Harvard scientists predict the future of the past tense

Thanks for sharing. I think their study is self-contradictory. On the one hand, they seem to assume that all verbs are "evolving" towards regularization. On the other, they note that the most used verbs, which evolve the fastest, will take the longest to regularize. The problem with their study is that they assume that regularization is the "goal" of all verbs. However, they are faced with the linguistic reality that the most used verbs are the most irregular. This is true in any language (well, languages with verb conjugations--Chinese is exempted, I suppose). In Greek, for example, a less common used verb like "frontizw" (to care for) finds its subjunctive form easily "frontisw", following a z to s pattern. However, the words: eat (trow), say (lew), see (blepw), go in (mpainw), go out (byainw), etc. are quite irregular, and will probably stay so. trow becomes faw, lew becomes pw, blepw becomes dw, mpainw becomes mpw, and byainw becomes byw. Anyway, the point is that they are on to something, but they are misstating it slightly (in my opinion). It is when the words are LESS used that they will morph into the regular form. So, it is less of an EVOLUTION than it is a DEVOLUTION. The word "wed," from their example, is not commonly used as much as it used to be; it has been replaced mostly with "marry." So, they are right that wed will devolve into the standard past tense, because people stop using the word. Very interesting, overall.

Very interesting. I can't even comprehend how hard it must be to learn English as a second language and deal with all the irregular verbs and exceptions to various rules. It's hard enough when you are a native speaker. It would be interested to see more examples from their study of irregular verb forms that have changed into regular forms in English.

I've just heard that all verbs are irregular in Hungarian. Has anyone heard this before? Or is this just a misunderstanding?

Art, yeah, I saw this article yesterday and passed it around to other ChinesePod staff. I think it's fascinating. Pulosm, I think you misunderstood the article a bit. It said that the most-used verbs in English are all irregular and will stay so, precisely because they're the most-used verbs. I don't see anything wrong with the logic, and based on what you read, I think you actually agree with it. Regularization is not a "goal," but rather an empirically observable trend. Likewise, calling it "devolution" is a subjective label, when really we're just talking about "change." Evolution is not inherently "good." It is just the change that happens over time.

Joachim, This should answer your question: http://www.personal.psu.edu/faculty/a/d/adr10/hu3.html So, it looks like no. I think it's very unlikely that any language could have all irregular verbs. It's just too cumbersome for the human memory. If such a language ever existed, it would start regularizing itself immediately. The annoying thing about Hungarian verbs is vowel harmony. I have some experience with this concept through Korean and Turkish. I am not a fan.

Considering that Chinese is a very old language. It obviously reached the ultimate nirvana of verb regularization.

henning, And we are all very grateful that it did... :)