13

which of itself suffices to overthrow almost his whole system.

So much by way of preliminary observation. And now to enter fairly on our investigation, let us examine first the class of words already alluded to-mine, fine, &c., which may for convenience' sake be distinguished as the I words. In these our inquiry will be how the strongly accented vowel was formerly pronounced; and, as above remarked, the answer in which all our southern dialects concur is that the sound was (ǝi): they have all preserved traditionally a sound more or less closely approaching, if not identical with, that diphthong in all there is one or more of the (a) sounds followed by one or more of the (i) sounds, and words, Dutch the pair or series compressed into very nearly firm the evidence the same compound vowel. (On the exact of our own dia- nature of diphthongs see below, § 29.) But we are not left to the voice of English tradition alone; for in a considerable number of those words the root vowel is the same in Dutch and German as in English, the Dutch j and German ei both being (əi). Here is the list.

As to I (ǝi)

and German con

lects: list of

words.

To these may be added riche, which, though it has the in modern times, is always long in the Ormulum and Chaucer and Early English generally, and is the Du. rijk, Germ. reich.

*

Inasmuch then as the root vowel in all these words is pronounced alike as (ai) by the Dutch and the High Germans, as well as all the branches of the Engelcyn, while in the same words the vowel is (ii) in certain other Gothic languages (the Scandinavian, for instance, and the PlattDeutsch), and there is no satisfactory evidence of any change in the pronunciation of these words, the conclusion seems inevitable that the Angles and Saxons and Hollanders and High Germans constitute a separate division (of course capable of subdivision) of the Teutonic race, and that this sound of (əi) existed in their languages in common at a period prior to the divergence of these tribes from one another. In other words, it seems to me probable that the present vowel sounds of mine wine were common to the Englisc, Hollandsch, and Deutsch, from a period of remote antiquity long prior to historical times.

This point is further discussed in § 102.

14

these

others may be

The sound was written in A.S. 1, and in E.E. 1, y, ii, or By means of ij; and by the aid of rhymes these 35 words words suffice to fix all E.E. words similarly spelt, and determined. others. Others, thus: if the pronunciation of wide and side can be determined, that of espide, which rhymes with these, is fixed, and therefore that of espy. If we fix shine with (əi), its preterit in such rhymes as those on p. 3 of the Allit. Poems-schynde, kynde, ynde, tynde, grynde, blynde-determines these others: a result, I may add, contrary to Mr. Ellis's view (p. 276), but confirmed by Orm's spelling of kinde, grindesst, and blind; while elsewhere these words rhyme with find, which Orm spells with the single n, as also we have findan with the accent in the Gloucester Fragments. I am aware that in MSS. the accent on the i when it stands next to a ¤, an н, or an m, often serves the purpose only of the later dot, that is, to show which of the upright strokes is the vowel. But in the Glouc. Fr. the writing is so large and clear, and the n is so distinct from the i, that I believe the accent to be fully intended as such. 15 This positive Whately lays it down as an important rule evidence not to in reasoning, that where there exists a body of moment an ob- positive evidence in favour of any conclusion, jection appears. such body of evidence is not to be set aside the moment we meet with an objection which we do not see how to surmount. Now we have a mass of such positive evidence as to the sound of i in the words I am discussing, and to the conclusion to which it leads, I should be prepared to hold even if from imperfect knowledge (for that is often all that an objection appeals to) I were unable to get over the difficulty that presents itself from a certain quarter, and upon which almost exclusively Mr. Ellis fixes his gaze. I refer to the rhyming of many of our English i words with French words containing the same written letter, which it is affirmed was sounded (i). But after all the objection seems by no means insurmountable; for on turning to Palsgrave, whose evidence is very "perplexing" to Mr. Ellis, he states most distinctly (as quite correctly quoted * Compare § 60.

be flung aside the

by Mr. Ellis) that the French i has two diverse sounds, one of them like the Italian i, and as we sound e in bee an insect, fee a reward; while as to the second he says, “If i be the first letter in a frenche word or the laste, he shall, in those two places, be sounded like as we do this letter y in these words with vs, by and by, a spye, a flye, awry, and suche other."

grave's expres

sions.

16 An objection Very good, then if any one asserts that these based on Pals words were sounded (bii and bii), (spii), (flii), (arii), he has Palsgrave dead against him, affirming as Palsgrave does that the y here had not the sound of the Italian i. What sound then had it? The English dialects answer with one voice, declaring first how these words are sounded themselves, and secondly how others are sounded which rhyme with these in Chaucer. First, they declare the words themselves to be sounded with (ǝi). Secondly, by and by rhymes with why, with the adverbial termination -ly (i.e. like, as it is lijk in Dutch; and this -ly always kept the vowel long till about the middle of the 17th century*); and the single by, which is the Dutch bij and German bei—the very same soundrhymes with nigh, sky, and I (Dan. jeg, pronounced Jǝi). Then spy or aspy or espy, rhymes with eye, high, dry, I, hie, sly, cry; and its past tense rhymes with betide, side, wide, abide, and these again with hide, chide, slide, glide, &c.; and the pronunciation of these the Dutch and German Tijd, Zeit, glijden, gleiten, &c., as we have already seen, confirm. Palsgrave's fly and awry in like manner rhyme with by and aspy and all the others with which these rhyme. Surely all this evidence is not to be pooh-poohed. 17 Further evi- And then again as to the sound of i as (əi) in not (ii) in French, Mr. Ellis himself-whose honesty and candour in argument deserve to be both admired and imitated-mentions a statement made by Mons. Le Héricher that the pronunciation of joli as jolaï (that is, nearly or quite with our English i) is still Since this sentence was written, our Mill Hill carrier, a Middlesex man, has told me he would do so and so "accordinglī" (əi).

dence that i final

was French.

(əi)

18

known in Normandy, and Dr. Le Taillis of Montebourg near Cherbourg states that this sound is "très-généralement usité" in Montebourg and the neighbourhood. Discredit has been thrown upon these statements because many people have not heard these sounds. Just so, when I resided in France in the neighbourhood of Boulogne, one day when a wasp was buzzing in the room, I noticed that my hostess-I was lodging at a village inn-called it “une vêpe;" and I have since, when mentioning the fact, been seriously assured by well-informed Frenchmen that I was mistaken, and that there is no such word in existence. But I heard it. I noticed instantly that it was the Latin vespa without the prefixed guttural. To make assurance doubly sure, I got her to repeat the word. And though millions may not have heard vêpe, I did; and that I should maintain not a whit the less tenaciously, even if I had not recently discovered that my observation is confirmed by Duméril. The conclusion then to which the testimony of these witnesses conducts us is that what in modern French is enem(i) was in Chaucer's time enem(əi), merc(i), (merc(əi), and Gu(i) Gu(əi); and Chaucer's rhymes show clearly that the final e made no difference after this vowel, as companye, flatterie, curtesye, tyrannye, melodie, contrarye, Lumbardye, rhyme indifferently with the same words with which enemy and merci and Guy rhyme. From many or most of these modern English fashion has removed the final accent; but that the vowel in Chaucer's time was sounded full and strong, as we still sound it in glorify, magnify, prophesy, multiply, lullaby, &c., is clearly evident.

And however strange such words may now sound to our ears, this termination is very common in Dutch, as in

* But God forbede but men shulde leve

Wel more thing than men han seen with eye!

Men shal not wenen every thing a lye

But yf himselfe yt seeth, or elles dooth;

For, God wot, thing is never the lasse sooth,

Thogh every wight ne may it not ysee.

The Prologe of Nine Goode Wymmen.

Many have made similar observations since Chaucer's time.