last edition of him; for he would make us believe the fault is in our ears, and that there were really ten syllables in a verse, where we find but nine. But this opinion is not worth confuting; it is so gross and obvious an errour, that common sense (which is a rule in every thing but matters of faith and revelation,) must convince the reader, that equality of numbers in every verse which we call heroick, was either not known, or not always practised in Chaucer's age. It were an easy matter to produce some thousands of his verses, which are lame for want of half a foot, and sometimes a whole one, and which no pronunciation can make otherwise. We can only say, that he lived in

+ Thomas Speght published the works of Chaucer in 1597, and again in 1602. His edition was reprinted in 1687, with some additions by the editor.

The position here laid down has been so ably contested by Mr. Tyrwhitt in his excellent edition of THE CANTERBURY TALES, that I am persuaded, after perusing the following observations on the subject of Chaucer's metre, every judicious reader will be of opinion that our author had not sufficiently studied our ancient language, and was mistaken in this assertion.

By far the most considerable part of Chaucer's works (says Mr. Tyrwhitt,) is written in that kind of metre which we now call the Heroic, either in distichs or in stanzas; and as I have not been able to discover any instance of this metre being used by any English poet before him, I am much inclined to suppose that he was the first introducer of it into our language. It had long been practised in France, in the Northern as well as the Southern provinces; and in Italy, within the last fifty

the infancy of our poetry, and that nothing is brought to perfection at the first. We must be

years before Chaucer wrote, it had been cultivated with the greatest assiduity and success, in preference to every other metre, by Dante, Petrarch, and Boccace. When we reflect that two of Chaucer's juvenile productions, the PALAMON AND ARCITE, and the TROILUS, were in a manner translated from the THESEIDA and the FILOSTRATO of Boccace, both written in the common Italian hendecasyllable verse, it cannot but appear extremely probable that his metre also was copied from the same original; and yet I cannot find that the form of his stanza in the TROILUS (consisting of seven verses) was ever used by Boccace, though it is to be met with among the poems of the King of Navarre, and of the Provençal Rimers. Whichever he shall be supposed to have followed, whether the French or Italians, it is certain that he coud not want in either language a number of models of correct and harmonious versification; and the only question will be, whether he had ability and industry enough to imitate that part of their excellency.

"In discussing this question we should always have in mind, that the correctness and harmony of an English verse depends entirely upon its being composed of a certain number of syllables, and its having the accents of those syllables properly placed. In order therefore to form any judgement of the Versification of Chaucer, it is necessary that we should know the syllabical value (if I may use the expression) of his words, and the accentual value of his syllables, as they were commonly pronounced in his time; for without that knowledge, it is not more probable that we should determine justly the exactness of his metres, than that we should be able to cast up rightly an account stated in coins of

upon

a

There was an

children, before we grow men. Ennius, and in process of time a Lucilius and a

former age, of whose current rates and denominations we are totally ignorant.

"Let us consider a moment, how a sensible critic in the Augustan age would have proceeded, if called upon to examine a work of Ennius. When he found that a great proportion of the verses were strictly conformable to the ordinary rules of metre, he would, probably, not scruple to conclude that such a conformity must have been produced by art and design, and not by mere chance. On the other hand, when he found, that in some verses the number of feet, to appearance, was either deficient or redundant; that in others the feet were seemingly composed of too few or too many syllables, of short syllables in the place of long, or of long in the place of short; he would not, I think, immediately condemn the old bard, as having all at once forgotten the fundamental principles of his art, or as having wilfully or negligently deviated from them. He would first, I presume, enquire whether all these irregularities were in the genuine text of his author, or only the mistakes of Copyists: he would enquire further, by comparing the genuine text with other contemporary writings and monuments, whether many things which appeared irregular, were not in truth sufficiently regular, either justified by the constant practice, or excused by the allowed licence, of the age: where authority failed, he would have recourse (but soberly) to etymology and analogy; and if after all a few passages remained, not reducible to the strict laws of metre by any of the methods above mentioned, if he were really (as I have supposed him) a sensible critic, he would be apt rather to expect patiently the solution of his difficulties from more correct manuscripts, or a more

Lucretius, before Virgil and Horace. Even after Chaucer, there was a Spencer, a Harrington, a

complete theory of his author's versification, than to cut the knot, by deciding peremptorily, that the work was composed without any regard to metrical rules.

I beg leave to pursue the same course with respect to Chaucer. The great number of verses sounding complete even to our ears, which is to be found in all the least corrected copies of his works, authorizes us to conclude, that he was not ignorant of the laws of metre. Upon this conclusion it is impossible not to ground a strong presumption, that he intended to observe the same laws in the many other verses which seem to us irregular ; and if this was really his intention, what reason can be assigned sufficient to account for his having failed so grossly and repeatedly, as is generally supposed, in an operation, which every balladmonger in our days, man, woman, or child, is known to perform with the most unerring exactness, and without any extraordinary fatigue?

"The offences against metre in an English verse, as has partly been observed before, must arise either from a Superfluity or Deficiency of syllables, or from the Accents being improperly placed.

With respect to the first species of irregularity, I have not taken notice of any Superfluities in Chaucer's verses, but what may be reduced to just measure by the usual practices of even modern poets. And this, by the way, is a strong proof of his real attention to metrical rules; for otherwise, if he had written without any restraint of that kind, a certain proportion of his deviations from measure must, in all probability, have been on the side of excess.

"But a great number of Chaucer's verses labour under

Fairfax, before Waller and Denham were in being; and our numbers were in their nonage till these last appeared.

an apparent Deficiency of a syllable or two. In some of these, perhaps the defect may still be supplied from Mss.: but for the greatest part I am persuaded no such assistance is to be expected; and therefore, supposing the text in these cases to be correct, it is worth considering whether the verse also may not be made correct, by adopting in certain words a pronunciation, different indeed from modern practice, but which, we have reason to believe, was used by the author himself.

For instance, in the genitive case singular, and the plural number of nouns, (which, as has been remarked above, in the time of Chaucer had the same expression,) there can be no doubt that such words as, shoures, ver. 1. croppes, ver. 7. shires, ver. 15. lordes, ver. 47, &c. were regularly pronounced as consisting of two syllables. Whenever they are used as monosyllables, it must be considered as a poetical licence, warranted however even then (as we may presume, from the natural progress of our language,) by the practice of inaccurate speakers in

common conversation.

"In like manner, we may be sure that ed, the regular termination of the past tense and its participle, made, or contributed to make, a second syllable in the words, perced, ver. 2. bathed, ver. 3. loved, ver. 45. wered, ver. 75, &c. The first step toward reducing words of this form to monosyllables, seems to have been to shorten the last syllable, either by transposing the final letters, as inwolde, ver. 144. sayde, ver. 763, &c. or by throwing away the d, as in-coste, ver. 1910. caste, ver. 2083, &c. In both these cases the words still remained of two syllables, the final e being sounded as an e feminine; but they were prepared to lose their last syllable by the easy licence of