It seems clear to me that questions like ‘How does this line scan?’ or ‘What is the metre of this poem?’ are not questions of the same sort as ‘What is bronze?’ This is a question about physical fact. But the only physical facts about which a metrical question could be put are presumably phonetic facts. And when we ask how a line scans we cannot be asking simply for the phonetic facts which occur when it is pronounced. For:
(a) If the scansion of a line meant all the phonetic facts, no two lines would scan the same way, for no two different lines are phonetically identical. If, on the other hand, we are asking only for some of the phonetic facts then we must want those which are relevant. But to what? Clearly, not to phonetic fact but to something else.
(b) Individuals differ in their pronunciation of a line. Even a single individual can hardly pronounce it twice in exactly the same way. If scansion meant physically phonetic fact no line would scan twice in the same way.
Are we then asking not how this or that individual reads the line aloud but how he ought to?
Unfortunately, even when we have ruled out gross barbarisms, there remain different and defensible ways of reading poetry aloud and they do not coincide with differences of opinion about metre. The two main schools may be called Minstrels and Actors. They differ about the proper relation between the noises they make and something else; that something else being the thing we are looking for, namely metre. Minstrels, singing or intoning, make their utterance conform to this, leaving you to imagine the rhythm and tempo which the words would have in ordinary speech. Actors give you that rhythm and tempo out loud, leaving you to imagine the metre. Yet both may be fully agreed as to what the metre is. They differ by deliberately making, or refusing to make, an imaginary archetype or paradigm actual. This paradigm is metre. Scansion is the conformity, made audible by Minstrels and concealed by Actors, of the individual line to this paradigm.
When we ask for the metre of a poem we are asking for the paradigm. But again, what sort of question is this? If one man describes our Blank Verse paradigm in terms of ‘feet’ and another in terms of crotchets and quavers, what sort of difference is the difference between them?
We cannot, or should not, be asking how the poet himself would have described it. As regards the greater part of the world’s poetry we do not know the answer to this. And even when poets have told us how they analysed their own metres, it is always open to us to say that their analysis was wrong, that their instinct or genius enabled them to produce what their often limited analytical powers did not enable them to understand.
Our results are so far discouraging. I am therefore going to suggest that metrical questions are profitable only if we regard them, not as questions about fact, but as purely practical. That is, when we ask ‘What is the metre of this poem?’, we are not, or should not be, asking which analysis of the paradigm is ‘true’ but which is the most useful. The utility of the analysis would, I submit, be in a direct ratio to the degree in which it gives those who adopt it the following powers:
(1) To say whether, so far as metre goes, a given line could or could not have occurred in a given poem.
(2) To quote any line, if not correctly, yet certainly without any metricidal error.
(3) Within any poem to distinguish normal from irregular lines.
(4) To detect textual corruption by the damage it has done to metre.
(5) To teach the metre quickly and easily to others.
That hardly any of our modern students possess these powers every university teacher knows.
Such is my conception of a good—or even ‘the right’—analysis. How are we to achieve it?
The first rule is ‘Avoid the Inductive Method’. It sounds very plausible to say: ‘Let us not be a priori. Instead of bringing to the actual lines some arbitrary idea of what is Regular, let us stick to facts—what the poet actually wrote. Let us, without any prejudice, tabulate all the types of line we find in the poem and then, inductively, construct the paradigm to cover them, to save the appearances.’ This commends itself to a scientific age. But surely it is quite fatal?
For if you proceed thus you will have no irregular lines at all. If your inductive paradigm ‘gets them in’, they have become regular. That is, they are specimens of alternatives, though rather rare ones, among those which the paradigm prescribes. In fact, they are like Virgil’s procumbit humi bos, which does not break the hexametrical paradigm at all but fulfils it in an unusual way.
But that is not what irregular lines do in English poetry. A poem in Latin hexameters where every single line ended in a monosyllable would be a very bad poem but it would still be unmistakably in hexameters. Similarly if Shelley’s ‘The weight of the superincumbent hour’* were really one of the alternatives allowed in Shelley’s paradigm, then a whole poem in such lines, though a bad poem, ought to be still a poem in the metre of Adonais. If this line fulfils the paradigm, no succession of such lines could break it.
Well, try:
The weight of the superincumbent hour,
The blows of a darkly returning power,
The roll of the breakers and (while we speak)
The glare of the sun on a faded flower,
The blight of the moon on a fevered cheek.
This is not the metre of Adonais at all. Worse still, it is a metre: a quite different one. So with Milton’s ‘Burnt after them to the bottomless pit’.* Add ‘Hell then received the unfortunate crew’ and a few more such lines and you will get a new and perfectly recognisable (though detestable) metre.
Inductively constructed paradigms thus fail because they ‘cover the facts’ too well. A formula which accommodates all the actual lines accommodates lines which, if repeated, would not be in the metre of the poem. We must adopt exactly the opposite procedure. We must not begin with individual lines, nor even with classified types of line. We must begin with the whole poem. That, if it is any good, will teach you the tune, the pattern, the paradigm. It is only in relation to this that the lines are lines at all.
The paradigm is theoretically, or ostensibly, or by legal fiction, or by make-believe, obeyed in every single line. In many lines it is actually obeyed. The continual approximations to and recessions from actual obedience, as of waves on a beach, make much of the excellence of any long poem. If you once start monkeying with the paradigm so as to ‘get in’ all the lines, this beauty is lost. One does not want the shore as well as the sea to be in motion.
The irregular lines are those in which the make-believe is strained to the utmost point. ‘The weight of the superincumbent hour’ is feigned to be, or deemed to be, ‘The weíght of thé supérincúmbent hóur’; and ‘Burnt after them to the bottomless pit’, to be ‘Burnt áfter thém to thé bottómless pít’. Of course not even the hardiest Minstrel would so read them. But equally the hardiest Actor is not appreciating them as verse at all unless his inner ear still hears the inner metronome ticking away. It has been set ticking by all the thousands of decasyllabic lines he has ever read.
If it ceased to tick, nearly all the merit of such lines would vanish. In Shelley’s, the very laboriousness of the suggested pronunciation (‘the supérincúmbent’) symbolizes the burden of the hour; in Milton’s, the denial, in fact, of the accents suggested by the metronome gives the sense of falling into a void.
When it comes to defining the paradigm, say, of decasyllabic verse, I do not see how we can avoid saying that each line contains five units of some sort. We do not of course mean, as some apparently think, that the poet ‘built them up’ out of such units, as a bricklayer builds a wall out of pre-existing bricks. Nor do we mean that any reader makes pauses between these units in pronunciation. We mean that wherever the paradigm is completely obeyed, analysis cannot help finding that certain phonetic configurations occur five times in the line.
Much metrical controversy is concerned merely with nomenclature; whether we should talk about these units as ‘feet’ or in some other language. Here again I maintain that we should be guided by utility.
Musical notation I would rule out at once. Book-lovers will not like the look, nor publishers the cost, of a page all spotted over with musical notation. Nor will it be of any use to readers of poetry unless they are also musicians.
The stock argument against calling the units by classical names (feet—or, in particular, iambi, trochees, and the rest) is that they are not really the same as the units of ancient verse. And even if we warn the student that the so-called English iambus is not to be confused with the quantitative iambus, he will in fact be encouraged to read Latin poetry in the wrong way. This has certainly done much harm to schoolboys in the past. They have been allowed to think that Virgil’s hexameters ended with the tempo of ‘strawberry jam-pot’, when they were probably most often like ‘All men have idols’—in fact, more like the slow movement of the Seventh Symphony and less like the Walkürenritt.
But it will be noticed that the whole danger here is not to the student’s English, but to his Latin, reading. This was certainly so in my own experience. Metrics of this type spoiled my appreciation of Latin poetry for years; I have never been able to find that they did my English studies anything but good. But if this is so, then the main objection to the classical nomenclature in English poetics is already out of date. We need not be afraid of encouraging our pupils to read Latin wrongly because we know they are not going to read Latin at all. We need no precaution against corns in a man who has already had both his legs amputated at the hip.
This being so, it is surely time to re-avail ourselves of the enormous advantages which the classical terms offer. They are as follows:
(1) If you talk of feet everyone knows what you mean. Do not be deceived by those purists who will reply, ‘I never know what people mean by a foot in connection with English verse.’ That’s only their fun. They know perfectly well that you mean the things which come seven times in a fourteener, five times in a blank verse line, and four times in an octosyllabic. If, on the other hand, you devise what you take to be a more scientific language, you will never in discussing this or that line (outside your own book) be able to use it without explanation. Even if it won universal acceptance it would be swept away by the next, and even purer, purist. Almost any agreed terminology is better than a perpetual reformation.
(2) If I am allowed to use all the classical names I can describe shortly and clearly all the metrical phenomena in English verse. If I may not speak of Choriambics I must take endless trouble to write, and you to read, any reference to what is happening in ‘This, this is he; softly awhile . . . or do my eyes misrepresent?’ (Samson Agonistes, 115–24).
(3) It enables us to present paradigms hard, jejune, dry, as paradigms ought to be, uninfected by questions of beauty. We need metrics if we are to become fully sensitive to poetry, as we need grammar before we can enjoy Homer and anatomy before we can draw. But the less these studies get mixed up with ‘sensibility’ the more they will ultimately do for it.
(4) Have you ever had a pupil not brought up on this scheme who was aware of metre at all? We are coming to acquiesce in a hair-raising barbarism on this subject. I have met an undergraduate who, after reading it, thought The Prelude was written in Spenserian stanzas. Hardly one out of five Honours candidates can quote three or four lines of blank verse without false lineation. Every day we hear Donne praised for startling metrical audacity in passages where the metre is as regular as anyone else’s. It is plain that our present methods do not work. Might we not go back to one that did?