H1: Multilingualism as Cause and Effect of Language Change: Historical Syntax of Romance Languages

Principal Investigators:Prof. Dr. Dr. h.c. Jürgen M. Meisel
Prof. Dr. Esther Rinke

Research Assistants:
Dr. Martin Elsig

Former Research Assistants:
Dr. Gisella Ferraresi

Dr. Marc-Olivier Hinzelin
Prof. Dr. Georg A. Kaiser
Dr. Ioanna Sitaridou
Anne-Kathrin Preißler

Student Assistants:
Mariano Junge

Kathrin Konrad
Xavier Protat

Research framework:

The analyses carried out in the project focus on diverse phenomena of syntactic change within the Romance language family. Taking as a theoretical basis the framework of Universal Grammar, we investigate in particular those changes which affect core areas of syntax and which might indicate changes of parametric properties. Our analyses are conducted against the backdrop of the latest insights gained from research in language acquisition. The primary research question relates to the role of language contact and of information structure as possible causes of such changes. By incorporating into our analyses the most recent empirical procedures which are applied in variationist sociolinguistics, we aim at making a significant contribution to a more general theory of language change.

Research in language acquisition has provided ample evidence that children acquiring more than one first language have no difficulties in assigning from early on the target values to the respective parameters of each of the languages involved (cf. Meisel 1989, 2000, 2001, 2007a, 2007b). Multilingualism in the child can therefore not be regarded as a sufficient condition causing parametrical change. Multilingualism in the linguistic input with which the child is confronted might, however, indeed play a crucial role in this regard. This could be the case, for instance, when speakers providing the linguistic input to the child, e.g. her parents, are second language learners of the language undergoing the change. For lack of L1 (first language) knowledge, they have not acquired the language-specific parametrical settings and will in all likelihood fail to produce target-like utterances as consistently as first language speakers do. It is reasonable to assume that such a situation may cause the child to diverge from the target in her patterns of first language acquisition. Rather than in situations which are characterized by a more or less stable coexistence of the multiple languages involved, this scenario is most likely to be found in cases of language contact exhibiting a considerable degree of social instability, e.g. when large-scale migratory movements take place.

Aside from language contact, an important influence on syntactic change is also exerted by different readings associated with different word order patterns in terms of information structure. In our analyses, information structure has not only turned out to account for the coexistence of syntactic variants within one language, whatever synchronic snapshot on the diachronic axis one considers, but it has also qualified as an active participant in influencing and even causing syntactic change, hence dispensing with the need to invoke parametric change or variation in each and every of the relevant cases (e.g. Gabriel & Rinke to appear, Kupisch & Rinke 2007, Rinke & Meisel 2007).

Research topics

As defined by the framework and the research questions outlined above, our analyses focus on specific diachronic phenomena of word order variation in Romance languages.

In the research literature, Old French has quite generally been attributed the label of a verb second language comparable to Germanic languages (cf. Adams 1987): it features a considerable number of postverbal subjects and null subjects in declarative matrix clauses whose first position is occupied by a non-subject constituent, e.g. an adverb or adverbial phrase. Although it has been made explicit that in addition to these apparent verb second effects, Old French also exhibits a non-negligible amount of verb first and verb third clauses which need to be considered as infractions of a verb second rule and which hence seriously challenge this analysis (Kaiser 2002), researchers continue adopting the verb second account for Old French (e.g. Labelle 2007, Mathieu 2006, 2007). Analyses conducted within our project strongly suggest that verb second effects in Old French are indeed only apparent, thereby confirming Kaiser (2002), in not being grammatically (or parametrically) conditioned (cf. Ferraresi & Goldbach 2002). The postverbal localization of the subject in declarative matrix clauses rather turns out to be driven by readings associated with information structure, the postverbal subject receiving focus reading and the preverbal one (as well as the null subject) topic reading (cf. Rinke 2005, 2006, Rinke & Meisel 2007). In this regard Old French rather parallels other Romance null-subject languages than Germanic verb second languages. These insights are now supplemented with research into the role of language contact between Old French and Old and Middle High German (Elsig 2008a). This is necessary in order to aim at a holistic account as to which language internal and external factors condition the occurrence of postverbal subjects and as to what structural and information structural interpretation the postverbal subject may receive in the different languages.
Our research has especially focused on phenomena of syntactic variation in the Romance languages. In their research on clitic doubling in Spanish, Gabriel & Rinke (to appear) have argued that synchronic variation of clitic doubling may be explained as a result of its diachronic evolution. Based on the observation that even though the lexical object generally receives a focus interpretation in clauses exhibiting clitic doubling and a topic interpretation in clauses exhibiting object dislocation, object doubling turns out to be most likely with constituents featuring a high degree of topicality. We have argued that this fact may be explained in diachronic terms under the assumption that doubling has evolved out of right dislocation. Therefore, the first elements to be doubled systematically are those with the highest degree of topicality.
Another case of syntactic variability is found in the Portuguese noun phrase: Whereas the definite article occurs almost obligatorily with the possessive pronoun in Modern Portuguese, it is very often omitted in Old Portuguese. Kupisch & Rinke (2007) have argued that this diachronic variation does not mirror a parametric change, but rather needs to be considered as the reflex of a general expansion of the article into possessive contexts and its subsequent grammaticalization towards the spell-out of the definiteness feature in D°.
The French interrogative system is characterized by a multitude of variants involving different word order patterns. Variants with the canonical subject-verb word order coexist with variants featuring subject-verb inversion. The diversity is not less striking when it comes to the placement of the wh-word or to the involvement of different interrogative markers and particles. In collaboration with Prof. Dr. Shana Poplack and the Sociolinguistics Laboratory at the University of Ottawa, a large-scale empirical assessment of data stemming from actual language use covering a time frame from the late Middle French time period until contemporary French which have been analyzed according to the methodological guidelines of the variationist approach, provides evidence that Pronominal inversion, the quantitatively most productive type of subject-verb inversion, has indeed become restricted, both to certain high frequency lexical contexts and to a limited class of linguistic contexts (i.e. second person singular and plural in Québec French). Complex inversion is virtually absent from language use, perhaps due to its invoking a high degree of formality. Its close relative, the use of the postverbal interrogative particle -tu, however, is highly productive in Modern Québec French. Free inversion has disappeared from the system, as has Simple inversion. The only contemporary remnant of subject-DP inversion, Stylistic inversion, is restricted to predicative clauses featuring a copula verb and, marginally, to unaccusative and passive contexts. We interpret these findings in terms of a structural change involving loss of verb movement from T° to C° after the passage from Middle to Classical French as well as loss of wh-movement to SpecCP with the emergence of early Modern French (Elsig 2008b, Elsig to appear).

Bibliography

Adams, M. (1987). From Old French to the theory of pro-drop, in: Natural Language and Linguistic Theory 5, 1-32.
Elsig, M. (2008a). Verb Second Effects in Old French: Evidence for a Verb Second Grammar? Ms. Universität Hamburg.
Elsig, M. (2008b). Variability within the French interrogative system: A diachronic perspective. In: P. Siemund & N. Kintana (eds.) Language Contact and Contact Languages. (Hamburg Studies on Multilingualism 7), Amsterdam: John Benjamins.
Elsig, M. (to appear). Diachronic Aspects of Syntactic Variation in the Interrogative System of Québec French. (Studies in Language Variation, ed. by P. Auer, F. Hinskens & P. Kerswill), Amsterdam: John Benjamins.
Ferraresi, G. & M. Goldbach (2002). V2 Syntax and Topicalisation in Old French, in: Linguistische Berichte 189, 3-25.
Gabriel, C. & E. Rinke (to appear). Information packaging and the rise of clitic doubling in the history of Spanish, in: G. Ferraresi & R. Lühr (eds.) The Role of Information Structure in Language Change. Berlin: Mouton de Gruyter.
Kaiser, G. (2002). Verbstellung und Verbstellungswandel in den romanischen Sprachen. Tübingen: Niemeyer.
Kupisch, T. & E. Rinke (2007). Italienische und portugiesische Possessivpronomina im diachronischen Vergleich: Determinanten oder Adjektive? Universität Hamburg: Arbeiten zur Mehrsprachigkeit N° 78.
Labelle, M. (2007). Clausal architecture in early Old French, in: Lingua 117 (1), 289-316.
Mathieu, E. (2006). Stylistic fronting in Old French, in: Probus 18, 219-266.
Mathieu, E. (2007). À propos des propriétés germaniques de l’ancien français, in: Cahiers linguistiques d'Ottawa / Ottawa Papers in Linguistics (CLO/OPL) 35, 107-136.
Meisel, J. M. (1989). Early differentiation of languages in bilingual children, in: K. Hyltenstam & L. K. Obler (eds.) Bilingualism Across the Life Span. Aspects of Acquisition, Maturity and Loss. Cambridge: Cambridge University Press, 13-40.
Meisel, J. M. (2000). The Simultaneous Acquisition of two First Languages. Early Differentiation and Subsequent Development of Grammars. Universität Hamburg: Arbeiten zur Mehrsprachigkeit No 7.
Meisel, J. M. (2001). The Simultaneous Acquisition of two First Languages: Early Differentiation and Subsequent Development of Grammars, in: J. Cenoez & F. Genesee (eds.) Trends in Bilingual Acquisition. Amsterdam: John Benjamins, 11-41.
Meisel, J. M. (2007a). On autonomous syntactic development in multiple first language acquisition, in: H. Caunt-Nulton, S. Kulatilake & I.–H. Woo (eds.) Proceedings of the 31st Boston University Conference on Language Development. Somerville, MA: Cascadilla Press, 26-45.
Meisel, J. M. (2007b). The weaker language in early child bilingualism: acquiring a first language as a second language?, in: Applied Psycholinguistics 28 (3), 495-514.
Rinke, E. (2005). Subjekt-Verb-Inversion im Frühromanischen: Altportugiesisch und Altfranzösisch im Vergleich, Ms. Universität Hamburg.
Rinke, E. (2006). Der Verlust der vP-internen Subjektposition im Französischen, Ms. Universität Hamburg.
Rinke, E. & J. M. Meisel (2007). On the Syntax of Subject-Verb Inversion and the Role of Information Structure in the History of French, Ms. Universität Hamburg.