There is a tendency in language history for changes to accumulate in such a way that their combined result is to produce a number of difference syntactic structures, different surface phonotactics, or even a typologically different language.  During this century, a tradition has arisen whereby successive changes which produce a clearly defined effect are explained by teleology, which motivates a change by reference to the effect that it will produce.  In most instances, the effect is produced several hundred years in the future.  Most teleological accounts require us to attribute to a grammar either an unconscious rationality or conscious intent.

    I will present several examples of teleological explanations, some of them well-known examples from English, whose many innovations over the past thousand years (phonological, morphological, and syntactic) have changed it from a highly inflected, synthetic language to an analytic one with little inflection.  A specific example of this is the development of the preterit-present verb class, which in Old English were full verbs, to the separate modal class in Contemporary English, which lacks most morphological and syntactic properties of verbs. Lightfoot (1979, 1981), who has presented the best-known historical analysis of these changes proposes a teleological account, but calls it the Transparency Principle.  Several other principles of change accepted by historical linguists are likewise teleology in disguise: rule conspiracies, rule reordering, push chains, typological universals, and invisible hand phenomena, to name a few.

    I argue that all such theories of change should be abandoned in favor of one which does not attribute to grammars any kind of conscious intent. I will elaborate the abductive/deductive model of language change (Andersen 1973) which accounts for such long-term changes by proposing that an initial reanalysis takes place which creates a new morphological category, new surface phonotactics, or a new syntactic structure.  Deductive changes result from this new status, often playing out over several centuries.  It is deductive change which produces successive changes which give the illusion of a grammar planning for its own future. In reality, it is the initial change which determines the subsequent course of events.

    Phonology is an ancient field of study and also one of intense contemporary activity. Several phonological theories have been proposed in the last 2 decades, and they differ in some fundamental respects. This talk is from the perspective of a consumer who must choose among these theories in the hope of understanding two practical problems--phonological development in children and phonological disorders in children and adults. The central question is: How can theories help?

Ray Kent, Professor of Communicative Disorders and Psychology at the University of Wisconsin-Madison, has published nearly 150 journal articles and book chapters on normal speech production, speech development in children, and various communication disorders.  He also has written or edited 12 books pertaining to communication sciences and disorders.  His current research emphasizes neurogenic speech disorders and speech development in infancy.  A focus common to both of these interests is the assessment of speech intelligibility using perceptual and acoustic methods.  For more information, you can visit the Waisman website at: http://www.waisman.wisc.edu.

    Despite extensive research over the past four decades, a number of major issues concerning the historical and current development of AAVE remain unresolved.  This study seizes upon a unique sociolinguistic situation involving a longstanding, isolated bi-racial community situated in a distinctive dialect region of coastal North Carolina since the early 1700s to address questions of  localized dialect accommodation and ethnolinguistic distinctiveness in earlier African American English.  A comparison of diagnostic phonological and morphosyntactic variables for speakers from four different generations of African Americans and a baseline European-American group shows that there was considerable accommodation of the localized dialect in earlier African-American speech.  Nonetheless, certain dialect features such as copula absence and 3rd person verbal *s marking were distinctively  maintained by African Americans in the face of localized dialect accommodation, suggesting long-term ethnolinguistic distinctiveness.  Contact-based linguistic and sociolinguistic explanation is offered for the selective retention of distinctive patterning in earlier in AAVE.

Modern Icelandic, a language which has developed in relative isolation from the rest of Germanic, is a language known for its unique consonant system, one very different from English in a number of ways.  It shares certain characteristics with Danish, its Scandinavian relative, but also differs from it in a number of important ways.  Most of the peculiarities relate to the stops and fricatives.  Among them are the complete lack of distinctive voice, strong aspiration in all word positions, and so called preaspiration as a distinctive feature of stops.  There are other unique features to the obstruent system as well as the sonorant system.  Due to these features, Icelandic consonantism stands out among the Germanic languages.  Some of the features are archaic, but others are owing to the Icelandic consonant
shift, a large scale innovation with many characteristics shared by the German and Germanic consonant shifts.

In this paper, I explore the position of anterior markers as they occur as pre-verbal, post-verbal independent markers or suffixed morphemes (suffixed to a verb stem) in a variety of Creole languages.  The Creoles under consideration may have European lexifiers (i.e., Haitian, Louisiana, Guinea-Bissau and Capeverdean Creoles) or non-European ones such as Chinook Jargon (cf. Thomason1983).

I argue in particular that the three positions underlined above (that I label Type 1 (pre-verbal), Type 2 (post-verbal independent marker) Type 3(post-verbal suffixed marker) respectively) have syntactic ramifications that lead Creoles to be classified into different typological verbal systems.

On this matter, I address three crucial issues in the realm of anterior markers and typological variation: Given that Creoles display different positions for Anterior markers, and that a few even develop inflection (meaning Tense inflection), what are the syntactic effects of those various positions? Furthermore, can parametrized variation be predicted between the Creoles with inflectional verbal morphology (Capeverdean Creole and Louisiana Creole for instance) and those without?  Finally, how do natural languages such as Creoles develop Tense inflection?

In this paper, I try to show that the position of anterior markers may be symptomatic of a different clausal architecture for the Creoles under investigation.  At the theoretical level, I argue that Creoles with inflectional tense markers may have additional heads and specifiers in their clausal structure accounting for uncommon syntactic constructions (in the realm of Creole languages) such as V-raising, subject-verb inversion and post-Neg subjects.  In summary, I explore a constellation of uncommon syntactic constructions that I correlate to the presence of an inflectional anterior marker.   Following Bobaljik and Thráinsson (1998), I will argue that from the perspective of learnability, the child has to rely on the detectable properties at PF (morpho-phonological properties) to set the parameters of a given language correctly.  As a result, different types of clausal architectures maybe generated.

Furthermore, while syntacticians have been mostly preoccupied with the syntactic effects of the loss of verbal morphology (as with the English language), the reverse focus of this paper is to examine what happens when languages such as Creoles develop Tense inflection (such as Capeverdean and Louisiana Creoles).  On this issue, I compare specifically the post-verbal independent Anterior morpheme ba in Guinea-Bissau Creole (Kihm,1980 &1994) and its suffixed counterpart ba in Capeverdean Creole (Silva, 1985 &1990; Baptista,1997) and show that the position of these markers have important ramifications in terms of verbal behavior and verbal constructions that I link to the different clausal architectures of these and other Creoles.

he decay and ultimate loss of Old English grammatical gender in the transition period between Old English and Middle English is usually attributed to the decay of nominal inflectional endings in conjunction with the rise of an uninflected definite article (i.e. the) and the functional redeployment of the Old English weak and strong demonstratives as deictics (proximal this/these, distal that/those). These were undeniably important factors, but they have to be integrated into the general restructuring of the morphology that took place during this period. This involved the following aspects: gradual generalisation of word-based instead of word and stem-based morphology; dissolution of inflectional classes in favour of a default system; dissociation of case and number as independent categories
marked by separate exponents instead of by joint ones with number becoming the dominant, class-defining category; loss of adjectival inflection (agreement). The loss of grammatical gender thus is a consequence of the general restructuring of the Noun Phrase as a whole, and not just the result of the loss of inflectional endings.

Friday, 29 October. 1999
Gambrell 152

Friday, 5 November, 1999
Gambrell 152

By tracing the history and the development of sign languages, this paper attempts to answer a deceptively simple yet important question:  why, within countries that speak the same language, are there so many distinct sign languages?  Although linguists have examined universal aspects of sign
language, no studies have yet compared sign languages in Spanish-speaking countries.  A detailed examination of the history of deaf education since the 16th century and the predominance of the oral method of teaching from 1880 to the late 1970s explains the geographic isolation of sign languages.
Discouraged and often forbidden, sign languages nevertheless continued to develop within deaf communities resulting in a multiplicity of sign languages, sometimes within the same nation.  While evolving in isolation from one another, these sign languages, and their historical analysis, provide us with an insight into their shared aspects which are key to understanding language itself.

The focus of this talk is the acquisition of Spanish main stress by native speakers of English. Recent studies focusing on L2 acquisition of other languages have emphasized that acquiring stress consists in resetting universal metrical parameters (Pater 1997, Archibald 1993a, 1993b, Mairs 1989, Anani 1989). In this talk I will show that in the case of English speakers learning Spanish the idea of monolithic parameter resetting falls short of the real complexity of the data.
If we put side by side the settings of metrical parameters in English and Spanish it turns out that learners may have nothing to learn.

(1)
 

Of course, as any EFL or Spanish instructor in the US knows, this is not the case. Learners of English or Spanish do make errors and there is a correlation between proficiency level and degree of error.  Where does this improvement come from? A priori, it does not seem to be due to parameter resetting.

In the talk I will consider separately three areas of stress assignment that need to be acquired:

1) The main metrical system. Controlled by the settings of metrical parameters or the ranking of alignment constraints.
2) Stress subsystems. Systematic exceptions probably controlled by other constraints dominating the main system.
3) Irregular words. Items outside the stress system and stored in the lexicon.

For each type of stress I will start by contrasting the L1 and L2 stress systems and will then make predictions about possible sources of problems for learners. Then I will present results from a recent experimental study conducted in collaboration with Barker and Cruz. The findings of this study will be used to assess the appropriateness of current theoretical positions. The goal here is to investigate the role of transfer in the acquisition of stress. First language learners are supposed to start from a default setting that is common to all learners. The factors that can affect L1 development are Universal Grammar, social context and individual differences. What makes L2 special is transfer: In L2 there is an additional factor that can complicate the picture: learners do not start from a common default state. Instead, they seem to transfer aspects of their L1 linguistic system into the new L2. For years, studies of L2 acquisition have been trying to determine what is transferred and under what circumstances. What can be inferred from the study of L2 Stress acquisition of Spanish is the following:
 

References:
Anani, M. 1989. “Incorrect stress placement in the case of Arab learners of English.” IRAL 27: 15-21.
Archibald, J. 1993a. “Metrical Phonology and the Acquisition of L2 Stress.” Confluence. Linguistics, L2 Acquisition and Speech Pathology. Philadelphia: John Benjamins.
Archibald, J. 1993b. Language Learnability and L2 Phonology: The Acquisition of Metrical Parameters. Dordrecht: Kluwer.
Mairs, J. L. 1989. "Stress assignment in interlanguage phonology: an analysis of the stress system of Spanish speakers learning English." in Linguistic Perspectives on SLA. (eds) S. Gass and J. Schachter. Cambridge: Cambridge University Press. 260-283.
Mazurkewich, I. 1984. “The acquisition of the dative alternation by second language learners and linguistic theory.” Language Learning 34:91-109.
Morales, A., L. Barker and G. Cruz. (manuscript). “Stages in the Acquisition of L2 Stress: From English to Spanish”. To appear in the proceedings of the Third Hispanic Linguistics Symposium. Cascadilla Press.
Pater, J. V. 1997. “Metrical parameter missetting in SLA.” ROA.
Phinney, M. 1987. “The pro-drop parameter in second language acquisition.” In Roeper & Williams (eds.) Parameter setting. D. Reidel.
White, Lydia. 1985. “The pro-drop parameter in adult second language acquisition.” Language Learning 35:47-62.

This talk discusses the acquisition of complex English determiner phrases of the type a toy factory.  Why do adult learners of English as a second language produce forms such as the toys factory?  Production data from L1 speakers of Mandarin Chinese, Czech, and German inform the analysis offered: failure to recognize the status of the regular plural s inflection in the English determiner system.  The discussion will then explore the role of L1 properties with respect to this failure - revealing some of the excitement and some of the pitfalls inherent in matching acquisitional problems with up-to-date frameworks from linguistic theory.

It is not uncommon to hear Barbadians speak of their existence in terms of the aphorism, "All O' We Is One," implying a shared culture and identity. In this talk, I explore this notion of oneness by examining the social and linguistic behavior within Fishtown, a small, rural fishing community in Barbados, comprised primarily of poor blacks and poor whites.  I operate on the premise that within the Caribbean, Creole vernaculars are imbued with social expressions of their speakers, giving us insights into individual and community cultures, identities and relationships.
The inter-relatedness of race and class in Fishtown is examined by exploring the extent to which older black and white people manifest similar linguistic behavior.  I draw on slave and bond-servant ancestral relations dating back to the seventeenth century to discuss the social dynamics present today.  Using Le Page and Tabouret-Keller's 'acts of identity', I suggest that subpopulations within Fishtown express themselves in linguistically complex manners that, in the case of poor whites, for example, says, "I'm Barbadian, I'm poor and I'm white," all at the same time, but in very discrete social and linguistic ways.

This presentation argues that the process of creole formation has to be understood within the context of various kinds of contact-induced change, as influenced by the nature of the contact setting, the linguistic inputs, and various constraints on mixture and restructuring. Creoles are adult creations, involving substantial inputs from both superstrate and substrate sources which can be identified, by and large. The linguistic mechanisms and constraints which operate in creole formation also operate in a wide range of contact situations which have produced outcomes as varied as, for example, bilingual mixed languages, "indigenized varieties," and cases of SLA. For instance, phenomena such as substratum transfer (or L1 retention), reanalysis, simplification, leveling, and so on can be found in all these kinds of contact. The task facing creolists is to explain what processes and mechanisms of change are peculiar to creole formation, and how they relate to those which operate in other contact situations. To illustrate the methodology, a brief account of the origins of the tense/aspect system of Sranan is presented. I will demonstrate that the semantics and use of specific categories (Imperfective, Completive/Perfect) can be explained in terms of reanalysis of superstrate forms in terms of substrate semantic categories. This process is reminiscent of the kinds of "relexification" that Lefebvre and others have argued for in the case of Haitian Creole. Some parallels with processes of change in other contact situations will be discussed, if time permits.

This talk addresses the issue of how second language learners resolve areas of conflict between what they know (L1) and what  they are learning (L2). These conflicts arise, for example,  when an L1 grammatical system requires syntactic constraints to dominate discoursal constraints, but the L2 system requires discoursal constraints to dominate.

We present a developmental account of L2 Spanish acquisition of topic-connected null subjects by native English speakers, languages where, unlike English, discourse requirements (topic-drop) override syntactic requirements (Case, EPP).

Subjects were native English speakers in three proficiency levels of Spanish. A discourse completion task involved 40 short dialogues written in the target language: 20 distractors, 20 target items (10 null occurrence of topic-connected subjects, 10 non-null occurrence of non-topic subjects.) Responses were analyzed for non-target-like performance, and error rates for each condition were computed.

Results show statistically significant developmental effects for the acquisition of topic-connected null subjects across the three proficiency groups.  Results also show gradual improvement for all levels on non-topic errors.

These results find a natural explanation i the Optimality learning theoretic constraint demotion mechanism of Tesar and Smolensky (1998). Assuming that L1 ranking serves as the initial-state for L2 acquisition (Schwartz and Sprouse, 1996)¯Parse >> Subject >> Drop-Topic¯learners restructure their interlanguages by successively demoting Subject, and then Parse, below Drop -Topic, as evidenced in their interlanguage error-rates, converging finally to the steady-state target ranking: Drop-Topic >> Parse >> Subject.

For quite some time now, linguists have speculated about the putative relationship between Gullah (a creole language spoken along the Sea Islands of South Carolina and Georgia) and African-American Vernacular English (AAVE). The relationship between these two varieties has been of particular interest to creolists, who have argued that certain distinctive features of the AAVE grammar are representative of creole influence, either through direct descendance from Gullah or through language shift by speakers of a Gullah-like creole. These theories have been put forth in opposition to earlier theories held by dialectologists, who have argued that AAVE derived from early British English sources. At the forefront of this debate have been studies of copula variability in AAVE, which have revealed marked divergences from patterns found in other English dialects while showing parallels with certain features characteristic of creole varieties. This talk will focus on the results of a study comparing the systems of copula variability in Gullah and AAVE and the sociohistorical conditions believed to have resulted in the development of the two varieties. A number of striking parallels are revealed in this study that offer support for the theory of an AAVE-Gullah connection.

Psycholinguistic research in second/foreign language (L2) learning/acquisition has recently made great inroads in becoming an essential component of teacher-education  programs nationwide (e.g., Bardovi-Harlig and Hartford 1997; Leow 1994a, 1994b, 1995a, 1995b). Indeed, the findings from this line of research investigation (e.g., processing instruction) have, to a certain extent, already impacted on language instruction in many L2 classrooms.  Underlying these studies is the premise that some level of attention to, and in some studies awareness of, form or linguistic data is crucial for L2 processing to take place.  To promote a further understanding of the roles of attention and awareness in foreign language learning I will (1) briefly describe current theoretical approaches to the roles of attention and awareness in language learning, accompanied by empirical studies that have tested these approaches, (2) describe and situate recent focus on form studies conducted under an attentional framework, and (3) provide, based on the review, some general suggestions for teachers to consider in their role of language facilitators in the classroom setting.

As a result of the spread of English and of the scattering of South Asians across the globe, the creation of new non-Western cultural identities has emerged and multiple social identities of English have developed.  In particular English has become the new language of gender identity in the circle of South Asian bilingual women writers who have elected to write in English for various reasons.  The pluralistic nature of English is represented in different forms: the universal tongue, the language of rootlessness, the voice of freedom, the language of defiance.  English, therefore, has altered the linguistic behavior of bilingual women writers and their attitudes toward the language.  These multiple social identities of English have certainly left their marks on demythologizing the notion of one English language, one English literature, and one kind of English speaker to advocate a more inclusive model of many Englishes and multicanons of English literatures.

The results of Latin intervocalic -n-, including the reflexes of nouns ending in -ONE, -ANU and -ANA, vary according to the Romance descendent. While the Portuguese and Spanish data (e.g., Sp. pan ~ panes, Ptg. pão ~ pães) have been extensively studied and are well understood (though analyses of the Portuguese data are not without controversy), there are other dialects related to both Spanish and Portuguese that show different results. The main dialect group addressed in this paper is that of Galician, whose many sub-dialects show a variety of sg./pl. alternations of words that derive from -ANU and -ANA.

The data shows a range of forms from nasalized variants identical to those of Portuguese (A: irmão, irmã, arguably the most conservative stage from which the other dialects evolved either directly, B-D, or indirectly, E-H), to denasalized masc. forms, irmao(s) (B-D). The feminine examples, however, are even more interesting: forms range from exhibiting a total loss of nasality (original A: irmã (arguably, < UR /-an/) > B: irmá(s)), to realization of the nasal as a consonant in both singular and plural (C: irman(s), < UR /-an/ + /-s/), to evidencing total loss of nasality only in the plural (D: irmán ~ irmás; likewise for F forms). (These data are treated in Perez 1982.)

Galician thus provides us with a rich and unique testing ground for the hypothesis that all variation is merely minimal reranking of (relevant) constraints, which in this case hold across all Ibero-Romance languages. Other important concepts that are implemented in this OT approach are simplification (e.g., -ns- > -s, which also occurred in Vulgar Latin: MENSA > mesa), a shift away from markedness (e.g., -ão > -ao; -ã > -an) and reanalysis by the hearer-speaker (e.g., GERMANU > [irmãw], [irmaw] > /irmao/).

To conclude, such complex Romance data of this sort have not been treated in OT before, and the analysis given here will show in detail how minimal reranking of a small number of constraints and lexicon optimization accounts for both historical change and present dialectal variation. This paper thus provides a theoretical implementation of the phenomenon that underlies the geographic distribution of singular/plural nasal alternations across the dialect continuum that extends along northwestern Spain, and in so doing, it provides a more internally-consistent approach to the Galician data than Perez' previous account, including a reasonable alternative to his family history/tree of the eight dialects.

                Data (Perez, 209)
                                        Galician dialects:
                                        (reflexes of Latin GERMANU, GERMANA 'brother, sister')
 

References
Girelli, Carl Anthony. 1988. Brazilian Portuguese Syllable Structure. Ph.D. dissertation. University of Connecticut.
Herculano de Carvalho, José Gonçalo C. 1952. Porque se falam dialectos leoneses em terras de Miranda? Revista Portuguesa de Filologia (Coimbra). Vol. 5, 265-280, 508.
Inkelas, Sharon. 1994. The consequences of optimization for underspecification. NELS 24 and Rutgers Optimality Archive [http://ruccs.rutgers.edu/roa.html].
Leite de Vasconcellos, J. 1900. Estudos de Philologia Mirandesa. Lisboa: Imprensa Nacional.
Mattoso Câmara, Joaquim. 1972. The Portuguese Language. Trans. by Anthony J. Naro. Chicago: Chicago UP.
Moraes Ferreira, Albino J. de. 1898 [sic]. Dialecto Mirandez. Lisbon: Libraria da Silva.
Morales-Front, Alfonso and D. Eric Holt. 1997. On the interplay of morphology, prosody and faithfulness in Portuguese pluralization. In Fernando Martínez-Gil and Alfonso Morales-Front, eds., Issues in the Phonology and Morphology of the Major Iberian Languages. Washington, DC: GU Press. 393-437.
Padgett, Jaye. 1994. Stricture and nasal place assimilation. NLLT 12, 465-513.
Padgett, Jaye. 1995. Partial class behavior and nasal place assimilation. ROA-113.
Perez, Jose I. 1982. Observaciones en torno a la desaparición de la -N- intervocálica en gallego. Verba 9, 201-213.
Vásquez Cuesta, Pilar and Maria Albertina Mendes da Luz. 1980. Gramática da língua portuguesa. Lisboa: Edições 70.
Yip, Moira, 1994. Morpheme-level Features: Chaoyang syllable structure and nasalization. ROA-81.