The Short and the Long of it
A BRIEF HISTORY OF THE WORDS AGAIN AND AGAINST,
WITH AMIDST, ACROSSED, AND WHILST BEING THROWN IN FOR GOOD MEASURE
By Anatoly Liberman
There are two questions here. First, why does again rhyme with den, fen, ten rather than gain? Second, where did -t in against come from? I’ll begin with against.
Old English had a ramified system of endings. The most common ending of the genitive was -s, which also occurred as a suffix of adverbs, or in words that, by definition, had no case forms at all (adverbs are not declined!). It is easy to detect the traces of the adverbial -s in such modern words as always, (must) needs, and nowadays, and its obscure origin will not interest us here. Some adverbs ending in -s were used with the preposition to before them, for example, togegnes “against” (read both g’s as y) and tomiddes “amid.” They competed with similar and synonymous adverbs having no endings: ongegn and onmiddan. As a result, the hybrid forms emerged with -s at the end and a “wrong” prefix: ageines and amides. Initial a- in them is the continuation of on-; hence our modern forms against and amidst. So far, everything is clear. The tricky part is final t.
Obviously, this -t has no justification in the early history of either against or amidst. According to the usual explanation, both words so often preceded the definite article the that -sth- was simplified (“assimilated”) to st. This explanation is plausible. By way of analogy, it may be added that a similar process has been postulated for the verb hoist. It surfaced in English texts in the 16th century in the form hoise, and all its native predecessors and cognates elsewhere in Germanic look like it. Perhaps the infinitive was changed under the influence of the preterit and the past participle (as in the now proverbial to be hoist with one’s own petard), but, not inconceivably, in the phrase hoise the flag the same process occurred as in ageines the/amides the. (I have seen the conjecture that -st goes back to the superlative degree of adjectives. This reconstruction is fanciful.)
However, t also developed in words that did not always precede the definite article. Thus, earnest “pledge money,” a noun with a long an intricate history, was first attested in the form erles. Here the influence of the all-important adjective earnest, from Old Engl. earnost, should not be ruled out. Tapestry is another borrowing from French (tapisserie; compare Engl. on the tapis, literally, “on the table cloth,” a calque of French sur la tapis: tapis and tapestry are of course related). The inserted t in tapestry is called a parasitic sound that developed between s and r—not much of an explanation, even though it is possibly true. Whilst is from whiles, which, like against, had no historical -t. Both words seem to have developed along similar lines, but the troublesome consonant may be “parasitic.” The same holds for -st in amongst. What matters is that when such words as against and amidst struck root in the language, they were analyzed by speakers as again and amid plus a suffix, and such “illegitimate” adverbs as dialectal onct (that is, once-t) and the well-known acrossed emerged. Their history is instructive from the point of view of what the Standard accepts and what it rejects. Against is a bona fides adverb, for it has a meaning different from again. Amidst and amongst are bulky and rather useless synonyms of amid and among. Whilst, which is much more common in British than in American English, sounds, I think, a bit snobbish even in England. Acrossed, though common, remains an object of mild derision. Once-t is regional and nearly incomprehensible. This shows that the norm (the Standard) is capricious, for from a phonetic point of view all the words mentioned above make sense and are easily pronounceable. (Here I would like to answer one of our correspondents who questioned the idea of the Standard and asked: “But does the Standard exist?” Oh yes, it does. Just try to teach English, whether to native speakers or foreigners—especially foreigners, and you will immediately discover that you have a clear idea of what is right, what is wrong, and what is optional in “good English,” that is, in your own educated variety, which may not be too different from that of you neighbor.)
We can now turn to again. The relevant Old English forms were ongægn and ongegn, both with a long vowel in the root. As noted, the letter g in them had the value of Modern Engl. y, and in Middle English we find agen, ayen, again, and agein, among others. In all of them g designated y. Later the native forms were ousted by their Scandinavian close cognates, which had g rather than y. But for the Vikings’ raids and the Danish domination over Britain in the Middle Ages, the Modern English verbs get and give, to cite just two most anthologized examples, would have begun with the sound of yield, which has managed to retain its ancient pronunciation. Let it be remembered that Old Engl. ongægn and ongegn had long vowels in the root. The processes that led to the shortening of long e are not quite clear. Perhaps the word’s occurrence in an unstressed position played a role. Perhaps the vowel first lost its length in the preposition against because it stood there before three consonants (even two would have been enough!), and again followed suit. In any case, in the 18th century the so-called orthoepists (reformers who tried to make English spelling correspond more closely to the current pronunciation—apparently, a hopeless endeavor) unanimously recommended the variant agen. Shakespeare used “rhyme to the eye” so often that his practice is not a good guide to the details of his pronunciation, but in some cases there is no doubt that he said agen. It seems that the variant with -ai-, far from continuing Middle Engl. ayen, was the product of a new lengthening and should therefore be listed among innovations. Many 20th-century dictionaries give both ai and e in again/against as equally acceptable, but in American English, which, like any colonial language, tends to preserve a more archaic norm, only agen and agenst seem to be heard. Some older dialectal variants of again give one pause, especially the one traceable to the vowel of father, but they may be ignored in the context of the present discussion.
By way of conclusion, here is a short etymological note. The root of again is the same as in Old Icelandic í gegn and Modern German gegen ~ entgegen “opposite,” all pronounced with hard g. The Old English adverb meant “in the opposite direction” and “all the way back.” In Middle English, it came to mean “in return.” But the only now familiar sense “once more, anew” goes back to the 14th century at the latest.
Anatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to him care of email@example.com; he’ll do his best to avoid responding with “origin unknown.”