Reviewed by Philipp Brandenburg, University of Puerto Rico, Mayagüez

In an investigation of the principles that underlie accentuation in ancient Greek, Philomen Probert focuses on nouns and adjectives with specific suffixes and concludes that the distribution of accents indeed follows a finite, although complex, set of rules.

In Part 1 (Chs. 1–4), P discusses the obstacles of research in a field where direct evidence is missing. The system of accentuation, however—as introduced by ancient grammarians in the second century BC and continuing in medieval manuscripts through the introduction of minuscule writing in the ninth century AD—is corroborated by papyri and ancient music. A refreshingly nontraditional overview of Greek accentuation, a summary of historical changes in accentuation, and a brief history of scholarship make these chapters an informative introduction to the subject. However, I do not accept the existence of an oral tradition of accentuation from Homeric times down to the grammarians (33; 44) nor that of the ‘Homeridae’ (cf. Fehling 1979).

Part 2 (Chs. 5–12) investigates how individual morphemes influence the position of accent. Recessive accentuation is the default (it is more frequent, the norm in Lesbian, applied to loanwords from Latin, and found on words that in Indo-European had a final accent). Derivational morphemes, having inherent accentual properties, may thus demand a nonrecessive accent. Adjectival suffixes often have nonrecessive accentuation, whereas nouns with the same suffixes more often than not have recessive accentuation. Exceptions to these principles occur in either highly frequent or extremely infrequent words with -ro-, -to-, and -no-. With -lo- and -mo- the situation is complicated by other factors (e.g. animacy or synchronic transparency). High frequency allows for storage of individual (i.e. irregular) forms in the mental lexicon, whereas a rare word, P says, ‘may not lose its status as a complex form and its derivational suffix will retain its synchronic identity’ (233). But where should synchronic knowledge about the morpheme’s identity derive if not from the default (i.e. mid-frequent) forms? P holds that highly frequent nondefault accentuation is (historically) correct: speakers (or scribes) could equally well apply a hypercorrect accentuation to an infrequent word exactly because of its rarity (e.g. as German wohlgesonnen currently replaces wohlgesinnt ‘well-disposed’).

This book is a valuable contribution to our understanding of accentuation in Ancient Greek and will attract both classical philologists and historical linguists. Appendices with commented word lists as well as several indices guarantee accessibility, although acquaintance with the Greek alphabet is a prerequisite. P’s vast erudition in the field of Greek literature stands beyond any doubt.

REFERENCE

FEHLING, DETLEV. 1979. Zwei Lehrstücke über Pseudo-Nachrichten. Rheinisches Museum 122.193–210.