A few months ago, I was asked by a TV programme to make up a language for their monsters to speak, and with that, my linguistics life completed a cycle. When I was about 11 or so, I grew fascinated with language, mainly from reading Ursula Le Guin’s A Wizard of Earthsea, a book I still completely love. Le Guin envisaged a world where the words actually created the reality, and every single piece of existence had its own particular name. Fascinated by this idea, and already developing my inner language geek, I started making up languages to explore whether they could work like that. To do this, I had to learn how real languages actually worked. At school, they just taught French and German (and later some Latin), but my local library (sadly closed this year because of government cuts) was full of teach-yourself books on weird and wonderful languages, as well as some pretty impenetrable linguistics books. I think the librarians were a bit perplexed by a twelve year old taking home tomes on philology and grammar he couldn’t possibly understand. They were right, I didn’t understand them at all, but I was so hooked by that point, that I read them anyway, and I guess some stuff sunk in. I remember winning a competition for local schools at St Andrews University, when I was about 16, and buying, with my £20 prize, second hand copies of Chomsky’s Syntactic Structures and Aspects of the Theory of Syntax (which I still have). Again, I wasn’t really able to understand these books in much depth, but the idea that you could use rigorous, mathematical, means to try to get under the skin of language was, and still is, just endlessly fascinating to me.
I went to Edinburgh University to do Linguistics with Artificial Intelligence, a course that, in classical UK university style, didn’t really exist (they were borrowing the final year of the course from a Master’s programme). There Mark Steedman took us, in ten short weeks, on a whirlwind tour of syntax: from Syntactic Structures, through the context-free debates, to the Dutch Infinitival Complement construction, Combinatory Categorial Grammar and Winograd’s Eliza program. It was, intellectually, one of the most exhilarating experiences of my life, and I remember a couple of cases where the security guards had to come throw us out because we were all still pestering Mark with questions an hour after the lecture was meant to have finished.
From then I was really a budding categorial grammarian, and, given that it was Edinburgh, and unification-based frameworks were de rigeur, my undergrad theses (we had to do two) were a Unification Categorial Grammar analysis of purpose clauses and an analysis of fluid-ergativity in Eastern Pomo that suggested unification graphs should be allowed to be cyclic—both of these now lost in the fire that destroyed the old Artificial Intelligence department library in Edinburgh. A natural progression, as I moved to doing a Masters in Cognitive Science, was to HPSG (I was really a bit overly dismissive of the GB research program, thinking it relied too much on filters, and not really realizing that HPSG was really a giant set of filters!) At the same time, though, the power of unification based approaches began to make me feel very unsatisfied as to how good a job they were doing of explaining why languages seemed to be cut from very similar cloths.
Just about then, Chomsky’s Minimalist work began to appear, and I thought to myself: hey, this is like categorial grammar with movement, that’s very cool. I’d already begun a PhD in Edinburgh working with Elisabet Engdahl on whether it was possible to graft a Discourse Representation Theory type semantics to a minimalist syntax in order to try to understand the semantic effects (or lack of them) of various object movement operations in Scottish Gaelic as well as other languages. Elisabet suggested I spend some time at UMass Amherst, where I got a taste of what a North American graduate programme was like (way more intensive than the British one!), and there I really learned how to think like a syntactician in Hagit Borer’s syntax course. My thesis ended up arguing that Agr nodes (all the rage then) were associated with a Heimian style semantics (constraining the elements in their specifiers to be discourse familiar). Cross linguistic variation emerged from conditions on which bit of a movement chain was fed to the semantic mechanisms, so in some languages Agr’s semantic effect was masked by other aspects of the syntax. Then Chomsky decreed Agr dead the following year!
Agr or no Agr, I was very lucky and got a job at the University of York, where I slightly neglected syntax for a short period as I panicked about teaching and found myself dangerously capable at administration. York was strong in syntax, and I had great fun working on all sorts of problems with George Tsoulas and Bernadette Plunkett; but York also had a growing sociolinguistics group, and one of Sali Tagliamonte’s students, Jen Smith, and I ended up sharing a spooky old 15th Century townhouse, with George, in York. I began to get interested in thinking about how to understand the morphosyntax of the sociolinguistic variation Jen was documenting for the Buckie dialect of Scots.
It was frustrating, but ultimately massively rewarding, to try, with Jen, to bring together two quite distinct parts of linguistics, and, this collaboration was crucial in leading to my next job, at Queen Mary University of London. My partner had been working in London since I moved to York, so, in classic junior academic fashion (he was a post-doc at UCL in environmental science) we were living long-distance lives. Then a job came up at QMUL for a sociolinguist. With my newly found credentials from working with Jen, I went down to QMUL and told them all about the syntax of negative concord in Buckie (I even had some chi-squared test results, in amongst the uninterpretable features and movement chains!). Although QMUL really wanted a sociolinguist, they ended up appointing a dyed-in-the-wool minimalist syntactician. QMUL also appointed Colleen Cotter at the same time, and together with Jenny Cheshire, we began to build the department at Queen Mary.
A lot of the next dozen years involved trying to cram in some syntax in between Chairing the Department, developing new degree programmes, trying to raise funding for PhDs and postdocs, and becoming Head of School (which involved trying to understand what colleagues in Film, Comparative Literature or Mediaeval French wanted). I wrote a textbook I’d never intended to, and Daniel Harbour and I began to work on Kiowa, a language he’d been investigating for his PhD, which led to a slightly impenetrable book, and to me being convinced that our theories of phrase structure were both too complex and too lax (echoes of my earlier worries about unification, I think). I also spent a quiet few months on the Isle of Skye working on Gaelic Noun Phrases (thank you, the Leverhulme Trust!), with further fieldwork trips over the Scottish Hebrides to Barra and Lewis, where I pestered the local populations with what they obviously thought were bizarre questions about their language, and discovered many a fine whiskey.
London is an amazing place to do linguistics. There are hundreds of languages to work on, fantastic colleagues, both at Queen Mary and in the larger linguistics community here, and it’s probably because I’m in London that that TV company called me to ask me to make up a language for them, completing this cycle of my linguistic life.