Friday, June 1, 2012

Sherlock or Watson: Advice for linguists

Occasionally I get a reader asking for advice about studying linguistics. Recently, I responded to such an email and I post some of my advice here. Ignoring my advice is generally wise...

Standard Advice = Follow your dreams. Do what you're passionate about. If you want it bad enough, you can make it happen.

My tough question: Are you more Sherlock or Watson? Here's why I ask...

Reality = Be realistic about what you are willing to do (and suffer through). Everyone has dreams, but most people find it difficult to persevere through the boredom of other people's tedium to grab their brass ring. If you are a very focused person, you can accomplish your goals regardless of anything, but if you lack that near sociopathic ability to flat ignore other people and their constant distractions (think Sherlock), then you should probably have a plan B (think Watson); a good plan B involves developing a standarized set of skills that are obvious, valued and marketable. Sherlock is a "consulting detective", whatever the hell that is. Watson is a medical doctor.

Here are some specific responses to what was brought up in this specific email:

Q: You say you are interested in "semantics, morphology, philosophy of language, and possibly language acquisition."
A: Great enthusiasm, but calm down. Overreaching is the death of success in grad school. Studying everything is for retirement. Careers are built on specializations. My grad adviser once said, "there are two kinds of linguists: those who study sound, and everyone else." You seem to fall on the "everyone else" side of the house (congratulations, those sound folks are weird).

Q: "...most of the PhD linguistics students have had significant exposure to linguistics before arriving."
A: False. This is a common assumption. But remember that there are very few schools that offer a linguistics undergrad degree (as opposed to English, math, business, etc, which is offered at every college on the planet). Many people come to linguistics in at least as round-about way as you. Don't judge them, and they won't judge you.

Q: So you've "never taken an actual linguistics class"?
A: Not a problem. Buy a used copy of Language Files for $6 and work through the first 8 chapters (the rest is crap). Don't worry about the edition number, Ohio State has been recycling the same crap with new colors since 1997.

Q: "I need to get some linguistics training."
A: Not really, not yet, that's what the grad program is for. But check out the tour of BYU's excellent online corpora; Also, work through the NLTK book. I admit a computational bias, but this will give you some great tools and practice. Also try Coursera for NLP and Machine Learning. If you work through Language Files, NLTK, and Coursera's NLP class this summer, you are gold.

Q: You need "a strong piece of research to submit as a work sample."
A: True, you need a good peice of writing, but it doesn't necessarily have to be linguistics; but it should be data based analysis. You could work through one of the NLTK exercises and do a technical write up. That's a possible approach. Also, see if BYU's corpora site suggests research projects.

Q: You point out that I wrote "Linguistics is no field for mediocrity."
A: I stand by this wrt academic linguistics (professors at universities); but there are options in industry, which I discussed in this guest post on Zoltan Varju's blog.

Some additional posts to wade through:

10 easy ways to fail a Ph.D.

Tough love: An insensitive guide to thriving in your PhD (more hard science related, but some good advice).

TV Linguistics - Pronouncify.com and the fictional Princeton Linguistics department

 [reposted from 11/20/10] I spent Thursday night on a plane so I missed 30 Rock and the most linguistics oriented sit-com episode since ...