Text: Edgar Allan Poe, “The Rationale of Verse” [Part II], Southern Literary Messenger, November 1848, 14:673-682


[page 673, column 1:]



[Concluded from our last Number.]

One of our finest poets, Mr. Christopher Pease [[Pearse]] Cranch, begins a very beautiful poem thus:

Many are the thoughts that come to me

In my lonely musing;

And they drift so strange and swift

There's no time for choosing

Which to follow; for to leave

Any, seems a losing.

“A losing” to Mr. Cranch, of course — but this en passant. It will be seen here that the intention is trochaic; — although we do not see this intention by the opening foot, as we should do or even by the opening line. Reading the whole stanza, however, we perceive the trochaic rhythm as the general design, and so, after some reflection, we divide the first line thus:

Many are the | thōughts thăt | cōme tŏ | me [[mē]]. |

Thus scanned, the line will seem musical. It is — highly so. And it is because there is no end to instances of just such lines of apparently incomprehensible music, that Coleridge thought proper to invent his nonsensical system of what he calls “scanning by accents” — as if “scanning by accents” were anything more than a phrase. Whenever “Christabel” is really not rough, it can be as readily scanned by the true laws (not the supposititious rules) of verse, as can the simplest pentameter of Pope; and where it is rough (passim) these same laws will enable any one of common sense to show why it is rough and to point out, instantaneously, the remedy for the roughness.

A reads and re-reads a certain line, and pronounces it false in rhythm — unmusical. B, however, reads it to A, and A is at once struck with the perfection of the rhythm, and wonders at his dulness in not “catching” it before. Henceforward he admits the line to be musical. B, triumphant, asserts that, to be sure, the line is musical — for it is the work of Coleridge — and that it is A who is not; the fault being in A's [[A's]] false reading. Now here A is right and B wrong. That rhythm is erroneous, (at some point or other more or less obvious,) which any ordinary reader can, without design, read improperly. It is the business of the poet so to construct his line that the intention must be caught at once. Even when men have precisely the same understanding of a sentence, they differ and often widely, in their modes of enunciating it. Any one who has taken the trouble to examine the topic of emphasis, (by which I here mean not accent of particular syllables, but the dwelling on entire words,) [column 2:] must have seen that men emphasize in the most singularly arbitrary manner. There are certain large classes of people, for example, who persist in emphasizing their monosyllables. Little uniformity of emphasis prevails; because the thing itself — the idea, emphasis, — is referable to no natural — at least to no well comprehended and therefore uniform law. Beyond a very narrow and vague limit, the whole matter is conventionality. And if we differ in emphasis even when we agree in comprehension, how much more so in the former when in the latter too! Apart, however, from the consideration of natural disagreement, is it not clear that, by tripping here and mouthing there, any sequence of words may be twisted into any species of rhythm? But are we thence to deduce that all sequences of words are rhythmical in a rational understanding of the term? — for this is the deduction, precisely to which the reductio ad absurdum will, in the end, bring all the propositions of Coleridge. Out of a hundred readers of “Christabel,” fifty will be able to make nothing of its rhythm, while forty-nine of the remaining fifty will, with some ado, fancy they comprehend it, after the fourth or fifth perusal. The one out of the whole hundred who shall both comprehend and admire it at first sight — must be an unaccountably clever person — and I am by far too modest to assume, for a moment, that that very clever person is myself.

In illustration of what is here advanced I cannot do better than quote a poem:

Pease porridge hot pease porridge cold

Pease porridge in the pot — nine days old.

Now those of my readers who have never heard this poem pronounced according to the nursery conventionality, will find its rhythm as obscure as an explanatory note; while those who have heard it, will divide it thus, declare it musical, and wonder how there can be any doubt about it.

Pease | porridge | hot | pease | porridge | cold |

Pease | porridge | in the | pot | nine | days | old. |

The chief thing in the way of this species of rhythm, is the necessity which it imposes upon the poet of travelling in constant company with his compositions, so as to be ready at a moment's notice, to avail himself of a well understood poetical license — that of reading aloud one's own doggerel.

In Mr. Cranch's line,

Many are the | thoughts that | come to | me, |

the general error of which I speak is, of course, very partially exemplified, and the purpose for which, chiefly, I cite it, lies yet further on in our topic.

The two divisions (thoughts that) and (come to) are ordinary trochees. Of the last division (me) we will talk herafter. The first division (many [page 674:] are the) would be thus accented by the Greek Prosodies (māny ăre thĕ) and would be called by them s αστρολογος. The Latin books would style the foot Pæon Primus, and both Greek and Latin would swear that it was composed of a trochee and what they term a pyrrhic — that is to say a foot of two short syllables — a thing that cannot be, as I shall presently show.

But now, there is an obvious difficulty. The astrologos, according to the Prosodies’ own showing, is equal to five short syllables, and the trochee to three — yet, in the line quoted, these two feet are equal. They occupy precisely the same time. In fact, the whole music of the line depends upon their being made to occupy the same time. The Prosodies then, have demonstrated what all mathematicians have stupidly failed in demonstrating — that three and five are one and the same thing.

After what I have already said, however, about the bastard trochee and the bastard iambus, no one can have any trouble in understanding that many are the is of similar character. It is merely a bolder variation than usual from the routine of trochees, and introduces to the bastard trochee one additional syllable. But this syllable is not short. That is, it is not short in the sense of “short” as applied to the final syllable of the ordinary trochee, where the word means merely the half of long.

In this case (that of the additional syllable) “short,” if used at all, must be used in the sense of the sixth of long. And all the three final syllables can be called short only with the same understanding of the term. The three together are equal only to the one short syllable (whose place they supply) of the ordinary trochee. It follows that there is no sense in thus ( ˘) accenting these syllables. We must devise for them some new character which shall denote the sixth of long. Let it be ( ‹) — the crescent placed with the curve to the left. The whole foot (māny are the) might be called a quick trochee.

We come now to the final division (me) of Mr. Cranch's line. It is clear that this foot, short as it appears, is fully equal in time to each of the preceding. It is in fact the cæsura — the foot which, in the beginning of this paper, I called the most important in all verse. Its chief office is that of pause or termination; and here — at the end of a line — its use is easy, because there is no danger of misapprehending its value. We pause on it, by a seeming necessity, just so long as it has taken us to pronounce the preceding feet, whether iambus [[iambuses]], trochees, dactyls or anapæsts. It is thus a variable foot, and, with some care, may be well introduced into the body of a line, as in a little poem of great beauty by Mrs. Welby:

Here we dwell on the cæsura, son, just as long as [column 2:] it requires us to pronounce either of the preceding or succeeding iambusses [[iambuses]]. Its value, therefore, in this line, is that of three short syllables. In the following dactylic line its value is that of four short syllables.

I have accented the cæsura with a (~~~) by way of expressing this variability of value.

I observed, just now, that there could be no such foot as one of two short syllables. What we start from in the very beginning of all idea on the topic of verse, is quantity, length. Thus when we enunciate an independent syllable it is long, as a matter of course. If we enunciate two, dwelling on both equally, we express equality in the enumeration, or length, and have a right to call them two long syllables. If we dwell on one more than the other, we have also a right to call one short, because it is short in relation to the other. But if we dwell on both equally and with a tripping voice, saying to ourselves here are two short syllables, the query might well be asked of us — “in relation to what are they short?” Shortness is but the negation of length. To say, then, that two syllables, placed independently of any other syllable, are short, is merely to say that they have no positive length, or enunciation — in other words that they are no syllables — that they do not exist at all. And if, persisting, we add anything about their equality, we are merely floundering in the idea of an identical equation, where, x being equal to x, nothing is shown to be equal to zero. In a word we can form no conception of a pyrrhic as of an independent foot. It is a mere chimera bred in the mad fancy of a pedant.

From what I have said about the equalization of the several feet of a line, it must not be deduced that any necessity for equality in time exists between the rhythm of several lines. A poem, or even a stanza, may begin with iambuses, in the first line, and proceed with anapæsts in the second, or even with the less accordant dactyls, as in the opening of quite a pretty specimen of verse by Miss Mary A. S. Aldrich:

The wa | ter li | ly sleeps | in pride |

Dōwn ĭn thĕ | dēpths ŏf  thĕ | āzūre | lake. |

Here azure is a spondee, equivalent to a dactyl; lake a cæsura.

I shall now best proceed in quoting the initial lines of Byron's “Bride of Abpdos [[Abydos]]:”

Know ye the land where the cypress and myrtle

Are emblems of deeds that are done in their clime —

Where the rage of the vulture, the love of the turtle

Now melt into softness, now madden to crime?

Know ye the land of the cedar and vine,

Where the flowers ever blossom, the beams ever shine,

And the light wings of Zephyr, oppressed with perfume, [page 675:]

Wax faint o’er the gardens of Gul in their bloom?

Where the citron and olive are fairest of fruit

And the voice of the nightingale never is mute —

Where the virgins are soft as the roses they twine,

And all save the spirit of man is divine?

’Tis the land of the East — ‘tis the land of the Sun —

Can he smile on such deeds as his children have done?

Oh, wild as the accents of lovers’ farewell

Are the hearts that they bear and the tales that they tell.

Now the flow of these lines, (as times go,) is very sweet and musical. They have been often admired, and justly — as times go — that is to say, it is a rare thing to find better versification of its kind. And where verse is pleasant to the ear, it is silly to find fault with it because it refuses to be scanned. Yet I have heard men, professing to be scholars, who made no scruple of abusing these lines of Byron's on the ground that they were musical in spite of all law. Other gentlemen, not scholars, abused “all law” for the same reason: — and it occurred neither to the one party nor to the other that the law about which they were disputing might possibly be no law at all — an ass of a law in the skin of a lion.

The Grammars said something about dactylic lines, and it was easily seen that these lines were at least meant for dactylic. The first one was, therefore, thus divided:

Knōw yĕ thĕ | lānd whĕre thĕ | cyprĕss ănd | myrtle [[myrtlĕ]]. |

The concluding foot was a mystery; but the Prosodies said something about the dactylic “measure” calling now and then for a double rhyme; and the court of enquiry [[inquiry]] were content to rest in the double rhyme, without exactly perceiving what a double rhyme had to do with the question of an irregular foot. Quitting the first line, the second was thus scanned:

Arē ĕmblĕms | ōf deĕds thăt | āre dŏne ĭn | thēir clĭme. |

It was immediately seen, however, that this would not do: — it was at war with the whole emphasis of the reading. It could not be supposed that Byron, or any one in his senses, intended to place stress upon such monosyllables as “are,” “of,” and “their,” nor could “their clime,” collated with “to crime,” in the corresponding line below, be fairly twisted into anything like a “double rhyme,” so as to bring everything within the category of the Grammars. But farther these Grammars spoke not. The inquirers, therefore, in spite of their sense of harmony in the lines, when considered without reference to scansion, fell back upon the idea that the “Are” was a blunder — an excess for which the poet should be sent to Coventry — and, striking it out, they scanned the remainder of the line as follows:

—— ēmblĕms ŏf | deĕds thăt ăre | dōne ĭn thĕir | clĭme. |   [column 2:]

This answered pretty well; but the Grammars admitted no such foot as a foot of one syllable; and besides the rhythm was dactylic. In despair, the books are well searched, however, and at last the investigators are gratified by a full solution of the riddle in the profound “Observation” quoted in the beginning of this article: — “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.” This is enough. The anomalous line is pronounced to be catalectic at the head and to form hypermeter at the tail: — and so on, and so on; it being soon discovered that nearly all the remaining lines are in a similar predicament, and that what flows so smoothly to the ear, although so roughly to the eye, is, after all, a mere jumble of catalecticism, acatalecticism, and hypermeter — not to say worse.

Now, had this court of inquiry been in possession of even the shadow of the philosophy of Verse, they would have had no trouble in reconciling this oil and water of the eye and ear, by merely scanning the passage without reference to lines, and, continuously, thus:

Know ye the | land where the | cypress and | myrtle Are | emblems of | deeds that are | done in their | clime Where the | rage of the | vulture the | love of the | turtle Now | melt into | softness now | madden to | crime [[ |]] Know ye the | land of the | cedar and | vine Where the | flowers ever | blossom the | beams ever | shine Where [[And]] the | light wings of | Zephyr op | pressed by per | fume Wax | faint o’er the | gardens of | Gul in their | bloom Where the | citron and | olive are | fairest of | fruit And the | voice of the | nightingale | never is | mute Where the | virgins are | soft as the | roses they |twine And | all save the | spirit of | man is di | vine. ‘Tis the | land of the | East ‘tis the | clime [[land]] of the | Sun Can he | smile on such | deeds as his | children have | done Oh | wild as the | accents of | lovers’ fare | well Are the | hearts that they | bear and the | tales that they | tell.

Here “crime” and “tell” (italicised) are cæsuras, each having the value of a dactyl, four short syllables; while “fume Wax,” “twine and,” and “done Oh,” are spondees which, of course, being composed of two long syllables, are also equal to four short, and are the dactyl's natural equivalent. The nicety of Byron's ear has led him into a succession of feet which, with two trivial exceptions as regards melody, are absolutely accurate — a very rare occurrence this in dactylic or anapæstic rhythms. The exceptions are found in the spondee “twine And” and the dactyl, “smile on such.” Both feet are false in point of melody. In “twine And,” to make out the rhythm, we must force “And” into a length which it will not naturally bear. We are called on to sacrifice either the proper length of the syllable as demanded by its position as a member of a spondee, or the customary accentuation of the word in conversation. There is no hesitation, and should be none. We at once [page 676:] give up the sound for the sense; and the rhythm is imperfect. In this instance it is very slightly so; — not one person in ten thousand could, by ear, detect the inaccuracy. But the imperfection [[perfection]] of verse, as regards melody, consists in its never demanding any such sacrifice as is here demanded. The rhythmical must agree, thoroughly, with the reading, flow. This perfection has in no instance been attained — but is unquestionably attainable. “Smile on such,” a dactyl, is incorrect, because “such,” from the character of the two consonants ch, cannot easily be enunciated in the ordinary time of a short syllable, which its position declares that it is. Almost every reader will be able to appreciate the slight difficulty here; and yet the error is by no means so important as that of the “And “ in the spondee. By dexterity we may pronounce “such” in the true time; but the attempt to remedy the rhythmical deficiency of the And by drawing it out, merely aggravates the offence against natural enunciation, by directing attention to the offence.

My main object, however, in quoting these lines, is to show that, in spite of the Prosodies, the length of a line is entirely an arbitrary matter. We might divide the commencement of Byron's poem thus:

Know ye the | land where the. |

or thus:

Know ye the | land where the | cypress and. |

or thus:

Know ye the | land where the | cypress and | myrtle are. |

or thus:

Know ye the | land where the | cypress and | myrtle are | emblems of. |

In short we may give it any division we please, and the lines will be good — provided we have at least two feet in a line. As in mathematics two units are required to form number, so rhythm, (from the Greek αριθμος, number,) demands for its formation at least two feet. Beyond doubt, we often see such lines as

Know ye the —

Land where the —

lines of one foot; and our Prosodies admit such; but with impropriety; for common sense would dictate that every so obvious division of a poem as is made by a line, should include within itself all that is necessary for its own comprehension; but in a line of one foot we can have no appreciation of rhythm, which depends upon the equality between two or more pulsations. The false lines, consisting sometimes of a single cæsura, which are seen in mock Pindaric odes, are of course “rhythmical” only in connection with some other [column 2:] line; and it is this want of independent rhythm which adapts them to the purposes of burlesque alone. Their effect is that of incongruity (the principle of mirth;) for they include the blankness of prose amid the harmony of verse.

My second object in quoting Byron's lines, was that of showing how absurd it often is to cite a single line from amid the body of a poem, for the purpose of instancing the perfection or imperfection of the line's rhythm. Were we to see by itself

Know ye the land where the cypress and myrtle,

we might justly condemn it as defective in the final foot, which is equal to only three, instead of being equal to four, short syllables.

In the foot (flowers ever) we shall find a further exemplification of the principle in the bastard iambus, bastard trochee, and quick trochee, as I have been at some pains in describing these feet above. All the Prosodies on English verse would insist upon making an elision in “flowers,” thus (flow’rs,) but this is nonsense. In the quick trochee (māny are the) occurring in Mr. Cranch's trochaic line, we had to equalize the time of the three syllables [[(]]ny, are, the,) to that of the one short syllable whose position they usurp. Accordingly each of these syllables is equal to the third of a short syllable, that is to say, the sixth of a long. But in Byron's dactylic rhythm, we have to equalize the time of the three syllables (ers, ev, er,) to that of the one long syllable whose position they usurp or, (which is the same thing,) of the two short. Therefore the value of each of the syllables (ers, ev, and er) is the third of a long. We enunciate them with only half the rapidity we employ in enunciating the three final syllables of the quick trochee — which latter is a rare foot. The “flowers ever,” on the contrary, is as common in the dactylic rhythm as is the bastard trochee in the trochaic, or the bastard iambus in the iambic. We may as well accent it with the curve of the crescent to the right, and call it a bastard dactyl. A bastard anapæst, whose nature I now need be at no trouble in explaining, will of course occur, now and then, in an anapæstic rhythm.

In order to avoid any chance of that confusion which is apt to be introduced in an essay of this kind by too sudden and radical an alteration of the conventionalities to which the reader has been accustomed, I have thought it right to suggest for the accent marks of the bastard trochee, bastard iambus, etc., etc., certain characters which, in merely varying the direction of the ordinary short accent ( ˘) should imply, what is the fact, that the feet themselves are not new feet, in any proper sense, but simply modifications of the feet, respectively, from which they derive their names. Thus a bastard iambus is, in its essentiality, that is to say, in its time, an iambus. The variation lies only in the [page 677:] distribution of this time. The time, for example, occupied by the one short (or half of long) syllable, in the ordinary iambus, is, in the bastard, spread equally over two syllables, which are accordingly the fourth of long.

But this fact — the fact of the essentiality, or whole time, of the foot being unchanged, is now so fully before the reader, that I may venture to propose, finally, an accentuation which shall answer the real purpose — that is to say what should be the real purpose of all accentuation — the purpose of expressing to the eye the exact relative value of every syllable employed in Verse.

I have already shown that enunciation, or length, is the point from which we start. In other words, we begin with a long syllable. This then is our unit; and there will be no need of accenting it at all. An unaccented syllable, in a system of accentuation, is to be regarded always as a long syllable. Thus a spondee would be without accent. In an iambus, the first syllable being “short,” or the half of long, should be accented with a small 2, placed beneath the syllable; the last syllable, being long, should be unaccented; — the whole would be thus (co2 ntrol.) In a trochee, these accents would be merely conversed, thus (manl2 y.) In a dactyl, each of the two final syllables, being the half of long, should also be accented with a small 2 beneath the syllable; and the first syllable left unaccented, the whole would be thus (happi2 nes2 s.) In an anapæst we should converse the dactyl thus, (in2 th2 e land.) In the bastard dactyl, each of the three concluding syllables being the third of long, should be accented with a small 3 beneath the syllable, and the whole foot would stand thus, (flower3 s  e3 ve3 r.) In the bastard anapæst we should converse the bastard dactyl thus, (i3 n th3 e re3 bound.) In the bastard iambus, each of the two initial syllables, being the fourth of long, should be accented, below, with a small 4; the whole foot would be thus, (i4 n th4 e rain.) In the bastard trochee, we should converse the bastard iambus thus, (many4 a4.) In the quick trochee, each of the three concluding syllables, being the sixth of long, should be accented, below, with a small 6; the whole foot would be thus, (many6   a6 re  th6 e.) The quick iambus is not yet created, and most probably never will be; for it would be excessively useless, awkward, and liable to misconception — as I have already shown that even the quick trochee is: — but, should it appear, we must accent it by conversing the quick trochee. The cæsura, being variable in length, but always longer thanlong,” should be accented, above, with a number expressing the length, or value, of the distinctive foot of the rhythm in which it [column 2:] occurs. Thus a cæsura, occurring in a spondaic rhythm, would be accented with a small 2 above the syllable, or, rather, foot. Occurring in a dactylic or anapæstic rhythm, we also accent it with the 2, above the foot. Occurring in an iambic rhythm, however, it must be accented, above, with 1 1/2; for this is the relative value of the iambus. Occurring in the trochaic rhythm, we give it, of course, the same accentuation. For the complex 1 1/2, however, it would be advisable to substitute the simpler expression 3 2 [[3/2]], which amounts to the same thing

In this system of accentuation Mr. Cranch's lines, quoted above, would thus be written:

Many6   are6   the6 | thoughts tha2 t | come to2 | me3/2

In my2 | lone2ly | musin2g, |

And the2y | drift s2o | strange an2d | swi3/2ft

There's n2o | time fo2 r | choos2 ing |

Which t2o | follo2w | for t2o | leav3/2e

An2y, | seems a2 | losin2g. |

In the ordinary system the accentuation would be thus:

Māny arĕ thĕ | thōughts thăt | cōme tŏ | mē |

In my | lōnely | mūsing, |

ānd thĕy | drīft sŏ | strānge ănd | swīft |

Thēre's nŏ | timē fŏr | choōsing |

Whīch tŏ | fōllŏw, | fōr tŏ | lēave

āny, | seēms ă | lōsĭng. |

It must first be observed, here, that I do not grant this to be the “ordinary” scansion. On the contrary, I never yet met the man who had the faintest comprehension of the true scanning of these lines, or of such as these. But granting this to be the mode in which our Prosodies would divide the feet, they would accentuate the syllables as just above.

Now, let any reasonable person compare the two modes. The first advantage seen in my mode is that of simplicity — of time, labor, and ink saved. Counting the fractions as two accents, even, there will be found only twenty-six accents to the stanza. In the common accentuation there are forty-one. But admit that all this is a trifle, which it is not, and let us proceed to points of importance. Does the common accentuation express the truth, in particular, in general, or in any regard? Is it consistent with itself? Does it convey either to the ignorant or to the scholar a just conception of the rhythm of the lines? Each of these questions must be answered in the negative. [page 678:] The crescents, being precisely similar, must be understood as expressing, all of them, one and the same thing; and so all prosodies have always understood them and wished them to be understood. They express, indeed, “short” — but this word has all kinds of meanings. It serves to represent (the reader is left to guess when) sometimes the half, sometimes the third, sometimes the fourth, and sometimes the sixth, of “long” — while “long” itself, in the books, is left undefined and undescribed. On the other hand, the horizontal accent, it may be said, expresses sufficiently well, and unvaryingly, the syllables which are meant to be long. It does nothing of the kind. This horizontal accent is placed over the cæsura (wherever, as in the Latin Prosodies, the cæsura is recognized) as well as over the ordinary long syllable, and implies anything and everything, just as the crescent. But grant that it does express the ordinary long syllables, (leaving the cæsura out of question,) have I not given the identical expression, by not employing any expression at all? In a word, while the Prosodies, with a certain number of accents, express precisely nothing whatever, I, with scarcely half the number, have expressed everything which, in a system of accentuation, demands expression. In glancing at my mode in the lines of Mr. Cranch, it will be seen that it conveys not only the exact relation of the syllables and feet, among themselves, in those particular lines, but their precise value in relation to any other existing or conceivable feet or syllables, in any existing or conceivable system of rhythm.

The object of what we call scansion is the distinct making [[marking]] of the rhythmical flow. Scansion without accents or perpendicular lines between the feet — that is to say scansion by the voice only — is scansion to the ear only; and all very good in its way. The written scansion addresses the ear through the eye. In either case the object is the distinct making [[marking]] of the rhythmical, musical, or reading flow. There can be no other object and there is none. Of course, then, the scansion and the reading flow should go hand in hand. The former must agree with the latter. The former represents and expresses the latter; and is good or bad as it truly or falsely represents and expresses it. If by the written scansion of a line we are not enabled to perceive any rhythm or music in the line, then either the line is [[un]]rhythmical or the scansion false. Apply all this to the English lines which we have quoted, at various points, in the course of this article. It will be found that the scansion exactly conveys the rhythm, and thus thoroughly fulfils the only purpose for which scansion is required.

But let the scansion of the schools be applied to the Greek and Latin verse, and what result do we find? — that the verse is one thing and the scansion quite another. The ancient verse, read aloud, is [column 2:] in general musical, and occasionally very musical. Scanned by the Prosodial rules we can, for the most part, make nothing of it whatever. In the case of the English verse, the more emphatically we dwell on the divisions between the feet, the more distinct is our perception of the kind of rhythm intended. In the case of the Greek and Latin, the more we dwell the less distinct is this perception. To make this clear by an example:

Mæcenas, atavis edite regibus,

O, et præsidium et dulce decus meum,

Sunt quos curriculo pulverem Olympicum

Collegisse juvat, metaque fervidis

Evitata rotis, palmaque nobilis

Terrarum dominos evehit ad Deos.

Now in reading these lines, there is scarcely one person in a thousand who, if even ignorant of Latin, will not immediately feel and appreciate their flow — their music. A prosodist, however, informs the public that the scansion runs thus:

Mæce | nas ata | vis | edite | regibus |

O, et | præsidi’ | et | dulce de | cus meum |

Sunt quos | curricu | lo | pulver’ O | lympicum |

Colle | gisse ju | vat | metaque | fervidis |

Evi | tata ro | tis | palmaque | nobilis |

Terra | rum domi | nos | evehit | ad Deos. |

Now I do not deny that we get a certain sort of music from the lines if we read them according to this scansion, but I wish to call attention to the fact that this scansion and the certain sort of music which grows out of it, are entirely at war not only with the reading flow which any ordinary person would naturally give the lines, but with the reading flow universally given them, and never denied them, by even the most obstinate and stolid of scholars.

And now these questions are forced upon us — “Why exists this discrepancy between the modern verse with its scansion, and the ancient verse with its scansion?” — “Why, in the former case, are there agreement and representation, while in the latter there is neither the one nor the other?” or, to come to the point, — “How are we to reconcile the ancient verse with the scholastic scansion of it?” This absolutely necessary conciliation — shall we bring it about by supposing the scholastic scansion wrong because the ancient verse is right, or by maintaining that the ancient verse is wrong because the scholastic scansion is not to be gainsaid?

Were we to adopt the latter mode of arranging the difficulty, we might, in some measure, at least simplify the expression of the arrangement by putting it thus — Because the pedants have no eyes, therefore the old poets had no ears.

“But,” say the gentlemen without the eyes, “the scholastic scansion, although certainly not handed down to us in form from the old poets themselves (the gentlemen without the ears,) is nevertheless deduced, Baconially, from certain facts [page 679:] which are supplied us by careful observation of the old poems.[[“]]

And let us illustrate this strong position by an example from an American poet — who must be a poet of some eminence, or he will not answer the purpose. Let us take Mr. Alfred B. Street. I remember these two lines of his:

His sinuous path, by blazes, wound

Among trunks grouped in myriads round.

With the sense of these lines I have nothing to do. When a poet is in a “fine phrensy” he may as well imagine a large forest as a small one — and “by blazes!” is not intended for an oath. My concern is with the rhythm, which is iambic.

Now let us suppose that, a thousand years hence, when the “American language” is dead, a learned prosodist should be deducing from “careful observation” of our best poets, a system of scansion for our poetry. And let us suppose that this prosodist had so little dependence in the generality and immutability of the laws of Nature, as to assume in the outset, that, because we lived a thousand years before his time and made use of steam-engines instead of mesmeric balloons, we must therefore have had a very singular fashion of mouthing our vowels, and altogether of hudsonizing our verse. And let us suppose that with these and other fundamental propositions carefully put away in his brain, he should arrive at the line,

Among | trunks grouped | in my | riads round.

Finding it in an obviously iambic rhythm, he would divide it as above, and observing that “trunks” made the first member of an iambus, he would call it short, as Mr. Street intended it to be. Now farther: — if instead of admitting the possibility that Mr. Street, (who by that time would be called Street simply, just as we say Homer) — that Mr. Street might have been in the habit of writing carelessly, as the poets of the prosodist's own era did, and as all poets will do (on account of being geniuses) — instead of admitting this, suppose the learned scholar should make a “rule” and put it in a book, to the effect that, in the American verse, the vowel u, when found embedded among nine consonants, was short. What, under such circumstances, would the sensible people of the scholar's day have a right not only to think, but to say of that scholar? — why, that he was “a fool, — by blazes!”

I have put an extreme case, but it strikes at the root of the error. The “rules” are grounded in “authority” — and this “authority” — can any one tell us what it means? or can any one suggest anything that it may not mean? Is it not clear that the “scholar” above referred to, might as readily have deduced from authority a totally false [column 2:] system as a partially true one? To deduce from authority a consistent prosody of the ancient metres would indeed have been within the limits of the barest possibility; and the task has not been accomplished, for the reason that it demands a species of ratiocination altogether out of keeping with the brain of a bookworm. A rigid scrutiny will show that the very few “rules” which have not as many exceptions as examples, are those which have, by accident, their true bases not in authority, but in the omniprevalent laws of syllabification; such, for example, as the rule which declares a vowel before two consonants to be long.

In a word, the gross confusion and antagonism of the scholastic prosody, as well as its marked inapplicability to the reading flow of the rhythms it pretends to illustrate, are attributable, first to the utter absence of natural principle as a guide in the investigations which have been undertaken by inadequate men; and secondly to the neglect of the obvious consideration that the ancient poems, which have been the criteria throughout, were the work of men who must have written as loosely, and with as littla [[little]] definitive system, as ourselves.

Were Horace alive to day [[today]], he would divide for us his first Ode thus, and “make great eyes” when assured by the prosodists that he had no business to make any such division:

Mæce2 na2 s | at2 avi2 s | edi2 te2 | regib2 u2 s |

O e2 t præ2 | sid3 iu3 m et3 | dulce2 de2 | ous [[cus]] me1 u2 m [[me2 u2 m]] |

Sunt qu2 os cu2 r | ricu2 lo2 | pulve3 re3 m O3 | lympi2 cu2 m |

Colle3 gi3 sse3 | juvat | me2 taqu2 e | fervi1 dis2 [[fervi2 dis2]] |

Evi3 ta3 ta3 | rotis | palma2 qu2 e | nobi2 lis2 |

Terra2 ru2 m | domi2 no2 s | eve2 hi2 t | ad | De2 os2. |

Read by this scansion, the flow is preserved; and the more we dwell on the divisions, the more the intended rhythm becomes apparent. Moreover, the feet have all the same time; while, in the scholastic scansions, trochees — admitted trochees — are absurdly employed as equivalents to spondees and dactyls. The books declare, for instance, that Colle, which begins the fourth line, is a trochee, and seem to be gloriously unconscious that to put a trochee in apposition with a longer foot, is to violate the inviolable principle of all music, time.

It will be said, however, by “some people” that I have no business to make a dactyl out of such obviously long syllables as sunt, quos, cur. Certainly I have no business to do so. I never do so. And Horace should not have done so. But he did. Mr. Bryant and Mr. Longfellow do the same thing every day. And merely because these gentleman [[gentlemen]], now and then, forget themselves in this way, it would be hard if some future prosodist should insist upon twisting the “Thanatopsis,” or the “Spanish [page 680:] Student,” into a jumble of trochees, spondees, and dactyls.

It may be said, also, by some other people that in the word decus, I have succeeded no better than the books, in making the scansional agree with the reading flow; and that decus was not pronounced decus. I reply that there [[can be]] no doubt of the word having been pronounced, in this case, decus. It must be observed that the Latin case, or variation of a noun in its terminating syllables, caused the Romans — must have caused them to pay greater attention to the termination of a noun than to its commencement, or than we do to the terminations of our nouns. The end of the Latin word established that relation of the word with other words, which we establish by prepositions. Therefore, it would seem infinitely less odd to them than it does to us, to dwell at any time, for any slight purpose, abnormally, on a terminating syllable. In verse this license, scarcely a license, would be frequently admitted. These ideas unlock the secret of such lines as the

Litoreis i ngens inventa sub illicibus sus,

and the

Parturiunt montes [[et]] nascetur [[nascitur]]  ridiculus mus,

which I quoted, some time ago, while speaking of rhyme.

As regards the prosodial elisions, such as that of rem before O, in pulverem Olympicum, it is really difficult to understand how so dismally silly a notion could have entered the brain even of a pedant. Were it demanded of me why the books cut off one vowel before another, I might say — it is, perhaps, because the books think that, since a bad reader is so apt to slide the one vowel into the other at any rate, it is just as well to print them ready-slided. But in the case of the terminating m, which is the most readily pronounced of all consonants, (as the infantile mama will testify,) and the most impossible to cheat the ear of by any system of sliding — in the case of the m, I should be driven to reply that, to the best of my belief, the prosodists did the thing, because they had a fancy for doing it, and wished to see how funny it would look after it was done. The thinking reader will perceive that, from the great facility with which em may be enunciated, it is admirably suited to form one of the rapid short syllables in the bastard dactyl (pulve3 re3 em O3) — but because the books had no conception of a bastard dactyl, they knocked it in the head at once — by cutting off its tail.

Let me now give a specimen of the true scansion of another Horatian measure; embodying an instance of proper elision. [column 2:]

Int2 ege2 r | vitæ | scele3 r3 isqu3 e | purus |

Non e2 ge2 t | Mauri | jacu3 li3 s ne3 | que arcu |

Nec ve2 ne2 | natis | gravi3 da3 sa3 | gittis,

Fusce2, pha2 | retrâ.

Here the regular recurrence of the bastard iambus [[dactyl]], gives great animation to the rhythm. The e before the a in que arcu is, almost of sheer necessity, cut off — that is to say, run into the a [[a]] so as to preserve the spondee. But even this license it would have been better not to take.


Had I space, nothing would afford me greater pleasure than to proceed with the scansion of all the ancient rhythms, and to show how easily, by the help of common sense, the intended music of each and all can be rendered instantaneously apparent. But I have already overstepped my limits, and must bring this paper to an end.

It will never do, however, to omit all mention of the heroic hexameter.

I began the “processes” by a suggestion of the spondee as the first step towards verse. But the innate monotony of the spondee has caused its disappearance, as the basis of rhythm, from all modern poetry. We may say, indeed, that the French heroic — the most wretchedly monotonous verse in existence — is, to all intents and purposes, spondaic. But it is not designedly spondaic — and if the French were ever to examine it at all, they would no doubt pronounce it iambic. It must be observed that the French language is strangely peculiar in this point — that it is without accentuation and consequently without verse. The genius of the people, rather than the structure of the tongue, declares that their words are, for the most part, enunciated with an [[a]] uniform dwelling on each syllable. For example, we say “sylla bifica tion.” A Frenchman would say syl-la-bi-fi-ca-ti-on; dwelling on no one of the syllables with any noticeable particularity. Here again I put an extreme case, in order to be well understood; but the general fact is as I give it — that comparatively, the French have no accentuation. And there can be nothing worth the name of verse, without. Therefore, the French have no verse worth the name — which is the fact, put in sufficiently plain terms. Their iambic rhythm so superabounds in absolute spondees as to warrant me in calling its basis spondaic; but French is the only modern tongue which has any rhythm with such basis; and even in the French, it is, as I have said, unintentional.

Admitting, however, the validity of my suggestion that the spondee was the first approach to verse, we should expect to find, first, natural spondees, (words each forming just a spondee,) most abundant in the most ancient languages, and, secondly, we should expect to find spondees forming the basis of [page 681:] the most ancient rhythms. These expectations are in both cases confirmed.

Of the Greek hexameter, the intentional basis is spondaic. The dactyls are the variation of the theme. It will be observed that there is no absolute certainty about their points of interposition. The penultimate foot, it is true, is usually a dactyl; but not uniformly so; while the ultimate, on which the ear lingers is always a spondee. Even that the penultimate is usually a dactyl may be clearly referred to the necessity of winding up with the distinctive spondee. In corroboration of this idea, again, we should look to find the penultimate spondee most usual in the most ancient verse; and, accordingly, we find it more frequent in the Greek than in the Latin hexameter.

But besides all this, spondees are not only more prevalent in the heroic hexameter than dactyls, but occur to such an extent as is even unpleasant to modern ears, on account of monotony. What the modern chiefly appreciates and admires in the Greek hexameter is the melody of the abundant vowel sounds. The Latin hexameters really please very few moderns — although so many pretend to fall into ecstasies about them. In the hexameters quoted, several pages ago, from Silius Italicus, the preponderance of the spondee is strikingly manifest. Besides the natural spondees of the Greek and Latin, numerous artificial ones arise in the verse of these tongues on account of the tendency which case has to throw full accentuation on terminal syllables; and the preponderance of the spondee is farther ensured by the comparative [[in]]frequency of the small prepositions which we have to serve us instead of case, and also the absence of the diminutive auxiliary verbs with which we have to eke out the expression of our primary ones. These are the monosyllables whose abundance serve to stamp the poetic genius of a language as tripping or dactylic.

Now paying no attention to these facts, Sir Philip Sidney, Professor Longfellow, and innumerable other persons more or less modern, have busied themselves in constructing what they supposed to be “English hexameters on the model of the Greek.” The only difficulty was that (even leaving out of question the melodious masses of vowel,) these gentlemen never could get their English hexameters to sound Greek. Did they look Greek? — that should have been the query; and the reply might have led to a solution of the riddle. In placing a copy of ancient hexameters side by side with a copy (in similar type) of such hexameters as Professor Longfellow, or Professor Felton, or the Frogpondian Professors collectively, are in the shameful practice of composing “on the model of the Greek,” it will be seen that the latter (hexameters, not professors) are about one third longer to the eye, on an average, than the former. The more abundant dactyls make the difference. And it is [column 2:] the greater number of spondees in the Greek than in the English — in the ancient than in the modern tongue — which has caused it to fall out that while these eminent scholars were groping about in the dark for a Greek hexameter, which is a spondaic rhythm varied now and then by dactyls, they merely stumbled, to the lasting scandal of scholarship, over something which, on account of its long-leggedness, we may as well term a Feltonian hexameter, and which is a dactylic rhythm, interrupted, rarely, by artificial spondees which are no spondees at all, and which are curiously thrown in by the heels at all kinds of improper and impertinent points.

Here is a specimen of the Longfellow hexameter:

Also the | church with | in was a | dorned for | this was the | season |

In which the | young their | parents’ | hope and the | loved ones of | Heaven |

Should at the | foot of the | altar re | new the | vows of their | baptism |

Therefore each | nook and | corner was | swept and | cleaned and the | dust was |

Blown from the | walls and | ceiling and | from the | oil-painted | benches. |

Mr. Longfellow is a man of imagination — but can he imagine that any individual, with a proper understanding of the danger of lock-jaw, would make the attempt of twisting his mouth into the shape necessary for the emission of such spondees as “parents,” or such dactyls as “cleaned and the” and “loved ones of?” “Baptism” is by no means a bad spondee — perhaps because it happens to be a dactyl; — of all the rest, however, I am dreadfully ashamed.

But these feet — dactyls and spondees, all together, — should thus be put at once into their proper position:

“Also, the church within was adorned; for this was the season in which the young, their parents’ hope, and the loved ones of Heaven, should, at the feet [[foot]] of the altar, renew the vows of their baptism. Therefore, each nook and corner was swept and cleaned; and the dust was blown from the walls and ceiling, and from the oil-painted benches.”

There! — That is respectable prose; and it will incur no danger of ever getting its character ruined by any body's mistaking it for verse.

But even when we let these modern hexameters go, as Greek, and merely hold them fast in their proper character of Longfellowian, or Feltonian, or Frogpondian, we must still condemn them as having been committed in a radical misconception of the philosophy of verse. The spondee, as I observed, is the theme of the Greek line. Most of the ancient hexameters begin with spondees, for the reason that the spondee is the theme; and the ear is filled with it as with a burden. Now the Feltonian dactylics have, in the same way, dactyls for the theme, and most of them begin with dactyls — which is all very proper if not very Greek — [page 682:] but, unhappily, the one point at which they are very Greek is that point, precisely, at which they should be nothing but Feltonian. They always close with what is meant for a spondee. To be consistently silly, they should die off in a dactyl.

That a truly Greek hexameter cannot, however, be readily composed in English, is a proposition which I am by no means inclined to admit. I think I could manage the point myself. For example:

Do tell! | when may we | hope to make | men of sense | out of the | Pundits |

Born and brought | up with their | snouts deep | down in the | mud of the | Frog-pond?

Why ask? | who ever | yet saw | money made | out of a | fat old —

Jew, or | downright | upright | nutmegs | out of a | pine-knot? |

The proper spondee predominance is here preserved. Some of the dactyls are not so good as I could wish — but, upon the whole, the rhythm is very decent — to say nothing of its excellent sense.


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

As for part I, this essay is a nightmare for the typesetter, and there are complications that are not easily resolved in HTML. Of particular trouble is the broad use of characters with special marks or accents. These have been repeated here as closely as possible, but by using a broader than standard font, may or may not display properly on every computer or device. Additional marks have not been possible to reproduced using native font characters. For example, in the original text, Poe places a squiggly line (~~~) over several words. Due to HTML limitations, these words are underlined in this text. The numeric accents used throughout this essay appear, in the original text, under the characters they mark. Again due to HTML limitations, these have here been set as subscripts following the character. 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 II] (B)