We live in a golden age of Ancient Greek accentuation, which Dieu’s excellent tome helps us celebrate. For our knowledge of Aelius Herodian (c. 2 AD), “La source principale” (p.8) of our knowledge on Greek accentuation, no longer are we bound to Lentz’s edition—or prose composition, such was his mosaic of the splintered tradition—since a spate of recent activity has ushered in new critical editions of the epitomes to Herodian, Johannes Alexandrinus Praecepta Tonica (ed. Xenis 2015) and “Pseudo”-Arcadius Περὶ καθολικῆς προσῳδίας (ed. Roussou 2018). In the last decades, West’s editions of Homer (1998; 2017) have brought in fresh and provocative perspectives on accentuation in epic (not uncontroversially so: see the review of West’s Odyssey in BMCR 2019.01.05 by Graziosi and Haubold). New light comes from the study of phonology: already since Dieu’s monograph, Sandell & Gunkel (2024) challenge widely held views on the acute accent, and as this review goes to press I have become aware of a significant new publication on the phonology of Greek accentuation, Probert 2025. Likewise, as new works on Indo-European accentuation and diachronic prosody (e.g., Sandell 2023) have us reconsider received doctrine on the prehistory of Greek accent and ablaut as represented by, e.g., Rix (1992), a return to the philological sources is salutary. All these works speak to the subdiscipline’s vigorousness. How does Dieu’s monograph match up? In brief, Dieu’s Traité sets the standard as a reference work on Ancient Greek accentuation, for which no English-language equivalent even competes. The name Traité conjures up the treatise by Vendryes (1904) but Dieu embraces a broader range: this is not a work of pedagogy but of reference, not for the neophyte but the expert. Like Vendryes, Dieu offers a major resource of the philological facts and an exploration of prehistory—an “exposé descriptif et enquête diachronique” (Avant-propos p. v)—but at more than double the length of his predecessor, Dieu’s Traité makes for an all but exhaustive account. In this stated goal, Dieu’s monograph manifestly succeeds. Dieu is judiciously critical of the labyrinthine sources of our knowledge and demonstrates a mastery of comparative linguistics and textual philology alike. This new Traité, already fairly crowned with the 2023 “Prix Alfred Croiset” of the Académie des Inscriptions et Belles Lettres, deserves to be at hand for anyone working on the Ancient Greek language.
The book is neatly organized in the following chapters. 1. Sources (manuscripts, music, lexicography, comparative linguistics); 2. Nature of the accent; 3. Values of the accentual signs; 4. General laws relating to the accent (law of limitation, σωτῆρα-rule, and the rhythmic laws Wheeler’s and Vendryes’s); 5. Proclitics; 6. Enclitics; 7. Accentuation of verbs; 8. Accentuation of nouns, the meatiest portion dividing into “base accent” (7-10) and “case accent” (11), with a preliminary section on accentual properties of morphemes, then organized according to suffix classes: nouns in -ο-, -μο-, -νο-, etc.; 9. Accentuation of nouns continued (compounds); 10. Accentuation of nouns continued again (proper names, anthropo- and toponyms); 11. Accentuation of nouns concluded (inflectional accent); 12. Pronouns, numerals, adverbs; 13. Accentuation of words in phrases (anastrophe, crasis, elision, etc.); and 14. Accentuation in the dialects, including the Homeric Kunstsprache. A rich bibliography and indices, including a helpful index of grammatical and musical vocabulary, round out the tome. Amidst such a diacritic thicket, typos are miraculously few.
Perhaps the quality that most shines through is Dieu’s fairmindedness. As he puts it (p. viii), he strives to avoid every form of dogmatism, a laudable move when the subfield of Indo-European accent and ablaut has at times entrenched into just the kind of cantankerous rigidity that Dieu avoids. Dieu regularly reports views promulgated by a single school even when those views are judged improbable or unlikely. Conversely, he seldom promotes a view, and when the matter isn’t clear, he says so; I didn’t attempt to count how many paragraphs end with the tagline non liquet. That’s not a criticism so much as Dieu’s honest admission of limits on knowledge, and it should be taken as a spur for us all towards further research: the time is ripe to reassess the evidence, to make newly informed analyses, and to keep an open mind to new ways of reconstruction. I will probably be one of the few who enjoys reading a work like this cover to cover but the histories of the field are engagingly written, from the accentual disputes in Alexandria to Meillet’s mulling Lithuanian comparanda. Indeed, one great strength of the book is Dieu’s command of prior scholarship, as he delights in revisiting older works for what they might offer now. A typical example comes on p.122 where we are referred to the reflections of the great nineteenth-century Sanskritist Benfey (writing in 1880 on anastrophe), antiquated though these ideas may be, for yet they remain “quelquefois encore stimulantes.” I would say the same of many reflections that Dieu recovers: they remain stimulating.
Although no topic in Greek grammar so quails the heart as accentuation (perhaps), yet the accents are vital to the meaning of words and anyway make for the sound and music of the Greek language. Many corners of grammar that this work investigates might appear arcane, but these seemingly minor matters open out to the realms of literary criticism and historical reconstruction and so merit a broader readership. Did Homeric bards belt out the opening of the Odyssey as ἄνδρά μοι ἔννεπε, Μοῦσα (printed by West) or ἄνδρα μοι (with many editors)? Which is likelier that Homer sang or Plato heard? Typical of Dieu (159-73), he brings to this question acumen, marshalling a medley of manuscripts, grammarians, and editors past to assess the material, turning also to comparative evidence (diphthongs in the Baltic languages) and more generally the question of whether Ancient Greek had resonant diphthongs. From the philological fact that ἄνδρά μοι is possible, one could turn to linguistic analysis: what would each reading of the host and clitic dyad represent for prosodic domain formation and recursion therein?[1] Another example (p.247), in a great acme of our Iliad, when Hector cuts down Patroclus, the son of Menoitios who has himself killed (πεφνοντα) so many, is the participle accented πέφνοντα with Aristarchus or πεφνόντα with Tyrannio? In a typically informed treatment, Dieu sifts the scholia (ad Π 827), who suggest Aristarchus interprets πέφνοντα as a present vs. Tyrannio’s πεφνόντα an aorist, akin to the fluctuation in the nom.sg. καταπέφνων or καταπεφνών (Ρ 539). Herodian—and Dieu’s reference is rightly now to Ps-Arc. 339.6 Roussou (though generously noting also the older edition, 201.5-6 Schmidt)—takes this form as a second aorist, agreeing with the morphological analysis by Aristarchus, but actually still accenting πέφνων (as also with the exceptional inf. πέφνειν), which would not be an analogy to the present participles but indicative of an Aeolic origin. This verb is ancient and inherited—it is the signature verb of the Indo-European dragon-slayer (Watkins 1995: ch. 49, this passage p. 482)—but how was it heard in Homer’s day, in scholarly circles from antiquity to Byzantium? Ultimately Dieu leaves the question open. Small stuff, but philology is founded on these bricks that build to a grander edifice. This example touches on verbal aspect, maybe also the mind of Hector (how does he envision the deeds of Patroclus?), the antiquity of the thematic vowel as an inherently accented suffix in reduplicated aorists, and the linguistic layering of Aeolisms in epic diction. At times, the matters are truly minor but new questions might arise: Is the infinitive ζευγνύμεν (Π 145), with metrical lengthening of the suffix -νυ- (surely aided by the allomorph in the singular, ζεύγνῡμι), to be treated as inducing the σωτῆρα-rule, so ζευγνῦμεν (with West; see p. 558 for discussion)? From this evidence could be posed new questions about the imposition of the σωτῆρα-rule onto poetic forms. At nearly 700 pages the work will not be for the faint of heart, but then again the book is conceived as one of reference, not perusal.
Anyone steeped in Greek historical linguistics will carp some about missing items. Mine are minor but to record just a few from a sample of pages: p. 214 on the Ionic contract τιθεῖσι etc., reference should be to Willi (2012); p. 242, brief excursus on the “Urtext” of Homer and the thematic aorist infinitives like βαλέειν, see Nikolaev (2013). In the same vein, some forms could of course be handled differently: p. 282 πῑ́πτω with a long vowel is dubious (West 1998: xxi “iota breve habuit”); p. 341, on τέρυς, see Nussbaum (1997). A presentational quibble is that the Caland system, informing much of Dieu’s discussion of nominal derivation (and well it should), is introduced but obliquely and late in the day (p. 298 n. 507). This is small stuff. More substantially on the Indo-European end, when Dieu has recourse to Anatolian (surveyed p. 13, and sprinkled throughout) he is unaware of the major revisions by Yates (2016; 2017, i.a.). Anatolian is particularly relevant because that branch is widely held to be first to hive off from the family, thus providing a precious window onto the oldest stages of the proto-language. Moreover, Yates has proposed several Anatolian reconstructions that directly impact our picture of Ancient Greek (cf. Yates 2019 on *-ói-stems, like φειδώ ‘sparing’, stem φειδοῖ-). This agreement of Ancient Greek, Vedic, and now Anatolian, corroborated by Germanic and Balto-Slavic and gainsaid by none, could be taken to evince that such a “lexical accent system” or something very like it ought to be reconstructed for the proto-language.
Indeed, it is just this step, shy of reconstructing specific accentual properties of morphemes and an explicit phonological account of how they combine, clash, and resolve, that seems to me a shortcoming. Dieu explicitly eschews generative grammar but really operates without any formal phonological framework. He cites several recent studies but hardly engages with them. In a work that devotes reams to wrestling with what scholars of the late nineteenth and especially early twentieth century made of the Ancient Greek accents, I felt a few pages at minimum could have been spared for formal approaches from phonology of the last fifty years. Citations to works like Steriade 1988 are dutiful but superficial (more recent works like Steriade 2018 do not make it in, and Steriade 2013 unfortunately remains unpublished and was presumably unavailable to Dieu); important items in this vein, those seeking to capture interactions of accentual properties, syllabification, and morphology (e.g., Kiparsky 2003), are absent.
The closest we come is Chapter 8, which opens with a helpful discussion of the accentual properties of morphemes and how they combine, basically all of which I agree with (clear survey in Gunkel 2014). There’s “dominance” (ability to determine the place of the accent) and “force” (ability to bear the accent), these two and other properties mixing in morphemes to deliver the surface position of the accent. Dieu makes an exception of his eschewal when he draws from the toolkit those theoretical insights that seem expedient. He mentions first an inherently accented morpheme (think -ρό-, -εύς, etc.). Second, Dieu cites a default accent in the case of no dominant accented morphemes, which for Greek means the recessive accent (following Probert 2006). Third, he uses “demorphologization,” a term Probert (2006: see index p. 440 s.v.) introduces and puts to good use, describing a process where a word stem that was once segmentable ceases to be so, subsequently treated as an unanalyzed stem, or root, with repercussions for the accent (Probert 2006: p. 412 for a definition). Dieu endorses work in this vein. Accentual dominance is described as particularly pertinent—and yet it will not be used (pp. 253-7). Although Dieu acknowledges that this type of analysis systematizes facts observed long ago, and systematizing is one mark of a good phonological theory (i.e., it can generalize), yet Dieu declines to use it because “elle ne permet pas forcément de mettre en évidence la genèse des systèmes linguistiques” (p. 255). This, to me, is strange. A description of Greek accentuation is in the first place synchronic, the rules or constraints holding across the phonology of the language; an adequate theory of accentuation does not need to explain the system’s origin. What we need is a maximally explanatory framework for prosodic structures, including a classification of suprasegmental categories—a synchronic analysis. Diachronically, we need a plausible (i.e., learnable) pathway from prehistory to the attested stages. Now, learnability and change in prosodic phonology are particularly challenging, admittedly; Kiparsky (2015: pp. 82–83), for one, locates the “locus of morphophonological variation and change” not in the word accents themselves but the system that assigns them, thus correlating prosodic change not with segmental but nearer to syntactic change. Tricky indeed—but we have to try. Absent a phonological framework, we lose generalizations over the otherwise disparate facts; diachronically, “analogy” becomes the sole arrow in Dieu’s quiver.
Space forbids me from diving far into details, but an example of this drawback comes from his treatment of the recessive accent in (almost all) finite verbs. This major morphological class (verbs) falls within the law of limitation, for which Dieu does not offer an explicit phonological account. Dieu favors a “genèse” that begins from an enclisis of the verb, following a proposal by Wackernagel (of 1877), based primarily on Vedic Sanskrit, where finite verbs in matrix clauses lack a word-level accent (bharanti 3pl. ‘they carry’ but in a subordinate clause bháranti). This picture would be corroborated by certain facts of the Germanic languages. But diachronically this is a far from self-evident account. The Vedic verbal accent much likelier represents a lowered tone at the level of sentence intonation, not word-level prominence. As Weiss (2020: p. 117, with reff.) puts it, “[t]he non-accentuation of verbs in main clauses in Vedic has nothing to do with enclisis of the more general sort.” Similarly, the Germanic metrical evidence speaks for a relatively weaker prominence of the verb in a verb-complement string (Getty 2002: pp. 52–69, 139–47), but not enclisis. Moreover, the evidence from grammatischer Wechsel unequivocally reflects a retained verbal accent in Germanic (e.g., Old English 3sg. wearþ ‘became’ but 3pl. wurdon < *wurðón, cf., e.g., Fortson 2010: 346). An open-class lexical category like “verb” is not a likely category for a broad enclisis; more plausible is that nouns are prosodically privileged over verbs, a cross-linguistically established trend (e.g., Haspelmath 2012). If this diachronic sketch tracks, we should not start from a proto-Greek clitic verb. Now, the one account Dieu pits himself against, that by Probert (2012; see now 2023 and 2025) precisely does not begin from this position but instead from a verbal accent that instantiates the law of limitation avant la lettre. In this view, Proto-Greek *phérō, *phéresi, *phéreti (or *phérei), 3pl. *phéronti would all appear to embody the law for childhood learners as they acquired their Greek verbs. From this data, Probert argues, learners converged on an accentually restrictive window at the right edge of the word, which she formalizes, following Steriade (1988), in terms of metrical phonology. In a plain difference of approach, Probert (2023:129) begins by establishing an account of the law of limitation because the law must have a “reasonable synchronic analysis in order to have existed at all.” This analysis Dieu does not provide; Probert’s proposal seems to me the more fruitful starting point. If many verbs adhered to the budding law, learners could create a constraint on the accentable window from this body of evidence. Strong support for this could come from the fact that very few (any?) verbal suffixes in Proto-Greek would have been lexically accented, in contradistinction to nominals, and so have offered vanishingly few forms that “break” the analysis by overriding the default recessive accent. Dieu (p. 200) chides Probert for failing to recognize the extent of tudáti presents but this is a red herring: tudáti presents in Proto-Greek would be scarce at best (cf. Willi 2018: pp. 351–3). The few finite verbs that are not recessively accented (e.g., aorist imperative εἰπέ, with infinitive εἰπεῖν, participle εἰπών) show that a verbal category with a lexically accented suffix was not clitic and could in fact retain its accent. Whereas many nominal classes are recessively accented but, crucially, not those with a lexical accent (e.g., adjectives like μαθηματικός, λιγυρός, or nouns like βασιλεύς), so few verbal classes would have been inherited with inherently accented suffixes that learners would not parse their verbs as anything but recessive.
Might this be the origin of the recessive accent in verbs? I don’t know for sure and a review is the place neither to articulate nor to adjudicate in full a proposal. I suggest it here to reveal a deeper split that will, I predict, continue to shape the field in coming years. The account above could not be suggested without reference to key concepts from contemporary linguistics: metrical phonology, accentual properties of morphemes, and the critical role of learners in diachronic prosody. I suggest this different pathway because I find such an account more explicit and explanatory. However, let me underscore that I have begun thinking through this issue because Dieu’s monograph is so solidly grounding, his reflections so stimulating. To eyeball scores of forms so lucidly arranged, so expertly assessed in terms of their philological worth, readily inspires new thoughts. I am confident that future studies of Greek accentuation will engage with and just enjoy Dieu’s meticulous repository of facts on the ground.[2]
References
Fortson IV, Benjamin W. (2010). Indo-European Language and Culture: An Introduction (2nd edn.). Oxford / Maldon, Mass.: Wiley-Blackwell.
Getty, Michael. (2002). The metre of Beowulf: A constraint-based approach. Berlin: de Gruyter.
Gunkel, Dieter. (2014). Accentuation. In Encyclopedia Ancient Greek Language and Linguistics (Vol. 1, A-F, pp. 7–10). Leiden / Boston: Brill.
Haspelmath, Martin. (2012, July 6). Nouns are prosodically privileged over verbs [Billet]. Diversity Linguistics Comment. https://doi.org/10.58079/nss3 (https://dlc.hypotheses.org/210)
Kiparsky, Paul. (2003). Accent, Syllable Structure, and Morphology in Ancient Greek. In Elizabeth Athanasopoulou (ed.), Selected Papers from the 15th International Symposium on Theoretical and Applied Linguistics (p.81-106). Thessaloniki: Aristotle University of Thessaloniki.
Kiparsky, Paul. (2015). New perspectives in historical linguistics. In Claire Bowern & Bethwyn Evans (eds.), Routledge Handbook of Historical Linguistics (pp. 64–102). London: Routledge.
Nikolaev, Alexander. (2013). The Aorist Infinitives in -EEIN in Early Greek Hexameter Poetry. The Journal of Hellenic Studies, 133, 81–92.
Nussbaum, Alan J. (1997). A Note on Hesychian τέρυ and τέρυας. In Douglas Q. Adams (ed.), Festschrift for Eric Hamp: Vol. II (pp. 110–119). Washington, D.C.: Institute for the Study of Man.
Probert, Philomen. (2006). Ancient Greek Accentuation: Synchronic Patterns, Frequency Effects, and Prehistory. Oxford: Oxford University Press.
Probert, Philomen. (2012). Origins of the Greek Law of Limitation. In Philomen Probert & Andreas Willi (eds.), Laws and Rules in Indo-European (pp. 163–181). Oxford: Oxford University Press.
Probert, Philomen. (2023). Review of DIEU, ERIC: Traité d’accentuation grecque. Kratylos, 68(1), 123–132.
Probert, Philomen. 2025. Open Questions in Ancient Greek Phonology: Some New Evidence from Enclitics. In Jesús de la Villa & A. Striano (eds.), Advances in Ancient Greek Linguistics (pp. 33–64). Berlin: De Gruyter.
Rix, Helmut. (1992). Historische Grammatik des Griechischen: Laut- und Formenlehre (2nd edn.). Darmstadt: Wissenschaftliche Buchgesellschaft.
Roussou, Stephanie. (2018). Pseudo-Arcadius’ Epitome of Herodian’s De Prosodia Catholica. Edited with an introduction and commentary. Oxford: Oxford University Press.
Sandell, Ryan. (2023). Towards a Dynamics of Prosodic Change: Corpus-Based and Computational Studies in the Synchronic and Diachronic Prosodic Phonology of Indic, Greek, and Germanic [Habilitationsschrift]. Munich: Ludwig-Maximilians-Universität München. [Online: https://www.ryan-sandell.com/_files/ugd/812729_c91236174de14eedb80822a5fa58462e.pdf]
Sandell, Ryan, & Gunkel, Dieter. (2024). On the representation and realization of the Ancient Greek acute. Journal of Greek Linguistics 24, 78–183.
Steriade, Donca. (1988). Greek Accent: A Case for Preserving Structure. Linguistic Inquiry, 19, 271–314.
Steriade, Donca. (2013). A Synchronic Analysis of Ancient Greek Accent [Harvard-MIT lectures, 2012-13].
Steriade, Donca. (2018). Quantitative Rhythm and Saussure’s Tribrach Law. In David M. Goldstein, Stephanie W. Jamison, & Brent Vine (eds.), Proceedings of the 28th Annual UCLA Indo-European Conference, Los Angeles, November 11–12, 2016 (pp. 231–266). Bremen: Hempen.
Vendryes, Joseph. (1904). Traité d’accentuation grecque. Paris: Klincksieck.
Watkins, Calvert. (1995). How to Kill a Dragon: Aspects of Indo-European Poetics. Oxford: Oxford University Press.
Weiss, Michael. (2020). Outline of the Historical and Comparative Grammar of Latin (2nd edn.). Ann Arbor: Beech Stave.
West, Martin L. (1998). Homeri Ilias (2 vols). Stuttgart and Leipzig: Teubner (vol. 1), Munich and Leipzig: Saur (vol.2).
West, Martin L. (2017). Homerus Odyssea. Berlin: De Gruyter.
Willi, Andreas. (2012). Kiparsky’s Rule, thematic nasal presents, and athematic verba vocalia in Greek. In Philomen Probert & Andreas Willi (eds.), Laws and Rules in Indo-European (pp. 260–278). Oxford: Oxford University Press.
Willi, Andreas. (2018). Origins of the Greek Verb. Cambridge: Cambridge University Press.
Xenis, Georgios. (2015). Iohannes Alexandrinus: Praecepta Tonica. Berlin: De Gruyter.
Yates, Anthony. (2016). Stress Assignment in Hittite and Proto-Indo-European. Proceedings of the Linguistic Society of America, 1, 25/1-15.
Yates, Anthony D. (2017). Lexical Accent in Cupeño, Hittite, and Indo-European [PhD Thesis]. University of California, Los Angeles.
[Online: http://www.adyates.com/uploads/1/1/2/0/112011417/yates2017diss.pdf]
Yates, Anthony D. (2019). Suffixal *o-vocalism without “amphikinesis:” On Proto-Indo-European *–oi-stems and ablaut as a diagnostic for word stress. In David M. Goldstein, Stephanie W. Jamison, & Brent Vine (eds.), Proceedings of the 30th Annual UCLA Indo-European Conference (pp. 199–221). Bremen: Hempen.
Notes
[1] On both questions see now Probert 2025, who argues for the accentuation ἄνδρά μοι but against the analysis as continuant diphthongs. On the question of clitics and a recursive accentuation in clitic sequences, I also refer to her article.
[2] Acknowledgments: I am grateful to the editors of BMCR for their patience, as this review was delayed; the book’s arrival was postponed by the publisher, and at 700-plus pages took me a while to read. I am thankful to Brent Vine for reading a draft; of course, I alone am responsible for the contents.