By Anatoly Liberman
Why don’t good and hood rhyme with food and mood? Why are friend and fiend spelled alike but pronounced differently? There is a better way of asking this question, because the reason for such oddities is always the same: English retains the spelling that made sense centuries ago. At one time, the graphic forms we learn one by one made sense. Later the pronunciation changed, while the spelling remained the same. Therefore, the right question is: What has happened to the pronunciation of the words that give us trouble?
Everybody has heard that vowels can be short or long, but those who have not studied Greek or Latin should beware of such terms. According to the formulation English speakers learn pat, pet, pit, and pot have short vowels, as opposed to pate, Pete, spite, and spoke. However, those are conventional terms inapplicable to today’s pronunciation. Some idea of what length or at least duration means can be gained from words like father and spa. They indeed have “long a.” Flaw and saw have “long o” in the pronunciation of the speakers who distinguish between Shah and Shaw; too has “long u.”
In ancient Indo-European (which includes Germanic), a long vowel functioned as a combination of two short ones. Then, not earlier than the thirteenth century but hardly later than the fifteenth, the situation in English changed, and vowels became indivisible, with their duration depending on how the postvocalic consonant adhered to them. If the contact was close, the vowel was pronounced short, but if it was loose or if there was no consonant following a vowel, length prevailed. Although untrained speakers of British English rarely associate the difference between bid and bead, cod and cord, let alone bud and bard, with vowel duration, the difference is significant. Speakers of American English have still more trouble here. Even Germans, in whose language the situation is roughly the same as in English, complain that the recent spelling reform disturbs them when they have to decide whether the previous vowel is short or long.
Linguistic revolutions occupy a few pages in elementary textbooks. In real life, they entail vacillation, acceptance of a new norm, returning to the conservative variant, and multiple inconsistencies. This is what happened in English. The choice of the vowel-consonant contact was not regulated by strict rules, though some tendencies can be detected even after so many centuries have passed. In Old English, good, hood, and mood had the same vowel (approximately as in Modern Engl. saw) and formed a perfect rhyme. That state of undisturbed harmony was ruined by the introduction of the new regime: every word had to choose its type of contact.
Apparently, the choice was often arbitrary. Good and hood acquired close contact, and the vowel shortened, while mood ended up with a loose contact and, consequently, a long vowel. Some words develop fast, others are slowcoaches. When the so-called Great Vowel Shift struck (this happened in early Modern English), the original vowel of good, hood, and mood (to repeat: it resembled aw in Modern saw) acquired the value of Modern English too. As is well known, in the north of England both words in the phrase shut up are pronounced with the vowel of put, that is, shoot oop, unlike what happens in the south. But Standard English also lacks consistency: put and cut do not rhyme; bull and bulb have different vowels. It seems that the vowel of blood changed from what we today hear as aw in saw to “long u” as in Modern woo so early that it became short in time for changing to the vowel of Modern cut. Therefore blood (similarly, flood) now rhymes with bud. The spelling of good, mood, and blood takes no cognizance of the changes that have happened in those words. On paper they still occupy the same chamber as they did at the court of King Alfred.
Scholars have been describing English pronunciation since the seventeenth century, and we have detailed evidence of the great vacillations in the way all such words were pronounced in the not too distant past. It always comes as a surprise that some variants, vastly different from those we now hear, existed not a thousand years ago, but in the days of Fielding, Dickens, and even later. In Middle English, friend rhymed with fiend. Yet, unexpectedly, close contact set in in it before nd, a consonantal group that at an earlier epoch caused lengthening (that is why bind and bound rhyme with lined and crowned: without lengthening they would have rhymed with dinned and the first syllable of Tundra). Close contact meant shortening. The two words are still spelled alike, though only fiend has retained loose contact.
With time, the visual image of the most familiar words becomes so familiar that we stop noticing the incongruities. Said, plaid, says, done, gone, one, and dozens of their linguistic cousins are hieroglyphs to be learned individually. Our reverence for the past will probably never allow us to switch to sed, sez, plad, or, God forbid, nun (the latter for none). What a devilish plan to separate friend and fiend, even though living speech has separated them quite successfully! Frendship will certainly be less precious without a superfluous letter.
By way of conclusion, here is a curious example of interplay between shortening in a dissyllabic word and folk etymology. Once the word Christscrosse existed. It meant “figure of a cross.” In compounds and long words the first vowel was regularly shortened, which explains the difference in the modern pronunciation of holy and holiday, south and southern, white and Whitsunday, Christ and Christmas. The first vowel in Christcrosse developed as in Christmas. That word also denoted “alphabet,” because the figure of the cross was used in front of the alphabet in hornbooks and primers (early primers consisted of a single page protected by a sheet of transparent horn). The word lost its religious significance to such an extent that the pronunciation and spelling crisscross obtained and the compound was understood as a reduplication like trictrac, flip-flop, and so forth; hence the change of sense.
Those who were impressed by the story told here may ask themselves how they pronounce the word primer. Does it rhyme with dimer or dimmer in their speech? Cogitation and an attempt to square accounts with their phonetic conscience will allow them to spend a pleasant week in anticipation of the next post (which will have nothing to do with spelling).
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.”