back as the orthoepists will help us, we find these words already exceptional. Hart writes dubl, kuntre, tuch,* kura}, diskurs, with u = (u), I believe, and ụ = (uu); but he never writes pund, muth, pronuns (i.e. pronounce), kunsel, konfund, dut, for pound, mouth, &c. Gil writes trubl and flurish; but croun, doun, loud, proud, kloud. And in many other words the evidence of early orthography combined with that of modern pronunciation-one portion of the evidence coming part way to meet the other-inclines me to think that from some cause, not always easy to ascertain or even conjecture, it became the fashion to spell certain words in a manner which in reality at no period represented the sound. On would, should, could, I will remark below, as also on show, mow, blow, slow, &c.: see §§ 57 and 47. But besides these, there are other words in which I doubt whether the ou was ever sounded (ǝu), as youth, young, couple, souper (Chaucer, also soper supper), source, bourne, mourn, mould, &c. Saxon accents: the spoken words ed. = 45 Disuse of Anglo- As to the accents on i and u the facts appear to be these. There were in Anglo-Saxon classes still distinguish of words different, though containing the same written vowel, and frequently, though not habitually, distinguished, namely by accents;† and these appear even in the earliest MSS. we possess, the words being distinct from those with i, the ú words from those with u, and so on. In course of time these accents ceased to be written, which may not improbably have been because the scribes, accustomed to write Latin and French without any such diacritic signs, disliked the look of them, seeing clearly that accents at once stamped their fair calligraphy with an * In one place he writes touch, either from force of habit, or else it is a mere misprint. Unfortunately such misprints are but too common in most early works of this class. † Of course I do not mean that the accent was not very often omitted. It was in fact most commonly omitted in many or most MSS.; less no doubt through simple carelessness on the part of the writer, than because of the reader's supposed familiarity with the word. But still in certain words it very often appears; and it is a rare thing to find a word written with an accent which has no claim to one. appearance of vulgarity by showing so conspicuously that the writing was only in the language of the profanum vulgus. But whatever the cause was, the fact remains that accents disappeared: very few are to be found in the Ormulum, none in La3amons Brut, the Ancren Riwle, and all later writers. But before long some of the same classes of words are again found distinguished in writing, and even more regularly than before, though with a different distinction. No substitute was provided for the accented í; but we find the words which in Anglo-Saxon had ú afterwards written with ou or ow, apparently for no other reason than to distinguish them from the u words; this new orthography coming into use probably about the close of the 13th century. Who first introduced it, or in what part of England it arose, I have not had time to investigate, if indeed these questions can now be answered; but whenever it appeared, it was only a new mode of representing a difference of sound which itself was as old as the language. 46 We proceed now to consider two other classes of words. And let us approach them on the side of the symbols, thus. The A.S. & be- I turn to the A.S. dictionaries, Bosworth's came o. and Grein's, and going right through them I find the following list of words spelt with á-a complete list, I believe, of all the words with á (excluding derivatives) that survived to or beyond Chaucer's time, except acsian, which Grein is in error in accenting, and with five other apparent exceptions which I will mention : á, ác, án, ár, ágan, ágen, bán, bát, bláwan, brád, cláð, fá, fám, flá, gá, gár, gást, gát, gráf, gránian, grápian, hál, hálig, hám, hár, hás, hát, hláf, hálford, hwá, lá, lám, lár, láð, má, mára, máwan, ná, nón, rád, ráp, sár, sáwan, sáwel, sceáwian, snáw, stán, swá, swápan, tá, tácn, twá, þá, wá, wár, wráð; also certain vernal preterits as arás, bát, glád, sceán, wrát. The later forms of these words are-o (= ever), oak, ōne (proare—o nounced as we still sound it in the compounds alone, atone, and only; though it has passed through the form of oön into wun), oar, own (verb and adj.), bone, boat, blow (Lat. flare), broad (which we now call brawd), cloth (now 47 cloth), foe, foam, flo (= arrow), go, goar, ghost, goat, grove, groan, grope, whole, holy, home, hoar, hoce (as it is still called in Devonshire, though polite English has corrupted in into hoarse), hōt (which since Chaucer's time we have shortened into hot), loaf, lōrd (now lord), whō-now (Huu), but in Chaucer's time (whō) rhyming with mo,—lo, loam, lore, loth, mo, more, mow (vb.), no, nōne (now pronounced nun), road, rope, sore, sow, soul, show, snow, swope-now swoop-stone, so, toe, token, two-now (tuu)*-tho, woe, ore-weed (a term still used in Devonshire for sea-weed), wrōth (now more commonly wroth); and the preterits arose, bote (from bite), glode (from glide), shōne (now shon), and wrote. Apparent ex Old Norse words, The five words excluded from the list are ceptions really those which the Old Norse has modified or notAnglo-Saxon. superseded, just as the above words in the Scottish dialect,—ane, ain, aik, hame, rape, bane, stane, &c. -are not really modern forms of the Anglo-Saxon words above quoted, but of the Old Norse einn, eigin, eyk, heimi, reip, bein, steinn, &c., with ei = (eei). Our five words are spátl, which the Old Norse spýta—now pronounced (spiita) but of old probably (spyyta)—has transformed into spittle; and swán, swát, wác, wáfian, which the Old Norse sveinn, sveiti, veikr, and veifa have ousted altogether, becoming swain, sweat (sweet) now (swet), weak (week) now (wiik), and waive and waver.† dinavian element dialects. 48 Mr. Murray's Mr. J. A. H. Murray says, "There seems view of the Scan- ground to regard many of the characteristics in the Northern of the northern dialect which currently pass as Danish as having been original elements of the North Angle speech, due to the fact that this dialect was, like the Frisian, one which formed a connecting link between the Scandinavian and Germanic branches. Such * It is doubtless the influence of the w preceding that has changed the sound of (o) into (u) in who, two, swoop; while (oon) changed into (wən) finds its exact a alogue in wuts (wǝts) as the Devonian form of oats. + Slapan had the by-forms slæpan and slépan, the last of which alone has survived. D 49 50 characteristics would of course be strengthened and increased by the influx of Danish and Norwegian settlers, but the influence of these was necessarily at first confined to particular localities, and only gradually and at a later period affected the northern dialect as a whole."* These views are probably correct; but there can scarcely be a doubt that in England south of the Humber the forms spittle, &c. were due to the influence of the Danish invaders rather than to that of the Northern Angles, unless indeed we extend Mr. Murray's hypothesis to the whole of the Angles, instead of limiting it to the northern division. Inasmuch then as, with only these five exceptions so easily accounted for, all the Anglo-Saxon words in á which survived to or beyond the age of Chaucer are now pronounced, according to the tradition of all our Southern dialects (for I resolutely hold to this argument), with (0); and there is no reason to suspect that there has been any change since Chaucer's time; and in Chaucer too these The (0) sound words rhyme with French words like chose, or confirmed by words from the French like rose and suppose; Italian. nor is there any reason to suspect that the French chose, rose, &c.—especially as confirmed by the Italian cosa, rosa, &c.—have failed to preserve at least approximately the true ancient sound of their principal vowel; we seem to have pretty good ground for concluding that these words in the 14th century were sounded with (o); and there is no sufficient evidence that they were not sounded exactly the same in the earliest English. French and Chaucer's gat- (Consideration of the pronunciation of the toothed. Anglo-Saxon á will help us to decide the meaning of Chaucer's much disputed epithet of the Wif of Bathe-gattoothed; at least it enables us decisively to set aside the explanation of the word as signifying goattoothed, whatever that may mean. Gát (goot) would never be shortened into gat (gat), but into got (gɔt), whereas all the MSS. appear to have gat or gate. The true sense is gate-toothed, where however we must bear in mind that *Dialect of the Southern Counties of Scotland, Historical Introduction, p. 24. gate, from go, originally means, not a wooden barrier, but a passage: see my edition of the Castle of Love, Gloss. s.v. 3at. The compound signifies therefore that the "worthi womman" had teeth, not set in close rank, but with gateways, interstices, between them. I am glad to see that Dr. Morris similarly explains the word.) words in 6 sound ed with (uu), 00. 51 Anglo Saxon Let us go again to our dictionaries. Now we find another set of words with 6, of which modern English the following is, I believe, a complete list of such as reached Chaucer; blód, blówan, bóc bóg, bósm, bót, bróc, bróðor, cóc, cófa (?), cól, dóhtor, dóm, dón, eógoð, eóh, eów, flód, flór, flówan, fóðor, fóstur, fót glóf, glóm, gód, gós, grówan, hóc, hóf, hóp, hróf, hróst, hwópan, lócian, mód, móður, mónað, mór, nón, nósu (?), óðer, pól, ród, Róm, róse, sceótan, scólu, sóð, sófte (?), sóna, stól, stów, tó, tóð, wód, wóh, wrótan; and the preterites forsóc, sceóc, stód. Of twenty-six of these the 19th century representatives are-boot (“it boots not"), cool, doom, do, youth, yew, you, gloom, goose, hoof, hoop, roof, roost, whoop, mood, moon, noon, pool, rood, school ("a school of mackerel"), sooth, soon, stool, to, tooth, and root ("to root up"), all with (uu); twelve others we pronounce with oo (u)-book, bosom, brook, cook, foot, good, hook, look, shoot, forsook, shook, stood; seven others have the ŭ (→) of but-blood, brother, flood, glove, mother, month, other; and of these nineteen ten are found in the Ormulum, all with the long vowel. Of the remainder two (wód and wóg) are now obsolete; of two (bóg and dóhtor) the guttural following, which has now disappeared, has disturbed the vowel, so that from the sound of bough or daughter we can conclude nothing. The few that remain-blow (“full blown"), flow, grow, slow; fother, foster (child); floor, moor; cove, nose, Rome, rose, soft, I have not time to discuss, beyond observing that we know Rome, which was Rume also in the Chanson de Roland, to have continued as (Ruum) down to Shakespeare's time. The forty-five words already discussed, to which, judging from analogy, hód, hróc, and sceó should be added, are sufficient for |