This is not a book for the faint of heart, nor yet for a tyro in ancient comparative religion and myth. W.'s stated task is to illuminate ‘warrior-in-crisis’ (p. 90) patterns extrapolated from a set of Roman rituals and to suggest (I am not sure how strong this verb should be) their common conceptions and origins amongst ancient texts of Irish, Indic and Armenian sources. A common, primitive Indo-European ancestral source is presupposed pari passu.

In pursuit of his goal, W. musters an enormous quantity of textual, linguistic, archaeological and geographic data. Indeed, it is difficult to keep one's bearings in such a forest of variant material, despite W.'s genial efforts to post multiple signposts and reminders fore and aft of topics covered and topics to come. His footnotes, thankfully, are on the same page as the material footnoted. One comes also to appreciate deeply the numbering and titling of chapter sections, and the thorough table of contents and index.

W.'s Roman focal points are two rituals of early July, the Poplifugia and Nones Caprotinae, although he also gives considerable attention to rites of the Sororium Tigillum and Bona Dea. He maps these using two overlays; first, a ‘primitive Indo-European tripartite ideological conceptualizing’ (p. 263) developed by G. Dumézil and, second, a six-item matrix (crisis of the warrior, flight and a distant locale, feminine eroticism, clairvoyant woman, application of waters, order and inauguration) imported from W.'s earlier work (Indo-European Sacred Space [2006], pp. 211–19). The discussion shifts throughout from Roman to Indic to Irish (and Armenian) comparanda.

I must take the Irish, Caucasian and Oriental materials at W.'s word. The heroes CúChulainn and Indra can be construed to echo classical heroic patterns, and W. marshals a wealth of data piecemeal. But the argument for assimilated common origins of all similar stories falls flat. W. locates his most persuasive Roman ‘reflex’ (or Roman version of the Indo-European hypothetical prototype) in an account of Hercules and Cacus. W.'s general difficulty, however, is that the Roman reflex has so many variants which may or may not be linear. In order to fill the overlays adequately, several Roman warriors (not all of whom are fresh from the trauma of combat, as in the Irish and Indic versions) are needed. Much time and attention is given to Romulus, Camillus and Horatius, whose tales and rituals seem to fit the grid only in certain respects. Occasionally the evidence has to be massaged a good deal to effect alignment.

For example, W.'s exegesis of Livy's account of Horatius (pp. 181–3) puts some weight on the historian's use of ferox to describe the subject. W. observes that the adjective derives from an Indo-European term ‘denoting “beast-eyed”’. Within a paragraph, W.'s analysis of the primitive root of the adjective has become part of Livy's meaning: ‘The feminine gaze – the intense stare of Horatia – directed toward the warrior who has become a menace to society is answered by ferōx, a state characterized by the savage gaze of the wild beast: a scenario of reciprocal gazing is lexically set up by Livy's presentation of the tradition of the raging, dysfunctional warrior and the ocular feminine’ (pp 182–3). This assertion makes at least two petitiones: first, that at this point in Livy's story, prior to the sororicide, Horatius is openly a menace and, second, that Livy somehow, in some way, designed the scene so that the Indo-European root (not the Latin word) of the adjective carries the semantic load.

The texts of Propertius (4.9) and Juvenal (6) are also closely examined and cited as evidence to bring the Heracles–Cacus episode into line with the grid. Propertius' contribution is discussed, inter alia, in the chapter titled ‘Erotic Women and the (Un)Averted Gaze’ (item 3 in the 6-itemed matrix). Propertius presents the priestess of the grove of Bona Dea cautioning Hercules ‘Parce oculis’ (2.9.53) and then she reminds the hero of the price Tiresias paid for seeing what he should not. W. finds ‘The Greek blinding trope is here a synchronic expression of the diachronic averting-of-the-gaze component of the FEMININE EROTICISM feature of the primitive structural matrix – figuring prominently in the Irish reflex of the Indo-European myth – and surfacing elsewhere, as we shall see’ (p. 138). The phrase ‘synchronic expression of the diachronic averting-of-the-gaze component’, albeit clear to cognoscenti, is obscure to me. As an elegiac poet, Propertius' selection of mythic adornment fulfils his poetic aims. How comes it to be eo ipso evidence of primitive tradition? Further, Juvenal 6 is taken by W. as evidence for ancient Roman attitudes towards the rites of the Bona Dea, citing Brouwer (H.H. Brouwer, Bona Dea: the Sources and a Description of the Cult [1989], pp. 369–70) in support. Brouwer's ‘It is possible … something similar to Juvenal's ideas …’ (underlining mine, p. 173 n. 106) becomes for W. a ‘popular notion in Juvenal's time’ (p. 173). Later, discussing the Roman rites of early July, W. asserts Juvenal's picture of the rites as a datum: ‘… no overt expression of feminine nudity, of lewd displays of the feminine body, of the sort that we encounter in the Irish tradition of Cúchulainn and in the Italic tradition of Hercules/Semo Sancus’ (p. 244). Juvenal has many virtues as a satiric poet, but he cannot be interpreted literally without considerable argumentation.

Cavils of textual analysis aside, there is a great deal of information provided by W. that is quite interesting, but the eccentricity of vocabulary and phrasing is a difficulty, for example ‘eventive’ (p. 175), ‘straightforwardly disambiguate’ (p. 175) and possibly ‘indentify’ (p. 243) (in all honesty, I am uncertain whether the last is a typo or merely unusual usage). The patois precludes a generalist's entry into the conversation.

It is clear that many ancient myths and rituals, many of them Indo-European, present trauma as an upshot of combat. Was this an open question? Post-combat angst has been afoot at least since Homer. It does not necessarily follow that independent strands of the tripartite ideology or six-part matrix lead back to an entirely hypostatised Indo-European prototype. Nor need it in order for W.'s tracking of these strands to be heuristic. There is useful material in this book, whether or not the épée strikes home.