A. M. Devine and Laurence D. Stephens, The Prosody of Greek Speech. New York/Oxford: Oxford University Press, 1994. Pp. xxii + 565. ISBN 0-19-508546-9.
Reviewed by Gregory Dobrov, The University of Michigan (firstname.lastname@example.org).
The intersection of classical philology and contemporary linguistics implicit in The Prosody of Greek Speech (hence PGS) is so narrow that the scholars who might be expected to read the book cover-to-cover with interest and understanding may well be the very same group called upon to review it for professional journals. This restriction to a specialist readership has as much to do with presentation as with the difficulty of bridging demanding disciplines. As I read this impressive record of the long and productive collaboration between A. M. Devine and Laurence D. Stephens, I recalled being asked over the years by quite a few mystified readers of CP and TAPA if I understood the DS papers and, if so, what they were about. Separately, it is true, the installments have been hard to contextualize, much less enjoy. Within the expansive framework of PGS, however, their project becomes more coherent, accessible and interesting.
DS state their purpose clearly in the first sentence of the preface: "the aim of this book is to answer the question what did Greek prosody sound like?" My most general disappointment with PGS, then, is its reluctance to answer this question in a way that would be intelligible to a competent philologist of Greek with some linguistics, in the manner of, say, W. S. Allen's Vox Graeca and Vox Latina. I single out Allen as a linguist whose talent for elegance and communication has made his little books useful and famous; he is also, of course, the author of the well-written Accent and Rhythm (1973), an important piece of comprehensive scholarship in the field of theoretical Greek prosody. I fear that PGS may not receive the attention it merits as the authors have chosen to embed their original research in an unnecessarily digressive and derivative matrix advertised by a bibliography of nearly two thousand items and a layout peppered with "borrowed" illustrations of everything from stroboscopic views of vocal cords to prosodic enclisis in Kwakwala.
The reconstruction of Greek prosody in PGS follows a hierarchy of domains: syllable, word, appositive group, minor phrase, major phrase, utterance. The ten chapters of the book are, accordingly, devoted to: the physiology of prosody (1), the syllable (2), rhythm (3), pitch (4), word prosody (5), connected speech (6), the appositive group (7), the minor phrase (8), the major phrase and utterance (9), and topic and focus (10).
In the first chapter DS dally explaining the basics of speech production (i.e., "labial" vs. "dental") at a level of sophistication oddly incongruous with that required to appreciate the next nine chapters. At the other extreme is their refusal to translate crucial but admittedly "less than transparent" testimonia from the likes of Dionysius of Halicarnassus and Choeroboscus on Hephaestion. From a historical and applied linguistics point of view the omission of any diachronic or dialectal perspective from the prolegomena is astonishing. Surely the undifferentiated use of texts from Homer to Babrius as evidence for a homogeneous "spoken Greek" is a dubious foundation that needs more than mere mention (as on pp. 451-52), it needs a defense!
This question of "what exactly is PGS about?" is bound up with another: "to whom exactly is PGS addressed?" The authors' lack of interest in appealing to a more general Classics readership is evident throughout: having implied that they wish to engage the serious Hellenist, they wander all over the intellectual map conversing in turn with various members of an implicit audience consisting mostly of metricians and phonologists. But which phonologists, exactly? In distinction from modern work in the field such as D. Steriade's "Greek Prosodies and the Nature of Syllabification" (diss. MIT, 1982), DS exclude a general linguistics readership by leaving all the Greek untransliterated and untranslated. As for Classical metricians, they are a practical lot likely to be frustrated with PGS as offering little in the way of concrete, applicable results.
Thus W. S. Allen's warning in Accent and Rhythm (p. 16) that "only by becoming a linguist can the metrist be adequate to his task" is programmatic for PGS. On page 46, for example, DS explain that "there remains an enormous amount of variability in the actual duration of what, at a more abstract level, would be considered the same phonological unit.... This variability arises from the influence of the context in which the phonological unit is embedded. Large portions of the rest of this book are devoted to eliciting the complex of factors that make up that context." PGS, then, is a loosely concatenated series of technical studies in prosody along these lines and, as such, does have much to offer the interested reader. DS have developed a formulaic approach to each problem they tackle: they break an issue down into manageable bits which they process first on a general, cross-linguistic or comparative, level. From this perspective they work their way back to Greek, responding to relevant research and presenting their own findings. Consider, for example, DS at work at on a particularly interesting problem.
In their discussion of the syllable DS engage proponents of the new concrete phoneticism such as M. Rossi and M. L. West who argue that meter is sensitive to subtle submoraic durational distinctions. The view of the sequence biceps>longum>anceps>breve as a general functional hierarchy underlies most research that uses metrical distribution as evidence for moraic, submoraic, and segmental duration. The promised "careful empirical investigation" of this problem takes up a full half of the 63-page chapter and is broken down into discussions of method, syllable structure, and a host of specific moments: muta cum liquida, intrinsic vowel duration, contextual vowel duration, intrinsic consonant duration, contextual consonant duration, rime structure, light syllables, heavy syllables, superheavy syllables, and prepausal location. The typical sub-section follows the pattern outlined above, i.e., from cross-linguistic comparison to a focus on Greek. Thus, in their discussion of superheavy syllables (pp. 76-79), DS adduce evidence for phonologically significant hyperbinary difference (e.g., light-heavy-superheavy) in languages such as Estonian and Yavapi. They continue with glances at evidence from research on North German dialects, Japanese, Arabic, Hindi, Finnish, Sanskrit, Classical Persian, Urdu, Luganda, Hausa, Bolanci, Alabama, Tewa, Kaliai and Yokuts. Then comes the inevitable DS table illustrating an analysis of Greek verse: sampling Homer, tragedy, and the iambographers they test the metrical relevance of CVC syllables in a specific environment (long third anceps in the central syllable of the molossus-shaped word). The result is simply that the distinction heavy/superheavy is not metrically relevant in Greek. Caveat lector: the task of repeatedly following a complex route to a simple answer -- from paragraph to section to chapter and beyond -- requires chalcenteric determination. I confess that when I had worked my way through about fifty pages of this particular thicket, I cheated and looked ahead, weary of the ring-composition pattern seasoned with innumerable references to exotic research. On page 84 I found the scoop:Consequently, it is not possible to sustain either the idea that segmentally based submoraic distinctions in duration are metrically relevant or the overly restrictive position according to which the metre gives evidence for no durational distinctions other than the binary distinction between heavy and light syllable.... What these data suggest is a theory in which syllable sequences are phonologically processed into rhythmic structures for speech production.This is an important, if inelegantly stated, conclusion valuable indeed to anyone interested in metrical and prosodic theory. Like most moments of insight in PGS, this summary accords with common sense and is far from surprising. I wish only that the journey were not so tedious.
The contribution of PGS becomes more sketchy and tentative as the authors move outward from their field of mastery towards ever longer constituents. The chapter on the appositive group, for example, though spanning ninety pages (285-375), sounds in the end like an update to the discussion of proclisis in H. W. Chandler's Practical Introduction to Greek Accentuation (Oxford 1881). They do survey a fascinating assortment of material along the way, however, including Broca's aphasia, word games, figures of speech, and inscriptional evidence. "Proclitics are not atonic," they conclude. "Although they are part of the rising trajectory, they are more comparable to lexical grave accented words than to preaccentual word internal syllables ...." Again, this is interesting if impossible to incorporate into the actual pronunciation of Greek. Given DS's fascination with cross-linguistic comparison, I was surprised to find no mention of current work on anaphora in their discussion of function words (pp. 291-92). Syntax is not their strong suit, to be sure, and there are evidently limits even to a bibliography with global ambitions.
As a native speaker of Russian, I especially enjoyed the last chapter of PGS, "Topic and Focus," if not for conclusive results (an unfair expectation) at least for the way in which DS draw attention to a central aspect of discourse usually glossed over with general talk of "emphasis." This is an important phenomenon at the interface between syntax and semantics that often eludes students of ancient languages. One can only marvel at DS's intrepid and virtuosic attempts to use the Delphic Hymns, poetic texts, and inscriptional evidence to shake a "dead" language to life. Reversing E. Sapir's injunction, "study carefully the phonetic system of a language ... and you can tell what kind of verse it has developed" (p. viii), they work wonders hypothesizing the cadences and sonic contours of spoken ancient Greek. This is an interesting aspect of their work that defies simple description so I invite you to visit their gallery of diagrams (Chapter 9) where every picture tells a story.
If my yearning for more coherence and concreteness was left unfulfilled, I nevertheless appreciate how much greater a challenge DS have had to face in studying Greek than do those who work on modern languages. In the end, however, I am still haunted by the image of the diligent Hellenist who remembers W. B. Stanford's enthusiastic 1967 Sather Lectures on "The Sound of Greek" or who may have been intrigued by K. J. Dover's J. H. Grey Lectures on "Greek Word Order" (both ignored by DS) and who seeks an enhanced understanding of the language s/he teaches for a living. After reading pp. 376-455 s/he will be no doubt be dazzled by the long, detailed, and aggressive speculations on pitch accent catathesis; s/he will wonder how they can be so confident that the "melodies" of the Hymns have direct bearing on the language used in daily life (where and when, by the way?); and s/he will probably lay the book aside somewhat annoyed with linguistics in general and repeat the question with which it all began: "what did Greek prosody sound like?"