When I was at the MOT workshop in honour of Glyne Piggott this month, people would occasionally ask me, “So what are you working on these days?”
Now, lacking the neurotypical skill of telling when people really want to know the answer and when they’re just being polite, I usually just shrug off that question — though a couple of times I’ve started giving an answer which, after the first thirty seconds, probably bored the unfortunate asker into forgetting their own name.
So just in case anybody really is interested, here’s a description of what I’m working on lately. Now when anybody asks me, I can give my usual vague answer and tell them to read my blog for more details. Most people can continue to smile politely and get on with their lives. The tiny fraction of a percent who really want to know — well, you’re reading this now. I apologize in advance for boring you to the point where you forget your own name.
So what am I working on these days?
There’s the usual batch of small projects that might or might not turn into journal articles some day. There’s the pilot project for the humungous American Sign Language corpus project that I’m working on with Erin Wilkinson and Terry Janzen.
But most of my time is going to The Book…
The Book is a horrid, horrid creature who asked to stay over in my head one weekend and neglected to ever move back out. It leaves its laundry on the floor and dishes in the sink. And it’s sprouting new volumes so fast that I suspect it’s either mastered parthenogenesis or else it’s been sneaking out of my skull at night and consorting with other books, completely heedless of my lectures to it about the importance of protection.
It all started during the five-year soul-sucking prison sentence that was my term as department head. That’s honestly as positive as I can make the experience sound. Most days were spent on earth-shatteringly vital tasks. Like: “Justify why your staff criminally spent the money in that budget line buying a fax machine when it was clearly intended to be spent on sending individual faxes.” Or “Drop everything you’re doing and immediately count how many cubic metres of confidential documents your department keeps.” Or “Listen to this long, rambling speech by the director of human resources justifying why his computer system still doesn’t work this year, followed by a helpful outline of the extra work that’s going to mean for you.”
Every erg of creative energy that wasn’t annihilated by that black hole went into the teaching I still did. In sum, that was five years when the amount of writing I managed to get done was (measured to within any arbitrary precision): zero. The number of projects I managed to successfully plan was: zero. The amount of field work I got done was: pretty damn close to zero.
While I couldn’t do most of the things I actually liked about my job, what I could still do was read — not a lot, but some. About an article a day. So, since I clearly wasn’t about to accomplish anything else remotely resembling research, I resolved to do just that: read one article a day, every day, no matter what.
I’d seldom be able to fit into my head all the intricate argumentation in a typical formal phonology or syntax paper before my mental gas-tank hit empty, so I ended up reading a lot of articles in phonetics, psycholinguistics, child development, and communication disorders — areas I’d wanted to get more depth in for a long time but had never got around to.
Some days I’d only manage one of the four-pagers in Psychological Science or in the ICPhS proceedings. Usually it would be closer to 10 or 15 pages, like a typical article in JASA, Journal of Phonetics, or Journal of Memory and Language, or a chapter from one of the Laboratory Phonology books. Occasionally I’d have the time to squeeze in something far more substantial, like some gigantic two-column, 50-page review article or somebody’s dissertation — more often I’d have to break the longer things over several days (something I could never do for formal phonology or syntax), but that didn’t absolve me of my self-imposed duty to read another complete article on each of those days. As a matter of discipline (something I’ve never been all that big on), I had to read that article every day, even if it meant sneaking in half-page letter to Science between Christmas brunch and Christmas dinner.
Before I knew it, I’d read over a thousand articles. Whew.
Scarier still, they were starting to make a creepy sort of sense. Even though I couldn’t find three papers in a row that actually agreed on something, and though several of the writers clearly saw themselves as locked in to-the-death grudge-matches with each other, there were actually coherent themes emerging (at least in my fevered mind), odd connections popping up between this paper on dialect differences in perceptual phonetics and that paper on three-year-olds learning syllable structure and that other paper over there on the nonword repetition skills of dyslexic adults.
So that’s how the idea for The Book began — quite possibly the worst idea ever in the entire history of me.
Partway through the reading marathon, I’d begun to be able to tell the difference between good work and dreck, to figure out the point of an article, how exactly it fits into a long chain of debate between shifting coalitions of co-authors spanning decades and dozens of papers.
A tiny part of me got a little resentful: “I shouldn’t have had to wade through that much dreck to get to the point where I am now.” I started wishing there’d been a good introductory book that would have let me skip over the first couple of hundred papers or so. Which is a dangerous wish for me, because whenever I have a wish like it, some other thoroughly evil part of my brain starts whispering, ever so reasonably, “Well, why don’t you write it yourself?” And it just nags and nags and nags, day and night, whispering and nagging and whispering and nagging, till it’s a foregone conclusion that I will write the damned thing.
So The Book is about all the cool findings from the disciplines on the border of phonology that I think phonologists should know about and take into account when dreaming up their theories. It’ll hopefully be the book I wish had existed when I started.
Here’s the outline I’ve got so far.
memory: What kinds of stuff do people have to remember in order to learn and use the phonological system of their language? This chapter will cover both evidence for exemplars and evidence for abstract phonological categories independent of exemplars.
articulation: In order to speak or to sign, you have to coordinate lots of different pieces of your anatomy — moving them all at just the right time relative to each other. How do speakers and signers coordinate all that? And how do they learn to coordinate it all?
(Actually, you know what? I’m going to have to say “speakers and signers” really often, and say “listeners and, uh, people who see signs and have to understand what the signer meant” really often. And I’m lazy. And besides it sucks that there really isn’t good equivalent equivalent to the word “listener” for sign-language addressees. So, from now on, whenever I say “speaker” or “listener”, just assume that I’m talking about their sign-language analogues too. In writing The Book, I’m trying to take seriously the idea that a theory of phonology that can only explain spoken languages has done, at most, only half its job.)
perceptual categories: How do listeners go from hearing a bunch of sound waves to having decided, “Oh, that’s a [p] sound!” How do they learn to do that? In fact, how do they learn which perceptual categories are in their language in the first place?
perceptual categories, part deux: In order to make the decision “Oh, that’s a [p] sound rather than a [b] sound,” a listener can’t just pay attention to a single property of the sound waves. Rather, they have to integrate and juggle lots of different acoustic properties, some of them more reliable, some of them less, some of them only useful if some other property is also there — and also pay attention to visual input. How do they do that? How do they learn to do that?
(Actually, you know what? I’m going to be saying “And how do they learn to do that?” so often that you might as well just assume that I’ve added that sentence to the end of every chapter description from now on.)
word recognition: How do listeners go from sound waves to deciding “Oh, that’s the word pear” or “Oh, that’s the word bear”? How can those decisions go back and make listeners go back and revise earlier decisions about whether they heard a [p] or a [b]? What kind of information do they have to pay attention, and when do they pay attention to it? What kind of phonological knowledge do they need inside their heads to make those decisions? And how do they learn it? (Sorry, I promised I’d quit saying that.)
word production: Like word recognition in reverse. How do speakers go from an idea (“I want to talk about that furry meowing thing”) to having prepared the motor plan that they need for pronouncing the word cat. This chapter will include some weird findings, like the relationship between your planning system for producing words and your ability to remember lists of things for short periods of time and your ability to learn new words and foreign languages, and the stranger relationship between moving your arms around and planning words — even if you’re speaking and not signing.
Everything up to here forms a nice coherent whole — and by “nice coherent whole” I mean nothing more than that it’s mostly already written. (Except for the last chapter, and a couple of other sections, and a small country’s worth of footnotes, and a large country’s worth of proofreading and fact-checking and revisions, and… well, mostly.) So this marks a logical place for The Book to pause and call it a day, or at least to call it “Volume 1”.
But wait! There’s more! Order now and you can eventually get — with no money down and no payments for the next four years — a whole additional volume! Or two! They don’t actually exist yet (which is why you don’t have to pay for them yet), but that’s never stopped anything from ruining my life. The fantasized topics in the next cancerous excrescences of The Book are as follows:
computer modelling: How do psycholinguists and computer geeks make sure that the theoretical models they come up with actually make the predictions about human behaviour that their creators think they make? This chapter will be an overview of some of the most common techniques being used to model human language use and learning, including (but most definitely not limited to) the connectionist models so dreaded by formal phonologists who’ve heard of it. Most of the research covered in the following chapters makes heavy use of computer modelling, so this intro has to be up front.
morphology: How do speakers and listeners cope words with more than one piece to them? Do they bother learning cats as a new word, or do they always just break it down into the cat and -s pieces that they already know — and how can we tell? How does morphology influence phonology and vice versa?
reading: For a century, linguists have been proclaiming to everybody willing to listen (mostly just our intro students) that we don’t give a damn about written language, that it’s irrelevant, that everybody already knows everything interesting there is to know about the phonology of their language before they get to kindergarten and that’s the knowledge we’re interested in explaining — not the artificial writing nonsense that comes later.
A century ago, this was actually a necessary political stand for our newborn discipline to take, way back when we had to defend the relevance of our subject matter against arrogant Victorian prigs who assumed only fine art and high literature was worth studying. (Like, what could a bunch of children, ignorant savages, and working-class peons possibly teach us about what it means to be human anyway?) Okay, we won that battle (inside universities anyway). Now it’s time to admit that our century-old slogans aren’t actually completely entirely 100%, um… true. Turns out that writing can matter. Learning to read and write an alphabet profoundly rewires a person’s brain in ways that have only become clear in the last couple of decades, ways that affect the speaker’s phonology even when they’re just listening and nowhere near a printed word.
Even more ominous, turns out that many of our assumptions about how phonology works aren’t universal facts about humans, just facts about the subset of humans who read an alphabet. In short, we’ve been unconsciously suffering from exactly the privileging of written language that we’ve loud and long proclaimed we’re opposed to. Oops.
Hmm, that was a long chapter description. Aren’t you glad I stopped saying “And how do they learn all that”?
variation: the meeting of phonology and sociolinguistics, with some historical linguistics thrown in. How do phonologies differ between dialects? How do speakers use phonological choices to communicate social ideas — “I’m like you”, “I’m not like you”, “I’m nothing at all like my parents.” How do listeners cope with that hodge-podge? How do these “non-linguistic” things stick their fingers into the gears and mess around with how the “linguistic” processes from the last several chapters do their job? How can all this result in languages changing through time?
bilingualism and second-language phonology. The last few chapter descriptions have exhausted me — I’ll just leave it at that.
disorders: Phonologies don’t always work. How can it break down and what does that tell us about the normal system? What exactly is different about the processing systems we’ve talked about so far in people with aphasia, dyslexia, apraxia of speech, stuttering, and so on?
neural and genetic bases: What do we know about where and how the brain does all the previously discussed work? What do we know [short answer: not much] about the genetics that makes it possible?
phonological constituents: Finally, we can start getting back to the kinds of things phonologists have usually thought to be their jobs. Stuff like syllables, phonemes, metrical feet, features, segments, moras, onsets, phrases, morpheme boundaries — the whole menagerie. The last big part of the book will be a survey of the phonetic and psycholinguistic evidence relating to the existence of each traditional constituent and the role it plays in normal processing.
I’m still completely undecided about how I’m going to go about getting The Book out there and into your hands, both of my eagerly waiting readers.
I’m sorely tempted to just dump the PDF file on the internet when it’s done, letting it float free into the hands of friends, enemies, impoverished grad students, Google, and shady websites that will offer to sell pirated copies to the morally and intellectually challenged for “way less than retail”. The PDF file will have lots of cool features that a paper book wouldn’t — colour, hyperlinks for cross-references and citations, as many pages as I want (or rather, as many pages as The Book insists on sprouting). I’ve already got tenure, so the extra cred I’d get from a big name publisher would have a minuscule effect on my life.
On the other hand, I’m still Luddite enough and vain enough to appreciate a nice, solid tome with my name on the spine. Okay, honestly what I really like about a nice, solid tome is that it’s immutable — it will be physically impossible for me to waste the rest of my life fiddling with commas and releasing a new version every two weeks.
Maybe it’s possible to get the best of both worlds. Some publishers are fine with free PDFs floating around in parallel to the printed book. Maybe I can get a deal like what David MacKay got from Cambridge for Information Theory, Inference, and Learning Algorithms. But all those worries are months, years, decades in the future. Right now I have a small country’s worth of footnotes to finish.
So that’s what I’m working on these days. Wishing you hadn’t asked about now?
↑ “clearly” = mentioned in passing near the end of a long memo on a completely different topic sent to the wrong person halfway through the budget year.
↑ I’m sure Rob Hagiwara and a couple of other innocent bystanders would add: “I shouldn’t have had to waste so much time listening to Kevin rant dementedly about all the dreck he was reading.” So let this be a warning to all academics: Bad writing doesn’t just hurt the reader. Second-hand effects can ripple out through the entire society, damaging all they touch, and ultimately leading to the fall of western civilization.
↑ I can often trick these thoughts into going away by pleading other work: “Aw, I’m sorry, best-recruitment-brochure-ever-for-the-field-of-linguistics, I’d love to write you. Honest I would. Unfortunately I’ve got this conference paper that needs to be finished right now. But let’s do lunch some day.” But The Evil Book knew damn well I wasn’t writing anything else (for reasons I’ve already whined about) and it called my bluff. “Well, of course you can’t write me now, Kevin. But one day your prison sentence will be over and you’ll go on sabbatical, and when that day happens, I’ll be waiting for you. Mwa-ha-ha!”
↑ Keen-minded victims of the past few years of my phonological theory class might notice eerie resemblances between this chapter outline and the course syllabus. All I can say is: I’m sorry, guinea-pigs.
↑ It’s comforting to think that there’ll be literally dozens of copies of The Book sitting on library shelves spanning the globe, and that when the zombie apocalypse comes, my work will live on and contribute to the survival of the last remaining humans, who will be able to burn it for warmth like they could never burn dozens of copies of my immortal memo on the budget line for faxes.