Blogging the LSA: Epilogue.

The moment I published my ranty LSA blog part one I was given something else to chew on. I left my laptop and went to Mark Liberman's plenary talk on the future of linguistics. Now when Mark gives a talk on the future of linguistics, chances are that it's worth paying attention. And this did not disappoint.

Following a poker analogy that I'm 99% sure came from Alan Kors, Mark argued that linguistics as a discipline started off with pretty good cards. Once upon a time, every person of even a little schooling received serious education in grammar, rhetoric, and logic. Educated people studied foreign languages, both living and extinct. Reputable scholarly types from Descartes to Jefferson understood the importance of developing analytical skills to be applied to the study of language.

Fast forward to today. He gave a number of cute illustrations of the current status of linguistics as contrasted with the prevalence of other disciplines that started off with far less collateral, psychology being his main example. Looking at indicators such as undergraduate enrollments, LSA vs. APA membership numbers, and the presence of appropriately schlocky supermarket checkout magazines, he demonstrated what everyone knows. Psychology is hot! Linguistics, not so much. Indeed, the discourse in many domains relevant to the study or use of language, including technology, education, cognitive psychology, game theory, evolutionary anthropology, law, neuroscience, and many others, is today dominated by people with little or no background in linguistics.

So then he started to talk about why.

One of his main arguments was that disciplines such as psychology, history, and political science have adopted a sort of big tent approach, where anyone working on even vaguely connected work is invited to the discussion. Contrast this with the approach taken by many linguists, where you're only welcome at the table if you come from a narrow range of subfields in a narrower still range of departments in a narrower still range of universities. Do anything else, and you're not really doing linguistics.

Hey, this is starting to sound a little like the flip side of the first LSA blog post that I made!

Mark encouraged the audience to address this by reaching out to colleagues in other disciplines. If you have someone at your university interested in language and the law, or how reading might be more effectively taught, or rhetoric and composition, invite them to join your graduate group. Make a big tent, and include people, and you're not only going to raise awareness of your discipline, but you might also find your work enriched by the perspective of someone from somewhere else. This is really the right approach in so many respects, and it's not generally implemented very well.

But I started thinking about this, and there's really a whole lot further that we could take it. The thing is that until career paths outside academia are not only tolerated but also supported by faculty in the finest departments, progress will be limited. With regard to technology and education in particular, there is much valuable work being done in government and the private sector. Granted most of it today is not traditional academic research as such. However, one reason that this is so is that people in these branches do not typically bring a rigorous research background to their positions. And one reason that this is so is that people with rigorous research backgrounds are strongly discouraged from diverging from the prescribed path that most of their advisors know most well. But the fact is, if the field of linguistics wants to make inroads into other disciplines, the first thing that needs to happen is that it needs to be okay for promising students to choose research and professional paths other than ongoing direct contribution to theoretical linguistics. This is true both for students within academia and for students who are considering leaving it for a time.

So clearly I'm at Microsoft, so I have a personal story here, but it is a story that I find echoed by many of my colleagues both in academia and outside it. I know that when I was in school, even in a place that does this remarkably better than most places, I felt strongly discouraged from pursuing interests outside theoretical and computational linguistics narrowly defined, even though I had great interest in and passion for related areas within human-computer interaction and education. I was discouraged in part because of the track I was seen to be on, heading for a career in academic research in the syntax-pragmatics boundary. If I had found it easier to pursue some of these related interests as a student, I feel that my work would have improved – and you never know, I might not have been so quick to leave academia. Even once I had decided go into industry, I got real pushback from many people. Did I really want to go into software? How could that possibly be interesting? I believe the questions were genuine, and came from real consternation. I have heard very similar stories from colleagues who displayed interest in education, anthropology, even math or technology. Even at a department that is more genuinely interdisciplinary in its approach than maybe any other.

But in order for the discipline of linguistics to really succeed, you need classically trained linguists to do many things. You need them to develop interests in and do productive work in technology; in curriculum design from elementary schools on up through college; in government language policy positions; in the related fields of computer science, psychology, math, anthropology, and so on. The list really goes on. Although there are some common areas of overlap, it is difficult to delimit the set of possible crossover opportunities, because exactly what you should want are linguists who can take their training and find new ones. Even in publishing schlock magazines!

I think that encouraging students to pursue alternative paths where appropriate has a bunch of less noble but still useful effects as well. Students can get jobs, for one thing. And graduates well placed can help other students get fellowships or internships or jobs of their own. It gives linguists a great opportunity to shape what’s going on in related fields, in governments, and in the private sector. It can also result in formalized collaborations that in turn can result in greater visibility and funding for linguistics departments. Finally, I think the first department that really gets this right -- with alumni mentoring programs, with seminars by people doing related work, with appropriate curriculum flexibility, with government/industry partnerships for internships, etc -- is going to see a major positive response in their attrition rates, because students will see more possibilities for using their skills and engaging their interests rather than fewer. That's very important to maintaining the kind of positivity that is really required to get through the slog that is a PhD.

One of the really cool things about working in international software is that not everyone is like me. My background adds a certain kind of value here. I also have colleagues who have expertise in the more humanities-oriented side of linguistics, in classic software development, in graphic design, in marketing and business, in usability, in finance, in math, and in many, many other areas. They all end up being useful in the work that we do. For me, that's what makes this a great industry to work in, and it is the most striking change between my professional life a few years ago and my professional life today.

On that note, I return to your regularly scheduled discussions on international software. :)