In Harold Bloom's obituary in the New York Times, Adam Begley asked the question that was behind Bloom's sponsorship of literary canons, a question that avid readers have asked themselves over the centuries ‘What, in the little time we have, shall we read?’

Ivan Matijašić (henceforth IM) in this well-researched expansion of his Italian PhD thesis, deals with the formation and development of ancient Greek historiography. IM approaches it from an ancient rhetorical tradition as historiography was then regarded and judged as a branch of rhetoric, with most texts and fragments surviving as they became models to be copied and emulated by schoolchildren.

IM starts with a definition of the word canon. The word itself has religious connotations (‘rule’) and it was only in the 17th century that it started to be used in the sense of a list of books by the best authors in a given literary genre. IM prefers canons in the plural, meaning ‘the variety of selections by different individuals for diverse purposes’, as it encapsulates the paradigmatic nature of canons in that they are authoritative and prescriptive but also open, that is, bound to change with the needs and tastes of each era. Pinning down the definition of canon is not the only problem IM encounters, as the majority of non-canonical works and even a good number of canonical ones have not survived through the medieval tradition. This is particularly true of historians of the Hellenistic period. For this reason, IM had to rely on literary criticism, on what ancient rhetors and school teachers said about the canons of ancient Greek historiography.

The supremacy of rhetoric being therefore unavoidable, it is fitting that IM's starting point is Quintilian's influential Institutes of Oratory and Cicero's mentions of Greek historians in his letters, philosophical works and principally in his rhetorical treatises (On the Orator, Brutus and Orator). Although Cicero never wrote history, his deference to historiography is clear as he called it magistra vitae. On the top of Cicero's list, unsurprisingly, are the two authors IM calls hyper-canonical, Herodotus and Thucydides, but Cicero also refers to Philistus, Ephorus and Theopompus, his judgement being inspired by their ‘excellence in style’. A notable absence is Xenophon, seen more as a princeps philosophorum than a historian by both Cicero and Quintilian.

Chapters 3 and 4 are dedicated to the canonical lists of Dionysus of Halicarnassus (with a detour to Isocrates in the fourth century BC, justified by his presence in the works of Dionysus as one of the first to attempt to write a canon), which are the same as Cicero's except for two details – Dionysus does include Xenophon in the canons but is, a fact that might be astonishing to ancient and modern historiographers alike, very critical of Thucydides whom he considers inferior to Herodotus in choice of subject matter, style and narrative. Thucydides’ style obfuscates and his narrative requires an interpreter, says Dionysus. IM cites Thomas Hobbes in his dismantling of Dionysus’ conclusions – Dionysus was being patriotic, having come from Halicarnassus, birthplace of Herodotus; but most importantly, Dionysus was trying to establish his own authority and to promote his own work. In this, for those of us who have studied Herodotus, he is very similar to his Halicarnassian predecessor who famously set out to establish his Histories over the works of Homer.

Chapter 5 is where IM struggles in sustaining an argument the most, due to a lack of Hellenistic sources. The only conclusion to be drawn from this chapter is the unquestionable supremacy of Herodotus and Thucydides. Chapter 6 is an extension of the former chapter, but more interesting and fruitful as IM analyses the Greek historiographical canons from lesser known rhetorical treatises and the progymnasmata (school texts, which served as rhetorical exercises). These literary and papyrological sources from the Roman Imperial Age have contributed decisively to the shaping of the canons of Greek historiography, most authoritatively via Hermogenes’ On the Categories of Style, in which the second century AD rhetorician endorsed the imitation of Herodotus’ ‘fabulous narrative’ and ‘use of poetic language’, Thucydides’ ‘powerful and solemn style’ and Hecataeus of Miletus’ ‘plain, yet genuine and sweet style’ and the avoidance of Theopompus, Ephorus, Hellanicus and Philistus.

Chapter 7 discusses how diverse authors reacted to the canons of previous ages, and how each canon influenced the next only to a certain extent, each new list having its own purposes and objectives and, in this way, shaping the canons themselves. IM considers Dio Chrysostom, Roman emperors of the fourth century AD, such as Julian the Apostate, and important literary figures such as the eminent grammaticus and rhetor Ausonius and Saint Jerome. IM also analyses papyrological documents from Imperial Egypt and lists of authors preserved in a small number of Byzantine manuscripts. This period is crucial as the move from papyrus rolls to manuscripts in the Christian Era doomed many texts to forgetfulness. Religious zeal led Ammianus Marcellinus to say that imperial libraries were ‘being shut like tombs’. The library of Alexandria was perhaps the most illustrious victim of that age, but texts had to contend with natural disasters, predators like rats and bookworms, continued use and lack of interest and investment. Preservation was costly, requiring the services of a scribe, materials and storage space. This range of factors is perhaps why even canonical texts have not stood the test of time or survived only in fragments.

The conclusions are much more tentative than the meticulous treatment IM has given to his sources, but I would disagree with other reviewers in that I do not find IM's research at any point tedious, as IM acknowledges he ‘does not aspire to completeness’ but to stimulate ‘further research’ (p.6). This book is quite academic, and is therefore more informative to teachers of Ancient Greek and scholars interested in Greek historiography than secondary school students. The thoroughness and span of IM's research was valuable reading, particularly in contextualising the different canons. In the end, what is constant in the Ancient Greek Historiographical canons in this book and elsewhere is the supremacy of Herodotus and Thucydides, embodied in the janiform (looking both ways) herm now in the Naples Archaeological Museum, but originally discovered in Hadrian's villa. These two historians are not just the two authors that have survived to represent historians in Ancient Greek literature, but they are themselves part of the Western literary canon at large.