, one sees the difficult
task of establishing what constitutes a bilingual speaker. Insofar as Y
"labels" these participants as (child) bilinguals, the issue which might
have been addressed more fully is the interaction of VOT with the one
grammar versus two grammar bilingual debate. Given the empirical findings,
can VOT shed any light on how the bilingual phonological system may differ
from that of the monolingual? Are younger bilinguals predicted to be
equivalent to two monolinguals in terms of VOT production? Of substantial
interest, then, would be the assessment of the subjects' VOT values for
both languages; since in order to make more informed predictions
concerning the definition of grammar boundaries for the bilingual speaker,
neither language can be viewed in isolation. As the above questions
demonstrate, Y's inquiry potentially offers an excellent analysis of
bilingual phonological acquisition within a formal theoretical framework. 
Further studies resulting from this project are eagerly awaited. The issues
of age and phonological acquisition re-appear in 'Bilinguals in Little
Havana' (143-150). This work grows out from a 1982 investigation examining
the patterns of Spanish phonological variation found in the English of
Cuban teens in Little Havana. The earlier study showed age of English
exposure to be the significant factor in the degree of Spanish influence
detected. While author MARGUERITE MACDONALD looks to a new generation of
Cuban subjects in 1988 to bolster the previous data, it is nearly
impossible to replicate the 1982 study and make subsequent comparisons, as
is her primary objective. First of all, as M herself notes, Little Havana
is a much changed community, and the new subjects, immigrants of the Mariel
Boat Lift, encounter a different sociolinguistic climate from those of the
original experiment. This said, M later discounts the claim by arguing
(unpersuasively) that phonology is less influenced by social factors than
are other forms of the grammar (syntax or morphology). Secondly,
replication is unworkable since the two groups of informants are not
identical in terms of age or initial socioeconomic background. Unhappily
for the reader, M chooses to overlook the meaning of true scientific
replication, in pursuit of the current analysis. Speakers arriving to the
U.S. at 6-10 years of age (mean age 7.8) in the original study are thus
compared to a grouping of Mariel speakers arriving at 8-10 years (mean age
9.4). In a flurry of statistics, M determines that both informant groups
exhibit the same patterns of variation, based on age classifications. 
Since replication has been flawed from the outset, one wonders if there is
any validity in this conclusion As with YAVAS (this volume), a problem
with M's findings is also that the high levels of individual variation
among subjects cannot be related to age alone. Upon careful observation
one notes that in the younger subjects, M takes into account not only age
of arrival, but "distance from the social mainstream" as well. For older
arrivals, only age is considered relevant. In any case, it is difficult to
control the amount of exposure that an individual child receives to the L2,
unless the exposure time is held constant for every child in a given study
for a given period of years. Moreover, in the current work there is no
explanation offered as to why in the Mariel (8-10 years) group, the average
variation produced for some sounds should show a lower degree of Spanish
influence when compared to the same variation produced from the 1982 group
which contains slightly younger subjects (6-10 years). Simple comparison
aside, M implements an impressive array of statistic devices to advance her
claim. Unfortunately the MANOVA results are unclear, and those tables listed do
not appear to relate to the tables referred to in the discussion. In the
final analysis, the evidence presented cannot be said to inconclusively
support the hypothesis. What M brings to bear here is a rather
statistically sophisticated, thought- provoking project: but it is not a
replication, and with this failure, it does little to provide a convincing
explanatory view of phonological variation among bilinguals. As stated
previously, of equal interest to bilingual phonology studies would be the
degree of English phonological variation found in the subjects' Spanish
speech, especially given the "native" English patterns of the "mainstream"
bilinguals. If we are to conduct accurate studies of bilingualism, it
behooves us to examine the phonological inventories, as the case may be, of
both languages, instead of analyzing the grammars in isolation. 'Spanish
in Contact with Itself...' (151-157) makes virtually no mention of Spanish
language contact in the U.S, (at this writing, Mexico City and the
Caribbean do not qualify), however JORGE GUITART'S treatment of dialects 
of the same language in contact situations is certainly timely, given the
recent expansion on variational themes in the field. To begin, G makes use
of a disparate contingent of theoretical devices (all of which have
appeared before in the literature) in order to accommodate this previously
"inexplicable" phenomenon of bilectalism. The concept of interlanguage,
for instance, has been invoked in relatively recent L2 studies and in
various theoretical approaches. It is also the pivotal focus in this
phonological investigation, in the case of the speaker who functions with
two dialects, but exhibits "imperfections" in one lect or the other. Also
figuring prominently in the theoretical stockpile is the categorization of
rhyme consonants in terms of "radicalness" versus "conservativeness." 
These characterizations are claimed to manifest themselves in Low
(non-standard) dialect or High (standard) dialect, respectively. The
advantage of this classification is shown through the prediction of certain
processes that may affect postnuclear consonants in a given language
(dialect) style. In this light, such a device may be useful in explaining
the diglossic nature of dialects. The author utilizes yet another
theoretical tool to formalize phonological variation within language
dialects. By adapting Labov's "switching" principle, G extends the idea of
variation as linguistic behavior to encompass the differences between
radical and conservative lects. To this end, G manages to weave these
diverse theoretical parts into a surprisingly coherent whole, providing a
clear picture of those elements which may, in principle, guide phonological
variations in Spanish dialects. More than anything, this paper displays
theory design in action, as evident from the minute quantity of linguistic
evidence or cited references. As with any theory in progress, one may
expect to uncover weaknesses. First, the author depends primarily on
anecdotal commentary from Spain to support his claims. G's familiarity
with Andalusian lects allows for a concise representation of the
distinctive dialects to the extent that he is able to discern degrees of
variation between radical, conservative, and "interlect" patterns based on
the individual phonemic inventories posited for each dialect. On the other
hand, G seems at a loss in his examination of Caribbean Spanish. The
difficulty may stem from a failure to flesh out one small area within the
Caribbean, since throughout the entire region, there are countless dialects
of spoken Spanish. Nevertheless, G claims that Spanish of the Caribbean
does not exhibit clear diglossic tendencies, due to the close similarity
between the phonemic inventories of "High" and "Low" speech. In effect,
these speakers are said to utilize radical forms or interlect forms, but
rarely conservative forms. Notwithstanding the dearth of empirical support
for this type of speculation or the tacit assumption that unlike Peninsular
Spanish, Caribbean lects fit easily under one simplistic umbrella, G's
indirect pronouncement that the education level of the Caribbean speaker
determines the lect in play is uninformed (and forgetful). Not more than
four pages earlier, he rejects the idea of schooling as a facilitator of
conservative lects and interlects, referring to Andalusian and Castilian
Spanish as prime examples. Given such absence of scholarly rigor, one
seriously questions the validity of the research in general. All things
considered, it does have its value. G places himself firmly in the camp of
those attempting to give bilingual analyses more principled footing. His
treatment of the bi-dialectal perspective certainly brings into the mix a
range of unique and provocative notions that should be of interest to
monolinguals and bilinguals alike. In generative syntax, a fundamental
paradigm has long been established concerning grammars such as Spanish,
which admit the occurrence of phonetically unrealized elements ('pro') as
subjects of a tensed clause, and those grammars which generally do no allow
'pro,' such as English. A central issue of the "pro-drop" discussion in
'Patterns of Pronominal Evolution in Cuban-American Bilinguals' (159-186)
is the interaction of these quasi-contrasting pronominal behaviors within
bilingual speech patterns. Focusing on Spanish usage, JOHN LIPSKI draws
from non-syntactic (discourse) domains as well as syntactic (including LF)
domains to pursue this line of research via a multi-phase study. First, a
large-scale data collection effort results in a corpus of monolingual and
bilingual Spanish representations of null and overt subject pronominals. 
Second, a pilot study is undertaken th obtain grammaticality judgements on
sentences extracted from the phase-one corpus. Wisely taking into account
the degree of both Spanish and English available to the individual
participant, the study examines three groups of Spanish-speakers,:
monolingual, "balanced" bilinguals, and English-dominant bilinguals
(speakers who retain only superficial fluency in Spanish). This two-part
categorization of bilinguals also allows L to capture certain hitherto
unexplained features of real world bilingual speech. The idea that
unbalanced bilinguals are the most informative sources of data seems
definitely to be on the right track, allowing a plausible alternative to
the currently established, yet highly idealized, model of the so-called
balanced bilingual. Based on the findings, L tentatively concludes that
Spanish-English bilinguals predictably differ from monolingual speakers of
Spanish in that the bilinguals (particularly, the Spanish-recessive group)
do not maintain the same rigid distinctions in their use of Spanish lexical
and 'empty' pronouns. The bilinguals demonstrate a tendency to de-focus
overt pronominals, on a par with English speech. However, it does not
appear to be the case that these informants comply fully with English
patterns, since there is still a native-like preference for Spanish
pronominal behavior. L suggests that bilinguals in his study have acquired
a relatively stable dual language- system with respect to the null subject
parameter. Thus, while the theoretical possibility exists, no doubt as
outlined in the Old French pattern of null subject loss (Adams 1987), it
seems unlikely that the Cuban-American bilinguals as a whole are following
an attrition pattern as a result of contact with a non-null subject grammar
such as English. The hypotheses advanced by L are largely solid and
original. Given the conceptual appeal of Principles and Parameters Theory,
it is not surprising, then, that L situates his bilingual study within this
framework; although the irony is that it is here where his approach
requires further improvement. One has the impression that central
questions of the theory such as parameter setting (or re-setting) actually
play a minor role in L's own study. This impression is magnified, since L
offers not so much as a glimpse into the current developments and issues on
parameter theory in general, nor on null subjects in particular. Given
this most salient of linguistic topics, a more contemporary presentation of
works (for example, Hyams 1994, Rohrbacher and Roeper 1995, among others)
would be beneficial as background information available before the 1996
publication date of this collection. Furthermore, there is even an
independent study available (Satterfield 1995) which treats the issue of
the bilingual parameter-setting, as will be sketched in the following
section:

 S draws largely from data concerning English- Spanish-speaking
children in order to advance a UG-based alternative to language acquisition
which builds on Clark's (1990a and b) computational model for parameter
setting. The proposed model is aptly called the Extended Parameterization
Hypothesis (EPH), as it has the distinction of accommodating both
monolingual and bilingual L1 learning tasks within one unified Principles
and Parameters (PPT) framework. Taking the behavior of the Null Subject
Parameter across languages as illustration, S begins by raising important
questions about the adequacy of standard parameter setting models in the
face of compelling evidence from both bilingual and monolingual child
speech facts. It is demonstrated that "universal" premises which guide the
well-known parametric approaches greatly complicate any attempt to
construct an economical bilingual analysis. Moreover, given a case such as
the English-Spanish learner, previous accounts render simultaneous
acquisition of contrasting null subject settings theoretically implausible.
To complement this argument, S also shows that recent developments in
linguistic theory (i.e., the Minimalist Program), as well as studies on
language learnability (Clark 1992, Clark and Roberts 1993) similarly
motivate the view that, while initially convincing, the standard parameter
models are potentially more costly and less effective in terms of
monolinguals as well. With support for a reformulation of the parameter
setting process thus established, the second part of the paper formalizes a
system of language learning which takes as its point of departure Clark's
application of the genetic algorithm. Briefly stated, the EPH is able to
emulate parameter setting by evaluating various forms of input and
cataloging the data. The model is constructed so as to make use of changing
conditions in the "environment," allowing for the differentiation of input
text in the bilingual system with progressively greater accuracy, while
maintaining the stable population(s) of the target grammar(s). Successive
generations of hypotheses represent the most fit parametric values relative
to the ambient language(s), subsequently driving the child toward the final
state grammar(s). When consistently supplied with input from two
languages, the EPH demonstrates that two hypotheses can survive, thus
providing the dual parameter settings that the bilingual learner may
require . As claimed, the EPH produces interesting results: (a) it further
demonstrates the conceptual force of the PPT framework; (b) it dispenses
with many standard assumptions, thus avoiding the bottleneck of high
deductive costs; and relevantly, (c) it represents a significant step
toward explaining the factors which constrain the acquisition of syntactic
knowledge without inhibiting the child's ability to acquire multiple
languages simultaneously.

Despites its oversights, L succeeds in providing a theoretically relevant
and thoroughly documented account of bilingual null subject patterns. 
Given the pilot study, appendix, and the ample (though not up-to-date)
bibliography, bilingual studies scholars pursuing their own "pro- drop"
research will find an arsenal of theoretical and field method information. 
In 'Spanish- English Code-Switching: Conditions on Movement' (187-201),
FRANCESCO D'INTRONO presents a detailed overview and extension of the
influential GB syntactic analyses by Woolford (1983, and subsequent works)
concerning Spanish-English code-switching phenomena. D argues that his
proposed model not only covers Woolford's examples of grammatical instances
of code-mixing, it additionally explains those ungrammatical instances of
mixing. D builds his analysis through the use of invariant principles
which are supplied by the (biological) UG endowment and "fine-tuned" for
each language according to syntactic parameters of variation. The idea is
that modules such as X-bar Theory, for instance, can be used to account for
word order without requiring more complex machinery such as phrase
structure rules. With said framework in place, D offers a unifying
alternative that reaches to explain instances of ungrammatical
code-switching overlooked in previous bilingual literature as differences
in the configurations of Spanish and English syntax, produced, for example,
by differing cases of Move-alpha. Unfortunately, several of the accounts of
movement-induced ungrammaticality are stipulative at best: D offers no firm
footing for his claims that the Spanish infinitive raises to COMP position;
 or that all subordinate clauses in Spanish are dominated by an NP-node and
thus must be Case-marked, whereas English subordinate clauses lack this
structure. One more interesting movement question involves a WH-phrase of
grammar A moves to [Spec, CP], or cluase initial position. The problem
arises if the WH-trace is not governed by a lexical element also belonging
to grammar A. The main portion of the discussion depends on the premise
that constraints on extraction in monolingual utterances and bilingual
code-switched sentences are regulated by the Empty Category Principle (ECP)
and other formal principles of UG. D incorporates the "same lexicon
condition" into the definition of proper government within the ECP, with
the result that the ECP still states that traces of moved phrases must be
properly governed; however, proper government is held to entail lexicon
co-indexation, plus government (lexical or antecedent) which occurs either
at S-Structure, or at LF, in the case of multipleWH- questions and WH- in
situ phrases. Given the intricate GB architecture, D does a good job of
showing that select syntactic distinctions and broad principles such as the
ECP may provide an adequate basis for explaining the unacceptability of
different code-switching samples of movement. This said, there are
criticisms to be leveled at this study, these include: a) D's failure to
explicitly cite DiSciullo et al.'s (1986) GB-based analysis which so
obviously influenced his subsequent analysis; b) The assumption that the
audience is conversant in GB Theory; although to his credit, D skillfully
guides the reader through the complexities of his analysis; c) The Spanish
data is not given English glosses; and finally, d)More care should have
been taken with the type of judgements considered to be code-switching
data, since some of the examples provided are extremely murky. In effect,
one must wonder if examples such as (44) and (45) (p. 198) are appropriate
if seen as mixed constructions:

 (44) *Arrived Juan last night.
 (45) Juan arrived last night. 

D implies that these are instances of code-switching which are accounted
for by the proper government conditions. The problem, of course, is that
proper names are often shared or borrowed between both lexicons, so that
the above sentences as purely English constructions would obtain the same
grammaticality judgements as D lists for the "code-mixed" version. 
In spite of some questionable theoretical stipulations and a few empirical
inadequacies in this analysis, D manages to provide a deeper analysis of
code-switching phenomena. While current reformulations in generative
syntactic theory have obviated the need for government and, hence, the ECP,
in another era it would be of interest to take D's line of explanation a
bit farther. Given these configurational foundations and ECP
modifications, an analysis that would constrain the behavior of
codeswitched adjuncts would be most welcome. The final article of the
collection is well-organized and provocative. TORIBIO AND RUBIN (203-226)
offer a pithy overview and critique on the syntactic constraints proposed
in previous studies with an end to systematically accounting for
intrasentential code-switching patterns. Citing Belazi et al. (1994), they
present a nearly convincing argument that these well-known analyses fail,
on theoretical and empirical grounds, to explain the bilingual phenomena
for which they were intended. Among other things, T&R demonstrate that
there is good reason to believe that certain extralinguistic components
must be taken into consideration in any serious study of code- mixing. 
Thus, the degree of bilingual proficiency for a given speaker is a relevant
factor for understanding code-switching behaviors, as are the (bilingual)
community to which the speaker belongs and the speaker's attitudes toward
language mixing. As for other aspects of investigation, T&R note that
techniques in data collection and data selection are often technically
flawed in that they lack consistent methods for analysis of crucial
components such as linguistic competence or syntactic restrictions. To the
extent that the authors effectively show the inadequacies of "competing"
analyses, they advance a Minimalist-based approach to account for
syntactic switching constraints. Drawing on reasonable conclusions
outlined in DiSciullo et al. (1986) T&R invoke essentially the same
condition, given here as the Functional Head Constraint (henceforth FHC).
The restriction is held to be independently motivated by virtue of the
unique relation placed on head and complement pairings. The FHC blocks a
language switch from occurring between a functional head f and its
complement c in the following manner: On analogy with other abstract
features (e.g., Case, agreement) which are drawn from the lexicon and
subsequently checked/matched per Chomsky (1993), there is also a language
feature L which must be matched at some phase of derivation. UG is said to
require fand c to carry the same L-feature in order for the derivation to
converge. If f and c differ with respect to L, as in code-switching, then
due to the FHC, the derivation "crashes" and the construction is considered
unacceptable. One aspect of T&R's treatment of the FHC that is problematic
is their failure to justify their assumptions concerning the status of
code-mixing. A concrete example of this criticism is reflected in (1):

 (1) El hijo de Juan me dio [NP este book]
 The son of Juan me-dat. he gave this book
 "Juan's son gave me this book."

Here, the final DP/NP consists of two codes, but there are various
interpretations of this fact. According to Woolford (1984) and D'Introno
(1996 this volume), the NP represents an instance of acceptable
"categorial" code-switching, whereas under the FHC approach, it is
ungrammatical as a code-mixed item, falling instead under the heading of
acceptable 'lexical borrowing." Luckily for T&R, the main thrust of this
paper concerns the role that the operation of language feature matching
plays in identifying bilingual code-switching proficiency. In short, they
introduce a general hypothesis that is relevant for L1, L2, and L1 or L2
bilingual speech: as a speaker develops greater linguistic competence,
s/he becomes more "sensitized" to the UG constraints that apply in language
mixing. This postulation is actually quite elegant, as it provides a
flexible explanation to the wide scope of variations found in language
switching constructions. For instance, the young bilingual child gradually
establishes the L-features for each language so that his code-mixing
becomes increasingly driven by the proper syntactic constraints with no
"errors.' The authors attribute ill-formed utterances (those which violate
the FHC) to less-balanced bilinguals or to monolinguals who use loanwords
from a foreign grammar. T&R suggest that unacceptable code-switching forms
emerging in L2 speech spring from the L2 learner's inability to fully
access UG principles such as the FHC after a critical period of time. 
Good news for adult L2 learners may possibly be derived from the findings
of T&R's study which indicates that the amount of "sensitivity" to UG
constraints is correlated to linguistic proficiency, with more advanced L2
learners showing greater awareness of acceptable code-mixing, even after
the so-called "critical period" for language learning. T&R deserve kudos
not only for the care with which they substantiate their own claims, but
also for the rigor with which they scrutinize the possible alternatives,
building on effective elements of previous works. As this paper represents
one of the first serious code-mixing studies to come out of the Minimalist
framework, it will be interesting to see the impact that 'Code-Switching in
Generative Grammar' has on future investigations.

While the essays contributed vary in level of linguistic analysis and even
quality, the collection does include some substantive discussions, which
may serve to more rigorously frame conceptual and theoretical notions of
bilingualism which have been conspicuously lacking in the larger context of
linguistic theory. On this front, there is a wealth of material here which
can promote and motivate future theoretical analyses of bilingual grammar. 
This volume can also serve as an excellent resource manual given its
in-depth empirical and descriptive research which is useful to the
mushrooming studies of a sociolinguistic nature. These points make it a
welcome addition to the study of Spanish bilingualism in particular and of 
languages in contact overall. 


REFERENCES
Adams, M. 1987. From Old French to the theory of pro-drop. Natural
Language and Linguistic Theory. 5:1-32.

Appel, R. and P. Muysken. 1987. Language Contact and Bilingualism. 
London: Edward Arnold.

Belazi, H. et al. 1994. Code-switching and X-bar theory: the functional
head constraint. Linguistic Inquiry. 25: 221-237. 

Chomsky, N. 1993. A minimalist program for linguistic theory. In Hale,
K. and S.J. Keyser, eds.: The view from building 20. Cambridge, MA: MIT 
Press.

Clark, R. 1990a and b. Papers on learnability and natural selection.
Technical Reports in Formal and Computational Linguistics. No. 1. 
Universite de Geneve.

Clark, R. and I. Roberts. 1993. A Computational Model of Language 
Learnability and Language Change. Linguistic Inquiry. 2: 299-345.

Corvalan, G. 1988. Current status of Guarani in Paraguay. In Paulston,
C., ed.: International handbook of bilingualism and bilingual
education. New York: Greenwood Press. 

DiSciullo, A. et al. 1986. Government and code-mixing. Journal of
Linguistics. 22:1-24.

Fishman, J. 1966. Language Loyalty in the U.S. The Hague: Mouton.

Haugen, E. 1953. The Norwegian Language in America. (2 vols.) 
Philadelphia: University of Pennsylvania Press. 

Haugen, E. 1972. The analysis of linguistic borrowing. In Firchow, E. et
al., eds: Studies by Einar Haugen. The Hague: Mouton. 

Hualde, J. and J. Ortiz de Urbina, eds. 1993. Generative studies in
Basque linguistics. Amsterdam: John Benjamins.

Hyams, N. 1994. Nondiscreteness and variation in child language. In Y.
Levi, ed: Other children, other languages. Hillsdale, NJ: Lawrence 
Erlbaum Associates.

Klee, C. and L. Ramos-Garcia, eds. 1991. Sociolinguistics of the 
Spanish-Speaking World: Iberia, Latin America, United States. 
Tempe, AZ: Bilingual Review Press. 


Posner, R. and J. Green, eds. 1993. Trends in Romance linguistics and 
philology, vol. 5: Bilingualism and linguistic conflict in Romance. 
Berlin: Mouton de Gruyter.

Roeper, T. and B. Rohrbacher. 1995. Null subjects in early child English
and the theory of the economy of projection. Penn Working Papers in 
Linguistics. 2: 83-119.

 Rubin, J. 1968. Bilingualism in Paraguay. Anthropological Linguistics. 
 4/4:52-8.

Satterfield, T. 1995. The Bilingual Selection of Syntactic Knowledge: The
 extended parameterization hypothesis. PhD dissertation. 
University of Iowa.

Woolford, E. 1983. Bilingual code-switching and syntactic theory. 
Linguistic Inquiry. 14/3:520-536.