Bryn Mawr Classical Review 03.01.14

RESPONSE: Daitz on Wyatt on Daitz

Stephen G. Daitz
The City College, CUNY

I am delighted that Wyatt (see BMCR 2.7.9) has given me an opportunity to clarify my views on the question of pause in the oral rendition of Homeric verse. I hope to show that despite divergences in detail, our overall views are actually closer than might at first be apparent.

First our divergences, addressed in the order of Wyatt's "Response".

1. Wyatt quotes John Crowe Ransom's statement, "I do not myself know how poetry should be read". This statement is in a sense a truism, for obviously there is no one way that all poetry should be read. But could we not accept a partial converse to this statement: We do know how classical poetry (and most other poetry) should not be read, e.g. unrhythmically (assuming that our reading aims for some degree of poetic authenticity).

2. Wyatt posits a dilemma in reading a verse from Donne's tenth elegy: "So if I dream I have you, I have you", the dilemma being that a reading of the verse as five consecutive iambs would distort the meaning. True, but it is a common prosodic feature of English iambic poetry to substitute occasionally a choriamb for two iambs, e.g., "To bé or nót to bé, thát is the quéstion." And so forth, with a choriamb substitution in exactly the same position. Donne's verse can be satisfactorily rendered: "So íf I dréamed I háave you, I háve you."

3. Wyatt believes that line end of the hexameter is sufficiently marked without pause because only one syllable is permitted after the longum of the sixth measure and because the phenomena of liaison, elision, and correption are allowed within the line but not at line end. However, these phenomena, correctly noted by Wyatt, would be more of a compositional parameter for the poet in the act of composition than they would be a signal of line end for the listener. It seems to me that the normal rhythm of the hexameter's code (- ญญ - -) plus the pause of at least one mora after the last syllable is the true signal to the listener that the rhythmical promise of the hexameter has been fulfilled. (The pause factor would be crucial in lines ending with a double spondee.)

Wyatt is prepared to assume that in Iliad 1.51-2 (... E)FIEI\S / BA/LL' AI)EI/ ...) a rhapsode would not pause after E)FIEI/S because of the enjambement. This assumption is presumably based upon the belief that the listener might lose the meaning if a pause of one mora after E)FIEI/S were made. (What other reason would there be not to pause?) I think that such an assumption of possible listener confusion is not warranted, particularly if the listener has been trained through long experience to expect a regular external (end of line) pause.

4. Concerning internal pause (pause within the line), Wyatt states that the various conventions of epic verse (caesura, elision, liaison, etc.) are not obligatory, for the text shows exceptions. I believe Wyatt is correct for most of these conventions, but how do we know that the principle of syllabic liaison (a syllable will always begin with a consonant if a consonant is available) is not obligatory? Incidentally, syllabic liaison is not only a convention of epic verse, but is a basic phenomenon of the Greek language.

As to a pause between the first two words in Iliad 1.52 BA/LL' AI)EI/, the only reason to posit such a pause is again the assumption that the listener might lose the meaning if no pause were made. As before, I believe such an assumption unwarranted, for the rhapsode can make the listener aware of a sense boundary after BA/LL' by slowing the tempo and by changing the pitch, without interrupting the rhythmic flow.

5. Likewise, in Odyssey 1.2, Wyatt posits a pause after PLA/GXQH (- u)since "poet and audience would know that because the vowel was short, another vowel would follow". However, the Homeric audience was not composed primarily of grammarians listening for the application of rules of prosody, but of a hoi polloi highly trained to listen for rhythmical regularity. In quantitative rhythm, time marches on. A pause after a short syllable in effect lengthens that syllable and in the case of correption (PLA/GXQH), defeats the purpose of correption which is to reduce a long syllable to a short. Besides, the traditional comma in our texts after PLA/GXQH represents German grammatical punctuation and is not even reflected in normal German speech.

Since the publication of my article in AJP, a piece of additional evidence has come to my attention which I believe corroborates my conclusion that internal pause was normally avoided in oral rendition of Homer and of other classical poets. A fragment of Philodemos indicates that the words in musical settings were rendered in continuous (uninterrupted) fashion in contrast to the natural (normal, grammatical) fashion.

The stimulation arises ... through the words being rendered with an uninterrupted delivery rather than with a natural delivery (i.e. with pauses).
Philodemos, Peri Mousikês XXVIII Kemke
Philodemos, it is true, is speaking of musical settings of words, but in ancient Greece, poetry and music were so closely intertwined, that I think it safe to assume that what applied to one form of verbal rendition applied equally to the other.

6. Wyatt states that Nicanor's rendition of Odyssey 1.1, with its long pauses before and after the vocative, Mousa, was an accepted pronunciation. But how do we know? Nicanor's statements may represent purely theoretical prescription, not description.

7. Wyatt states that "we have no evidence for or reason to suspect a glottal catch in Greek". However, in Homer we have many words with uncontracted vowels, e.g. NHLEE/S. Two extreme cases are Iliad 12.318 A)KLEE/ES (- u u -) and Iliad 14.271 A)A/ATON (- u u -). Each of the underlined vowels represents a separate syllable. It seems to me that in a rendition of such words, to avoid an aural smear of alphas or epsilons, and to maintain syllabic identity one must make at least a mini-glottal stop between the consecutive vowels. Can Wyatt suggest any other way?

Despite the above divergences on details, Wyatt's concluding statement that ancient performance of Homer "may have involved a pause here and there in the verse, and ... will have introduced ... signals appropriate to the syntactic structure..." is actually quite close to the conclusions given in my article. I feel quite certain the evidence strongly indicates that external pause was the normal procedure in Homeric performance, though it would be unprovably dogmatic to assert that external pause was never omitted. Likewise, I believe the evidence adduced in my article indicates that internal pause was normally not practiced in Homeric performance, particularly after a short vowel, after elision, or in syllabic liaison. But, as I stated in my article, there undoubtedly were exceptions, e.g. for purposes of metrical lengthening, to avoid hiatus, for literary expressiveness. (In my own recent recording of the Iliad, such occasional pauses, although relatively rare, have been introduced.) However, in a large majority of cases, the punctuation of the OCT and other standard texts of Homer, suggesting frequent internal pauses, should be resolutely ignored in oral performance of Homer. The semantic flow of the poem, including sense boundaries, can be adequately conveyed to the listener by variations of tempo and pitch without the rhythmic confusion that frequent internal pauses would produce. And so, in the large view, I believe that Wyatt's thoughts on pause and my own are highly compatible.