PETER TRUDGILL on what lies behind our irregular everyday phrases.
The most usual way to make past-tense verb forms in English is to add -ed to the basic form of the verb: want–wanted, love–loved; play–played. So why is the past tense of go not goed but went? Even with verbs which are irregular and do not add -ed, like sing–sang, run–ran, leave–left, you can still see a clear link between the present-tense and the past-tense forms. But go and went do not seem to be even remotely connected.
This is a nice illustration of how languages are not necessarily entirely logical systems. Languages have to have large amounts of regularity or infants would not be able to learn them. But they do not have to be totally regular – they can tolerate a certain amount of irregularity, and most of them do.
It is not always possible to explain why a particular word behaves in a non-systematic way, but in the case of go we understand the process that produced this situation: the modern verb go was originally two separate words. Went used to be the past tense of wend, and only came to function as the past tense of go from about 1500 onwards. To compensate for this, wend then had to acquire the new past-tense form wended, as in ‘he wended his way home’.
We can see the same kind of irregularity in the case of the adjective good. If we say nice–nicer–nicest, why don’t we say good–gooder–goodest? The same question arises for all the Germanic languages: Norwegian has god–bedre–best; in Dutch it is goed–beter–best; and in German gut–besser–best. This irregular pattern must have been present in our common parent Germanic language. It is obvious that we are again dealing with two originally different words here, one which gave us modern good and another which gave us better and best, but the peculiarity of the pattern is so ancient that we do not know what those words were.
In the case of the adjective bad, we know a bit more about the process, (though we don’t actually know where the word bad itself came from – it appears to be peculiar to English). Medieval English speakers said bad–badder–baddest, but gradually speakers started to use worse–worst instead, and this was actually from an old Germanic root: in Danish worst is værst.
When unrelated forms are used for different versions of a single word like this, linguistic scientists call it ‘suppletion’. You can see that the same process is currently beginning to happen with the noun person. We say one person but, while it is possible to say two persons, the normal plural of person is people, so we usually say two people. If you hear somebody trying to book a table in a restaurant for ‘four persons’, you can be almost certain that English isn’t their native language.
The most extreme form of suppletion in English is provided by the verb to be. This has numerous different forms, many of which look and sound nothing like one another: be, been, being; is; are; was, were. In older forms of English there was also art and wast.
Go and went came from two different original verbs about 500 years ago, but the different forms of English be came from three or more different original word-forms in the ancient Indo-European language – and more like 5,000 years ago than 500. English been, and German bin and Dutch ben ‘am’, come from an Indo-European verb which meant ‘to grow’. English was, and German (ge)wesen and Dutch (ge)weest, ‘been’, come from an ancient root meaning ‘to live’. We can see, too, that English is, German ist, and Latin est have come from a common root. And it is not a coincidence that English am corresponds to jam in Albanian and eimai in Greek.
Originally coincidence simply referred to a situation where two events co-incided – happened at the same time. But the most common modern meaning is probably ‘a notable concurrence of events having no apparent causal connection’. This is relatively recent, though: until the 1800s, people seeking to convey the modern sense typically wrote of ‘unexpected’, ‘strange’ or ‘singular coincidences’.