Unabashed Naïveté

Steven Pinker: Words and Rules

leave a comment »

Cover of Words and Rules The kind of wordplay Pinker sometimes engages in is mindblowing (read to the end):

A governor-general is a general governor, namely, one who has several governors under him. The puzzle is, why didn’t they simply call him a general governor? After all, the adjective comes before the head noun in English, not after it. The answer is that these words, together with many other terms related to government, were borrowed from French when England was ruled by the Normans in the centuries after the invasion of William the Conqueror in 1066. In French, the adjective can come after the head noun, as in États-Unis (United States) and chaise longue (long chair, garbled into the English chaise lounge). The earliest citation in the Oxford English Dictionary is from 1292: “Tous attorneyz general purrount lever fins et cirrographer” (All general attorneys may levy fines and make legal documents). Anyone who insists that we eternally analyze (hence pluralize) these words as they were analyzed in the minds of the original speakers of Norman French also should insist that we refer to more than one major general as majors general, because a major-general was once a general major (from the French major-général). Long ago our linguistic foreparents forgot the French connection and reanalyzed general from a modifying adjective to a modified noun. So if you are ever challenged for saying attorney-generals, mother-in-laws, passerbys, RBIs, or hole-in-ones, you can reply, “They are the very model of the modern major general.”

(If reading that was not a profound, life-changing moment for you, then perhaps you are unaware of the “Major-General’s Song”, in which case you should look it up. It is widely parodied; for example in this xkcd.) I can only wish he were still at MIT.

Words and Rules is probably the least popular of Pinker’s books written for a wide audience, only because it is the most technical. Indeed, in his C-SPAN In-Depth interview, Pinker laughs at himself for thinking that popular audiences would care to read hundreds of pages on the topic of irregular verbs (though the book takes a good look at the inflection of both nouns and verbs, regular and irregular). But I think if you have any interest in the topic at all, the book really very readable.

And intellectually satisfying: zooming in at the process of inflecting words (even in a language as boring inflection-wise as English) one finds an unexpected sort of mathematical depth. The major-general example is actually pretty shallow compared to the other linguistic phenomena examined in this book. There’s no way I could do justice to the intricacy and formal beauty the book illuminates in a mere blog post, but here’s a glimpse (I probably make some inaccurate statements below; please correct me if you see a mistake): In English, to form the plural of a noun, we add -s. To conjugate a verb in the present singular third person, we also add -s. This is a coincidence we can exploit to reveal something about the mental processes underlying language production. See, we don’t exactly just add -s, sometimes we add a -z sound instead (birds), or even something that sounds like -iz (kisses). (Pinker avoids IPA in the book, but in no way does that dumb down the point.)

How do we decide which ending to use? Native speakers certainly do not memorize which ending it is for every noun and verb separately. There is a pattern, roughly as follows:

  • If the word ends in something that already sounds like -s, use -iz, as in misses, boxes, inches, slices, minimizes
  • Otherwise, if the word ends in an unvoiced consonant sound (p, t, k, f), use -s, as in jumps, books, pits, laughs (knives and leaves and company are exceptions: note that they change the f to its voiced friend, v)
  • Otherwise, the word ends in a voiced consonant or a vowel; use -z, as in trees, hogs, fills, nods, kebabs

To convince you that these are real rules, not just patterns that occur most of the time, test them on made-up nouns like the famous wug or made-up verbs (I plose, you plose, it ___).

Now an interesting thing happening here is that the rules work for both pluralizing nouns and conjugating verbs in the third person. It would be rather unparsimonious for the rule to actually be implemented twice in the brain. We conclude therefore that the rule is actually a phonological one, rather than morphological. When a word is inflected, our brains first tack on the ending (Pinker argues that the default ending in this case is actually -z) without regard to the rules. Then, closer to the time that we actually pronounce the word, the brain (another part of it, perhaps) looks for incongruities like jumpz, and devoices the -z into an -s. The general rule is that you can’t have in the same syllable (I think) two consonants that are differently voiced right next to each other (sonorants m, n, l, r are not thus restricted). This rule also is in effect when we form the past tense with -d: after unvoiced consonants as in walk the -d too must be devoiced, so that the -d in walked actually sounds like a -t.

Pinker uses investigations in the same spirit to build up his words-and-rules thesis, which I won’t describe here. The book draws evidence not only from the above type of introspective reasoning (does X sound correct to me, a native English speaker?), but also in experiments involving both normal speakers and speakers with various impairments who are asked to perform various language tasks, sometimes while their brains are being scanned. It’s “social science”, so you don’t expect it, but the thrill of seeing how Pinker draws conclusions from the experimental data really is the thrill of empirical science, surely the same thrill that Galileo must have felt when he first realized that the distance covered by a falling body varies as the square of the time passed.

The unfortunate thing is that it appears Pinker’s conclusions, not just in Words and Rules but also in The Language Instinct and his academic work, are not shared by all linguists. They praise him for bringing linguistics to the public, but spurn they way he tosses competing linguistic theories aside. Indeed, if there is widespread disagreement about the validity of Pinker’s claims, he gives too little space to them in his popular books, in which alternative viewpoints are explained only so that they can be shot down. I might be one of his victims: with few competing theories of language so lucidly and stylishly presented to a popular audience, I may only be so excited about him because he has been my first and only tutor in the subject. The main controversy seems to be concerning the view that the specific ability to learn and use language is innate in humans, or at least a view that is closer to that than its alternative, which considers our linguistic abilities a by-product of humans’ general cognitive abilities. The former appears also to be associated (I think!) with “generative grammar”, and the latter with “connectionism”, but I won’t embarrass myself trying to say what I think those mean. I guess I won’t know for sure who’s right until after some formal training.

In the mean time, I have been looking at some of Pinker’s critics. Here are some critical reviews of The Language Instinct:

And then here’s an interesting debate about mice-eaters that I haven’t gotten around to reading:

So that’s that.


Written by leonxlin

July 20, 2012 at 8:14 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: