Durational approaches to Ancient Greek metre have been rejected (in my opinion, with good reason) by Devine and Stephens (e.g. in "The Homeric Hexameter and a Basic Principle of Metrical Theory", Classical Philology, 1976, 141-163 and more thoroughly in their 1984 book Language and Metre), since they are inconsistent and unnecessary to explain the rules that have been discovered for different types of verse (something that they, together with scholars like Van Raalte, Sicking and Beekes, have done using rhythmical or rhythmical-linguistical analysis). I would like to know, if possible, how is the “V-to-V distance” concept different to (others?) durational approaches and what (yet unexplained?) rules do you expect might it help to explain.
Thank you for your question and my apologies for the late answer. Devine and Stephens took it for granted in their 1976 article and later work that weight distinctions are binary in stress and meter. Binarity for them was a point of doctrine before it was an empirical issue.
Now we know better for stress: see Matt Gordon's book on the cross-linguistic typology of weight "Syllable Weight" 2006, Routledge, mostly on stress. Gradient weight in stress is discussed p.126. Regarding meter see Kevin Ryan's 2011 article "Gradient syllable weight and weight universals in quantitative metrics" in Phonology 28 (2011) 413–454. doi:10.1017/S0952675711000212 and in his 2011 UCLA thesis. Ryan shows that effects of gradient weight are found in all the quantitative systems he has analyzed, including in Greek and Latin.
That said, I have not as yet worked through Devine and Stephens' evidence against Irigoin and West's views on multiple degrees of weight in Greek, to see who went wrong where. But Ryan's theory of gradient weight is intelligible, persuasive and fully general (in the sense of being also consistent with the stress evidence), so I'm putting my money on that.
Note also that it's possible that some arguments for gradient weight (e.g. Irigoin's) could be wrong without the basic idea of gradient weight in meter being wrong. What suffices for now is that Ryan's evidence shows there is clear empirical evidence for it.
Note finally that the V-to-V interval idea does not directly predict that every weight sensitive phenomenon will display effects of each step on the weight scale.
Having read Ryan’s paper, I begin by sharing your opinion that his theory is “intelligible, persuasive and fully general”. His methodology is probably flawless and I should also say that his regression analysis have left me eating humble pie.
That said, his interpretation of the data has a problem: he concludes that the hexameter is unique in the metrics of the world in having weak positions preferring heavier syllables than the strong positions. Also, he takes that as a confirmation of the theory of the ratio 1:1,2 (longum:biceps), which contradicts most of ancient testimony and, more than that, suggests than in hexameters of the (fairly common) type sssdd (or ssddd or ssssdd – “s” being spondee and “d”, dactyl) there was no problem in distinguishing strong from weak positions, which is at least strange since positions 2, 4 and 6 would be longer and most likely heavier than positions 1, 3 and 5.
I believe the problem is meta-theorical: he goes from metrical facts to phonological interpretation to metrical rules; that is, from the distribution of syllable types to weight sensitivity to longum:biceps ratio. But metric has other layers, and the final jump has to ignore the fact that, while syllable type distribution is clearly skewed, it is only skewed, that is, it is not clearly regulated. There are hundreds (well, to be honest, I am fairly confident there are hundreds) of hexameters with only VC types in biceps and only VV types in thesis (which of course is in clear contradiction with a 1:1,2 ratio).
I believe his data for the hexameter can be better explained from a rhythmical point of view: there was a tendency to place heavier syllables in biceps because the biceps is an ambiguous element and did not have the listener's expectations for a heavy syllable with which the longum counts. I could even say that the problem was precisely the 1:1 ratio between the constituents (something that could be verified by studying anceps:longum ratio in the trimeter): if the biceps had been longer [I am sorry if this is not the right tense] than the longum, it would have had [idem] no reason for placing heavier syllables there. It is because it was not that the poet was inclined to place heavier syllables in order not to risk the proper perception of the rhythm in spondees. It might be seen as a case of overcompensation: in order not to risk the 1:1 ratio (which, of course, should have been quite clear in dactyls), there was a tendency to place heavier syllables in spondees.
I feel this explanation is too brief, but I hope it is enough as to sense what I am going for with it. I would like to go on, but perhaps it would be a little too much for the “researchnet dialogue” genre. Thank you for your answer and for making me aware of the existence of Ryan’s stimulating paper.
- Badges
- Science topic
- Similar topics
- Psychology