Morphemes

The most obvious iregularities in some of the commonest ‘words’ in English squeeze more than one element of meaning into one ‘word‘. Man and  men both refer to adult human males, but the first singles out just one, while the second refers to more than one. Go and went both signify motion away from the speaker’s frame of reference, but the second uses a different root and shifts the time frame to the past. Two elements of meaning in each case, number or tense and the word itself. Contractions like don’t and can’t denote negation by the n’t. Am, is and has involve three elements, the ‘person’, first or third, the singularity, and the present tense. These minimal elements of meaning are known as ‘morphemes’, in these cases with two or three ‘fused’ together. The study of these and more complex cases like the different semantic roles of the -er morpheme in slipper, sleeper, and so on, is known sometimes as ‘morphology’ and sometimes as ‘morpho-phonemics’.

The handful of irregular forms like men, women, and children contrast with the ‘regular’ forms like babies, uncles, aunts, witches, where the two morphemes are obvious and transparent, with a root on the left and a plural morpheme, denoting more than one, on the right, subject to a rule without an exception in the language that changes the form according to the last sound of the stem, as in pats, pans, and patches.

In verbs like be, have, go, fall, there is a contrast with regular cases like exist, possess, travel, which add a different ED morpheme, again on the right, and again without exception, to denote some reference to a time frame other than the immediately verifiable present, as in “If I had existed in the past, I would probably have died before I was this old” – with a specification of counterfactually in the had as well as its syntactic role.

In be and go and the past forms, was, were and went, the change is in the root. In the other irregular cases, of which there are several hundred, mostly very common, as in fell, as opposed to fall, the vowel changes, in spent as opposed to spend, the voicing of the final consonant changes, in slept as opposed to sleep, both the vowel changes and a voiceless form of the regular D is added. The only way to generalise across all of these cases is to say that the tense, and in a handful of cases the number, is expressed by the form as a whole with two morphemes – whether these are separable, as in the regular cases, or not.

The morphology is difficult for children. For children learning English, there is a learnability hurdle in the irregularities in slept, ate, drank, fell, and so on, where English morphology is at its most complex with around 300 relevant cases. Some of the errors involve just missing out the relevant morpheme -s  or -ed. Others involve an overgeneralised regular form as in sleeped, eated, drinked, falled. Rarely, but in way that is highly noticed, the regular morpheme is added to the irregular form as in wented. Usually, in these unusual forms, there is just a doubling of forms. But very occasionally the overgeneralisation is multiplied. One of my children once said “I think I might have misunderstoodended that.” It occurs to me that with four markings, one irregular past, one participial form, and two regular forms of the past, he might have been looking for what is known as an ‘evidential’ form to express a category which is regularly encoded in languages, though not in English, to encode a degree of uncertainty. It would appear that in the unrecorded early history of Proto-Indo-European, this was the meaning of what became fossilised syntax in subjunctive forms which English speakers find hard to learn in Spanish, Italian and French. As children get older, they sometimes say things like sleepering instead of sleeping. But once again, it may be that they are trying to express a category which happens not be encoded in English.