The Department of Linguistics, in association with the Departments of Computer Science and Psychology, is planning the following Collaborative Research Centre, (hopefully) to be funded by the German Science Foundation.  If the evaluation goes well, research in this programme would commence in July 2011.

The general theme of the CRC is the temporal dynamics of language – how language occurs in time and how linguistic structures and processes are shaped through their temporality.  Assuming that languages can only be what they could become over time, the particular concern is the relationship between transience and pertinacity.   The vision is to re-orient the field of historical linguistics, in its relation to universals research and typology, through innovatively combining linguistic, neuroscientific, and computational approaches.

The focal question is:  Are kinds of forms and constructions, categories and relations, rules and constraints, structural principles and processes transient or pertinacious, or in other words, unstable in time (changing fast) or stable in time (changing slowly or resisting change altogether, thus appearing static)?  Moving from the ascertaining of temporal scenarios to conditions and causes of developments, the explanatory question is:  Why is something linguistic transient and something else pertinacious?

The structural domains whose transience and pertinacity is scheduled for in-depth differential study are (i) those parts of syntax and semantics which are implicated in the organisation of discourse;  (ii) the structure of morphological systems, comprising both inflection and derivation;  and (iii) prosody (the melody and rhythm of speech) and its interfaces with morphology, syntax, and semantics.

Linguistic structures and processes can be transient/unstable or pertinacious/stable in time taken in any of these four senses:  (i) the time course of the planning and execution and of the processing of speech events;  (ii) the life span available for individuals to acquire and reorganise a lexicon-and-grammar, and also for linguistic abilities to be impaired or undergo attrition with age;  (iii) the history of generations of members of speech communities who have the same or a different lexicon and grammar as the preceding generation whose speech events they have been witnessing or as the members of other speech communities they have come to be in contact with;  (iv) the evolution of man.  Our position as to the relationship between these senses is that in order to understand generational (= diachronic) dynamics it is crucial to understand speech-event dynamics and life-span dynamics, each with a dialectic of pertinacity and transience of its own.  The transition from pre-language to language (“origin of language”) will not be our concern here.

The approach we take is mentalistic:  languages are not things or just behavioural patterns, but mental lexicons and grammars.  The reason for giving centre stage to mental representations is that it is only through representations that an explanatory angle can be gotten on change and the resistance to change.  The temporal relationship between “the same” forms-in-constructions at a time t₁ and a later time t₂ is one of diachronic correspondence, not change in any strict sense.  The locus of linguistic change is the individual mind/brain;  for linguistic change to succeed, several minds/brains (those of all or some members of a speech community) need to be in agreement over doing things differently.

It is sometimes assumed that it is only language contact that is seriously destabilising grammars and lexicons.  But it really needs more systematic research to distinguish transience and pertinacity as due to inherent factors or due to the historical contingencies of substrate, superstrate, or adstrate influences.

A better understanding of pertinacity and transience is not only a major desideratum for historical linguistics, as grounded in the temporality of comprehension and production and of acquisition and attrition.  It also has important implications beyond:  most importantly, it promises to throw light on central questions concerning linguistic diversity and its patterns, insofar as what is pertinacious will, ceteris paribus (such as population-historical contingencies), therefore show less diversity across dialects, languages, families, and areas than what is transient.

The methods to be used are those of historical and comparative linguistics, informed by, as well as informing, contemporary linguistic theory, and crucially enriched by psycholinguistic and neurolinguistic experimentation and by computational modelling and visual analytics.  Hence the three pillars of this cooperative enterprise:  linguistics, cognitive neuroscience, computer science.  Cooperation between these particular disciplines, which crucially defines our approach, is very unusual for research in temporal-historical linguistic issues and aims to re-orient the field of historical linguistics.  The data from different subfields of linguistics (prosody, morphology, syntax, semantcis, pragmatics) will be integrated into a single systematic model of language, which can be computationally visualised and tested by experimental methods.  Conversely, as part of our undertaking, the sciences of the mind/brain and of computational modelling and visual analytics have a unique opportunity to expand into the challenging domain of the temporal dynamics of language and to thus advance the state-of-the-art with respect to these domains of inquiry.