Announcement for Workshop 3

"Formalising the dynamics of semantic systems in dialogue" - Staffan Larsson, Gothenburg

I will discuss some ideas towards formalising various aspects of the dynamics and coordination of semantic systems in dialogue, including:

• adopting ontologies and fragments of ontologies to talk about domains
• interleaving ontologies in dialogue, with examples from the map task corpus
• modifying ontologies by adding, removing and modifying concepts and features
• modifying individual concepts ("semantic plasticity")

I will relate these aspects to issues of cognitive representation (including meaning/usage potentials and their role in situated interpretation) and interactional processes and tools for semantic coordination in dialouge, such as feedback, negotiation and accommodation.

Ruth Kempson and Ronnie Cann

"Syntactic Change as Calcifying of Pragmatic Strategies" - Ruth Kempson and Ronnie Cann, Kings College London, and Edinburgh

In this talk, we argue that production pressures constitute a driving force in syntactic change. The case study is the emergence of clitic pronouns in Medieval through to Modern Spanish, whose distribution is, we argue, a calcification of strategies that determined the relatively free constituent ordering of Latin scrambling, as driven by production least-effort considerations (Sperber and Wilson 1995). The point of departure for this analysis is the modelling of dialogue using Dynamic Syntax (Kempson et al 2001, Cann et al 2005, which is a grammar formalism that has the dynamics of a parser, with syntax modelled as the building up of semantic representations from the linear sequence of words. The language of illustration will be Latin. According to this model, both speakers and hearers have to use the same mechanism of building up such structures relative to context, and the result is that parsing and production are tightly coordinated. In particular, we show how production is just as incremental and context-dependent as parsing, and how the major characteristics of dialogue (ellipsis, alignment,and given-new default ordering) can be seen as an immediate consequence of the constraint on speakers to minimise cognitive costs. From this sketch of a dialogue model for Latin, we will go onto show how the development of clitic pronouns and their changing position can be seen as routinisation of actions induced by the least-effort constraint on production, with the shift in constraints through medieval to Renaissance Spanish constituting a shift from one form of routinisation to another,. Finally we will argue that the current diversity of constraints on Romance clitics can be seen as different calcified reflexes of the earlier Latin strategies of tree growth that underlie scrambling. The result is that we have a formal reconstruction of the functionalist grammaticalisation account.

Zoran Macura and Jonathan Ginzburg

"Lexicon Convergence in a Population With and Without Metacommunication" - Zoran Macura and Jonathan Ginzburg, KCL

How does a shared lexicon arise in population of agents with differing lexicons, and how can this shared lexicon be maintained over multiple generations? In order to get some insight into these questions we present an ALife model in which the lexicon dynamics of populations that possess and lack metacommunicative interaction (MCI) capabilities are compared. We suggest that MCI serves as a key component in the maintenance of a linguistic interaction system. We ran a series of experiments on mono-generational and multi-generational populations whose initial state involved agents possessing distinct lexicons. These experiments reveal some clear differences in the lexicon dynamics of populations that acquire words solely by introspection contrasted with populations that learn using MCI or using a mixed strategy of introspection and MCI. Over a single generation the performance between the populations with and without MCI is comparable, in that the lexicon converges and is shared by the whole population. In multi-generational populations lexicon diverges at a faster rate for an introspective population, eventually consisting of one word being associated with every meaning, compared with MCI capable populations in which the lexicon is maintained, where every meaning is associated with a unique word.

Gerhard Jaeger and Anette Rosenbach

"Priming and Language Evolution" - Gerhard Jaeger and Anette Rosenbach University of Bielefeld and University of Duesseldorf

In this paper we connect two strands of research which have so far run in parallel in the linguistic literature, largely ignoring each other in the past, i.e. psycholinguistic research on priming and research on language change. In so doing we hope to shed light on some notorious issues in theories of language change. First and foremost, we hope to provide a psycholinguistic explanation for the empirical observation that (most) grammaticalization processes are not reversible. Based on insights from recent psycholinguistic research we also suggest that priming may provide a key to the understanding of what has been called 'the problem of linking', i.e. the puzzle of how performance preferences may come to be encoded in grammars (i.e. on the competence level) over time. We finally argue that priming is the 'missing link' in evolutionary models of language change in that it provides for a plausible linguistic replicating mechanism. This is a programmatic paper in which we try to demonstrate that combining psycholinguistic considerations with issues of language change opens up for a promising research programme that will ultimately allow us to test present-day speakers for mechanisms that have driven past changes.

Elizabeth Closs Traugott

"Grammaticalization, Emergent Constructions, and the notion of "newness"" - Elizabeth Closs Traugott, traugott@stanford.edu

Work in the last year has suggested ways in which two lines of research, one on grammaticalization, which is largely diachronic in orientation, the other on construction grammar, which is largely synchronic in orientation, can enhance each other. Focus has been on evidence that constructions can be emergent (e.g. Trousdale 2005, 2006, Traugott 2006, Bergs and Diewald Forthcoming). I will continue the discussion with suggestions about how grammaticalization and construction grammar studies, especially those of the type proposed in Croft (2001), complement each other with respect to the questions:

• In what sense do constructions “emerge”?
• How do new subtypes of a construction arise?
• Do grammatical constructions grammaticalize?
• How should “new grammatical structure” be conceptualized, and what implications does the answer have for the role of reanalysis and analogy in grammaticalization?

My example will be the development of some binominal Partitive Constructions into Degree Modifier Constructions with nominal complements, cf. a lot of furniture (as in an auction) > a lot of fun, followed by the extension of nominal complements to other Degree Modifiers, e.g. It’s very fun/pretty cowboy. These are “emergent” in the sense of the “logical outcome” (Dahl 2004) of local changes, but, as in all cases of grammaticalization, they are not “necessarily” emergent: some Partitive Constructions do not develop into Degree Modifiers, or only marginally so, e.g. a deal of, a piece of. My hypothesis is that there may be no totally “new” structures; however, there are new crystallizations of constructions. From this perspective, analogy should play an increased role in work on grammaticalization (see also, from different perspectives, Kiparsky 2005, Fischer forthcoming).

References

• Bergs, Alexander & Gabriele Diewald, eds. Forthcoming. Constructions and Language Change. Berlin/New York: Mouton de Gruyter.
• Croft, William. 2001. Radical Construction Grammar: Syntactic Theory in Typological Perspective. Oxford and New York: Oxford University Press.
• Dahl, Östen. 2004. The Growth and Maintenance of Linguistic Complexity. (Studies in Language Companion Series, 71). Amsterdam/Philadelphia: Benjamins.
• Fischer, Olga. Forthcoming. Approaches to Morphosyntactic Change from a Functional and Formal Perspective. Oxford: Oxford University Press.
• Kiparsky, Paul. 2005. Grammaticalization as optimization. HYPERLINK "http://www.stanford.edu/~kiparsky/" http://www.stanford.edu/~kiparsky/
• Traugott, Elizabeth Closs. 2006. Constructions and language change revisited: Constructional emergence from the perspective of grammaticalization. Paper presented at DELS, Manchester April 6th - 8th.
• Trousdale, Graeme. 2005. Words and constructions in grammaticalization: The end of the English impersonal construction. Paper presented at SHEL 4, Flagstaff, AZ, Sept. 30th – Oct. 1st.
• Trousdale, Graeme. 2006. Constructions and grammaticalization: Evidence from an English light verb. Paper presented at DELS, Manchester April 6th - 8th.

Regine Eckardt

"How does it feel when language is changing?" - Regine Eckardt, Goettingen

In this talk, I will discuss several instances where a certain word or construction in contemporary German and English seems to undergo a change presently. On basis of the assumption that the mechanisms of language change remain constant (uniformity principle), such cases can help to understand the nature of language change. Note that the examples at issue are cases of variation which clearly have not as yet reached the point where sociolinguistic methods could fruitfully be applied. We wittnes the emergence, not the spread, of new structures in language. Specifically, I will discuss German 'irgend' - where attestedly, uses that match English free choice 'any' are on the rise but not yet fully established: Are we waiting for actualization? German why-question pronouns - where a novel indefinite use can be proved to emerge: What are the initiating factors in contexts of use? English unselected embedded 'if' clauses - where a formerly well-defined division of labour between whether and if appears to be blurred to the gain of the shorter 'if'? German modals as main verbs - how many generations might it require for an elliptical use of modals (full verbs elided) to be reanalyzed as full verb + DP argument? The talk mainly intends to contribute to an emergent pool of examples of items under change. More research will be required in order to come to any comprehensive theory of language change.

Devyani Sharma

"Variation in individual grammars: Typological constraints and social embedding" - Devyani Sharma, King’s College London, devyani.sharma@kcl.ac.uk

The study of dialect change at the group level has found that (i) at the micro-level, unconscious speech accommodation plays a key role (Giles & Powesland 1975; Labov 2001; Trudgill 2004) and (ii) at the macro-level, diffusion through the group may proceed both geographically and socio-hierarchically (Trudgill 1983, Kerswill 2002). Despite the fundamental link between the individual and the group (“variation within the speech of a single speaker derives from the variation which exists between speakers”, Bell 1984), the boundary between these descriptions of sociolinguistic processes of change and the formal description of individual grammars remains unclear. This paper aims to relate variation in individual grammars to both social and typological constraints using the framework of Stochastic Optimality Theory (Boersma & Hayes 2001). The analysis implicitly links three of Weinreich, Labov & Herzog’s (WLH, 1968) problems in the study of language change: constraints (e.g. ‘what are possible changes?’), transition (e.g. ‘change proceeds by small steps’), and embedding (e.g. ‘how do changes relate to the social context?’).

The paper (based on a current project with Joan Bresnan and Ashwini Deo) will examine variation in subject-verb agreement and synthetic negation for the verb ‘be’ in the Survey of English Dialects (Orton et al. 1962-71). Extracting partial grammars of individuals, we first confirm leveling patterns across person, number, and negation (Ihalainen 1991; Cheshire, Edwards & Whittle 1993). We then find that individual alternations of forms bear striking structural resemblances to invariant dialect paradigms, and also reflect typologically observed markedness properties (Aissen 1999). In Stochastic OT, variable outputs of individual speakers are expected to be constrained by the same kinds of typological and markedness generalizations found cross-linguistically. In addition, stochastic evaluation reranks constraints by perturbing their ranking values, frequently matching the individual grammar to variable frequencies encountered in the environment. We thus find that individual variation (WLH’s ‘transition’) samples the typological space of possible grammars (WLH’s ‘constraints’), yet also crucially reflects features of grammars it comes into contact with (WLH’s ‘embedding’). The data reflect both regional and social sources of individual variation/change, the latter appearing to be stronger. The analysis relates individual and group dialectal variation to typological variation (Kortmann 1999; Anderwald 2003), and suggests that an individual grammar is sensitively tuned to frequencies in the linguistic environment, leading to the possibility of isolated loci of variability in the grammar rather than complete alternations of competing grammars.

References

• Aissen, Judith. 1999. Markedness and subject choice in optimality theory. Natural Language and Linguistic Theory 17(4): 673-711.
• Anderwald, Liselotte. 2003. Non-standard English and typological principles: The case of negation. In G. Rohdenburg and B. Mondorf (eds.), Determinants of Grammatical Variation in English. Berlin: Mouton de Gruyter, pp. 507–529.
• Bell, Allan. 1984. Language style as audience design. Language in Society, 13: 145–204.
• Boersma, Paul, and Bruce Hayes. 2001. Empirical tests of the Gradual Learning Algorithm. Linguistic Inquiry 32: 45-86.
• Cheshire, Jenny, Viv Edwards, and Pamela Whittle. 1993. Non-standard English and dialect levelling. In J. Milroy and L. Milroy (eds.), Real English: The Grammar of English in the British Isles. London: Longman, pp. 53–96.
• Giles, Howard, and Peter F. Powesland. 1975. Speech styles and social evaluation. London: Academic Press.
• Ihalainen, Ossi. 1991. On grammatical diffusion in Somerset folk speech. In P. Trudgill and J. Chambers (eds.), Dialects of English: Studies in Grammatical Variation. London: Longman, pp. 104–19.
• Kerswill, Paul. 2002. Models of linguistic change and diffusion: new evidence from dialect levelling in British English. Reading Working Papers in Lingusitics 6.
• Kortmann, Bernd. 1999. Typology and Dialectology. In B. Caron (ed.), Proceedings of the 16th International Congress of Linguists, Paris.
• Labov, William 2001. Principles of Linguistic Change. Volume 2: Social factors. Oxford: Blackwell.
• Orton, Harold, et al. 1962-1971. The Survey of English Dialects. Leeds: University of Leeds, E. J. Arnold.
• Trudgill, Peter 1983. On Dialect: Social and Geographical Perspectives. Oxford: Blackwell.
• Trudgill, Peter. 1986. Dialects in contact. Oxford: Blackwell.
• Trudgill, Peter. 2004. New Dialect Formation: The Inevitability of Colonial Englishes. Edinburgh: Edinburgh University Press.
• Weinreich, Uriel, William Labov, and Marvin Herzog. 1968. Empirical foundations for a theory of language change. In W. Lehmann & Y. Malkiel (Eds.), Directions for historical linguistics: A symposium. Austin: University of Texas Press.

Simon Kirby

"The Mechanisms of Adaptive Linguistic Evolution" - Simon Kirby, Edinburgh

HYPERLINK "http://www.ling.ed.ac.uk/%7Esimon" www.ling.ed.ac.uk/~simon

A decade ago the evolutionary biologists Eors Szathmary and John Maynard Smith identified eight major evolutionary transitions that have taken place during the history of life on Earth. They list the emergence of language as the last of these transitions because of its evolutionary significance in giving rise to human culture. Although this is clearly true, I argue that it is not the most significant feature of language from the point of view of evolutionary transitions. Utterances not only convey semantic (and ultimately cultural) information, they also carry information about their own construction. The consequences of this latter type of information transmission (from speaker to language learner) are profound. In this talk, I will argue that by virtue of its mechanism of transmission, language is itself an evolutionary system. Furthermore, a number of structural properties of language are best seen as adaptations, but not in the normal Darwinian sense. Rather than being biological adaptation of our species, they are evolutionary optimisations by language itself to the problem of getting transmitted from generation to generation of language learners. To illustrate how the process of linguistic transmission from speaker to learner can lead to cumulative structural adaptations, I will present the results of an experiment (carried out with Hannah Cornish and Kenny Smith) in which experimental subjects attempt to learn and reproduce as faithfully as possible an artificial "alien" language. What the subjects do not know is that the language they are being trained on is actually the output of the previous subject in the experiment. Despite initially seeding this experiment with a randomly generated unstructured language, we observe cumulative structural changes in the language transmitted from subject to subject over time. Furthermore these changes are adaptive - the language evolves to become easier and easier for subjects to learn. This improvement in learnability is achieved through the emergence of simple structural regularities. ??Although these experiments are at best tiny toy models of cultural evolution, they raise the possibility that linguistic structure in real human language may similarly be the result of cumulative adaptation through linguistic transmission. Suggestions from the audience will be sought for modifications to our experimental paradigm to explore the evolution of other features of linguistic structure.

Pat Healey

"The Role of Repair in the Development of Semantic Co-ordination" - Pat Healey, QMUL, London

Empirical and computational models of dialogue have primarly focused on explaining successful communication. For example, the `grounding model' focuses on explaining how successful referring expressions are progressively adapted during conversation (e.g., Clark, 1996). The priming mechanism used in the Interactive Alignment model focuses on how co-ordination develops when differences in interpretation are not detected (Pickering and Garrod, 2004).

Nonetheless, miscommunication is a ubiquitous phenomenon. Conversation shows a variety of structural and procedural features that are used to diagnose, signal and address communication problems (Sacks et. al., 1974; Schegloff, 1992). Hesitations, restarts and revisions of utterances are pervasive. Although conventionally termed disfluencies they frequently aid comprehension (Brennan and Schober, 2001; Clark and Fox Tree, 2002). Approximately 5\% of conversational turns are explicitly used to request clarification of other utterances. Repair and clarification mechanisms also play a key role in first language acquisition (Chouinard and Clark, 2003; Saxton 1997)

We present evidence from two 'maze' task experiments involving interaction mediated by a chat-tool. This technique enables relatively precise experimental manipulations of dialogue allowing us to interfere with the sequential coherence of the exchange and the apparent origin of turns. The results of these studies indicate that miscommunication or `misalignment' is a key resource for the development of semantic co-ordination. In particular, people must be able to detect and resolve problems with mutual-intelligibility before higher levels of co-ordination can develop. These results have implications for the design of natural language dialogue systems and suggest the need for systems that can dynamically adapt their semantic models in response to misalignment.

Paul Kiparsky

"On (uni)-directionality (provisional title)" - Paul Kiparsky, Stanford

Simon Garrod

"Where do graphical symbols come from?" - Simon Garrod, Glasgow

The origin of symbolic sign systems of the kind discussed by Peirce (1931- 58) is somewhat puzzling. Consider for example graphical signs (e.g., different kinds of writing system). It is easy to understand how iconic and indexical signs arose because there is a transparent relationship between sign and object (c.f., the early Japanese Kana ideographs on the left of the figure). But, how do abstract symbolic signs arise? (c.f., the later ideographs on the right of the figure). One suggestion has been that iconic graphical signs evolve into symbolic graphical signs through repeated usage (Tversky, 1995). I will report a series of interactive graphical communication experiments using a ‘pictionary’ task to establish the conditions under which such an evolution might occur. The first experiment rules out a simple repetition based account in favour of an account that requires feedback and interaction between communicators. The second shows how the degree of interaction affects the evolution of signs according a process of grounding. The final experiment confirms the prediction that those not involved directly in the interaction have trouble interpreting the graphical signs produced in Experiment 1. On the basis of these results I will argue that icons will naturally evolve into symbols as a consequence of the systematic shift in the locus of information from the sign to the users’ memory of the sign’s usage supported by an interactive grounding process.

References

• Peirce, C. S (1931-58) Collected Papers of Charles Sanders Peirce, 8 vols. Edited by Charles Hartshorne, Paul Weiss, and Arthur Burks. Harvard University Press, Cambridge, Massachusetts.
• Tversky, B. (1995). Cognitive origins of graphic conventions. In F. T. Marchese (Ed.). Understanding images. (pp. 29-53). New York: Springer-Verlag.

Miriam Bouzouita

"Clitic placement in the history of Spanish" - Miriam Bouzouita, KCL

In this paper I trace clitic placement through the history of (Mexican) Spanish from a Dynamic Syntax perspective (Kempson et al. 2001, Cann et al. 2005), a grammar formalism that reflects the left-to-right order of parsing through interaction of lexical, computational and pragmatic rules. In this system syntax is seen as strategies for the progressive building up of semantic representations following linear order. Synchronic accounts for the 13th, 16th, 18th and 20th century will be provided in order to model the diachronic changes and outline the progressive shift from a clitic system governed by processing (parsing/production) strategies (where the choice of strategy can be seen as dictated by pragmatic considerations) to a system in which clitic placement is fully determined by the verbal form along which the clitics appear. Furthermore, I will show that processing factors can be not only the source of syntactic variation but also a motive for diachronic change.

References

• Cann, Ronnie, Ruth Kempson and Lutz Marten (2005) The Dynamics of Language. Oxford: Elsevier.
• Kempson, Ruth, Wilfried Meyer-Viol and Dov Gabbay (2001) Dynamic Syntax: the Flow of Language Understanding. Oxford: Blackwell Publishing.