Skip to main content

Hello. It looks like you’re using an ad blocker that may prevent our website from working properly. To receive the best experience possible, please make sure any blockers are switched off and refresh the page.

If you have any questions or need help you can email us


The language of leaping to conclusions

American jazz musician Sonny Simmons plays a cor anglais - Credit: Getty Images

PETER TRUDGILL on how easy it is to make false assumptions about connections between words 

Etymology is the study of the origins of words and the development of their forms and meanings. It is a scholarly activity which needs considerable knowledge of large numbers of languages, plus language history, linguistic change, historical linguistics, semantics (the study of meaning) and phonology (the study of sound systems). It is no job for amateurs.

Without a great deal of study, research and erudition, there are many traps which amateur etymologists can fall into.

It seems rather obvious that German haben, ‘have’, Dutch hebben, ‘have’, and English have are all related to Latin habere, ‘have’, and hence to Anglo-Norman aver, Italian avere and French avoir.

It may seem rather obvious, but it is totally wrong – there is no historical connection between haben and habere. The resemblance is a complete coincidence: these words do not have a common origin.

It also seems rather obvious that the verb have cannot have any historical connection to heave – the two words have such very different meanings, even if they do look rather similar. Once again, however, this is completely the wrong conclusion to draw: have and heave are genuinely related in a way that have and habere are not.

Professional etymologists are very well aware that haben and habere cannot be related because they know that Ancient Germanic words beginning with h correspond to Latin words beginning with c (pronounced k), as well as to words beginning with the k sound in other Indo-European languages, as in English horn, Dutch hoorn versus Latin cornu, Welsh corn.

This is because of an ancient sound change which took place in the Proto-Germanic language whereby original k-sounds changed to h, as also in Latin cordis versus English heart. We can see this too in English-Latin-Welsh triplets such as hound – canis – ci, and hundred – centum – cant.

On the other hand, again illustrating the ancient change from k to h, English have truly is historically connected to Albanian kap, ‘grab’, as well as to Latin capere, ‘to take’, from the original Indo-European kap- ‘grasp’.

Capere is the form which our word capture comes from, showing that it is important for etymologists to be able to distinguish between words which are inherited and those which have been borrowed. Have is a modern English word because we inherited it from Indo-European kap- via Proto-Germanic habanja, which was passed down and changed through natural transmission from one generation to another over millennia.

Capture was a French word because French had inherited it from Latin captura, as this too was passed down from one generation to another. But it is a modern English word because, in the 16th century, anglophone people decided to borrow it from French and started using it in English.

Have and heave are related because they both go back to the same Indo-European kap-, with a meaning-shift in the case of heave from ‘take’, via ‘take hold of’, to ‘take hold of and lift’.

The moral is that etymologists must avoid guesswork based on flights of fancy and chance similarities. In his fascinating book Word Origins and How We Know Them, Anatoly Liberman shows that, although Latin cura, ‘care’, and English care are very similar, they are only coincidentally so.

These words cannot be historically related to each other because of the k to h sound-change, although they could of course have been connected if English had borrowed the latter from the former – but that is not what happened. Rather, care was inherited from Proto-Germanic karo, ‘sorrow, cry’, which goes back to the same Indo-European root as Latin garrio, ‘chatter’, hence Latin garrulus, ‘talkative’, which English has borrowed – not inherited  – as garrulous ‘too talkative’. So care does not have any connection to cura, however promising that hypothesis might seem; but it does have a link, however improbable, to garrulous.

Cor anglais

Cor anglais and French horn are the English names for two different orchestral instruments, the former a double-reed woodwind instrument, the latter brass. Cor anglais is literally ‘English horn’ in French, and it is also the French name for the instrument. But the French for ‘French horn’ is not cor français but cor d’harmonie.

What do you think? Have your say on this and more by emailing