This book is a survey of the morpho-syntax of time expressions in Ancient Greek. The aim is both empirical – to illuminate a variety of temporal constructions in investigating the uses of cases and preposition phrases in Ancient Greek – but also theoretical – to illuminate our understanding of expressing time in language. This is an impressive piece of work, thorough and meticulous. The empirical investigation is both quantitative and qualitative in nature where a great deal of care is placed not only in establishing strong generalisations but also in looking at individual, often ‘problematic’, examples. Moreover, G.'s methodological choices (for example the choice of authors, as well as the inclusion or not of genitive absolute with other genitives of time) are considered very thoroughly.

The scope of the work is impressive; it includes a thorough investigation of time expressions in Thucydides and then a diachronic excursus in both directions: other Attic authors, Hellenistic authors, non-literary Papyri and finally Biblical texts but also a look backwards to Homer to investigate the diachronic origin of the patterns seen in Classical and post-Classical prose. The book is structured as follows: first expressions of time are considered in great detail in Thucydides. This enables G. to establish the ‘Classical Greek pattern’ which is then compared with what is found in other Attic authors, in order to test its consistency across authors, genres and periods. The differences and similarities are considered and then the pattern is further investigated into the diachrony of Greek in order to establish trends of language change.

In Chapter 1, G. sets the scene in terms of his aims and his theoretical assumptions. As already mentioned, the main theme of his investigation is to identify the use, meaning and distribution of genitive vs dative of time in Ancient Greek. G. identifies the factors that affect the choice of one construction over the other: (a) the lexical semantics of the noun itself (he:mera ‘day’, nykta ‘night’, etc.); (b) the event type of the verb modified by the temporal expression; (c) the modifiers of the time noun. G. discusses the theoretical issues surrounding the choice of temporal expressions in terms of event type, aspect and the role of modifiers in temporal expressions. While the first two are fairly straightforward, the third one is based on Jiménez (1998) who argues that the choice of case in temporal expressions is also dependent on the modifiers of the time noun (e.g. the accusative occurring with polys ‘much’, oligos ‘little’ and cardinals and so on). G., however, dismisses Jiménez's proposal. The final section of this chapter is devoted to defining his data set in addressing a common problem for anyone working on Ancient Greek: how can one be certain that a genitive noun phrase is a genitive of time and not, for example, an adnominal genitive or a genitive absolute and so on (cf. p. 34). A careful examination of these preliminary data gives a taste of the level of detail that is to come.

Chapter 2 is devoted to the expressions of time in Thucydides. Initially G. clarifies his choice of author as the standard of Ancient Greek prose: based on both the genre of history as appropriate and also on his less idiosyncratic personal style. He starts by looking into the accusative of time, which is easier to understand than datives and genitives, and essentially is reserved for modifying durative and atelic events and habitual actions. Even this construction has counter-examples that are discussed at length. G.'s strength is not just in the exhaustive character of his investigation and the overwhelming generalisations that stem from that; it is also found in the profound way that he discusses counter-examples without any desire for swift dismissal, indeed with a painfully thorough eye, always driven by the desire of accuracy. He then moves on to address the factors that motivate the use of prepositional constructions like dia + genitive or epi + accusative instead of simple accusative. He then moves on to discuss punctual and limitative expressions with different nouns like day, month and year vs night and the seasons. These sections are again impressively devoted in detailed discussion of standard pattern and counter-examples that lead G. to propose that ‘perhaps the line between the punctual and the modal should be redrawn’ (pp. 94–5). He then moves on to discuss the noun chronos ‘time’ and its unique behaviour before moving on to distributive and approximative expressions. The overall conclusion from Chapter 2 is that while things are not as clear-cut as one might like, ‘we must set up finer semantic categories to elucidate the nuances conveyed by all different constructions’.

The aim of Chapter 3 is to extend the Thucydidean generalisations to different authors, genres and periods to gain an insight to the diachrony of time expressions. Essentially the chapter involves a side-by-side comparison of the generalisations already set for Thucydides and the use of the various constructions by Xenophon. Despite the ‘relative strangeness of [Xenophon's] syntax’ the Thucydidean case and prepositional usage detailed in Chapter 2 by and large holds for Xenophon as well. Their chief difference is the use of additional constructions owing to the somewhat more extensive set of temporal circumstances that need to be described. The similarities and the differences between the two authors are discussed at length and given that the similarities outweigh the differences, G. decides not to go through the data thoroughly in the other authors but instead focus only where they diverge from the standard observed. He moves on first to other Attic authors (Plato and Demosthenes) and then Herodotus. He then continues with later authors (Polybius, Diodorus, Plutarch and Epictetus) who show that temporal expressions are fairly stable diachronically. Non-literary papyri are also discussed at the end of this chapter, and the biggest issues are presented from the outset: as they do not form a uniform corpus, in terms of author, genre or style, the discussion will be mostly qualitative. Despite the main similarities among authors and texts, a main diachronic trend is established: that of a resurgence of the temporal dative in Roman times.

Chapter 4 moves onto Biblical Greek, focusing on the language of the Septuagint and the New Testament, two crucial but problematic texts. The main aim is to check the diachronic trends established in the previous chapters but also to make sure of the cause of any further trends, as to whether they originate from within Greek or whether they are Hebrew influences. The language of both the New Testament and the Septuagint shows small but significant differences from Attic – some directly related to Hebrew, such as the increase in the frequency of constructions with en, parallel to Hebrew bǝ, and some due to general Greek trends, such as the spread of accusative at the expense of the dative (and genitive) – in accordance with the overall loss of dative from Greek.

Chapter 5 is an interesting excursus backwards into Homeric Greek, in order to establish the diachronic origin of the patterns seen in Classical and post-Classical prose. As the data are scarce, they are therefore consulted in their entirety. Again this chapter involves a detailed discussion of examples with the occasional side-by-side comparison with the canonical uses that we have already seen in Attic prose. Specific Homeric idiosyncrasies are discussed, such as the personification of nyx ‘night’, and its use in the dative. The overall conclusion is that the dative of time and possibly also the genitive were used more freely than in Classical prose.

Finally, in Chapter 6 G. provides the main summary and diachronic trends, not only for classicists but also for linguists. According to G. the system of temporal expressions is remarkably stable, but there are two questions: (a) how the split marking of punctual constructions arises; and (b) how the decline of the dative affects its use in punctual constructions. His ideas relate to the meaning and use of calendrical vs non-calendrical nouns, and the brief dative renaissance, i.e. its increase in use with limitative constructions in Polybius and Diodorus under the influence of Latin.