The Cambridge History of English and American Literature in 18 Volumes (1907–21).>br>Volume I. From the Beginnings to the Cycles of Romance.
XIX. Changes in the Language to the Days of Chaucer§ 9. Development of Sounds
The changes in spelling that we have thus far noticed are merely changes in the manner of representing sound. There were others that were the result of altered pronunciation. It very often happens that very considerable changes take place in the sounds of a language without affecting the spelling, even when (as was apparently, the case in Middle English) there is no general prejudice against deviations from traditional correctness of orthography. Pronunciation, as a general rule, is not altered deliberately, but unconsciously. In the utterance of what is intended and believed to be one and the same vowel or consonant sound, each generation may very to an almost imperceptible extent from that which preceded it; and, if these slight changes are all in the same direction, the difference may, in the end, become indefinitely great. The normal result in such cases is that the letter comes to have a new phonetic value, and the spelling is not effected. The reason why there are exceptions to this normal course of things in Middle English was partly that sometimes two originally distinct sounds so developed as to become identical, and party that the orthography of French supplied a kind of external standard.
The history of the changes in English pronunciation down to the time of Chaucer is far too intricate to be treated here with any approach to completeness; but a few of its salient points may be briefly indicated.
The first remark to be made is that the course of development of several of the Old English sounds was quite different in different parts of the country. When we compare the modern English pronunciation of home, stone, with the Scotch and northern hame, stane, we see the last term of a divergent development (which began very early) of the Old English long a (pronounced as a in father). While the northern dialect progressively altered the sound in one direction, the midland and southern dialects progressively altered it in the opposite direction. We cannot precisely tell how far the change in the northern pronunciation had proceeded in the fourteenth century, because the spelling was not affected. But, in other dialects, as we know from various kinds of evidence, the sound was that of the “open [char]” as in lord, and it was expressed in writing by o or oo. The words “goad” (Old English g[char]d) and “good” (Old English g[char]d) are both written good in Chaucer’s spelling, but they were not pronounced alike; if the sounds had been confused they would not have been separated again in later pronunciation; and Chaucer never rimes a word that has the “open o” with one containing the “close o.” The latter retained its old pronunciation (that of the French o in rose), perhaps a little modified in the direction of its modern equivalent, the oo in cool.
The long e, like the long o, had an “open” and a “close” pronunciation, which Chaucer also keeps apart in his rimes. The open [char] comes from the Old English (Anglian) [char], [char]a, and the close [char] from Old English [char], [char]o. A word like chepe to buy (from Old English c[char]apian) which had the open [char], could not correctly rime with a word like kepe to keep (from c[char]pan) which had the close [char]. In northern dialects, the distinction was so slight that poets freely allowed the two sounds to rime with one another.
In all the dialects of Middle English, the short vowels [char], [char], [char], when ending an accented syllable, were lengthened, [char] and [char] becoming open [char] and open [char]. In Chaucer’s pronunciation, mete meat (Old English m[char]te) was an exact rime to grete, the plural of the adjective great (Old English gr[char]ate), but not to grete to greet (Old English gr[char]tan); [char]rote throat (Old English [char]rotu) rimed with hote to command (Old English h[char]tan), but not with b[char]te benefit (Old English b[char]t).
The Old English y (pronounced ü) kept its original sound in the south-west, and, perhaps, in parts of the west midland, being written u when short, and ui or uy when long; in Kent, it had become e before the Conquest; elsewhere, it was sounded exactly like i, and written, like that sound, indifferently i or y. The words “fire” “sin,” “knit,” have, accordingly, in the different localities the three types of form fuir, ver, fir; sunne, zenne, sinne; knutte, knette, knitte. Chaucer, whose London English was mainly east midland, uses occasionally a Kentish form like knette.
With regard to the pronunciation of consonants, there is little that needs to be said, as, for the most part, the Old English sounds not only continued unchanged down to the end of the fourteenth century, but remain so to the present day. The pronunciation of initial f and s as v and z (“vather came vrom Zummerzet”), which sounds so strange to visitors to the south-western counties, was, in the fourteenth century, current all over the south; in fact, the Kentish Ayenbite of Inwyt, of 1340, exhibits this pronunciation in the orthography with greater regularity than any other extant book. The gh sound of the letter [char] gradually changed into that of w, and this change was represented in the spelling. In the earlier of the two MSS. of the poetical chronicle called the Brut, written at the beginning of the thirteenth century, the author’s name appears as “La[char]amon,” but, in the later MS., written before 1300, it is turned into “Laweman.” On the other hand, in 1340, the Kentish Ayenbite has still forms like zor[char]e (sorrow) instead of Chaucer’s sorwe.