The pronunciation of the weak'e in name, &c., is uncertain, and it is safest to pronounce it like the e of men; it probably varied between the e of father and that of fishes. When -es rhymes on ys, &c., it must, of course, be pronounced -is.

The distinction between ei, ei, gi is not quite certain, and it is possible that they were all levelled under the first. So also with ễu, gu, and with ōu, qu.

The distinction of narrow and wide is inferred from phonetic

transcriptions of the beginning of the sixteenth century—the Welsh hymn to the Virgin (in Trans. Phil. Soc. 1880-1) being apparently still olderin which wide i is identified with the Welsh y. These transcriptions show that i was narrow, =(i), before the back conss. k, g, ŋ (as in sing) and before gh, and when final, and was wide, = (i), elsewhere. Hence we may with some confidence assume that in Chaucer's pronunciation the i was narrow in such words as thikke, six, big, kyng, nyght, many, and probably also in such a word as pitee, where an unaccented i is followed by a cons. beginning an accented syllable; and that it was wide in such words as wille, is, sitten, snibben. There is no direct evidence of a parallel distinction in u and ü, but these vowels no doubt followed the same law as i.

The distinction between ai and ei, as in day and wey, was probably still kept up in Chaucer's pronunciation, but the two diphthongs were beginning to be confused, probably through the a of ai being modified nearly to the sound of our vowel in man. ei probably had the broad sound of the diphthong in the Cockney pronunciation of name.

ē and ō were probably pronounced very close, approximating to ī and i respectively.

?= lengthened Old English o, as in broken, bifore, perhaps had a different sound (closer ?) from that of 9=O. E. a, as in hom.

noght, &c., is often written nought; the former spelling is adopted here to prevent confusion with ou= =(uu), this diphthong ou only occurring before gh.

As regards the consonants, it is necessary to pronounce the doubled ones really double, sonne 'sun' as in 'penknife' not as in 'penny.' So also in alle, sitten. ssh between vowels, as in fresshe, seems to mean (ss), but the doubling of the s when the sh is final, as in fissh, was unmeaning. The doubled cons. in French words, such as tirannje, do not necessarily imply that they were pronounced double: at any rate they were not pronounced double in French.

Most of the consonants were pronounced as in the present English. c had the two sounds of (k) and (s) as at present, and was silent in the combination sc in such words as science, conscience. In all cases where g had the 'soft' sound (dg) it is marked & in this book. g in ng seems to have been always distinctly sounded; as it still is in such words

(

as longer. Double gġ, as in egge edge,' was probably =(ddz), cch being (ttf), as in cacchen.

=

h was silent in honest, honour, hümble and some other French words, just as it is in the present English. It was further dropped in unaccented words, such as he, hath, how, these words keeping the h when emphatic and accented.

r was always a strong trill, as in Scotch.

The vowel i, y was often pronounced as the cons. y when preceded by certain conss., especially 7, n, s, and followed by a vowel, as in many a man, glorie, famüliër, condicioun, such a word as the last having in poetry sometimes four, sometimes only three syllables. ll in French words, such as boillen, vitaille, was perhaps sometimes pronounced ly (y cons.) but it was generally simple 7. French gn, as in compaignje, Boloigne, was perhaps sometimes pronounced ny (y cons.), but in familiar pronunciation it was= simple n, preceded either by a diphthong, as in the two instances just cited, or by a long vowel, as in signe=(siine), dignitee (diinitee).

gh had two distinct pronunciations. After the front vowels e, i it had the sound of German ch in ich, as in streight, high, bryght. This front gh was vocalized into consonantal y before a vowel, and then generally dropt, as in the plural hye. The other gh had the sound of German ch in auch, = the German ch in ach rounded. Hence it is always preceded either by (uu), as in ynough, plough, or by u forming the second element of a diphthong. This u is always written after a, as in taughte, laughter, while after o it is sometimes written, sometimes left to be inferred from the following gh, as in trogh, trough, thoghte, thoughte. This gh was often dropped before a vowel, especially after (uu), as in the plural ynowe. Both ghs were occasionally dropped finally or before 1, thus we find both streit and streight.

s had the two sounds (s) and (z), as in the present English. Such words as has, his, was were probably pronounced with (z) when unaccented, but when accented they took the sound of (s). So also the unaccented -es in houses, &c. had the zsound, while in monosyllabic plurals such as tres it probably had the s-sound, just as in ges' geese.'

The distribution of ƒ and v was analogous, and there can be no doubt that the accented adverb of='off,' was distinguished from the unaccented preposition of by the former having the f-, the latter the v-sound, just as at present. It is probable that emphatic and unemphatic if were distinguished in the same way.

It would, of course, have been practically impossible to write *ou=ov, &c.

th had the sound of () in such words as that, this, then, thou, as at present. Unaccented th in loveth, &c., and in unemphatic words such as hath, with probably had the same sound, the last two assuming the (p)-sound when emphatic. In some foreign words, such as apothecarie, it probably had the sound of t.

s, f, and th in accented syllables were always voiceless finally, and voiced between vowels or vowels and voiced consonants: voiceless in hous, wÿf, gth, voiced in houses (both ses), wives, othes; voiced also in such words as brethren.

As a general rule every written consonant must be pronounced. This applies especially to the k in kn-, and the w in wr-.

Stress.

There are three main degrees of stress or accent: strong, half-strong, and weak. In such a word as martyrdom the first syllable is strong, the second weak, the last half-strong. In poetry the strong and half-strong stresses were allowed to

change places where this was required by the metre, so that martyrdōm could be pronounced with the strong stress on the last syllable. So also creature in poetry may have the strong stress either on the first or the third syllable, the second and fourth being always weak. Other examples are resoun, nātūre. The popular tendency in these words was no doubt to throw the stress back as much as possible. Even weak syllables often take the strong stress away from an adjoining strong syllable in poetry. Such weak syllables are called accentuable. Final e, as in name, is never accentuable. Examples of accentuable weak syllables are afforded by the second syllables of worthy, syngynge, frendshipe. Stress-shifting on to a naturally weak syllable at the beginning of a line or a metrical pause (cæsura) should be softened down as much as possible by giving the two syllables a nearly equal stress, as in Pite 59 sheweth, Prol. 125 after. Stress-shifting on to the rhyme-syllable must, of course, be marked by a more emphatic stress, as in Pite 32 sodeynly. These are the three places in which stress-shifting is most frequent.

PHONOLOGY.

Quantity.

The most important change that took place in the fourteenth century was the lengthening of accented short vowels before a single consonant followed by a vowel, as in nāme, çten, byfore. i and u, however, were not lengthened: writen, sẵne. Monosyllables generally kept their short vowels, which are lengthened when a vowel is added in inflection or otherwise : smal, plur. smāle, bath, plur. bathes, bathen verb. If the cons. is doubled when a vowel is added, there is, of course, no lengthening of the preceding vowel: al, plur. alle.