Professor Nikolas Gisborne
Details of Professor Nikolas Gisborne's inaugural lecture.
Event details
Lecture title: "What’s Grammar For?"
Date: 29 April 2014
Time: 5.15pm
Venue: Lecture Theatre 175, Old College, South Bridge, Edinburgh, EH8 9YL
Lecture abstract
If you listen to how people talk about grammar in the press you could be forgiven for thinking that it is little more than an aesthetic matter, to do with ‘good’ and ‘bad’ writing. A hundred and fifty years after Henry Alford popularized the spurious ‘rule’ that the English infinitive should not be ‘split’, there are still people who—against all reason—insist, and trenchantly, that it’s wrong to put an adverb between TO and the bare form of the verb it’s associated with. But grammar is so much more than this.
When we study grammar, we’re studying the combinatorial patterns that allow us to build new, rich and complex ideas in someone else’s mind. Somehow, these patterns, and the generalizations that licence them, are acquired, stored and accessed. And yet there are several different theories of grammar to choose from; these might well agree about the data (though not always) but they often involve significant disagreements about the analyses. Typically, the disagreements are embedded in larger assumptions to do with our understanding of how language relates to other aspects of human experience. What is the relationship between language and society? What is the relationship between language and other cognitive systems?
Once we start asking these questions, we are posed a particular question: what is the job of grammar within the overall cognitive and social systems that language is part of? What’s it for? One possible answer is that it serves a function—it provides a structural skeleton that we build interpretations on. But why then is there so much diversity in human languages? Do humans not want to say the same sorts of things to each other, irrespective of the language they speak? We might also wonder whether asking what grammar is for helps us understand which analyses are better, and which are worse. Or are there better ways to test our analyses?