¹ Har. 20; to judge from Cicero's gloss on the sound later in the same chapter (‘horribilis fremitus armorum’), this was a clattering of arms; cf. the ‘fremitus armorum’ said to have been heard emerging from the depths of the earth in 100 b.c., Obsequens 45.

² Har. 9. For a recent discussion, see Hales, S., ‘At home with Cicero’, Greece and Rome 47 (2000), 44–55CrossRefGoogle Scholar.

³ The singular and plural forms are used interchangeably. In casual references to the speech, Quintilian (5.11.42) uses the plural, Asconius (70 C) the singular. The manuscript tradition uses the plural; I have followed Asconius and many recent writers.

⁴ The controversy is reviewed by Lenaghan, Har., 38–41; and, in reference to all four speeches, by Nisbet, R. G., M. Tulli Ciceronis, De Domo Sua ad Pontifices Oratio (1939), xxix–xxxivGoogle Scholar.

⁵ Corbeill, A., ‘The function of a divinely inspired text in Cicero's De haruspicum responsis’, in Berry, D. H. and Erskine, A. (eds), Form and Function in Roman Oratory (2010), 139–54Google Scholar; Gildenhard, I., Creative Eloquence: the Construction of Reality in Cicero's Speeches (2011), 326–43 (quote, p. 1)Google Scholar.

⁶ John North's work is a notable exception in focusing explicitly on the divinatory issues raised by the speech, in comparison with other evidence: especially ‘Diviners and divination at Rome’, in Beard, M. and North, J. (eds), Pagan Priests (1990), 51–71Google Scholar and ‘Prophet and text in the third century b.c.’, in Bispham, E. and Smith, C. (eds), Religion in Archaic and Republican Rome and Italy (2000), 92–107Google Scholar. I shall return to his arguments below. Otherwise, even lengthy modern accounts usually devote only a few — and those largely descriptive — pages to the speech; among the most significant of these are Rosenberger, V., Gezähmte Götter; das Prodigienwesen der römischen Republik (1998), 64, 80–1Google Scholar; Rasmussen, S. W., Public Portents in Republican Rome (2003), 186–91Google Scholar; Engels, D., Das römische Vorzeichenwesen (753–27 v. Chr.): Quellen, Terminologie, Kommentar, historische Entwicklung (2007), 639–42Google Scholar. The phrase ‘prodigy and expiation’ alludes to B. MacBain, Prodigy and Expiation: a Study in Religion and Politics in Republican Rome, Coll. Lat. 177 (1982), an influential discussion of prodigies — in which De Haruspicum Responso hardly figures.

⁷ Particularly important are, Platt, V., Facing the Gods: Epiphany and Representation in Graeco-Roman Art, Literature and Religion (2011)Google Scholar and Wiseman, T. P., ‘Visible gods, audible gods: epiphany and the Romans’, in Petridou, G. and Platt, V. (eds), Epiphany. Envisioning the Divine in the Ancient World (forthcoming)Google Scholar; I am very grateful to Peter Wiseman for showing me an advance copy of this article.

⁸ Other aspects of the intersection of Ciceronian philosophy and oratory are explored in Gildenhard, op. cit. (n. 5).

⁹ Dio 39.20. In a recent paper (as yet unpublished) John North has ingeniously attempted to reconcile the two accounts, rightly pointing out that the prodigies and their disputed interpretation are only incidental to Dio's main agenda in this section of his narrative. In my view, however, many difficulties remain. It is uncertain, for example, whether the speech of Clodius attacking Cicero noted by Dio is the same as that mentioned by Cicero himself. In Dio, Clodius’ speech appears to have taken place after the series of prodigies that Cicero does not refer to. Does Cicero's silence indicate that those prodigies had not taken place, or been reported, at the time of the delivery of De Haruspicum Responso — so suggesting (at least) two different speeches by Clodius? Or, are Dio's prodigies in fact those briefly mentioned at Har. 62 (‘an earthquake at Potentia and other fearful events not yet officially reported’)? We do not know. But, on any reconstruction, Cicero cannot be Dio's source; and no prodigies are recorded for this year by Obsequens.

¹⁰ Courtney, E. J., ‘The date of the De haruspicum responso’, Philologus 107 (1963), 155–6CrossRefGoogle Scholar; Wuilleumier, P. and Tupet, A.-M. (eds), Cicéron, Discours XIII, 2, Sur la réponse des haruspices (1966), 8–10Google Scholar; Lenaghan, Har., 22–8; T. P. Wiseman, Cinna the Poet (1974), 159–69, especially 162–6; R. Seager, Pompey the Great (2nd edn, 2002), 228. The fact that Cicero seems to address Cn. Cornelius Lentulus Marcellinus as presiding consul suggests that it was delivered in an ‘odd’ month of the year.

¹¹ V. Rosenberger, in Thesaurus Cultus et Rituum Antiquorum III (2005), 85–8, offers one such succinct, up-to-date account; likewise in ‘Republican nobiles: controlling the Res Publica’, in Rüpke, J. (ed.), A Companion to Roman Religion (2011), 292–303Google Scholar, especially 293. Yet more briefly, Beard, M., North, J. and Price, S., Religions of Rome (1998), vol. 1, 37Google Scholar.

¹² Events at the very beginning of 56 b.c. might suggest that the religious process could on occasion move quickly. If we trust Dio (39.15.1), a thunderbolt struck the statue of Jupiter on the Alban Mount ‘at the beginning of the year’; the Sibylline oracles had been consulted and a (supposedly fake) response produced by 13 January, that is, within two weeks (Cicero, Fam. 1.1). But this was in the hands of the XVviri in Rome itself, and did not involve any external consultation with the Etruscan haruspices.

¹³ On the possibility of specialist prodigy collections, see Rawson, E. D., ‘Prodigy lists and the use of the Annales Maximi’, Classical Quarterly 31 (1971), 158–69 (Reate, p. 164)CrossRefGoogle Scholar; reprinted in Roman Culture and Society (1991), 1–15; contested by MacBain, op. cit. (n. 6), 14–15, with further discussion by North, J., ‘Religion and politics: from Republic to Principate’, Journal of Roman Studies 76 (1986), 251–8CrossRefGoogle Scholar, especially 255.

¹⁴ Har. 20.

¹⁵ See, e.g., Livy 24.10; 32.1; 36.37.

¹⁶ North, op. cit. (n. 6, 2000), especially 93. For further discussion of Livy's literary treatment of prodigies, and the dangers of an uncritical reading of his accounts of chronology and procedure, see D. S. Levene, Religion in Livy (1993), especially 35–6; 62–9; 95–101; 104–17; Beard, North and Price, op. cit. (n. 11), vol. 1, 38–9; Davies, J. P., Rome's Religious History: Livy, Tacitus and Ammianus on their Gods (2004), 27–58Google Scholar.

¹⁷ North, op. cit. (n. 6, 2000), 94. Similarly Corbeill, op. cit. (n. 5), 141, claims that ‘most, if not all <of the response> can be reconstructed from his extant oration’; Rasmussen, op. cit. (n. 6), 190, only goes so far as to suggest that ‘the essence’ of the response ‘remains intact’.

¹⁸ Wissowa, G., Religion und Kultus der Römer (1902), 471Google Scholar.

¹⁹ John Lydus, De Ostentis 30.

²⁰ Har. 40.

²¹ Corbeill, op. cit. (n. 5), 143–4.

²² Har. 36; Clodius too is represented as reading the response from a written text, Har. 9.

²³ Har. 63; 62.

²⁴ Ando, C., The Matter of the Gods: Religion and the Roman Empire (2008), 125Google Scholar; the Roman gods — in historical time at least — are only slightly more voluble for Scheid, J., ‘La parole des dieux. L'originalité du dialogue des Romains avec leurs dieux’, Opus 6–7 (1987–89), 125–36Google Scholar and Dubourdieu, A., ‘Paroles des dieux’, in Dupont, F. (ed.), Paroles romaines (nd <1995>), 45–51),+45–51>Google Scholar.

²⁵ Dionysius of Halicarnassus, Rom.Ant. 8.55–6; Valerius Maximus 1.8.4; Plutarch, Cor. 37–8; see Schultz, C. E., Woman's Religious Activity in the Roman Republic (2006), 38–44Google Scholar. Wiseman, op. cit. (n. 7), offers useful observations on this and several of the divine voices that I discuss here.

²⁶ This account is drawn from Livy 5.32; 5.50; 5.52; a similar version is found in Cicero, Div. 1.101; 2.69. Also underlying this account is a clash between the individual non-élite citizen and the religious authorities of the state; the man who reported the voice, Marcus Caedicius (‘Disaster Teller’) was ignored in part because of his lowly rank (Beard, North and Price, op. cit. (n. 11), vol. 2, 42–3).

²⁷ Cicero, Div. 1.101; 2.69; Valerius Maximus 1.8.3.

²⁸ Obsequens 41; 45; 46; 48; 53; 59; Livy 31.12; 29.14 (‘cum horrendo fragore’); 24.44. As Greg Woolf reminded me, the portents on the death of Caesar evoked by Virgil at the end of the first Georgic include some striking noises: ‘armorum sonitum toto Germania caelo/ audiit’ (ll. 474–5); ‘vox quoque per lucos vulgo exaudita silentis/ ingens’ (ll. 476–7).

²⁹ Livy 35.21; 24.10.

³⁰ Such is my crude summary of Benveniste, E., Indo-European Language and Society (1973), 512–13Google Scholar.

³¹ Div. 2.69.

³² Livy 5.47; Dionysius of Halicarnassus, Rom.Ant. 13.7–8. The myth and history of this incident is discussed by Horsfall, N., ‘From history to legend: M. Manlius and the geese’, Classical Journal 76 (1979–80), 298–311Google Scholar; Ziolkowski, A., ‘Between geese and the auguraculum: the origin of the cult of Juno Moneta on the Arx’, Classical Philology 88 (1993), 206–19CrossRefGoogle Scholar.

³³ Rom.Ant. 8.56.3.

³⁴ op. cit. (n. 7).

³⁵ Rom.Ant. 8.56.1.

³⁶ It is treated straightforwardly as a prodigy by Engels, op. cit. (n. 6), 719.

³⁷ e.g. Livy 35.21; 41.13; Obsequens 27.

³⁸ Burkert, W., ‘Signs, commands and knowledge: ancient divination between enigma and epiphany’, in Iles Johnston, S. and Struck, P. T., Mantikê: Studies in Ancient Divination (2005), 29–49CrossRefGoogle Scholar (quote, p. 36); Platt, op. cit. (n. 7), 10.

³⁹ op. cit. (n. 7), 9.

⁴⁰ Har. 62.

⁴¹ Har. 63; 10. Similarly Har. 25: ‘Pro di immortales! qui magis nobiscum loqui possetis, si essetis vorsareminique nobiscum? ludos esse pollutos significastis ac plane dicitis.’

⁴² This (otherwise unattested) incident and its chronology is discussed by Lenaghan, Har., 114–17 and Wiseman, op. cit. (n. 10), who cuts Cicero's rhetoric down to size, and dissects the precise topography of these events. See also Salzman, M. R., ‘Cicero, the Megalenses, and the defense of Caelius’, American Journal of Philology 103 (1982), 299–304CrossRefGoogle Scholar, especially 303–4; Tatum, W. J., The Patrician Tribune: Publius Clodius Pulcher (1999), 211–12Google Scholar. Despite some confident claims to the contrary, the exact timetable of the Megalesia at this period is unknown.

⁴³ Har. 24.

⁴⁴ Har. 37

⁴⁵ Or, as Beard, North and Price stress (op. cit. (n. 11), vol. 1, 138), ‘when both Clodius and Cicero claimed as correct their own, partisan, interpretation of the prodigy, each was effectively attempting to establish his own position as the privileged interpreter of the will of the gods’.

⁴⁶ Har. 20. Later (Har. 56) he similarly appears to dispense with the intermediary rôle of the haruspices, suggesting that, grateful as they must be for the warning given, the populus Romanus has already taken action against Clodius ‘sua sponte’; though contrast Har. 61, where the warnings of haruspices supplement or replace human debate.

⁴⁷ Corbeill, op. cit. (n. 5), 151; Gildenhard, op. cit. (n. 5), 338.

⁴⁸ Har. 23; my argument here is indebted to Corbeill, op. cit. (n. 5), 152.

⁴⁹ Har. 25–6

⁵⁰ Gildenhard, op. cit. (n. 5), 327.

⁵¹ J. Rüpke, Religion of the Romans (2007), 228.

⁵² Engels, op. cit. (n. 6), 43–7. On different sides of this complex debate, see (e.g.): Rosenberger, op. cit. (n. 11, 2011), 296 (‘Prodigies did not foretell future disaster’); likewise Scheid, op. cit. (n. 24), 127 (on Cicero's intention (at Div. 2.71ff.) ‘de montrer que les auspices ne servent pas à prédire l'avenir’); conversely, Linderski, J., ‘Cicero and Roman divination’, La Parola del Passato 37 (1982–83), 12–38Google Scholar, repr. in Linderski, Roman Questions (1995), 458–84 (auspicia as ‘qualified prediction’); for Levene, op. cit. (n. 16), 4–5 ‘the events they <sc. prodigies> foretold were only conditional’ — though he sees the prodigy discussed in Har. as an exception, alluding ‘to an inevitable future’. Wallace-Hadrill, A., Rome's Cultural Revolution (2008), 251–2Google Scholar succinctly summarizes the argument that predictive elements became more important in the first century b.c.; but see the careful qualifications of North, op. cit. (n. 6, 2000). For ancient discussion, note Cicero's definition of divination at Div. 1.5.9 as praedictio atque praesensio ‘of those things which are thought to happen by chance’ (apparently equating Roman divination with Greek mantikê); and Festus 254L.

⁵³ Har. 29.

⁵⁴ Lenaghan, Har., 139

⁵⁵ M. Tulli Ciceronis quae vulgo feruntur Orationes Quattuor (1801), 349.

⁵⁶ Courtney, E., ‘Notes on Cicero’, Classical Review ns 10 (1960), 95–9, at 97CrossRefGoogle Scholar.

⁵⁷ This is the implication of, for example, Wiseman's chronology of events in 56 b.c., op. cit. (n. 10).

⁵⁸ Rasmussen, op. cit. (n. 6), 241–4 (quote, p. 244).

⁵⁹ Har. 4.

⁶⁰ See, e.g., Cicero, Prov. Cons 2; Sest. 38.

⁶¹ Har. 7. The noun vox occurs fifteen times in the course of this short speech.

⁶² Har. 17–18.

⁶³ Har. 1–2.

⁶⁴ Har. 61; for commovere or movere, in the context of the response, see also Har. 18; 25; 31.

⁶⁵ Har. 18.

⁶⁶ op. cit. (n. 24), 128; this point is broadly reflected by Lehoux, D., What Did the Romans Know? An Inquiry into Science and Worldmaking (2012), 37CrossRefGoogle Scholar.

⁶⁷ e.g. Div. 2.12 (prophecy vs conjecture); 2.54 (the necessity of interpretation of divine warnings).

⁶⁸ I am referring here even to some of the most sophisticated recent studies. For example, Feeney's, D. stress on ‘brain-balkanization’ — in Literature and Religion at Rome: Cultures, Contexts and Beliefs (1998)Google Scholar, drawing on Veyne's, P. discussion of ‘balkanisation des cerveaux’ in Les Grecs ont-ils cru à leurs mythes? (1983)CrossRefGoogle Scholar — has been very influential; and usefully so. But, with its stress, as Feeney puts it (p. 14), on the Greek and Roman capacity ‘to entertain different kinds of assent and criteria of judgement in different contexts, in ways that strike the modern observer as mutually contradictory’, this approach necessarily focuses on what appears different between Roman philosophical and other discursive practices — at the expense of what they share. I now regret that in ‘Cicero and divination: the formation of a Latin discourse’, Journal of Roman Studies 76 (1986), 33–46CrossRefGoogle Scholar, I was so concerned to challenge the view of Cicero's out-and-out philosophical scepticism that I did not also explicitly challenge the radical separation that is taken for granted between Cicero's philosophical and his more ‘public’ work (though as Lehoux, op. cit. (n. 66), 35 detects, some of my suspicions of that separation are occasionally visible under the surface). I still believe that my main challenge in that article was timely and well-made; but it would have been all the more powerful if it had explored in greater detail the important intellectual links between De Divinatione and the speeches.

⁶⁹ op. cit. (n. 5), 326–43.

⁷⁰ op. cit. (n. 5), 384, 389.

⁷¹ Ad Att. 7.1.8; M. Beard, The Roman Triumph (2007), 191.

⁷² Death and Renewal (1983), 41.

⁷³ Especially notable contributions include Moatti, C., La Raison à Rome: naissance de l'esprit critique à la fin de la République (1997)CrossRefGoogle Scholar and Wallace-Hadrill, op. cit. (n. 52).

⁷⁴ Lehoux, op. cit. (n. 66), seems to be moving in a similar direction to my own, albeit from a different intellectual starting point. I am struck by his emphasis on Roman dealing with the gods as an important aspect (and driver) of Roman ‘science’ (p. 12), and by his emphasis on the discursive intersections between Roman discussions of politics, law and nature (pp. 18, 21).