American Language

by H. L. Mencken

  Webster, in his “Dissertations on the English Language,” favored the pronunciation of either and neither as ee-ther and nee-ther, and so did most of the other authorities of the time, whether American or English. But the pronunciation of the words as eye-ther and nye-ther has been common in New England for a century or more, and at an earlier date they had been pronounced ay-ther and nay-ther, a usage still surviving in the English of Ireland. How the fashion for the eye-pronunciation arose is not known, but it was raging on both sides of the ocean by the middle of the Nineteenth Century, and is still in force. It was resisted stoutly by all the contemporary American virtuosi of language, including Edward S. Gould, W. D. Whitney and Richard Grant White. Said Gould in the middle 60’s:

  A common reply, in the United States, to the question, “Why do you say i-ther and ni-ther?” is, “The words are so pronounced by the best educated people in England.” But that reply is not true. That is to say, a majority of the best English usage is not on that side of the question. All that any man in the United States can gain by the pronunciation of i-ther and ni-ther is the credit, or the discredit, of affectation, or ostentation, — as who should say, “I know how they do it in England”; for assuredly, that pronunciation is not sanctioned by a majority of our best-educated men.63

  Whitney, in 1867, admitted the eye-ther and nye-ther were spreading in the United States, but denounced them as “the deliberate choice of persons who fancy that there is something more recherché, more English” in them.64 Seven years later he called the fashion for them “a relentless and senseless infection, which can only be condemned and ought to be stoutly opposed and put down,” and said that those Americans who had succumbed to it “ought to realize with shame the folly of which they have been guilty, and reform.”65 White, who was ordinarily something of an Anglomaniac, and strongly favored the broad a, nevertheless declared in 1870 that there was “no authority, either of analogy or of the best speakers, for eye-ther and nye-ther,” and called their use “an affectation, and in this country, a copy of a second-rate British affectation.”66 But they continued to make headway in both England and the United States. The Oxford Dictionary, in 1897, gave the preference to ee-ther and nee-ther, but admitted that eye-ther and nye-ther were “somewhat more prevalent in educated English speech.” H. W. Fowler, in “A Dictionary of Modern English Usage” (1926), predicted that they would “probably prevail,” though adding that they were “not more correct” than ee-ther and nee-ther. “Webster’s New International” (1934) held out for the latter, but gave eye-ther and nye-ther as variants. It used to be believed that they came into use in England on the accession of George I, who, speaking English badly, gave the diphthong its German value, but Dr. Louise Pound has demonstrated that this theory was nonsense.67

  Hilaire Belloc has said that “every vowel sound without exception” has taken in the United States “some different value from what it has” in England.68 This is an exaggeration, but there is sufficient truth in it to give it a certain plausibility. Even the a of such common words as cab, back and hand differs in the two countries: when Englishmen speak them rapidly they often sound, to American ears, like keb, beck and hend. In the United States, in keeping with our generally more precise habits of speech, they are pronounced more clearly. The differences between the English o in rock and the American o in the same word have long engaged phonologists. The former is described by Larsen and Walker69 as “a lightly rounded vowel, not usually found in General American Speech, though it is close to the short form of the American aw heard in the opening syllables of authentic and autocracy”; the latter is “the shortened ah-sound usually heard in what, not, as pronounced in General American Speech.” They go on:

  Cultivated English speakers do not recognize this ah-sound in the words commonly spelled in o, e.g., not, rod, rock, fog, hop, rob, pomp, on, beyond, novel; English phoneticians indeed condemn it as dialectal in these words, and recognize only the first sound described above. In American, on the other hand, both sounds are heard in all these words, the shortened ah-sound being preferred in all positions. Both sounds are heard in American speech also in the wa words, e.g., wander, want, wash, watch, swamp, swan, quarrel, squander, squalid; but here too the shortened ah-sound. is preferred.

  Yet a third sound is sometimes heard in these wa words — it is a downright aw. One encounters it in water, wash, swamp, swan and squalid, which become, roughly, wawter, wawsh, swawmp, swawn and squawlid. It also appears in God, which may be variously God (rhyming with nod), Gahd or Gawd. The first of the three, I believe, is commonly regarded as the most formal, and I have often noticed that a speaker who says Gawd in his ordinary discourse will switch to God (or maybe Gahd) when he wants to show reverence. Miss Sarah T. Barrows of the State Teachers College at San José, Calif., once determined the practice of 268 university students, all born in Iowa, in the pronunciation of watch, water and wash. She found that 156 preferred the ah-sound in watch, 96 the aw-sound and 16 the English o-sound. In water their preferences ran 128, 80 and 80, and in wash 165, 58 and 45.70 In the Eighteenth Century, as Krapp’s researches show,71 it was not unheard of for the a of care to be used in such words. Thus, the poets of the time rhymed war not only with care, but also with air, dare, glare, forbear, spare, share, blare, snare, despair, bear, bare and prepare. But the rhymes of poets are not always to be trusted, and it is to be noted that those examined by Krapp also occasionally rhymed war with car, mar, far, jar, scar and tar. In any case, it began to be rhymed with more after the beginning of the Nineteenth Century. At the present time its a is substantially equivalent to the o of story, but ah is also heard. In words containing au, aw or ou the sound is usually aw, but in the vulgar speech a flat a like that of land gets into haunt, jaundice and sometimes launch. Laundry may be lawndry, lahndry or londry. Aunt, of course, is ant to the plain people everywhere, save in the Boston area and parts of the South.

  Usage in the pronunciation of u still differs widely in the United States. Two sounds, that of oo in goose and that of u in bush, are used by different speakers in the same word. The oo-sound prevails in aloof, boot, broom, food, groom, proof, roof, rood, room, rooster, root, soon, spook, spoon and woof, and the u-sound in butcher, cooper, hoof, hoop, nook, rook and soot, but there are educated Americans who employ the oo-sound in coop, hoof and hoop. In hooves and rooves I have heard both sounds. Rooves seems to be extinct in the written speech as the plural of roof, but it certainly survives in spoken American. In words of the com-class, save company, Americans substitute ah for the u used by Englishmen; even compass often shows it. So do constable and conjure. The English are far more careful with the shadowy y preceding u in words of the duty class than Americans. The latter retain it in the initial position and in the medial position when unstressed, but omit it in almost all other situations. Nyew, nyude, dyuke, Tyuesday, enthyusiasm, styupid and syuit would seem affectations in most parts of the United States.72 The schoolmarm still battles valiantly for dyuty, but in vain. In 1912 the Department of Education of New York City warned all the municipal high-school teachers to combat the oo-sound73 but it is doubtful that one pupil in a hundred was thereby induced to insert the y in induced. In figure, however, Americans retain the y-sound, whereas the English drop it. Noah Webster was violently against it in all situations. The English lexicographer, John Walker, had argued for it in his “Critical Pronouncing Dictionary and Expositor of the English Language,” 1791, but Webster’s prestige, while he lived, remained so high in some quarters that he carried the day, and the older professors at Yale, it is said, continued to use natur down to 1839. In the South a y is sometimes inserted before a, i or o, especially following k, e.g., in cyard, Cyarter, kyind, cyow. This intrusion of the y was formerly common in New England also, and as Krapp says, “is not yet wholly extinct.” In lieutenant the Englishman pronounces the first syllable lef or lev; the American makes it loo. White says that the prevailing American pronunciation is relatively recent. “I never heard it,” he reports, “
in my boyhood.”74 He was born in New York in 1821. Nevertheless, it was advocated by Walker in 1793. The word was originally French, and loo comes closer to the original pronunciation than lef or lev. How the latter form arose is uncertain.75

  In Middle English the diphthong oi was pronounced like the oy of boy, but during the early Modern English period the pronunciation became assimilated with that of i in wine, and this usage prevailed at the time of the settlement of America. The colonists thus brought it with them, and at the same time it lodged in Ireland, where it still prevails. But in England, during the late Eighteenth Century this i-sound was displaced in many words by the original oi-sound, not by historical research but by deduction from the spelling, and that new pronunciation soon extended to the polite speech of America. In the common speech, however, the i-sound persisted, and down to the time of the Civil War it was constantly heard in such words as boil, hoist, oil, join, spoil, joist, pennyroyal, poison and roil, which thus became bile, hist, ile, jine, spile, jist, pennyr’yal, pisen and rile. Even brile for broil was sometimes noted. Since then the schoolmarm has combated the i-sound with such vigor that it has begun to disappear, and such forms as pisen, bile and ile are seldom heard. But in certain other words, perhaps supported by Irish influence, the i-sound still persists. Chief among them are hoist and roil. An unlearned American, wishing to say that he was enraged, never says that he was roiled, but always that he was riled. Desiring to examine the hoof of a horse, he never orders the animal to hoist but always to hist. In the form of booze-hister, the latter is almost in good usage. In the coal-mines of Southern Illinois hoist is pronounced correctly in hoisting-engineer, but he always hists the coal.76 Jine as a verb has retired to certain dialects, but the noun jiner, signifying a man given to joining fraternal orders, is still in common use. Most of the other vowel changes in vulgar American are also to be encountered in the British dialects. A flat a displaces the long e in rear (e.g., as a horse) and the short e of thresh and wrestle, producing rare, thrash and wrassle. In the days before the Civil War a short i displaced o in cover and e in chest and kettle, producing kiver, chist and kittle, but now only kittle is heard. Jedge for judge and empire for umpire survive more or less, but jest and jist for just are almost extinct. So are leetle for little, fust for first, sech for such, and tech for touch. But shet for shut is still in use and so is gal for girl. The substitution of guardeen and champeen for guardian and champion is very common. So is that of snoot for snout. So is that of muss for mess. In jaundice not only is the a flattened, but the final syllable becomes -ers. One stamps a letter but stomps with the foot. This last differentiation seems to have a number of parallels in English: the case of strap and (razor) -strop suggests itself at once. In vulgar American a horse chomps its bit, but champ remains a good shortened form of champeen. Similarly, a cow tromps her fodder, but a vagrant remains a tramp. By assimilation cartoon (a drawing) has been substituted for carton (a cardboard package). The last syllables of engine and genuine often rhyme with line. Webster said in his “Dissertations on the English Language” (1789) that mought for might was then “heard in most of the States, but not frequently except in a few towns.” It has now gone out, but the American Freemasons still use the archaic mote for may in their occult ceremonials.

  I have spoken of the American pronunciation of a few foreign words, e.g., piano, tornado, alma mater, the medical terms in -itis and the chemical terms in -ine. George O. Curme, a distinguished authority, says that in the plural of Latin words in -a the final -ae is “pronounced as e in react”77 but it is my observation that -ay is more often heard, and Larsen and Walker give it as a variant. Certainly vertebray is commoner than vertebree, and alumnay is at least as common as alumnee. On the level of refined discourse some effort is made in the United States to approximate the correct pronunciation of loan-words from living languages, and it would be unusual to hear an American medical man pronounce röntgenogram as if the first two syllables were runtgen — the pronunciation recommended to English radio announcers in “Broadcast English.” Even among the plain people loan-words brought in by word of mouth are commonly pronounced more or less plausibly, e.g., kosher, cabaret, buffet, chauffeur, chiffon, chef, négligé, frau, seidel, gesundheit, männerchor and café. I have, however, heard kaif for café, and among the words first apprehended in print brasseer is common for brassière, porteer for portière, jardeneer for jardinière, ratskiller for ratskeller, huffbrow for hofbräu, vawdvil for vaudeville, dash’und for dachshund, camoofladge for camouflage, shammy for chamois, fyancy for fiancée, massoor for masseur, de-bút for début, likkare for liqueur, née for née, premeer for première, meenoo for menu, switeser for schweizer, ródeo for rodéo, and coop for coupé. In the Hin-terwald the musical terms brought it by wandering performers undergo a radical transformation. Prélude becomes prelood, berceuse becomes bersoose, étude becomes ee-tude, scherzo becomes shirt-so, and träumerei becomes trowmerai. Some years ago the word protégé had a brief vogue in fistic circles, and was often used by announcers at prize-fights. They always pronounced it proteege. I once heard a burlesque show manager, in announcing a French dancing act, pronounce M. and Mile, as Em and Milly. And who doesn’t remember

  As I walked along the Boys Boo-long

  With an independent air

  and

  Say aw re-vore,

  But not good-by!

  Charles James Fox, it is said, called the red wine of France Bordox to the end of his days. He had an American heart; his great speeches for the revolting colonies were more than mere oratory. John Bright, another kind friend in troubled days, had one too. He always said Bordox and Calass.

  3. THE CONSONANTS

  The generally more distinct utterance of Americans preserves a number of consonants that have begun to decay in Standard English. The English have not only made a general slaughter of r; they also show a tendency to be careless about l, d, g and t, at least in certain situations, and even on the level of the best usage they drop a few h’s. An American always sounds the first l in fulfill; an Englishman commonly makes the first syllable foo. An American sounds the d in kindness; an Englishman doesn’t. An American sounds the final t in trait, and usually the t in often also; an Englishman makes the first word homologous with tray and reduces the second to off’n. In the United States the final g in the -ing words is usually sounded clearly, at least by speakers of any education; in England it often disappears, and indeed its omission is fashionable, and a mark of social status.78 Next after the use of the broad a, the elision of r before consonants and in the terminal position is the thing that Americans are always most conscious of in English speech. In 1913 the late Robert Bridges belabored the English clergy for saying “the sawed of the Laud” instead of “the sword of the Lord”79 and six years later he drew up a list of homophones, showing that the following pairs and triplets were pronounced exactly alike by his countrymen: alms-arms, aunt-aren’t, balm-barm, board-bored-bawd, hoar-whore-haw, lorn-lawn, pore-paw, source-sauce, saw-soar-sore, stalk-stork, taut-taught-tort, father-farther, ah-are, bah-bar-baa, taw-tore, raw-roar, more-maw, floor-flaw.80 “The majority of educated Englishmen,” says Robert J. Menner, “certainly do not pronounce the r before a consonant. Just as certainly the majority of educated Americans pronounce it distinctly.”81 John S. Kenyon estimates that two-thirds of all Americans do so.82 The violent Anglophile, Henry James, revisiting the United States after many years in England, was so distressed by this clear sounding of r that he denounced it as “a morose grinding of the back teeth,”83 and became so sensitive to it that he began to hear it where it was actually non-existent, save as an occasional barbarism, e.g., in Cuba-r, vanilla-r and California-r. He put the blame for it, and for various other departures from the strict canon of Oxford English, upon “the American school, the American newspaper, and the American Dutchman and Dago,” and went on piously: