ON EARLY ENGLISH PRONUNCIATION. 1 Introductory. IT is by no means an agreeable task to assail the conclusions at which the long and laborious and evidently con amore researches of a fellow-student have landed him; and it seems futile to attempt to uproot by this short essay those views which his large and learned book has caused to be so generally received in our Society. But as I have for some years had the subject of the present paper before my mind, and the results at which I have arrived differ entirely from Mr. Ellis's on some leading points, and the careful and candid consideration of his work has only to a very small extent modified my views, in the interests of philological truth I cannot consent to be silent. It is not a lack of industry with which Mr. Ellis can be charged; but I do impeach his logic, and seriously complain of the general conduct of the argument. spoken, not writ ten, language. 2 Our starting- It is impossible that a scholar who has point must be devoted many years to philological study, should really confound even for an instant language proper-that is the living voice-with the black marks on white paper which are the mere symbols of language; but it is quite possible that in dealing simultaneously with both language and its symbols, he may allow its symbols to occupy too prominent a position before his own mind and in his treatment of the subject. The question before us should, I apprehend, generally shape itself as follows: not, what sound did such or such a symbol represent? (though it may conveniently assume that form sometimes); but, how were such and such spoken B words of this 19th century spoken in the 14th or in the 9th? Mr. Ellis looks always to the symbol. 3 Mr. Ellis's max thography shows the sound." Now if we examine in Heywood's Proverbs im that "The or- and Epigrams the rhymes of words ending in -ear, -eare, -ere, -eere, and -eer, we shall soon find ourselves in inextricable confusion, if the letters alone are to guide us; but if we notice that the words which we now pronounce with (ii)-cleer, chere, here (adv.), here (vb.), neere, yeer, deer (adj.), deer (s), and appeare-rhyme with one another in Heywood, however he may spell them, but never rhyme with there, where, were, wear, swear, Edgeware, hair, hare, ear, spear, fear, answer, ere, bear (vb.), while these all rhyme, most of them repeatedly, with one another; and if examination of Sir Philip Sidney's poems leads (as it does) to precisely the same result, we may be warranted in drawing some conclusion from that fact. Besides, the former mode of putting the question has a tendency towards the assumption that each symbol, or group of symbols, stood only for one sound, or at most for one pair of sounds, one long and one short. Considering that our first vowel is at present the representative of at least four distinct sounds (as in fate, fat, father, fall), and our second vowel of at least three (as in we, when, were); we must not assume that it was entirely otherwise five or ten centuries ago. Mr. Ellis leans on the broken reed of the maxim that "The Orthography shows the sound." How untrustworthy the support is—though unhappily we some times have no other-will be abundantly proved further on. But besides trusting far too implicitly to this delusive maxim, Mr. Ellis in conducting his case exhibits singular partiality towards one class of witnesses, while others-by far the most important-he treats with undeserved disrespect: they are not indeed put out of court, but they are by no means allowed full, a patient, and impartial hearing. 4 Orthoepists not The too highly favoured witnesses are the grammarians and orthoepists, whose evidence may be impugned on the ground, not only that they are often as inaccurate observers as many of us moderns are, to be too much relied on. and on many points do not agree among themselves, in which respect doubtless their writings only the more exactly reflect the variety of popular usage,* but also that they too commonly are not content to let us know the simple facts that was not the object they had in view in writing, -but they endeavoured to guide usage to something different from what it was, and too frequently they mislead the modern reader by their assertion that the sound of a word is what it is not: they mean that it is so and so de jure, and the reader is apt to think they mean that it is so de facto. So Gil charges Hart with seeking rather "ducere quam sequi" our language by his mode of writing; Palsgrave again and again appeals to the speech of those "that pronounce the latine tonge aright,” i.e. in the manner that he approved; Erasmus, Cheke, Smith, all argued from written symbols that a written diphthong must represent a compound sound, and Smith in particular insisted on a distinction between ai and ei in English, which, though it may have existed in the dialects in certain words, his very insistence, as well as the rhymes of all the poets from Chaucer downwards, show not to have been observed in the received pronunciation; and Butler's language betrays the same tendency where he speaks of a "corrupt" usage. In this last case Mr. Ellis has very justly observed that "allowance must be made for the mode in which orthoepists speak of common pronunciations which differ from their own or from what they recommend-by no means always the same thing" (p. 124); as elsewhere (p. 139) he remarks on Gil's "anxiety to give prominence to the first element" in the diphthong ew. All such "anxiety" detracts from the value of a writer's evidence when it is the simple fact that the reader desires to ascertain; and probably many of the sounds which are vindicated by these older orthoepists may deserve to be characterized as "a theoretical pronunciation, which may be as false as that which Erasmus, Smith, and Cheke intro * Gil says: "In build ædificare, nondum iactum est fundamentum: pro suopte enim cuiusque ingenio, vnus buldeth per viλóv; alter beildeth per ei; tertius beeldeth per i longum : et adhuc quartus bildeth per i breve." duced into England for the Greek language” (see Academy for Apr. 15th, 1871). But Mr. Ellis is certainly not unaware of the weakness of his case in this important particular. There are however other objections to our relying too confidently on these authorities. One is that the earliest of them lived nearly a century and a half after Chaucer. I lay but little stress on this, not believing that any great change took place in the interval. A second is, that these orthoepists-comparative novices in their art-seem to have overlooked sounds which can be shown to have existed in common use in their day. A reader of Ben Johnson's account of a, would suppose that that was the symbol for only two sounds, apparently (a) and (A); but Gil twenty years earlier, and Hart seventy, had recognized three classes of words the vowel of which was written with a. Smith in his argument about the Greek ʼn points out only two es in English, as in whet (wheet), now wheat, and whet; yet he himself in his Index, which Mr. Ellis seems not to have discovered, recognizes another which for our purpose is evidently more important, for he calls it the e Anglica, of which breed and heel are his examples. And so he says elsewhere: "Recte etiam fortasse nunc Domine ne in furore, per e Italicum, non quemadmodum olim per illud .e. Anglicum, quod in bee cùm apis dicimus, aut me cùm què nostro more loquamur, obseruatur, &c." De Ling. Gr. Pron., p. 14 vo. 5 Statements of orthoepists, how But a yet graver objection is furnished by used by Mr. Ellis. Mr. Ellis's ingenuity, he having shown but too frequently the possibility of extracting from their words a sense totally at variance with what I believe they really meant; so that I prefer scarcely to draw any conclusions at all from premises which seem to be so doubtful. For instance-once more to anticipate the general argument— Salesbury represents the English words true, vertue, duke, Jesu, by truw, vertuw, duwk, tsiesuw; and Mr. Ellis, by a ratiocinative process which I cannot pretend to understand, concludes "that Salesbury's uw meant (yy)." I have submitted the words to several educated Welshmen, who all say that the u is (i) and the w (uu), and the diphthong is as nearly as possible the long English u-(iu) or (Juu) of tune, tube, union. It seems to me that Salesbury's description implies, in a manner than which nothing can be clearer, that these words were sounded in his time exactly as they are now, except that (trJuu) has become (truu), and the first syllable of virtue is no longer sounded (ver). As to these orthoepists, it must be confessed that their language, as manipulated by Mr. Ellis, is singularly unintelligible; and yet if, instead of studying only fragmentary quotations, and misleading explanations, and "transliterations" which assume every single point that is in dispute, we read the books themselves, and adopt the simple hypothesis that as a general rule our forefathers of those centuries pronounced their own language, and Latin and Greek too, just as we do now, almost every difficulty at once disappears. 6 Traditional pro nunciation our And who are the witnesses that are thrust main guide. aside? Our dialects as now spoken. Suppose we have to inquire concerning certain common and familiar words which we have inherited as part of the old English speech of our forefathers, for instance, those which we in our 19th century mode pronounce with (əi), such as mine, thine, fine, wine, shine, line, swine, wife, life, knife, &c. &c.-how these words, as to their strongly accented vowels, were pronounced several centuries ago; I contend that we have above all things to consider how these words are still pronounced in various English dialects—especially of course, for Chaucer, those south of the Humber. It is spoken language about which we are inquiring, and it is mainly language as now spoken that must furnish an answer to our question. The existing English dialects yield by far the most important evidence in the case, and their voice, in this particular part of the inquiry, Mr. Ellis scarcely suffers to be heard. If we listen to them, they with almost perfect unanimity assign to these words some such diphthongal sound as we still give them. There may be some discrepancy in their evidence as to the elements |