General abstract and Call for Papers

Abstract

The foundational theory of historical linguistics remains the “tree model” (Stammbaumtheorie), championed by Neogrammarians in the late 19th century, which posits a genetic relationship between languages and explains linguistic phenomena in terms of inheritance from or divergence from common ancestors. The isogloss is the stock-in-trade of this theory, where it represents a linguistic phenomenon—originally a word, but it could also be a phonological distinctive feature, an accentual pattern, a grammatical feature, or, less easily, a syntactic structure—that was inherited from a common ancestor and is shared by a subset of its descendants. The isoglosses, therefore, define the divergent branches of the genetic tree of such family of languages. But there is another kind of isogloss, which we can call a “convergent isogloss” in contrast to the “divergent isogloss” of the tree model. The “convergent isogloss” is a linguistic phenomenon that has not been inherited from a common ancestor, but nevertheless appears in several branches of the tree. This kind of isogloss is more closely associated with the “wave model” (Wellentheorie), once promoted as an alternative to the tree model by dialectologists, which is concerned with contact between languages rather than their slow divergence from a common source. The tree model has provided Indo-European studies, in particular, with its primary methodology and basic concepts and problems, but it is nevertheless clear that there has been contact between the Indo-European languages throughout their history and that some isoglosses must be considered the effects of contact (convergence) rather than inheritance (divergence). These “branch-crossing isoglosses” must be included in a general analysis of the development of the Indo-European family of languages. Although scholars have studied several kinds of contact-induced phenomena regarding particular Indo-European languages, there remains no general account of branch-crossing isoglosses in Indo-European studies.

The present workshop therefore aims at discussing this issue in a programmatic and methodological way. How do we recognize branch-crossing isoglosses? What differentiates them from divergent isoglosses? What role do they have in linguistic reconstruction? Do they have any broader methodological significance for historical linguistics? Can we arrive at a general typology, or even theory, of these isoglosses?

The first step will be to define these issues more clearly against a relatively restricted set of phenomena. Hence this workshop will focus on the “second generation” Indo-European languages. If the “first generation” Indo-European languages refer to those that existed between Proto-Indo-European and the various “common” languages that have been reconstructed for each significant subgrouping (Proto-Germanic, Proto-Indo-Aryan, Proto-Balto-Slavic, and so forth), the “second generation” refer to those languages that existed between these various “common” languages and the oldest languages of those subgroupings that are actually attested (Gothic, Vedic Sanskrit, Slavonic, and so on). The “second generation” does not refer to a particular time-period, but rather refers to a position on an Indo-Europe Stammbaum that has itself been constructed by formal analysis of divergent isoglosses: it is thus a stemmatic, rather than chronological “second generation.” (Hence both Old Prussian and Vedic Sanskrit, separated by millennia, both serve as endpoints for their respective “second generations.”) This is where we will look for and discuss convergence phenomena.

In this second generation convergent phenomena take on particular importance: indeed, while the defining features of the first-generation IE languages might be explained within the tree model as inheritances from a common ancestor, many characteristic features of the second-generation languages may have had their origin in contact across different branches and may thus be seen as convergent isoglosses.

The prominence of convergent phenomena may be connected to either the geography of these second-generation languages (which were more settled and localized than the first-generation languages, about which, strictly speaking, we do not possess any geographical data), or the societies in which they were spoken (which may have been multi-ethnic and characterized by a high degree of bilingualism).

Examples of the convergent isoglosses come from all the levels of analysis:
  • phonology (e.g. the increase in number of the fricatives in practically all the second-generation IE languages, or the change from a pitch accent to a stress accent);
  • morphology (e.g. the decrease in length of the inflections, the expansion of nominal derivation);
  • syntax (e.g. the development of the grammatical subject, the change from existential to transitive possessives predicates);
  • typology (e.g. the refinement of the parts of speech systems and the development of the adjective category).

The classical Comparative Method is not very effective in the identification and description of convergent isoglosses. That is because it privileges regular correspondences and disregards apparent similarities. Convergent isoglosses, however, are exactly the opposite: cases of apparent similarity which cannot be explained by recourse to inheritance from a common ancestor.

Another important heuristic question is the problem of distinguishing among different possible origins of the convergent isoglosses:
  1. casual natural parallel development;
  2. independent parallel drift;
  3. same substrate influence;
  4. contact proper.

These, and other, methodological issues have to be discussed during the present workshop.

For further examples of branch-crossing convergent isoglosses refer to Keidan (2013); for some similar considerations see also Alfieri (2011), Bauer (2000), Kulikov (2011).

Themes

Papers may address any issue relating to convergence and branch-crossing isoglosses; the following themes are listed in anticipation of the areas of general discussion.
  • General theoretical discussion of the applicability of the contact-oriented approach to the diachronic analysis of historical languages, including some of the terms and models that have been offered so far: “substrate”, “language drift”, Wellentheorie, and so on.
  • Theoretical discussion and formal definition of the notion of the convergent isogloss. Discussion of second-generation IE languages (including definitions thereof) and the associated “linguistic areas”.
  • Presentation of branch-crossing isoglosses within second-generation IE languages, or proposals for reconsidering divergent isoglosses as convergent isoglosses.
  • Theoretical discussion of the typology, classification and origin of the already known branch-crossing IE isoglosses.

References

  • Alfieri, L. 2011. “A Radical Construction Grammar approach to Vedic adjective”. Rivista degli Studi Orientali 84: 247–261.
  • Bauer, B. 2000. Archaic Syntax in Indo-European. Berlin; New York: Mouton.
  • Keidan, A. 2013. “Branch-crossing Indo-European isoglosses: a call for interest”. Indoevropejskoe âzykoznanie i klassičeskaâ filologiâ 17: 406–416.
  • Kulikov, L. 2011. “The Proto-Indo-European case system and its reflexes in a diachronic typological perspective: evidence for the linguistic prehistory of Eurasia”. Rivista degli Studi Orientali 84: 289–309.