Aarhus University Seal

Temporal aspects of language usage

Grammar, Cognition and Language Use Workshop

Info about event

Time

Wednesday 27 November 2013,  at 09:00 - 16:50

Location

Aarhus University, Nobelparken, building 1485 - room 316

Organizer

Prof. William McGregor, Linguistics, Department of Aesthetics and Communication, Aarhus University,

Workshop Programme:

 

Time

Speaker

Title

08.45-09.00

 

Welcome

09.00-09.45

Freek Van de Velde, University of Leuven

The tortuous trails of long-term language change: cyclic exaptation

09.45-10.25

Peter Bakker

The speed of change in creolization versus in normal intergenerational transmission

10.25-11.00

Coffee

 

11.00-11.45

Kasper Boye, Marie Herget Christensen, Nicoline Munck Vinther & Line Burholt Kristensen, University of Copenhagen

Psycholinguistic evidence for a usage-based theory of grammatical status

11.45-12.25

Mikael Parkvall, Stockholm University

Current Swedish dialect geography and its diachronic implications

12.25-13.30

Lunch

 

13.30-14.10

Bill McGregor

Synchrony in a usage based grammar

14.10-14.50

Jan Rijkhoff

Synchrony is determined by diachrony

14.50-15.30

Coffee

 

15.30-16.10

Jakob Steensig

Time as a hard fact in interaction and some implications for the description of a grammar of talk-in-interaction

16.10-16.50

Ethan Weed & Riccardo Fusaroli

Conversational scales: quantifying the dynamics of language use

 

Abstracts:

 

The tortuous trails of long-term language change: cyclic exaptation

Freek Van de Velde, FWO / University of Leuven

 

Compared to processes of biological evolution, the rate of language change is fast. This is due to the fact that language change operates through cultural evolution, rather than genetic evolution (Steels 2011), a process where replicators can rapidly spread from one host to the next. This lack of a genetic base for rapid transmission could misleadingly suggest that cultural evolution is driven by guided variation. In fact, guided variation is not exploited in full by language users. The reason is that language users can be seen to aim (subconsciously, of course) for ‘local generalizations’ (Van Marle 1990; Joseph 1992, 2004; Enger 2013; Van de Velde & Van der Horst 2013), capitalizing on superficial similarities between seemingly comparable constructions, without caring for the overall design of the grammar. As a result, the motivation behind constructions that are structurally and historically related can get obscured, and language change may occur by unexpected leaps, in a process of refunctionalization that Lass (1990) has called ‘exaptation’ – the opportunistic co-optation of a morpheme to express a new and often unrelated function.

One of the characteristics of exaptation is that it is an idiosyncratic, one-off process (Heine 2003: 172). In this talk, I will argue that the presumed non-recurrence of exaptation does not hold true. As will be shown by looking at several case studies in the history of the Germanic languages, both in the nominal and in the verbal domain, ‘cyclic exaptation’ is rampant, and warrants caution with regard to the overreliance on the universality of predestined pathways in language change. On a more methodological level, the case studies discussed show that due care is to be given to the nitty-gritty details of change, and that blind, large-scale corpus inquiry contains numerous pitfalls.

 

Enger, H.-O. 2013. ‘Morphological theory and grammaticalisation: the role of meaning and local generalisations’. Language Sciences 36: 18-31.

Heine, B. 2003. On degrammaticalization. In: B.J. Blake & K. Burridge (eds.), Historical linguistics 2001. Amsterdam: Benjamins. 163-179.

Joseph, B.D. 1992. ‘Diachronic explanation: putting the speaker back into the picture’. In: G.W. Davis & G.K. Iverson (eds.), Explanations in historical linguistics. Amsterdam: John Benjamins. 123-144.

Joseph, B.D. 2004. ‘Rescuing traditional (historical) linguistics from grammaticalization “theory”’. In: O. Fischer, M. Norde & H. Perridon (eds.), Up and down the cline – the nature of grammaticalization. Amsterdam: John Benjamins. 44-71.

Lass, R. 1990. ‘How to do things with junk: Exaptation in language evolution’. Journal of Linguistics 26: 79-102.

Steels, L. 2011. ‘Modeling the cultural evolution of language’. Physics of Life Review 8: 339-356.

Van de Velde, F. & J.M. van der Horst. 2013. ‘Homoplasy in diachronic grammar’. Language Sciences 36: 66-77.

Van Marle, J. 1990. Over de ongelijksoortigheid van synchronie en diachronie. Amsterdam: Meertens Instituut.

 

The speed of change in creolization versus in normal intergenerational transmission

Peter Bakker, AU

Perhaps close to 100 languages in the Americas, the Caribbean, Africa, India, the South Pacific (New Guinea, Vanuatu, Solomon Islands), the Philippines and Australia have been identified as creole languages.  These draw their lexicon from colonial languages, Arabic, Dutch, English, French, Portuguese and Spanish. In addition, some creoles developed from local languages as a consequence of forced migration of different groups in the same area, e.g. Grand Ronde Chinuk Wawa in Oregon and Naga in northeast India. All these languages draw their lexicon, especially the basic lexicon, from one language (normally the colonial language) but the grammatical system (syntax, morphology, phonology) is only partially similar to the language of the so-called lexifier, and not identical to any other pre-existing language. The combination of an inherited lexicon and a deviant and innovated grammatical system defines them as creole languages.

Earlier research using computational techniques based on large databases, has shown that creole languages not only deviate considerably in their typological make-up from all languages involved in the creation of creoles, but they resemble each other more than any other genetic or areal grouping of languages (Bakker et all 2011, Daval-Markussen et al 2012).

Current debates on the nature of creole languages revolve around two extreme and incompatible positions.  Some creolists on one side (Aboh, Ansaldo, DeGraff, Mufwene and most others) deny that there is nothing unusual with creole structures: the way Latin developed to French is exactly the same in nature, speed, etc., as French developed to Haitian Creole.  They consider themselves uniformitarianists.

The other party (Baker, Bakker, McWhorther, Parkvall and some others) have pointed out the systematic structural differences between creoles and non-creoles: some special process must be responsible for the special typological nature of creoles, which we can identify as creolization after a process of simplification/pidginization. These processes lead both to loss (of e.g. inflectional morphology) and reconstitution (of e.g. aspect marking, articles marking specificity, a specific position of the negator and TAM marking, SVO word order, etc.).

The synchronic typological difference between creoles and non-creoles, as argued for in McWhorter (2001, 2005) and proven in Bakker et al (2011), is the point of departure for the investigation of the next step: the process of change from a lexifier, via a pidgin, to a creole. For the lexicon, earlier results are contradictory, in that glottochronology sometimes indicated a rate of change that would split creoles from the lexifiers up to a 2000 years earlier than the actual split (Hall, Pentland), and other studies (McMahon) indicate a split close to the real point of genesis of creoles. These positions may be reconciled by taking the researchers methods into consideration. As for the rate of change in grammar, we lack as yet a yardstick needed for comparison, but I will point out some areas where processes of loss and development seem to have different rates in creoles compared to non-creoles. I will rely on recent research on stable features of languages.

 

Psycholinguistic evidence for a usage-based theory of grammatical status

Kasper Boye, Marie Herget Christensen, Nicoline Munck Vinther & Line Burholt Kristensen, University of Copenhagen

In this talk, we first outline the basics of a usage-based theory of grammatical status, and distinguish two basic ways in which grammar can develop from usage. Subsequently, we present two recent psycholinguistic experiments which support this theory.

 

Current Swedish dialect geography and its diachronic implications

Mikael Parkvall, Stockholm University

Swedish is unusual among major western European languages in that there exists no dialect atlas for the language. Individual dialect maps can be found dispersed in various publications, but tend – as dialect maps so often do – to depict the situation of more than a century ago. Moreover, they often deal with lexical aspects reflecting a pre-industrial society which is irrelevant to most of today’s speakers.

For that reason, I began making new dialect maps on the basis of blogs and web discussion forums, where at least some of the isoglosses cannot be more than a couple of decades old. What began as a hobby project has now developed into something more serious.

The method has plenty of advantages (and, of course, some disadvantages), which include the possibility of making predictions on developments that have already occurred, as well as – hopefully – future ones.

 

Synchrony in a usage based grammar

William B. McGregor, AU

Usage based grammars have become progressively more popular over the past couple of decades, and increasingly (contra Fritz Newmeyer) grammarians concur that grammar is not just grammar, and usage is not just usage. By and large, though, usage is conceived of (in usage based grammars) as an explanation for grammatical patterns. The causal link is mediated diachronically, and the rigid Saussurean duality of diachrony vs. synchrony is repudiated. Token frequency is generally considered a major synchronic factor supporting this association. In this paper I present arguments that usage is much more central to grammar than this: it is more than a mere explanatory device; it belongs to the guts of grammar, and is in some sense inherent to it. I present evidence for this claim via a case study of optionality of grammatical markers, as discussed in some detail in McGregor 2013 (‘Optionality in grammar and language use’. Linguistics 51 (6): 1147–1204.). It is shown that usage can constitute a semiotic system, indeed a semiotic system that lies within and is a part of grammar; I explore what this means to our conceptualisation of grammar and the division between semantics and pragmatics. I also raise the question of whether there are other domains where there is a grammar of usage; I answer in the affirmative, and discuss some possibilities. More generally, one wonders to what extent there might be isomorphisms in usage at different sized time slices, parallel with the situation for the notion of function in language. I will make some general remarks on this in concluding the paper.

Synchrony is determined by diachrony

Jan Rijkhoff, AU

A wide range of major grammatical phenomena currently observed in many languages are due to an ongoing diachronic process in which speakers, as their languages become increasingly more hierarchically structured, attempt to place together what belongs together (iconicity of distance). The synchronic result of this historical process was already captured in Behaghel’s First Law (1932: 4): Das oberste Gesetz ist dieses, daß das geistig eng Zusammengehörige auch eng zusammengestellt wird (‘The principal law is this: that what belongs together mentally is also placed close together syntactically’).

Three more specific iconic sub-principles can be formulated on the basis of Behaghel’s general observation: (i) the Principle of Domain Integrity, (ii) the Principle of Head Proximity and (iii) the Principle of Scope (e.g. Rijkhoff 1984, 1990, 2004, 2008). These ordering principles are manifestations of three cross-linguistically recurrent historical processes: grouping (e.g. Lehmann 1974), proximation (of head constituents), observing scopal relations in ‘syntagmatic extensions’ (Van de Velde 2009). Domain Integrity accounts for the fact that over time appositional constituents may develop into tight linguistic units (‘integral domains’). Head Proximity explains several syntactic phenomena, incl. certain Greenbergian word order correlations. The Principle of Scope accounts for ordering tendencies among modifiers (and affixes) in the noun phrase and the clause and is independently supported by metonymy. Ultimately these three ordering principles (i.e. formal reflections of cross-linguistically recurrent historical processes) appear to be motivated by cognitive considerations, which are deemed to facilitate language processing (cf. Moravcsik 2012: 243-275).

Time as a hard fact in interaction and some implications for the description of a grammar of talk-in-interaction

Jakob Steensig, AU

In talk-in-interaction, there is no left and right (as in ”left dislocation”). There is before, now and a projection/prediction about ”after”, what interaction linguistics and conversation analysis refer to as ”projectability”. This is one hard fact about time in talk-in-interaction that a description of a grammar for this kind of language use will have to come to terms with. The other is that there is no delete button. Once something is said or done, it is ”out there”, and interactants will have to deal with it and this is what ”repair organization” takes care of.

A grammar for talk-in-interaction is centrally built around projectability and the omni-relevance of repair. In my work with the grammar of Danish talk-in-interaction during the past 20 years, and, specifically, in the construction of the recent webpage samtalegrammatik.dk, which is a platform for a description of the grammar of Danish talk-in-interaction, I have tried to understand what these ”hard facts” mean for the description of this type of grammar. In my talk, I will give examples and discuss implications.

Conversational scales: quantifying the dynamics of language use

Ethan Weed & Riccardo Fusaroli, AU

People are surprisingly adept at adapting to each other, develop new ways of speaking, co-construct routines, even in the short span of a conversation. We present novel ways of quantifying these dynamics and the applications of these methods: identifying good patterns of interactions, better understanding conflict and deception, and even characterizing mental disorders and the social impairment they involve.