Text: Edgar Allan Poe, “The Rationale of Verse” [Part I], Southern Literary Messenger, October 1848, 14:577-585


[page 577, unnumbered, column 1:]



The word “Verse” is here used not in its strict or primitive sense, but as the term most convenient for expressing generally and without pedantry all that is involved in the consideration of rhythm, rhyme, metre, and versification.

There is, perhaps, no topic in polite literature which has been more pertinaciously discussed, and there is certainly not one about which so much inaccuracy, confusion, misconception, misrepresentation, mystification, and downright ignorance on all sides, can be fairly said to exist. Were the topic really difficult, or did it lie, even, in the cloud-land of metaphysics, where the doubt-vapors may be made to assume any and every shape at the will or at the fancy of the gazer, we should have less reason to wonder at all this contradiction and perplexity; but in fact the subject is exceedingly simple; one tenth of it, possibly, may be called ethical; nine tenths, however, appertain to the mathematics; and the whole is included within the limits of the commonest common sense.

“But, if this is the case, how,” it will be asked, “can so much misunderstanding have arisen? Is it conceivable that a thousand profound scholars, investigating so very simple a matter for centuries, have not been able to place it in the fullest light, at least, of which it is susceptible?” These queries, I confess, are not easily answered: — at all events a satisfactory reply to them might cost more trouble than would, if properly considered, the whole vexata quæstio [[questio]] to which they have reference. Nevertheless, there is little difficulty or danger in suggesting that the “thousand profound scholars” may have failed, first because they were scholars, secondly because they were profound, and thirdly because they were a thousand — the impotency of the scholarship and profundity having been thus multiplied a thousand fold. I am serious in these suggestions; for, first again, there is something in “scholarship” which seduces us into blind worship of Bacon's Idol of the Theatre — into irrational deference to antiquity; secondly, the proper “profundity” is rarely profound — it is the nature of Truth in general, as of some ores in particular, to [column 2:] be richest when most superficial; thirdly, the clearest subject may be overclouded by mere superabundance of talk. In chemistry, the best way of separating two bodies is to add a third; in speculation, fact often agrees with fact and argument with argument, until an additional well-meaning fact or argument sets every thing by the ears. In one case out of a hundred a point is excessively discussed because it is obscure; in the ninety-nine remaining it is obscure because excessively discussed. When a topic is thus circumstanced, the readiest mode of investigating it is to forget that any previous investigation has been attempted.

But, in fact, while much has been written on the Greek and Latin rhythms, and even on the Hebrew, little effort has been made at examining that of any of the modern tongues. As regards the English, comparatively nothing has been done. It may be said, indeed, that we are without a treatise on our own verse. In our ordinary grammars and in our works on rhetoric or prosody in general, may be found occasional chapters, it is true, which have the heading, “Versification,” but these are, in all instances, exceedingly meagre. They pretend to no analysis; they propose nothing like system; they make no attempt at even rule; every thing depends upon “authority.” They are confined, in fact, to mere exemplification of the supposed varieties of English feet and English lines; — although in no work with which I am acquainted are these feet correctly given or these lines detailed in anything like their full extent. Yet what has been mentioned is all — if we except the occasional introduction of some pedagogue-ism, such as this, borrowed from the Greek Prosodies: — “When a syllable is wanting, the verse is said to be catalectic; when the measure is exact, the line is acatalectic; when there is a redundant syllable it forms hypermeter.” Now whether a line be termed catalectic or acatalectic is, perhaps, a point of no vital importance; — it is even possible that the student may be able to decide, promptly, when the a should be employed and when omitted, yet be incognizant, at the same time, of all that is worth knowing in regard to the structure of verse.

A leading defect in each of our treatises, (if treatises they can be called,) is the confining the subject to mere Versification, while Verse in general, with the understanding given to the term in the heading of this paper, is the real question at issue. Nor am I aware of even one of our Grammars which so much as properly defines the word versification itself. “Versification,” says a work now before me, of which the accuracy is far more than [page 578:] usual — the “English Grammar” of Goold Brown — “Versification is the art of arranging words into lines of correspondent length, so as to produce harmony by the regular alternation of syllables differing in quantity.” The commencement of this definition might apply, indeed, to the art of versification, but not to versification itself. Versification is not the art of arranging &c., but the actual arranging — a distinction too obvious to need comment. The error here is identical with one which has been too long permitted to disgrace the initial page of every one of our school grammars. I allude to the definitions of English Grammar itself. “English Grammar,” it is said, “is the art of speaking and writing the English language correctly.” This phraseology, or something essentially similar, is employed, I believe, by Bacon, Miller, Fisk, Greenleaf, Ingersoll, Kirkland [[Kirkham]], Cooper, Flint, Pue, Comly, and many others. These gentlemen, it is presumed, adopted it without examination from Murray, who derived it from Lily, (whose work was “quam solam Regia Majestas in omnibus scholis docendam præcipit,”) and who appropriated it without acknowledgment, but with some unimportant modification, from the Latin Grammar of Leonicenus. It may be shown, however, that this definition, so complacently received, is not, and cannot be, a proper definition of English Grammar. A definition is that which so describes its object as to distinguish it from all others: — it is no definition of any one thing if its terms are applicable to any one other. But if it be asked — “What is the design — the end — the aim of English Grammar?” our obvious answer is, “The art of speaking and writing the English language correctly:” — that is to say, we must use the precise words employed as the definition of English Grammar itself. But the object to be obtained by any means is, assuredly, not the means. English Grammar and the end contemplated by English Grammar, are two matters sufficiently distinct; nor can the one be more reasonably regarded as the other than a fishing-hook as a fish. The definition, therefore, which is applicable in the latter instance, cannot, in the former, be true. Grammar in general is the analysis of language; English Grammar of the English.

But to return to Versification as defined in our extract above. “It is the art,” says this extract, “of arranging words into lines of correspondent length.” Not so: — a correspondence in the length of lines is by no means essential. Pindaric odes are, surely, instances of versification, yet these compositions are noted for extreme diversity in the length of their lines.

The arrangement is moreover said to be for the purpose of producing “harmony by the regular alternation,” &c. But harmony is not the sole aim — not even the principal one. In the construction of verse, melody should never be left out of view; yet this is a point which all our Prosodies have [column 2:] most unaccountably forborne to touch. Reasoned rules on this topic should form a portion of all systems of rhythm.

“So as to produce harmony,” says the definition, “by the regular alternation,” &c. A regular alternation, as described, forms no part of any principle of versification. The arrangement of spondees and dactyls, for example, in the Greek hexameter, is an arrangement which may be termed at random. At least it is arbitrary. Without interference with the line as a whole, a dactyl may be substituted for a spondee, or the converse, at any point other than the ultimate and penultimate feet, of which the former is always a spondee, the latter nearly always a dactyl. Here, it is clear, we have no “regular alternation of syllables differing in quantity.”

“So as to produce harmony,” proceeds the definition, “by the regular alternation of syllables differing in quantity,” — in other words by the alternation of long and short syllables; for in rhythm all syllables are necessarily either short or long. But not only do I deny the necessity of any regularity in the succession of feet and, by consequence, of syllables, but dispute the essentiality of any alternation, regular or irregular, of syllables long and short. Our author, observe, is now engaged in a definition of versification in general, not of English versification in particular. But the Greek and Latin metres abound in the spondee and pyrrhic — the former consisting of two long syllables; the latter of two short; and there are innumerable instances of the immediate succession of many spondees and many pyrrhics.

Here is a passage from Silius Italicus:

Fallis te mensas inter quod credis inermem

Tot bellis quæsita viro, tot cædibus armat

Majestas eterna ducem: si admoveris ora

Cannas et Trebium ante oculos Trasymenaque busta,

Et Pauli stare ingentem miraberis umbram.

Making the elisions demanded by the classic Prosodies, we should scan these Hexameters thus:

Fāllīs | tē mēn | sās īn | tēr qūod | crēdĭs ĭn | ērmēm |

Tōt bēl | līs qūæ | sītă vĭ | rō tōt | cædĭbŭs | ārmāt |

Mājēs | tās ē | tērnă dŭ | cēm s' ād | mōvĕrĭs | ōrā |

Cānnās | ēt Trĕbrĭ' | ānt'ŏcŭ | lōs Trăsy | mēnăqŭe | būstā [[ | ]]

ēt [[Ēt]] Pāu | lī stā | r'īngēn | tēm mī | rābĕrĭs | ūmbrām |

It will be seen that, in the first and last of these lines, we have only two short syllables in thirteen, with an uninterrupted succession of no less than nine long syllables. But how are we to reconcile all this with a definition of versification which describes it as “the art of arranging words into lines of correspondent length so as to produce harmony by the regular alternation of syllables differing in quantity?

It may be urged, however, that our prosodist's intention was to speak of the English metres alone, and that, by omitting all mention of the spondee and [page 579:] pyrrhic, he has virtually avowed their exclusion from our rhythms. A grammarian is never excusable on the ground of good intentions. We demand from him, if from any one, rigorous precision of style. But grant the design. Let us admit that our author, following the example of all authors on English Prosody, has, in defining versification at large, intended a definition merely of the English. All these prosodists, we will say, reject the spondee and pyrrhic. Still all admit the iambus, which consists of a short syllable followed by a long; the trochee, which is the converse of the iambus; the dactyl, formed of one long syllable followed by two short; and the anapæst — two short succeeded by a long. The spondee is improperly rejected, as I shall presently show. The pyrrhic is rightfully dismissed. Its existence in either ancient or modern rhythm is purely chimerical, and the insisting on so perplexing a nonentity as a foot of two short syllables, affords, perhaps, the best evidence of the gross irrationality and subservience to authority which characterize our Prosody. In the meantime the acknowledged dactyl and anapæst are enough to sustain my proposition about the “alternation,” &c., without reference to feet which are assumed to exist in the Greek and Latin metres alone: for an anapæst and a dactyl may meet in the same line; when of course we shall have an uninterrupted succession of four short syllables. The meeting of these two feet, to be sure, is an accident not contemplated in the definition now discussed; for this definition, in demanding a “regular alternation of syllables differing in quantity,” insists on a regular succession of similar feet. But here is an example:

Sīng tŏ mĕ | Isăbēlle.

This is the opening line of a little ballad now before me, which proceeds in the same rhythm — a peculiarly beautiful one. More than all this: — English lines are often well composed, entirely, of a regular succession of syllables all of the same quantity: — the first lines, for instance, of the following quatrain by Arthur C. Coxe:

March! march! march!

Making sounds as they tread,

Ho! ho! how they step,

Going down to the dead!

The line italicized is formed of three cæsuras. The cæsura, of which I have much to say hereafter, is rejected by the English Prosodies and grossly misrepresented in the classic. It is a perfect foot — the most important in all verse — and consists of a single long syllable; but the length of this syllable varies.

It has thus been made evident that there is not one point of the definition in question which does not involve an error. And for anything more satisfactory or more intelligible we shall look in vain to any published treatise on the topic. [column 2:]

So general and so total a failure can be referred only to radical misconception. In fact the English Prosodists have blindly followed the pedants. These latter, like les moutons de Panurge, have been occupied in incessant tumbling into ditches, for the excellent reason that their leaders have so tumbled before. The Iliad, being taken as a starting point, was made to stand in stead of Nature and common sense. Upon this poem, in place of facts and deduction from fact, or from natural law, were built systems of feet, metres, rhythms, rules, — rules that contradict each other every five minutes, and for nearly all of which there may be found twice as many exceptions as examples. If any one has a fancy to be thoroughly confounded — to see how far the infatuation of what is termed “classical scholarship,” can lead a book-worm in the manufacture of darkness out of sunshine, let him turn over, for a few moments, any one of the German Greek Prosodies. The only thing clearly made out in them is a very magnificent contempt for Leibnitz's principle of “a sufficient reason.”

To divert attention from the real matter in hand by any farther reference to these works, is unnecessary, and would be weak. I cannot call to mind, at this moment, one essential particular of information that is to be gleaned from them; and I will drop them here with merely this one observation: that, employing from among the numerous “ancient” feet the spondee, the trochee, the iambus, the anapæst, the dactyl, and the cæsura alone, I will engage to scan correctly any of the Horatian rhythms, or any true rhythm that human ingenuity can conceive. And this excess of chimerical feet is, perhaps, the very least of the scholastic supererogations. Ex uno disce omnia. The fact is that Quantity is a point in whose investigation the lumber of mere learning may be dispensed with, if ever in any. Its appreciation is universal. It appertains to no region, nor race, nor æra in especial. To melody and to harmony the Greeks hearkened with ears precisely similar to those which we employ for similar purposes at present; and I should not be condemned for heresy in asserting that a pendulum at Athens would have vibrated much after the same fashion as does a pendulum in the city of Penn.

Verse originates in the human enjoyment of equality, fitness. To this enjoyment, also, all the moods of verse — rhythm, metre, stanza, rhyme, alliteration, the refrain, and other analogous effects — are to be referred. As there are some readers who habitually confound rhythm and metre, it may be as well here to say that the former concerns the character of feet (that is, the arrangements of syllables) while the latter has to do with the number of these feet. Thus by “a dactylic rhythm” we express a sequence of dactyls. By “a dactylic hexameter” we imply a line or measure consisting of six of these dactyls. [page 580:]

To return to equality. Its idea embraces those of similarity, proportion, identity, repetition, and adaptation or fitness. It might not be very difficult to go even behind the idea of equality, and show both how and why it is that the human nature takes pleasure in it, but such an investigation would, for any purpose now in view, be supererogatory. It is sufficient that the fact is undeniable — the fact that man derives enjoyment from his perception of equality. Let us examine a crystal. We are at once interested by the equality between the sides and between the angles of one of its faces: the equality of the sides pleases us; that of the angles doubles the pleasure. On bringing to view a second face in all respects similar to the first, this pleasure seems to be squared; on bringing to view a third it appears to be cubed, and so on. I have no doubt, indeed, that the delight experienced, if measurable, would be found to have exact mathematical relations such as I suggest; that is to say, as far as a certain point, beyond which there would be a decrease in similar relations.

The perception of pleasure in the equality of sounds is the principle of Music. Unpractised ears can appreciate only simple equalities, such as are found in ballad airs. While comparing one simple sound with another they are too much occupied to be capable of comparing the equality subsisting between these two simple sounds, taken conjointly, and two other similar simple sounds taken conjointly. Practised ears, on the other hand, appreciate both equalities at the same instant — although it is absurd to suppose that both are heard at the same instant. One is heard and appreciated from itself: the other is heard by the memory; and the instant glides into and is confounded with the secondary, appreciation. Highly cultivated musical taste in this manner enjoys not only these double equalities, all appreciated at once, but takes pleasurable cognizance, through memory, of equalities the members of which occur at intervals so great that the uncultivated taste loses them altogether. That this latter can properly estimate or decide on the merits of what is called scientific music, is of course impossible. But scientific music has no claim to intrinsic excellence — it is fit for scientific ears alone. In its excess it is the triumph of the physique over the morale of music. The sentiment is overwhelmed by the sense. On the whole, the advocates of the simpler melody and harmony have infinitely the best of the argument; — although there has been very little of real argument on the subject.

In verse, which cannot be better designated than as an inferior or less capable Music, there is, happily, little chance for complexity. Its rigidly simple character not even Science — not even Pedantry can greatly pervert.

The rudiment of verse may, possibly, be found in the spondee. The very germ of a thought seeking satisfaction in equality of sound, would result [column 2:] in the construction of words of two syllables, equally accented. In corroboration of this idea we find that spondees most abound in the most ancient tongues. The second step we can easily suppose to be the comparison, that is to say, the collocation of two spondees — of two words composed each of a spondee. The third step would be the juxta-position of three of these words. By this time the perception of monotone would induce farther consideration: and thus arises what Leigh Hunt so flounders in discussing under the title of “The Principle of Variety in Uniformity.” Of course there is no principle in the case — nor in maintaining it. The “Uniformity” is the principle: — the “Variety” is but the principle's natural safeguard from self-destruction by excess of self. “Uniformity,” besides, is the very worst word that could have been chosen for the expression of the general idea at which it aims.

The perception of monotone having given rise to an attempt at its relief, the first thought in this new direction would be that of collating two or more words formed each of two syllables differently accented (that is to say, short and long) but having the same order in each word: — in other terms, of collating two or more iambuses, or two or more trochees. And here let me pause to assert that more pitiable nonsense has been written on the topic of long and short syllables than on any other subject under the sun. In general, a syllable is long or short, just as it is difficult or easy of enunciation. The natural long syllables are those encumbered — the natural short ones are those unencumbered, with consonants; all the rest is mere artificiality and jargon. The Latin Prosodies have a rule that “a vowel before two consonants is long.” This rule is deduced from “authority” — that is, from the observation that vowels so circumstanced, in the ancient poems, are always in syllables long by the laws of scansion. The philosophy of the rule is untouched, and lies simply in the physical difficulty of giving voice to such syllables — of performing the lingual evolutions necessary for their utterance. Of course, it is not the vowel that is long (although the rule says so) but the syllable of which the vowel is a part. It will be seen that the length of a syllable, depending on the facility or difficulty of its enunciation, must have great variation in various syllables; but for the purposes of verse we suppose a long syllable equal to two short ones: — and the natural deviation from this relativeness we correct in perusal. The more closely our long syllables approach this relation with our short ones, the better, ceteris paribus, will be our verse: but if the relation does not exist of itself, we force it by emphasis, which can, of course, make any syllable as long as desired; — or, by an effort we can pronounce with unnatural brevity a syllable that is naturally too long. Accented syllables are of course always long — but, where unencumbered [page 581:] with consonants, must be classed among the unnaturally long. Mere custom has declared that we shall accent them — that is to say, dwell upon them; but no inevitable lingual difficulty forces us to do so. In fine, every long syllable must of its own accord occupy in its utterance, or must be made to occupy, precisely the time demanded for two short ones. The only exception to this rule is found in the cæsura — of which more anon.

The success of the experiment with the trochees or iambuses (the one would have suggested the other) must have led to a trial of dactyls or anapæsts — natural dactyls or anapæsts — dactylic or anapæstic words. And now some degree of complexity has been attained. There is an appreciation, first, of the equality between the several dactyls, or anapæsts, and, secondly, of that between the long syllable and the two short conjointly. But here it may be said that step after step would have been taken, in continuation of this routine, until all the feet of the Greek Prosodies became exhausted. Not so: — these remaining feet have no existence except in the brains of the scholiasts. It is needless to imagine men inventing these things, and folly to explain how and why they invented them, until it shall be first shown that they are actually invented. All other “feet” than those which I have specified, are, if not impossible at first view, merely combinations of the specified; and, although this assertion is rigidly true, I will, to avoid misunderstanding, put it in a somewhat different shape. I will say, then, that at present I am aware of no rhythm — nor do I believe that any one can be constructed — which, in its last analysis, will not be found to consist altogether of the feet I have mentioned, either existing in their individual and obvious condition, or interwoven with each other in accordance with simple natural laws which I will endeavor to point out hereafter.

We have now gone so far as to suppose men constructing indefinite sequences of spondaic, iambic, trochaic, dactylic, or anapæstic words. In extending these sequences, they would be again arrested by the sense of monotone. A succession of spondees would immediately have displeased; one of iambuses or of trochees, on account of the variety included within the foot itself, would have taken longer to displease; one of dactyls or anapæsts still longer: but even the last, if extended very far, must have become wearisome. The idea, first, of curtailing, and, secondly, of defining the length of a sequence, would thus at once have arisen. Here then is the line, of verse proper.* The principle [column 2:] of equality being constantly at the bottom of the whole process, lines would naturally be made, in the first instance, equal in the number of their feet; in the second instance there would be variation in the mere number; one line would be twice as long as another; then one would be some less obvious multiple of another; then still less obvious proportions would be adopted: — nevertheless there would be proportion, that is to say a phase of equality, still.

Lines being once introduced, the necessity of distinctly defining these lines to the ear, (as yet written verse does not exist,) would lead to a scrutiny of their capabilities at their terminations: — and now would spring up the idea of equality in sound between the final syllables — in other words, of rhyme. First, it would be used only in the iambic, anapæstic, and spondaic rhythms, (granting that the latter had not been thrown aside, long since, on account of its tameness;) because in these rhythms the concluding syllable, being long, could best sustain the necessary protection [[protraction]] of the voice. No great while could elapse, however, before the effect, found pleasant as well as useful, would be applied to the two remaining rhythms. But as the chief force of rhyme must lie in the accented syllable, the attempt to create rhyme at all in these two remaining rhythms, the trochaic and dactylic, would necessarily result in double and triple rhymes, such as beauty with duty (trochaic) and beautiful with dutiful (dactylic. [[sic)[[.]]

It must be observed that in suggesting these processes I assign them no date; nor do I even insist upon their order. Rhyme is supposed to be of modern origin, and were this proved, my positions remain untouched. I may say, however, in passing, that several instances of rhyme occur in the “Clouds” of Aristophanes, and that the Roman poets occasionally employ it. There is an effective species of ancient rhyming which has never descended to the moderns; that in which the ultimate and penultimate syllables rhyme with each other. For example:

Parturiunt montes et nascitur ridiculus mus.

and again —

Litoreis ingens inventa sub ilicibus sus.

The terminations of Hebrew verse, (as far as understood,) show no signs of rhyme; but what thinking person can doubt that it did actually exist? That men have so obstinately and blindly insisted, in general, even up to the present day, in confining rhyme to the ends of lines, when its effect is even better applicable elsewhere, intimates, in my opinion, the sense of some necessity in the connexion of the end with the rhyme — hints that the origin of rhyme lay in a necessity which connected [page 582:] it with the end — shows that neither mere accident nor mere fancy gave rise to the connexion — points, in a word, at the very necessity which I have suggested, (that of some mode of defining lines to the ear,) as the true origin of rhyme. Admit this and we throw the origin far back in the night of Time — beyond the origin of written verse.

But to resume. The amount of complexity I have now supposed to be attained is very considerable. Various systems of equalization are appreciated at once (or nearly so) in their respective values and in the value of each system with reference to all the others. As our present ultimatum of complexity, we have arrived at triple-rhymed, natural-dactylic lines, existing proportionally as well as equally with regard to other triple-rhymed, natural-dactylic lines. For example:

Virginal Lilian, rigidly, humblily dutiful;

Saintlily, lowlily,

Thrillingly, holily


Here we appreciate, first, the absolute equality between the long syllable of each dactyl and the two short conjointly; secondly, the absolute equality between each dactyl and any other dactyl — in other words, among all the dactyls; thirdly, the absolute equality between the two middle lines; fourthly, the absolute equality between the first line and all the others taken conjointly; fifthly, the absolute equality between the two last syllables of the respective words “dutiful” and “beautiful;” sixthly, the absolute equality between the two last syllables of the respective words “lowlily” and “holily;” seventhly, the proximate equality between the first syllable of “dutiful” and the first syllable of “beautiful;” eighthly, the proximate equality, between the first syllable of “lowlily” and that of “holily;” ninthly, the proportional equality, (that of five to one,) between the first line and each of its members, the dactyls; tenthly, the proportional equality, (that of two to one,) between each of the middle lines and its members, the dactyls; eleventhly, the proportional equality between the first line and each of the two middle — that of five to two; twelfthly, the proportional equality between the first line and the last — that of five to one; thirteenthly, the proportional equality between each of the middle lines and the last — that of two to one; lastly, the proportional equality, as concerns number, between all the lines, taken collectively, and any individual line — that of four to one.

The consideration of this last equality would give birth immediately to the idea of stanza* — that is to say, the insulation of lines into equal or obviously proportional masses. In its primitive, (which was also its best,) form, the stanza would most probably have had absolute unity. In other words, the removal of any one of its lines would have rendered it imperfect; as in the case above, where if the last line, for example, be taken away, there is left no rhyme to the “dutiful” of the first. Modern stanza is excessively loose, and where so, ineffective as a matter of course.

Now, although in the deliberate written statement which I have here given of these various systems of equalities, there seems to be an infinity of complexity — so much that it is hard to conceive the mind taking cognizance of them all in the brief period occupied by the perusal or recital of the stanza — yet the difficulty is in fact apparent only when we will it to become so. Any one fond of mental experiment may satisfy himself, by trial, that, in listening to the lines, he does actually, (although with a seeming unconsciousness, on account of the rapid evolutions of sensation,) recognize and instantaneously appreciate, (more or less intensely as his ear is cultivated,) each and all of the equalizations detailed. The pleasure received, or receivable, has very much such progressive increase, and in very nearly such mathematical relations, as those which I have suggested in the case of the crystal.

It will be observed that I speak of merely a proximate equality between the first syllable of “dutiful” and that of “beautiful;” and it may be asked why we cannot imagine the earliest rhymes to have had absolute instead of proximate equality of sound. But absolute equality would have involved the use of identical words; and it is the duplicate sameness or monotony — that of sense as well as that of sound — which would have caused these rhymes to be rejected in the very first instance.

The narrowness of the limits within which verse composed of natural feet alone, must necessarily have been confined, would have led, after a very brief interval, to the trial and immediate adoption of artificial feet — that is to say of feet not constituted each of a single word, but two or even three words; or of parts of words. These feet would be intermingled with natural ones. For example:

ă brēath | căn māke | thĕm ās | ă breāth [[brēath]] | hăs māde.

This is an iambic line in which each iambus is formed of two words. Again:

This is an iambic line in which the first foot is formed of a word and a part of a word; the second and third of parts taken from the body or interior of a word; the fourth of a part and a whole; the fifth of two complete words. There are no natural feet in either lines. Again: [page 583:]

Cān ĭt bĕ | fānciĕd thăt | Dēĭty | ēvĕr vĭn | dīctĭvely |

Māde ĭn hĭs | īmăgĕ ă | mānnĭkĭn | mĕrely tŏ | māddĕn ĭt? |

These are two dactylic lines in which we find natural feet, (“Deity,” “mannikin;”) feet composed of two words (“fancied that,” “image a,” “merely to,” “madden it;”) feet composed of three words (“can it be,” “made in his;”) a foot composed of a part of a word (“dictively;”) and a foot composed of a word and a part of a word (“ever vin. [[sic]]”) [[.]]

And now, in our supposititious progress, we have gone so far as to exhaust all the essentialities of verse. What follows may, strictly speaking, be recorded as embellishment merely — but even in this embellishment, the rudimental sense of equality would have been the never-ceasing impulse. It would, for example, be simply in seeking farther administration to this sense that men would come, in time, to think of the refrain, or burden, where, at the closes of the several stanzas of a poem, one word or phrase is repeated; and of alliteration, in whose simplest form a consonant is repeated in the commencements of various words. This effect would be extended so as to embrace repetitions both of vowels and of consonants, in the bodies as well as in the beginnings of words; and, at a later period, would be made to infringe on the province of rhyme, by the introduction of general similarity of sound between whole feet occurring in the body of a line: — all of which modifications I have exemplified in the line above,

Made in his image a mannikin merely to madden it.

Farther cultivation would improve also the refrain by relieving its monotone in slightly varying the phrase at each repetition, or, (as I have attempted to do in “The Raven,”) in retaining the phrase and varying its application — although this latter point is not strictly a rhythmical effect alone. Finally, poets when fairly wearied with following precedent — following it the more closely the less they perceived it in company with Reason — would adventure so far as to indulge in positive rhyme at other points than the ends of lines. First, they would put it in the middle of the line; then at some point where the multiple would be less obvious; then alarmed at their own audacity, they would undo all their work by cutting these lines in two. And here is the fruitful source of the infinity of “short metre,” by which modern poetry, if not distinguished, is at least disgraced. It would require a high degree, indeed, both of cultivation and of courage, on the part of any versifier, to enable him to place his rhymes — and let them remain — at unquestionably their best position, that of unusual and unanticipated intervals.

On account of the stupidity of some people, or, (if talent be a more respectable word,) on account of their talent for misconception — I think it necessary [column 2:] to add here, first, that I believe the “processes” above detailed to be nearly if not accurately those which did occur in the gradual creation of what we now call verse; secondly, that, although I so believe, I yet urge neither the assumed fact nor my belief in it, as a part of the true proposition of this paper; thirdly, that in regard to the aim of this paper, it is of no consequence whether these processes did occur either in the order I have assigned them, or at all; my design being simply, in presenting a general type of what such processes might have been and must have resembled, to help them, the “some people,” to an easy understanding of what I have farther to say on the topic of Verse.

There is one point which, in my summary of the processes, I have purposely forborne to touch; because this point, being the most important of all, on account of the immensity of error usually involved in its consideration, would have led me into a series of detail inconsistent with the object of a summary.

Every reader of verse must have observed how seldom it happens that even any one line proceeds uniformly with a succession, such as I have supposed, of absolutely equal feet; that is to say, with a succession of iambuses only, or of trochees only, or of dactyls only, or of anapæsts only, or of spondees only. Even in the most musical lines we find the succession interrupted. The iambic pentameters of Pope, for example, will be found on examination, frequently varied by trochees in the beginning, or by (what seem to be) anapæsts in the body, of the line.

ŏh thōu | whătē | vĕr tī | tlĕ pleāse | thĭne eār |

Dĕan Drā | piĕr Bĭck | ĕrstāff | ŏr Gūl [[Gūll]] | ĭvēr |

Whēthĕr | thŏu choōse | Cĕrvān | tĕs’ sē | rĭoŭs ăir |

ŏr laūgh | ănd shāke | ĭn Rāb | ĕlaĭs’ eā | sy chaīr. |

Were any one weak enough to refer to the Prosodies for a solution of the difficulty here, he would find it solved as usual by a rule, stating the fact, (or what it, the rule, supposes to be the fact,) but without the slightest attempt at the rationale. “By a synæresis of the two short syllables,” say the books, “an anapæst may sometimes be employed for an iambus, or a dactyl for a trochee. ... In the beginning of a line a trochee is often used for an iambus.”

Blending is the plain English for synæresis — but there should be no blending; neither is an anapæst ever employed for an iambus, or a dactyl for a trochee. These feet differ in time; and no feet so differing can ever be legitimately used in the same line. An anapæst is equal to four short syllables — an iambus only to three. Dactyls and trochees hold the same relation. The principle of equality, in verse, admits, it is true, of variation at certai [[certain]] points, for the relief of monotone, as I have already [page 584:] shown, but the point of time is that point which, being the rudimental one, must never be tampered with at all.

To explain: — In farther efforts for the relief of monotone than those to which I have alluded in the summary, men soon came to see that there was no absolute necessity for adhering to the precise number of syllables, provided the time required for the whole foot was preserved inviolate. They saw, for instance, that in such a line as

ŏr lāugh | ănd shāke | ĭn Rāb | ĕlaĭs ēa | sy chāir, |

the equalization of the three syllables elais ea with the two syllables composing any of the other feet, could be readily effected by pronouncing the two syllable [[syllables]] elais in double quick time. By pronouncing each of the syllables e and lais twice as rapidly as the syllable sy, or the syllable in, or any other short syllable, they could bring the two of them, taken together, to the length, that is to say to the time, of any one short syllable. This consideration enabled them to effect the agreeable variation of three syllables in place of the uniform two. And variation was the object — variation to the ear. What sense is there, then, in supposing this object rendered null by the blending of the two syllables so as to render them, in absolute effect, one? Of course, there must be no blending. Each syllable must be pronounced as distinctly as possible, (or the variation is lost,) but with twice the rapidity in which the ordinary short syllable is enunciated. That the syllables elais ea do not compose an anapæst is evident, and the signs ( ̆ ̆ ̄) of their accentuation are erroneous. The foot might be written thus ( ̯ ̯ ̱) the inverted crescents expressing double quick time; and might be called a bastard iambus.

Here is a trochaic line:

The prosodies — that is to say the most considerate of them — would here decide that “delicate” is a dactyl used in place of a trochee, and would refer to what they call their “rule,” for justification. Others, varying the stupidity, would insist upon a Procrustean adjustment thus (del’cate) — an adjustment recommended to all such words as silvery, murmuring, etc., which, it is said, should be not only pronounced, but written silv’ry, murm’ring, and so on, whenever they find themselves in trochaic predicament. I have only to say that “delicate,” when circumstanced as above, is neither a dactyl nor a dactyl's equivalent; that I would suggest for it this ( ¯ ˆ ˆ) accentuation; that I think it as well to call it a bastard trochee; and that all words, at all events, should be written and pronounced in full, and as nearly as possible as nature intended them. [column 2:]

About eleven years ago, there appeared in “The American Monthly Magazine,” (then edited, I believe, by Mess. Hoffman and Benjamin,) a review of Mr. Willis’ Poems; the critic putting forth his strength, or his weakness, in an endeavor to show that the poet was either absurdly affected, or grossly ignorant of the laws of verse; the accusation being based altogether on the fact that Mr. W. made occasional use of this very word “delicate,” and other similar words, in “the Heroic measure which every one knew consisted of feet of two syllables.” Mr. W. has often, for example, such lines as

That binds him to a woman's delicate love —

In the gay sunshine, reverent in the storm —

With its invisible fingers my loose hair.

Here, of course, the feet licate love, verent in, and sible fin, are bastard iambuses; are not anapæsts; and are not improperly used. Their employment, on the contrary, by Mr. Willis is but one of the innumerable instances he has given of keen sensibility in all those matters of taste which may be classed under the general head of fanciful embellishment.

It is also about eleven years ago, if I am not mistaken, since Mr. Horne, (of England,) the author of “Orion,” one of the noblest epics in any language, thought it necessary to preface his “Chaucer Modernized” by a very long and evidently a very elaborate essay, of which the greater portion was occupied in a discussion of the seemingly anomalous foot of which we have been speaking. Mr. Horne upholds Chaucer in its frequent use; maintains his superiority, on account of his so frequently using it, over all English versifiers; and, indignantly repelling the common idea of those who make verse on their fingers — that the superfluous syllable is a roughness and an error — very chivalrously makes battle for it as “a grace.” That a grace it is, there can be no doubt; and what I complain of is, that the author of the most happily versified long poem in existence, should have been under the necessity of discussing this grace merely as a grace, through forty or fifty vague pages, solely because of his inability to show how and why it is a grace — by which showing the question would have been settled in an instant.

About the trochee used for an iambus, as we see in the beginning of the line,

Whēthĕr thou choose Cervantes’ serious air,

there is little that need be said. It brings me to the general proposition that, in all rhythms, the prevalent or distinctive feet may be varied at will, and nearly at random, by the occasional introduction of equivalent feet — that is to say, feet the sum of whose syllabic times is equal to the sum of the syllabic times of the distinctive feet. Thus the troches [page 585:] whēthĕr, is equal, in the sum of the times of its syllables, to the iambus, thŏu choōse, in the sum of the times of its syllables; each foot being, in time, equal to three short syllables. Good versifiers who happen to be, also, good poets, contrive to relieve the monotone of a series of feet, by the use of equivalent feet only at rare intervals, and at such points of their subject as seem in accordance with the startling character of the variation. Nothing of this care is seen in the line quoted above — although Pope has some fine instances of the duplicate effect. Where vehemence is to be strongly expressed, I am not sure that we should be wrong in venturing on two consecutive equivalent feet — although I cannot say that I have ever known the adventure made, except in the following passage, which occurs in “Al Aaraaf,” a boyish poem, written by myself when a boy. I am referring to the sudden and rapid advent of a star:

Dim was its little disk, and angel eyes

Alone could see the phantom in the skies,

Whĕn fīrst thĕ phāntǒm's cōurse wăs fōund tǒ bē

Hēadlǒng hī thĕr ward o’er the starry sea.

In the “general proposition” above, I speak of the occasional introduction of equivalent feet. It sometimes happens that unskilful versifiers, without knowing what they do, or why they do it, introduce so many “variations” as to exceed in number the “distinctive” feet; when the ear becomes at once baulked by the bouleversement of the rhythm. Too many trochees, for example, inserted in an iambic rhythm, would convert the latter to a trochaic. I may note here, that, in all cases, the rhythm designed should be commenced and continued, without variation, until the ear has had full time to comprehend what is the rhythm. In violation of a rule so obviously founded in common sense, many even of our best poets, do not scruple to begin an iambic rhythm with a trochee, or the converse; or a dactylic with an anapæst, or the converse; and so on.

A somewhat less objectionable error, although still a decided one, is that of commencing a rhythm, not with a different equivalent foot, but with a “bastard” foot of the rhythm intended. For example:

Here many a is what I have explained to be a bastard trochee, and to be understood should be accented with inverted crescents. It is objectionable solely on account of its position as the opening foot of a trochaic rhythm. Memory, similarly accented, is also a bastard trochee, but unobjectionable, although by no means demanded.

The farther illustration of this point will enable me to take an important step.

(To be continued.)


[The following footnote appears at the bottom of page 577, column 1:]

*  Some few passages of this article appeared, about four years ago, in “The Pioneer,” a monthly Magazine published by J. R. Lowell and R. Carter. Although an excellent work it had a very limited circulation.

[The following footnote appears at the bottom of page 581, column 1:]

*  Verse, from the Latin vertere, to turn, is so called on account of the turning or recommencement of the series of feet. Thus a verse, strictly speaking, is a line. In this sense, however, I have preferred using the latter word alone; employing the former in the general acceptation given it in the heading of this paper.

[The following footnote appears at the bottom of page 582, column 1:]

*  A stanza is often vulgarly, and with gross impropriety, called a verse.


This essay was originally published as “Notes Upon English Verse” in James R. Lowell's short-lived magazine The Pioneer for March 1843.

Poe's use of special accents makes this text a great challenge for the typesetter, and especially for a presentation in HTML. Although every effort has been made to replicate such markings, a few compromises have necessarily had to be accepted. Poe's use of numbers directly under or above some letters has also presented a problem here, and we have had to accept the slight variation of making these, respectively, subscripts and superscripts, meaning that they appear next to the letter rather than Poe's precise placement. As a result, an unauthentic spacing has been created, which does not fully recreate the text, but will have to suffice.


[S:2 - SLM, 1848] - Edgar Allan Poe Society of Baltimore - Works - Essays - The Rationale of Verse [part I] (Text-04)