Meter & Music: Willett v. West
I have been irritated for a while now with Steven J. Willett’s paper “Ancient Rhythmicians and Modern Prosodists: Searching for the Location of Meter.” (Available online here) As it is representative of many of my complaints about the study of versification in general, I have been wanting to tackle it in a Rob-style rant for a while now. After reading Rob’s rant about Positive Economics, I decided that here might be the right place to do so. As a counterpoint, I recently read a dazzling article about Akkadian meter by the eminent M. L. West (you may remember his withering comments in my previous post; the article is “Akkadian Poetry: Meter and Performance” from Iraq v. 59 pp. 174-87, available on JSTOR). His approach strikes me as a possible antidote to the unfortunate tendencies observable in Willett’s work — to which, however, they are by no means confined.
Willett starts out strongly enough; in fact, one of the things that initially drew me to his essay was that the complaint with which he begins is one of my own:
On the empirical level, it is obvious that those who write on the theory of versification as opposed to narrow technical analysis should have a solid grasp of the complete western tradition, beginning with Greek and Latin. Critical writings on theoretical metrics are strewn with errors that stem from a narrow provincial knowledge of English and widespread ignorance of Classical, Germanic, Romance and Slavic languages.
Certainly true — and he will shortly broaden the claim (as well he should) into an indictment of theoretical generalizations which ignore any meter, Western or non-Western, as well as “comparative metrics, historical linguistics and modern cognitive science.” (One wonders why he singles these three disciplines out in particular — it does strike one as a rather selective list.) Unfortunately, these flaws which he has done so well to single out in the others’ metrical scholarship strike me as nowhere more apparent than in his own. For example, immediately after the statement cited above, he begins listing claim which are common in metrical theory, but which he believes to be “factually, and quite demonstrably, false”:
(1) unrealized beats occur at the ends of the short lines in common measure, (2) metrical feet are not experiential, (3) the caesura is not a metrical phenomenon and (4) English is a stress-timed language
All right. Let us take them in reverse order, shall we? Of “stress-timing” Willett says:
[A]ll experimental attempts to find stress-timing in unconstrained English have proven futile. Cummins and Fort in fact offer direct phonetic timing data in support of more varied, music-like rhythms in human speech than simple stress-timing.
This is certainly the case. It is surely unnecessary to resort to the tape-recorder to see that stressed syllables in English are not actually equally spaced in time, or “isochronous.” But this disproves only the most naive understanding of stress-timing, and a modified notion of stress-timing continues to be used by linguists as well as literary scholars. Anthony Fox (one of the great experts in the prosody of natural languages) acknowledges “the negative results obtained in experimental studies,” but maintains that “this does not mean that either music speech have no regularity of rhythm, since rhythm is not a matter of absolute temporal equality,” and in fact “much of the temporal structure of English (and other comparable languages) is only explicable in terms of some form of isochrony, since the lengths of syllables must be determined at least in part in relation to the regular occurrence of stress” (Prosodic Features and Prosodic Structure, p.90) Of course, Fox might be wrong and Cummins and Fort might be right — but saying that an entire side of an argument that continues to rage within linguistic theory is “factually, and quite demonstrably, false” is both arrogant and symptomatic of the very ignorance of modern linguistics that Willett is attacking.
His response to the criticisms that have been leveled at the caesura is better founded:
Whether the caesura is a metrical reality depends on definition and poetic tradition. In English poetry the caesura is widely considered nothing more than a pause created by the syntax, and thus a matter for oral declamation, but in Classical poetry it is a word boundary that intersects the metron and constitutes a crucial rhythmic fact to which the Greeks were acutely sensitive, as Devine and Stephens document in their earlier study Language and Metre: Resolution, Porson’s Bridge, and their Prosodic Basis .<5> In Russian accentual-syllabic poetry, or syllabo-tonic poetry as they like to call it, the caesura is a word break that occurs at the same position in the line throughout the poem, often coinciding with a slight syntactical break.<6> It significantly affects the rhythm by forcing a stress on the following ictus to reestablish the rhythm after the break. Marina Tarlinskaja has even argued that the English caesura is not properly a pause but, like Russian, a word boundary that regularly occurs at the same position in the line, where it produces effects similar to those of Russian: the end of the first hemistich behaves metrically like line-end and the beginning of the second hemistich like line-beginning.
Certainly true. The caesura is metrical in some languages and not others; it would be absurd to argue that it is never metrical based on evidence from English alone. I have never seen anyone rise to such a degree of insularity — one would necessarily be unfamiliar even with French and Old English poetry — and I suspect him of attacking a straw man here, especially given that he provides no citation for the “commonplace.” Nonetheless, of the points he makes, this is certainly the most solid, and if anybody holds such notions, he does well to disabuse them. However, although I have not read Tarlinskaja’s work, it strikes me as rather strange — in most English poetry, there is no place in the line which is characterized by consistent word-end (in Nabokov’s words, Milton et al. “Gleefully sang it out of existence” [linebrick, could you make sure I have this right?])
About the “experientiality” of metrical feet, Willett says the following:
A large body of evidence from linguistics and cognitive science validates the concrete existence of prosodic feet as the source of metrical feet. On the grounds of historical linguistics, A. M. Devine and Laurence D. Stephens argue in their magisterial The Prosody of Greek Speech that “the rhythmic structures of Greek verse reflect, arise from and already exist in the rhythms of Greek speech and are not in principle the result of mapping the rhythms of Greek speech onto extraneous patterns, that is onto temporal patterns of nonlinguistic origin.”<3> Prosodic feet are, they demonstrate, the basic constituent of the rhythm of the Greek language (as in many other languages) and provide the basis for metrical feet. On the grounds of cognitive science, Fred Cummins and Robert F. Fort at Indiana University have in their recent experimental work established strong evidence for the reality of the metrical foot as a well-defined unit in the production of speech.
We are in high linguistic theory here. Although I am in much greater sympathy with this argument than in his argument about isochrony, mostly because I think Devine & Stephens are badasses, and it may be that “prosodic feet” are subject to more agreement among linguists than stress-, syllable-, and mora-timing, but the debate continues to be one of theory rather than facts. “Feet” are theoretical entities whose existence we posit in order to account for certain observed phenomena; their existence can never be “factually, and quite demonstrably” proven.
Furthermore, the existence of feet is essentially a matter of “phonology,” rather than “phonetics” — they exist to explain our experience of the sounds of language rather than directly describe them. Willett’s previous argument against stress-timing is an extremely hard-phoneticist one: we cannot measure isochrony in the sound-signal, and therefore it must not exist, ignoring the question of whether the assumption of isochrony structures the way we experience language. In fact, at least for Fox, the question of stress-, syllable-, and mora-timing is intimately related to the concept of the foot — he introduces the two at the same time and defines them in terms of one another. It is therefore extremely deceptive of Willett to switch levels between his discussions of such intimately related concepts in linguistics, citing phonetics disprove one and cognitive science to prove the other.
Finally, there is Willett’s point about “common measure,” or “ballad stanza” to its friends. One wonders whether Willett has listened to any ballads lately; he would surely have noticed that there is almost invariably a pause every other line, because (whatever we might say about meter in poetry) western music has a strong tendency to count in fours:
There is a house in New Orleans
they call The Rising Sun. [pause]
It’s been the ruin of many young girls
and God, I know I’m one. [pause]
One might point out here that the ballad I’ve chosen is actually in triple time (or, in later versions, compound time): even when the meter itself is triple, the phrase-structure comes in fours. How then, does Willett argue that this “unrealized beat” — on the face of things, factually and quite demonstrably present — is actually fallacious? He does so first of all by entirely ignoring music entirely, and secondly by abusing an actually quite brilliant observation by Gasparov:
Whether unrealized beats exist or not, comparative metrics shows that common measure evolved from the medieval Latin goliardic 7d+6f line with a tendency to trochaic rhythm. In English the dactylic ending at the midline pause changed to masculine, while each hemistich acquired an unstressed syllable at the start to create a rough iambic meter. The second hemistich then became the shorter three-stress line when the meter evolved into oral folk poetry and, as Gasparov notes, lost some of its accentual-syllabic clarity, turning into a 4-3-4-3-stress dol’nik <2>. Since the even lines of common measure spring from a three-stress hemistich and always contained three stresses in origin, they cannot be considered truncated four-stress lines, thus they never ended with unrealized beats.
This is the etymological fallacy itself. Now the etymological fallacy is one of my favorite fallacies, and I don’t believe it is quite fair to say (as most of the linguists in which Willett expresses such faith might say) that the origins of the common stanza have nothing to do with its synchronic structure, but the argument as it stands is precisely equivalent to amusing reductiones of the etymological fallacy as insisting that the “nice” can’t mean “pleasing” because it used to mean “ignorant” (source); whatever the source of the common stanza’s structure, it does not license us to ignore the evidence of actual ballads, much as we are likely to have a hard time of it if we get offended every time someone calls us “nice.” For example, only a conception informed by “suppressed beats” can explain why the morphology of the ballad stanza occasionally allows for the suppression of further beats, creating the so-called “short measure,” often in the very same ballad:
My mother was a tailor. [pause]
She sowed these new blue jeans. [pause]
My lover was a gamblin’ man
Down in new Orleans. [pause]
A little learning is a dangerous thing. Willett swoops down in order to save metrics with the his knowledge of “comparative metrics, historical linguistics and modern cognitive science,” but that knowledge is so partial, partisan, confused, and inappropriately applied that he often makes matters considerably worse than he found them. I remember when I entered Reed, one of my ambitions was to discover the clear and solid knowledge of the scientific approach to language — i.e., linguistics — and apply that knowledge to the study of literature. I was greatly disappointed to learn that that knowledge has not achieved any particular clarity or solidity — that any given topic, and particularly concerning prosody, is subject to fierce debate. Apparently, Willett has not realized this yet — one gets the feeling that he either has not been given the whole story, or is purposefully concealing it from us.
But what is worse, and what comes across clearly from our final example, is that Willett seems to have very little understanding of music — and so far as I can tell, he has a fairly low opinion of it. This is very problematic, because in the remainder of his paper he critiques various approaches to meter which depend on the “analogy” of music. Some of his criticisms are just.(1) But the problem, as we saw earlier, is that the relationship of music to meter is often much more than analogy; the ballad stanza is essentially a song meter, a trait it shares with a great number of verse-forms. Almost all cultures have traditions of spoken poetry as well; but almost everywhere spoken meters bear a close and productive relationship with sung ones, and the former have a tendency to develop from the latter. In fact, the appeal to “the experience of reading” as opposed to “the experience of listening” which dominates the later sections of Willett’s argument seems strange and insular when we reflect that most people in most places for most of history have listened to their poetry rather than reading it.
This brings us to M.L. West. By contrast to Willett, West is extremely knowledgeable about music — he literally wrote the book on Ancient Greek Music — and almost all of his metrical analyses are sensitive to the music which so often accompanied ancient poetry. Willett’s first criticism of the meter/music analogy — that meter frequently fails to function as “musical beating” is really a criticism of a particular Eurocentric conception of rhythm. To Willett’s rather garbled account of why Greek meter and “regular beating” are incompatible, compare West’s comparatively lucid critique of certain eighteenth-century tendencies in Greek metrics, which depends precisely on his awareness of music:
Furthermore, we have become accustomed in our western musical tradition to banal and repetitive rhythms. There is a danger that this may blunt our understanding of Greek metre, which sometimes, like the traditional folk music of eastern Europe and Asia, presents more intricate rhythms and changing bar-lengths. German scholarship in the last century devoted much effort to the rhythmical interpretation of asymmetrical cola on the erroneous premise that they must be divided into equal bars. (Greek Metre, p. 115)
Thus the musical analogy can actually help us overcome the very insularity of which Willett complains, provided we are aware that music is as diverse as language.(2) West’s paper on Akkadian meter is a particularly impressive example of what such an awareness can accomplish when combined with a modicum of imagination.
The first few pages of the article are tough going. Some of it was probably that I didn’t know any Akkadian at all; but nonetheless, there were paragraphs I had to read through several times before I could figure out what he was saying. I do wonder whether he might have been more clear in his exposition. But the point is fairly simple; most Akkadian poetry is written in what the Germans (it’s always the Germans) call a Vierhaber; that is, “a verse with four apparent accentual peaks, giving the sense of a balance of two against two” (West 175). He gives an example (the caesura is marked by a |):
īpulma Mummu | Apsû imallik,
sukkallum lā māgiru | milik mummīšu:
“Hulliqamma, abī, | alkata ešīta;
urriš lū šupšuhāt, | mūšiš lū sallāt.”
(Note: I don’t have any idea what this means.)
Now I don’t know where the stress lies in Akkadian; I’m not sure anybody does. But assuming that each major word gets one somewhere, the traditional account of the Vierhaber works fairly well. But there are a number of problems, not least of which is the not infrequent presence of a Dreihaber (a line containing three apparent accentual peaks) — and occasionally even longer or shorter lines.
But perhaps most problematic is the trouble identifying what counts as an “accentual grouping.” West attacks this problem through an analysis of the Theodicy, a poem which is not only a particularly regular one, but one of which the written tradition
makes an unusual acknowledgment of the metrical regimen. In three of the four Neo-Babylonian manuscripts each line is divided up not just into two segments, as happens in many poetic manuscripts, but into four. In two of the copies there are not only horizontal ruled lines separating the strophes but also vertical lines dividing the column into four cells, the words of the text being distributed carefully among them. (West 176)
From these manuscripts and one of the Enūma Eliš “in which the columns were divided into three segments,” West derives an astonishing variety of “unit-elements” which can be fitted into the four “apparent accentual peaks” of the Vierhaber:
(a) A single word;
(b) a group of one or more prepositives + accented word, e.g. ana Anunnakkī, ša ibnû, ša lā iqattû;
(c) a construct phrase, simple or compound, e.g. mālik ilāni, šādid nīr ili;
(d) a combination of (b) and (c), as in ina ūm lā šīmāti;
(e) a noun + qualifier, as in kala lumnu, rabi ahi;
(f) two words making a linked pair, as in abī u bāntī, eliš u šapliš;
(g) a dependent clause, as in mīnâ pakki ilīma, aššu lā īšû irītu. (West 178)
West goes on to apply his typology, with a certain amount of success, even to Akkadian poetry where the Vierhaber does not dominate, but the basic problem remains:
Well then, how are we to imagine a performer putting across verses in which there were generally four accentual peaks, but sometimes — unpredictably — three, or five? How could this not be as disconcerting to the hearers as it would have been if a Homeric singer had now and then delivered himself of a line with five or seven feet instead of six? Again, how can we make sense of the gross variations that exist in the length of the accentual units, that is to say in the distances separating one accentual peak from the next? They may occur on adjacent syllables, as in Agušaya A iv 10 šī ihsus qurdam, or there may be six or seven unstressed syllables between them. The variation is too great for the accentual peaks to have been aligned with a “beat” occurring at equal intervals.(3) Certain eminent scholars have found this so contrary to common sense that they have been driven to assume some quite different principle of accentuation, with bizarre results.
West’s problem here is much the same as Willett’s; the assumption of a “regular beat” in poetry is frequently unsustainable when applied to systems of versification other than English. His criticisms of “certain eminent scholars” who have attempted to find such a “regular beat” cannot help but call to our minds Willett’s criticisms of those scholars who find in Japanese poetry “a beating completely detached from both the linguistic prosody of Japanese and the traditional syllabic meter of the poem — nothing more than a distorting western habit imported into the poetry” (See note 1):
It is painful to see a man tying himself in such knots in public for the sake of a métrique alternante that has no basis whatever in the material evidence but derives entirely from his intuitive preconception. (West 182)
This is characteristically acerbic. But whereas for Willett, this objection is only a preliminary for two broader objections to treating poetry as something to be heard rather than read, West cannot ignore the fact that Akkadian poetry consistently presents itself as both transcripts of and scripts for performance, and even musical performance:
Many of the Akkadian hymns and narrative poems contain internal evidence of oral performance. They often begin with an azammur “I (will) sing (of) –” or a luzmur “let me sing (of) –“, and/or with a call to the people to “hear” the subject-matter of the song. In the epilogue to Enūma Eliš the written text is represented as having been made on the basis of older oral tradition, and it is to serve in its turn as the basis for future recitations; there is at least a pretence that it will be taught orally by seniors to their juniors. Kabti-ilani-Marduk, the poet of Erra and Ishum, looks forward to its indefinite preservation both in performance and in literary tradition (V 53-61). One or two passages in poetic texts allude to the accompaniment of a performance on a stringed instrument. (West 181; this is only a sampling of West’s evidence, which is in turn only a summary of the evidence collected in his The East Face of Helicon 590-9)
It is necessary, then, to take the question of performance very seriously. I cannot do justice here to the stunning argument that precedes from this point — it procedes through Ogden Nash and the cognate Ugaritic and Hebrew traditions (4) — but ultimately West ends up imagining a tradition of performance which is essentially non-rhythmic; the four “unit-elements” of the Vierhaber are not the product of an accentual schema, but rather of a “hierarchy of junctures” within the line. Although the line usually did not exceed 12 or so syllables, the length of each “unit-element” was variable; the cohesion of the line itself was preserved by the strong tendency of the penult to be occupied by a long syllable. The hierarchy of junctures was preserved not by a “regular beat” but by the melodic contour of the piece:
In its principle of progressive subdivision by greater and lesser disjunctions, and its commata of variable length, Babylonian versification appears analogous to the Hebrew. Both systems were probably based on a pattern of intonation with potentially musical implications, whether we think of actual singing or just of some sort of melodic chanting. (West 186)
In order to make clear what it would mean to have a “a musical structure with an identity based on a
characteristic melodic profile up and down the vertical axis while being freely adjustable along
the horizontal,” that is, one in which “the sequence of pitches is fixed, but the time-intervals between them can at some points be elastic,” West brings up the seemingly innocuous example of the singing of English psalms, which, although unmentrical, “acquire a perfectly clear profile from the formulaic melodies to which they are sung, the so-called psalm tones.” We can, then, imagine a musical performance without a steady rhythm without even going too far outside of our own experience. But it is then that he makes his truly dazzling move:
This is not a system invented for the English. In principle it goes back to the early Church; and the early Church took it over from the Synagogue. Scholars such as Abraham Idelsohn, Peter Wagner, and Eric Werner have established significant affinities between Gregorian chant and the psalmodie traditions of the Jewish diaspora. Idelsohn collected and compared melodies from widely separated Jewish communities in east and west, Persian, Yemenite, Babylonian, Sephardic and Ashkenazic, some of them thought to have separated off before or at the time of the Babylonian exile. The features that they share must, so it is plausibly argued, reach far back into antiquity.
So West’s apparently opportunistic example of English psalm-singing is not so otiose as it might seem; in fact, it provides important evidence for the kind of performance that Akkadian and Semitic meter in general was originally adapted to. His conclusion:
I suggest the following hypothesis to bring all this together. From sometime before 2000 BC Amorite and Akkadian poetry was chanted in a particular way, or in some cases sung with harp or lyre accompaniment. The performers had a small repertoire of conventional but elastic intonational or melodic matrices to which they could fit their verses. These matrices were structured by means of certain fixed pauses or cadences, major, minor, and minimal. The singer or reciter stretched or shortened the matrix from verse to verse so that the accentual peaks fell at the appropriate points of its profile.
The system did not depend on the number of accents being invariable. For example, if the poet, to express a particular idea, happened to hit on a form of words that contained five accent-bearing syllables, he could easily fit it into a matrix that normally supported four. Whichever of the five was the least important in terms of sentence-accent could be carried in the neutral section of the melody. If, on the other hand, his phrase contained only three accents, one of the nodal points of the melody could be allowed to pass unmarked. The final accent would coincide with the cadence in the usual way. (West 186; he then provides an example of the kind of scheme he means. Anyone who can access the article and read music is encouraged to visit the penultimate page and check it out.)
The value of West’s article is twofold. First, he presents a compelling hypothesis for how Akkadian and other early Semitic poetry was performed, a hypothesis he supports via strong comparative evidence and one that is able to motivate its observed peculiarities. Secondly, his argument is a reminder that our ideas about meter are inevitably linked to our assumptions about music; the decision to ignore music (like the decision to ignore ideology) is all too often the decision to accept those assumptions wholesale. This is an important response to Willett’s generalization about approaches to meter based on music:
In a system of poetic rhythm based on music theory, the identification of meter with beating is an obvious gambit to circumvent the difficult fact that meter is an artistic convention for the intentional selection and patterning of linguistic phenomena.
Meter is indeed an “artistic convention for the internal selection and patterning of linguistic phenomena,” as Willett, Gasparov, and many before them have claimed, but this “internal selection” does not exist in a vacuum from other factors, nor do our conceptions of it. In order to make sense of Akkadian meter, for example, it is important to consider that the kind of performance which is transcribed in Akkadian texts is a very different kind of performance from the musical performances we are used to, and that provide the basis for our own versification. For the “selection and patterning of linguistic phenomena” is in many cases executed in order to accommodate speech to music of a specific kind, and in the remainder of cases, the meter is most often developed by analogy to meters which are. How this mediation between speech and music is developed varies not only according to the specifics of the language in question, but also according to the musical traditions involved.
The remainder of Willett’s points, that “(2) reading is a radically different temporal experience than listening to music and (3) reader interactivity with the full semantic and pictorial abundance of the poetic text opposes the ceaseless forward movement of musical time,” essentially concern the difference between reading and any kind of performance. I do not think that these are worth touching on, because it seems obvious to me that meter belongs to the latter, not the former:
In origin, at any rate, metre and poetic form exist to please the ear, not the eye. When we read English verse that scans and rhymes, we may take it that the author intended us, if not actually to declaim it, at least to hear it mentally and appreciate it as we would appreciate a recitation. (West 181)
Or perhaps we should cite Willett himself (in a later article) on the same point:
Rhythm only exists when we hear a poem recited or we read it out loud. The silent articulation of a poem to ourselves by subvocalization also actualizes the rhythm, if in a more ghostly way. Absent the aural domain, rhythm is a mere abstraction. (Willett, Working memory and its constraints on colometry. QUCC 2002 N. S. N° 71 : p. 10)
I am sure that the temporal and interactive differences between reading and listening are an essential aspect of any theory of literature in general; but meter exists — where it exists in writing — precisely to occlude those differences. The metrical text is more oral than the unmetrical, although we read them both.
Having more or less concluded my little rant, I must admit that I am a bit ashamed of what I have done; whereas Rob took on one of the giants of economic theory, I have been beating up a relative nobody, and in a conference paper rather than a publication. By contrast, West (against whom I have pitted him) is relatively gigantic — hardly a fair fight. I was drawn to Willett because, although we harbor conflicting prejudices, our works are animated by similar concerns. He begins and concludes by insisting that his criticisms are meant “in a cooperative, not a destructive, spirit to advance our discipline by helping to clarify, point and tighten the most useful elements of temporal theories. A hard attack leads to a hard defense.” Inspired by the vigor of his attack, I have opted to strike back, and in the same cooperative spirit.
On the other hand, Willett is very representative of what we might call the “linguistic prejudice” that has dominated metrical studies of late. (So dominant, in fact, that one often feels that Willett is beating a dead horse.) Most of the best metrical theory of the last century or so has felt the need to insist that meter is primarily a matter of linguistic rather than rhythmic form — or if rhythmic, rhythmic as understood in language by linguists. This is certainly a healthful alternative to the frankly silly works — dominated by arithmetic gymnastics and highly subjective musical notation — that circulated during the previous century, but the problem with these works was not so much their engagement with music as the narrow-mindedness of their conception of music. Excluding music altogether only deepens this narrow-mindedness, as is everywhere in evidence when Willett generalizes about music.
But it may be simply that I find the majority of modern linguistic theory (and, for that matter, cognitive science) deeply irritating.
1. Others not so much; but as my primary claim concerns generalities rather than specifics, I will pass over them. But there is one which is so egregious that I feel the need to dispose of it in a lengthy footnote. Willett criticizes the notion that poetic meter might be somehow analogous to musical rhythm with reference to the purely syllabic (really moraic) meters such as the Japanese waka, which
is a continuous stream of syllables without lineation. Even when printed with artificial lineation, as waka sometime are, it is sonically meaningless. . .We find no phonological basis for beating, since the five- and seven-syllable segments have neither stresses nor repeating tonic sequences like Chinese, and cannot therefore elicit a four-tactus graded hierarchy. Might we nevertheless feel a musical beat, understood to be highly subjective and primarily physical/gestural in source, while reading the poem? If so, it would be a beating completely detached from both the linguistic prosody of Japanese and the traditional syllabic meter of the poem—nothing more than a distorting western habit imported into the poetry. A westerner might persuade himself that he feels a 1-2-3-4 binary beat, but no Japanese perceives or reads waka with one.
This was odd to me, because when I first read this paper, I was had just finished reading a weird little chapter on Japanese versification by Kōji Kawamoto (from his The Poetics of Japanese Verse: Imagery, Structure, Meter) who not only proposes such a “binary beat” but places himself in a long tradition of other Japanese scholars who have “persuaded themselves” that they felt such a thing. Of course, they take their inspiration from systems of Western prosody, and it might be argued that their response is thus contaminated and not representative of the “true” Japanese perception — a plausible if condescending argument — but this not one that Willett makes. Instead he claims that “no Japanese perceives or reads waka” with a binary beat, a claim which is factually, and quite demonstrably, false. The oversight is particularly strange when we notice that Willett teaches at a Japanese university, and here especially one wonders whether the evidence is simply being suppressed.
2. I was, for example, a bit put off when I was reading both Willett and Kawamoto on Japanese that neither acknowledged the existence of Japanese song.
3. West’s note here is of interest: “We can do this in English verse: when there is a variable number of unstressed syllables between the stressed ones, we make the stresses equidistant, speeding up the intervening syllables if there are more of them, slowing them down if there are less. But the number of these syllables is limited to between zero and two. If we try to put in more, a secondary accent automatically develops among them and the rhythm is ruined. In Akkadian verse no such constraint applies. We cannot here be dealing with a system of equally spaced stresses as in English. It is inconceivable that the same time-slot could be filled now by one, now by six or seven syllables: either the one would have to be unbearably dragged out, or the six gabbled at a ridiculous speed.” One might cite the exception of Hopkins; but I’m sure that anyone who has attempted to read “sprung rhythm” in perfect isochrony will find West’s observations here more rather than less compelling.
4. Important background: Akkadian, Hebrew, Ugaritic, and some other languages such as Arabic, are all Semitic languages deriving from the same proto-language. It is therefore plausible that certain aspects of their culture — especially verbal culture — might be cognate as well. Ogden Nash is valuable primarily for the amusement he provides and the possibilities he suggests rather than any direct relation.