Horton Heared a Who!

Kids say the darnedest things. "We holded the baby rabbits." "The alligator goed kerplunk." "Horton heared a Who!" These lapses, you might dimly recall, have something to do with irregular verbs. But please don't stop reading just yet. Children's errors are not just anecdotes for grandparents or reminders of long-forgotten grammar lessons. They're windows into the workings of language, history and the human mind.

Verbs in English come in two flavors. Regular verbs like walk and smell form the past tense by adding -ed: Today I walk, yesterday I walked. English has thousands of them, and new ones arise every day, thanks to our ability to apply rules instinctively. When people first heard to spam, to mosh and to diss, they did not run to the dictionary to look up the past tenses; they knew they were spammed, moshed and dissed.

Even children do it. Told that a man likes to wug, they will say yesterday he wugged. Children are not sponges; they're constantly creating sentences and words, never more clearly or charmingly than when they encounter the second flavor of verb, the quirky irregulars. The past tense of spring is sprang, but the past of cling is not clang but clung, and the past of bring is neither brang nor brung but brought. English has 180 irregulars, a ragtag list that kids simply must memorize.

But when an irregular word is still fresh in the mind, it is fragile. If a child's memory cannot cough up held quickly enough, he or she adds -ed by default and says holded instead.

Irregular and regular verbs embody the two underlying tricks behind the gift of articulate speech: words and rules. A word is a memorized link between a sound and a meaning. The word duck does not look, walk or quack like a duck. But we can use it to convey the idea of a duck because we all once learned to connect the sound with the idea.

We also combine words into bigger words and sentences, using the second trick of language, rules. Journalists say that when a dog bites a man, that isn't news but when a man bites a dog, it is. Rules let us convey news by reshuffling words.

Regular and irregular verbs today have their roots in old border disputes between words and rules. Many irregulars can be traced back over 5,500 years to a mysterious tribe that came to dominate Europe, western Asia and northern India. Its language, Indo-European, is the ancestor of Hindi, Persian, Russian, Greek, Latin, Gaelic and English. It had rules that replaced vowels: the past of senkw- (sink) was sonkw-.

Language as it evolves is like the game of Broken Telephone, in which a whispered phrase gets increasingly distorted as it passes from lip to ear. Eventually speakers no longer discern the rule behind a motley set of mangled verbs. They just memorize them as a list, as do subsequent generations. These are the irregulars, the fossils of dead rules.

The irregulars are vulnerable too because they depend on fallible memory. If a verb declines in popularity, speakers may not hear its irregular form often enough to fix it securely in memory. They fall back on -ed, changing the language for following generations. That is why forms from Chaucer's time such as chide-chid and writhe-wrothe turned into chided and writhed.

You can feel that force of history acting today. Smote, slew, throve and forsook sound odd, and few people use them. In a century, they'll probably go the way of chid and wrothe.

Do irregular and regular verbs really come out of a dictionary in one part of the brain and a grammar in another? Perhaps. Neuroimaging techniques suggest that regular and irregular forms may trigger signals in different parts of the brain. Some neurological patients seem to have damaged dictionaries: they strain to retrieve words but speak in fluent sentences; like children, they say heared and holded.

Why pay so much attention to the lowly irregular verb? I see these studies as part of a trend that biologist E.O. Wilson calls "consilience": the bridging of science and humanities through an understanding of how the mind works. A slip of the child's tongue may link the migrations of great prehistoric tribes to the brain-imaging technologies of the next millennium.

STEVEN PINKER is a professor at M.I.T. and author of How the Mind Works (Norton) and Words and Rules: The Ingredients of Language (Basic Books), from which this article is adapted