Oxford University Press's
Academic Insights for the Thinking World

Drinking Up Eisel, Or,
the Oddest English Spellings (Part 9)


By Anatoly Liberman

Bishop John Wilkins (1614-1672), a renowned man who regularly preached before the king himself, had multifarious sensible ideas, as one can judge by reading his works. A discovery of a new world, or, A discourse tending to prove that ‘tis probable there may be another habitable world in the moon: with a discourse concerning the probability of a passage thither… (we, postmodernists, love “discourse,” don’t we?) and Mercury, or, The secret and swift messenger shewing, how a man may with privacy and speed communicate his thoughts to a friend at any distance (this is what I do every Wednesday with the help of this blog). However, our readers are probably familiar only with his treatise Of the principles and duties of natural religion. Bishop Wilkins believed that English spelling is an appendix to the curse of Babel, and many wise and learned people shared his opinion. The very spelling shewing proves him right. (Shew survived the 19th century. Among the famous modern writers G. B. Shaw never wrote show. The reason for this strange spelling will be explained at some other time).

English has several diphthongs and a considerable number of digraphs. A digraph, it will be remembered, is a combination of two letters designating one sound, for instance, th, sh, ch, ee, and ea. Some English digraphs are particularly charming because they designate nothing: neither sound nor fury. Such is gh in eight: eit and ate would render the same pronunciation, but that would be depressingly easy. Eight is not an isolated example. Before gh, ei is the norm: consider weigh, neigh, inveigh, and their ilk. It also occurs before gn (with its useless g), l, m, and nt (in deign, feign, veil, vein, feint, and so forth), but there it has competitors, as way, whey, Dane, ordain, fain, fane, avail, vane, and faint “shew.” We can add obeisance (akin to obey) and sheik (a homonym of shake, transliterated from Arabic) for good measure: in both, ei turns up for a reason. Eisel is “vinegar” and is remembered, if at all, only because Hamlet asked Laertes what he would have done for Ophelia: “Woo’t drink up eisel? Eat a crocodile?” Eisel turned up in written English three centuries before Hamlet, and if it had survived, it would have been the only word of its structure with ei; but it did not.

Real trouble begins elsewhere. Why heifer, if it is pronounced heffer? In Old English (the earliest recording goes back to the year 900), heifer was spelled heahfore, heahfru, and heahfre. The word must have been transparent to those who coined it, but it is not to us. Later heyfer ~ hayfor (the continuation of heahfore) alternated with hekfore (apparently, its doublet, extant in nearly the same form in modern British dialects) and heffre, the result of vowel shortening. As far as I can judge, the most reasonable hypothesis explains heah- (from hag-, allied to hedge, and haw-, the first component of hawthorn) and heck- (still occurring in vernacular English) meaning “enclosure” and “rail” respectively, and -fore “occupant.” Allegedly, young female cows were kept separately from bulls. Be that as it may, the modern form reflects the pronunciation of heffre but the spelling of heyfer. Other monsters that combine the pronunciation of one dialect and the spelling of another are busy, bury, and build. Heffre is the product of vowel shortening. The same sound change happened in leisure, borrowed from French, but not universally. That is why in British English, leisure rhymes with pleasure, whereas in American English it still has the vowel of lesion. To the shortening of the old diphthong we owe the modern pronunciation not only of says, said, and again ~ against (in which ai is spelled) but also of Leister and nonpareil.

The digraph ei is familiar in French words ending in -ceive. Their redeeming quality is that the related nouns—reception, deception, etc.—tell us that we should spell ei, not ie in them. But when hearing sieve, with its unpredictable short vowel, and frieze, one would as lief write sive (on the model of give, live, talkative) and freeze. We have to reconcile ourselves to the difference between perceive and believe, though they have the same vowel under stress. Ceiling is a Romance word, but its origin is obscure; it has not always been spelled with ei (the earliest recorded forms are celing, siling, seeling, sieling, and others). Ceiling looks as though it were derived from French ciel “sky,” but this derivation is far from certain.

Perhaps the most vexing word with ei is either (neither is either with a negation), for ei seldom has the value of “long i.” It occurs in a few borrowed nouns and coinages by individuals (for instance, in eider, as in eiderdown, and eidolon “unsubstantial image,” Carlyle’s invention), and once again before gh (in height and sleight). Either is a special case. It started as a compound made up of three elements, a fact made especially clear by its Old High German cognate eo-gi-hweder. They correspond—element by element—to English ay “ever,” the extinct prefix ge- (one can see its obliterated remnant in the Chaucerian past participle y-clept “called,” a favorite of thick dictionaries, as well as in e-nough: compare it with German ge-nug), and whether. Later those elements grew together: the diphthong was monopthongized (“smoothed”) and the long vowel was sometimes shortened. Several variants emerged. The American one is, as usual, more archaic. The two pronunciations of either are among the best-known shibboleths of the main regional varieties of English. At present, some people in the United States use the British variant, but most do not. In 1870, an American linguist wrote of either with the vowel of I that it is “an affectation, and in this country, a copy of a second-rate British affectation.” We have come a long way from such strictures.

Thus we have ei in eight, height, ceiling, and heifer. In each case, it has a different value. This is why, when foreigners come across ceiling, Leister, and either, they cannot guess their pronunciation, and, conversely, if they learned the language by word of mouth, they have no idea how to spell veil, avail, vale, sleight, weight, and the rest. Poor foreigners, but poor native speakers.

Anatoly Liberman is the author of Word Origins…And How We Know Them. His column on word origins, The Oxford Etymologist, appears here each Wednesday. Send your etymology question to blog.us@oup.com; he’ll do his best to avoid responding with “origin unknown.”

Recent Comments

There are currently no comments.

Leave a Comment

Your email address will not be published. Required fields are marked *