PETER TRUDGILL assesses just how (il)logical certain languages are.
Over the years there have been plenty of people who have not hesitated to assert that French is a “very logical language”, the implication being that other languages are less so. But it is actually rather easy to point to aspects of French which do not on the face of it seem particularly “logical” in our usual understanding of this word.
For example, why would ‘I am going’ be je vais, but ‘we are going’, nous allons, and ‘we will go’, nous irons? Is there anything logical about the fact that la main, ‘the hand’, is grammatically feminine but le doigt, ‘the finger’, is masculine – and is there in fact anything logical about grammatical gender anyway? Why, if you put the definite article le, ‘the’, before the noun homme, ‘man’, do you get l’homme, while if you put le before hibou, ‘owl’, you get le hibou without the contraction?
French spelling is particularly difficult to regard as logical. This has got nothing to do with the language itself, of course: spelling systems can be rationalised without the language itself being changed at all – Norwegian had a big spelling reform in 1907, for instance. But any supporter of the merits of the French language who wishes to argue that it is logical for the sequence of letters in eaux, ‘waters’, to represent the sound “o” would be on rather tricky ground.
English is no better, of course: it was perfectly sensible to spell our word knight like that in 1300, when that spelling was an accurate representation of the pronunciation: (the k was sounded and the gh represented an h-sound). Today it makes no sense to use six letters to stand for three sounds.
Languages which we could point to as being apparently rather more logical than French might well include Finnish. It is instructive to see how Finnish nouns behave grammatically.
For example, the word for ‘car’ is auto. Autossa means ‘in the car’, with the -ssa ending representing a case known to grammarians as “inessive”, which signifies ‘being inside’. Note, then, that autoissa means ‘in the cars’ – the marker of the plural is -i, and here it is inserted between the noun and the case ending.
Because of the regular way Finnish works, if we add the possessive marker -si to ‘car’ it gives us autosi, ‘your car’, and we can easily guess that autossasi will mean ‘in your car’, with the structure of this word being auto-ssa-si, ‘car-in-your’. The plural form, ‘in your cars’ is autoissasi = auto-i-ssa-si, ‘car-plural-from-your’, again regular and transparent and – do we perhaps dare to say – logical?
Similarly, if we are told that autosta means ‘from the car’, and that -sta is the elative case-ending meaning ‘coming out from inside’, then we can easily work out that autoista means ‘from the cars’, and we can also quickly figure out what autostasi and autoistasi mean.
Turkish is another European language which works in a similar way. The Turkish word for ‘man’ is adam. Men, the (scarcely logical) plural of English man, is adamlar. Similarly, the plural of kitap, ‘book’ is kitaplar; and the plural of masa, ‘table’ is masalar. ‘Of the man’ is adam?n, employing the possessive marker -?n, so adamlar?n is transparently ‘of the men’. And, if you know that adama is ‘to the man’, then you will be able to correctly predict that adamlara means ‘to the men’.
The fact is, however, that languages are not based on logic. If they were, logicians would not have had to develop their own logical ‘languages’. Human languages have to be systematic and regular to an extent, or small children would not be able to learn them as quickly as they do, but they are not necessarily particularly rational. Languages are not logical systems; they are linguistic systems, with rules and exceptions and irregularities.