Anyone familiar with the French title (L’Étranger) of Albert Camus’s famous novel The Stranger, also known as The Outsider, would be right to assume an etymological connection between étranger and stranger. While Old French estrange became étrange in modern French, it lost the first ‘e’ entirely in English, to become strange.
This is an example of aphaeresis or apheresis: the dropping of an initial sound or sounds of a word. Despite its uncommon name, the process is familiar. It’s what lies behind the shortening of especially to ’specially, because to ’cause (also spelt cos), espy to spy, esquire to squire, and alone to lone. As you can see, what’s lost is often an unstressed initial vowel – this is a particular type of aphaeresis known also as aphesis.
Though it’s essentially a phonetic shortcut, what happens in speech tends to manifest in writing. Poets are fond of aphaeresis because it lets them manipulate prosody better. This is why in many poems you’ll see upon appear as aphaeretic ’pon, amid as ’mid and it was as ’twas. Skilled poets control metre with surgical precision, and indeed aphaeresis – from a Greek word for taking away – can also refer to both amputation and the removal of blood.
When linguistic aphaeresis is under way, an apostrophe may stand in for the missing letters, as in some of the examples we’ve seen. Sometimes it stabilises there, as with ’til for until and ’cept for except. (Note that although ’til is an abbreviation of until, till is not – it predates until.) And sometimes the apostrophe disappears outright, as in violoncello → ’cello → cello.
Sometimes the result is two distinct words, as in amend and its aphetic product mend. Ticket comes from etiquette, which once referred to a little note. And did you know fence comes from defence and once meant ‘the action of defending’? Another outcome is that a sound may fade through aphaeresis while the letter associated with it remains as a vestige of the older pronunciation, as with the silent ‘k’ of knife, knight, knock, and knee.
Other examples of aphaeresis include round from around, ’tention from attention, possum from opossum, drawing room from withdrawing room (a room to withdraw to, not draw in), and cute from acute, whose etymology I’ve noted in a post about gender-skewed words. Even that everyday contraction, I’ve, is an example of aphaeresis, according to David Crystal’s Dictionary of Linguistics and Phonetics.
Should we start calling them phesis and phaeresis?
Email this Post
Fascinating stuff Stan, as always, and a new term for me. I was wondering if the same term would apply to a word like orange (perhaps it’s the only one…), where the initial ‘n’ of the original naranj/narang has disappeared over the course of time. And what about ‘newt’, where the ‘n’ has detached itself from the article (it was originally an ewt, or so the etymological dictionaries assure me) and joined itself onto the noun? is there a name for that?
Thanks, Liz. The term I normally see used for that phenomenon is rebracketing; its Wikipedia page provides other names. I don’t think orange and co. count as examples of aphaeresis, because the “n” sound didn’t disappear the same way (and didn’t disappear at all when we say “an orange”).
I was going to offer metanalysis, which is another term for it, coined by Jesperson about 100 years ago. OED has an entry for metanalysis, though not for rebracketing. According to google ngrams, rebracketing as a term starts appearing in 1950, and by the 1980s was closing in on metanalysis, but round about 1989 metanalysis began a massive spike which is still climbing, while rebracketing began tailing off. The growth of metanalysis may be due to the word acquiring a wider range of meanings, or because google sees it as being the same as meta-analysis.
Thanks, Stephen. Metanalysis is also the term used in Burchfield’s revision of Fowler; he describes it as a philological term and gives the example of adder from Old English nǣdre. The Oxford Dictionary of English Grammar repeats Burchfield’s entry (or vice versa) and notes its coinage by Jesperson (1914) and the verb form metanalyse “alter by metanalysis”. Maybe I should make it my default term for the process rather than rebracketing, which I prefer solely for aesthetic reasons.
Interesting. I wonder why some prepositions got the aphaeresis treatment and not others? Why, for example, were ‘neath (from beneath) and ‘tween/’twixt (from between/betwixt) used, but not ‘low, ‘side or ‘yond?
Elizabeth: Good question. Maybe it has to do with clarity: while there’s no danger of misanalysing ‘neath or ‘twixt, clippings such as ‘low and ‘side are less distinctive and therefore more potentially confusing. The OED shows that yond (also yont) had some currency as an aphetic form, especially in poetry.
Just as estranger became stranger, the french napperon became the english apron. But do you know examples of aphaeresis going the other way round ie from english to french?
Bruno: I don’t think apron is a case of aphaeresis, for reasons I gave in a comment above.
That’s a good question about examples going in the other direction; I can’t think of any offhand, but maybe readers can suggest some.
If apron does not fit then what about esquif which turns skiff on reaching Dover?
Another example I am thinking of is the cartoon character Andy Cap. But is it really an aphaeresis as Andy would drop the h, wouldn’t he?
Bruno: If you mean the boat, skiff probably comes from French esquif, as you suggest, though the OED says it might also be from similar or identical words in Spanish, Portuguese, or Italian.
“Andy Cap” is more a play on words. I’ll be looking at h-dropping and h-adding on my own blog soon.
Estrange is still with us as a verb, of course, and still fairly close to its original meaning ‘to cause to be strange, or a stranger, or as a stranger (to)’ (OED sense 1).
John: That’s true, of course, and it’s a word I’ve always liked.
That may be stretching it a bit but what about “diamond”? As the adjective “adamant” recalls diamond stems from the greek word “adamans” which itself can broken in “a-” privative prefix and “damans” to tame. By some aphaeresistic move the word has lost its prefix in most languages except in Russian and Persan (almaz) (a souvenir of Alexander?) .
Bruno: Yes, diamond is related to adamant through Latin. Apparently diamond comes (via Old French) from Latin diamantem, which lost its initial ‘a’ under the influence of various technical Greek words beginning with dia-. I don’t know if it was aphaeresis.
[…] century. Apparently it comes from a Scottish variant of effect, so it’s a good example of aphesis (also aphaeresis): the loss of an initial sound in a word. Here are the three senses included in the […]
Several people asked about words like “newt” and “apron.” These are not examples of apheresis. This phenomenon occurred during the Middle English period, when the n was lost on an indefinite article before a consonant. So Old English min faeder became Middle English my fader, but myn eye remained. This caused a great deal of confusion for English speakers during this time. The n sometimes transferred from the article to the noun, or vice versa. Thus an ewte became a newt. A napron became an apron. This is also where we get nickname from–an ekename.