¹ In civilian life, the seriousness with which feasts were observed varied. Some Romans and (though this is not explicitly attested) possibly some Greeks were appalled at the laziness of Jews for having a weekly rest (cf. Seneca, , De Superstitioe, ap. 08 De Civ. Dei 6.11Google Scholar; Tac. Hist. 5.4), but Macrobius, (Sat. 1.16.9)Google Scholar none the less states that Roman religious feasts too were meant to be observed by rest. A fine had to be paid if any Roman was seen by a priest doing work on a rest day. Xenophon claims that no Athenian would dream of doing anything important on the Πλυντήρια (Hell. 1.4.12). Perhaps the degree of respect paid to such occasions was sometimes a rather personal matter, as with Christian days like Good Friday in Western Europe today. Xenophon was a notably pious man and had an axe to grind in stressing the importance of the Πλυντήρια. Alcibiades and his friends obviously thought that it would not do him any harm to return on that day and, as it turned out, they were right. The passions of the Hermokopid affair were not reawakened on this occasion, any more than in 412/11, although the Eumolpidae and Kerykes raised objections at first (Thuc. 7.53.2).

Parke, H. W., Festivals of the Athenians (London, 1977), 190Google Scholar, estimates the normal Athenian attitude at a rather low level. Macrobius hints at a Roman means of avoiding private inconvenience when he notes that according to Scaevola work was permitted whenever it was necessary to avoid loss (Sat. 1.16.11): public dies nefasti were by contrast carefully observed by the cessation of official public business. Although Jewish attitudes towards labour performed on the Sabbath were altogether stricter, they too varied considerably, as is easily seen from a comparison of the rabbinic laws preserved in the Mishnah (Shabbat 7.2, and passim) with the sectarian customs of the Essenes (cf. Jos. B.J. 2.147).

² Comparable scruples undoubtedly manifested themselves in other religions and societies. For example, the Hittite King Mursilus II returned to his capital in the middle of a war to celebrate the purulliyas festival (Gurney, O. R., The Hittites (Harmondsworth, 1953) 152Google Scholar). But the Greeks, Romans and Jews provide by far the richest material.

The section on Greece is primarily the work of A.J.H., that on Rome and the Jews of M.D.G.

³ Pritchett, W. K., The Greek State at War (California, 1979), iii. 185Google Scholar, notes that he can only find two instances of any attempt to organise celebration in the field — and of these one is unclear and the other is the work of Alexander the Great, who was something of a law unto himself: on another occasion he altered the calendar to postpone an unlucky month when launching a campaign (Plut. Alex. 16).

⁴ For the dates and duration of the various festivals, so far as they are known, see Popp, H., Die Einwirkung von Vorzeichen, Opfern und Festen auf die Kriegführung der Griechen im 5. und 4. Jahrhundert v. Chr. (Erlangen Dissertations, 1957), 75–144Google Scholar.

⁵ Thuc. 3.15.2; also Mardonius in Hdt. 7.9 (2) and 9.48.

⁶ E.g. Elis and other states in 480 b.c. (Hdt. 7.206). Their lack of enthusiasm for fighting north of the Isthmus in 479 b.c. weakens Popp's case for believing their sincerity (op. cit. 127). Cf. also Thuc. 5.54.4 and 8.9–10 for other cases.

⁷ Xen. Hell. 4.7.2.

⁸ Thuc. 5.54.

⁹ Thuc. 3.3 and 3.56; Xen. Hell. 5.2.29. How far purely local festivals were meant to be respected by outsiders is not clear.

¹⁰ Thuc. 5.49. After the Olympic truce was proclaimed in Elis there would be a lapse of time before heralds could proclaim it in other cities of Greece. This was the case in 420 b.c., when the Spartans were accused of violation and defended themselves by pointing out that the heralds from Elis had not yet made their proclamation in Sparta and asking why, if a proclamation in Elis was the relevant, and sufficient, one, the heralds came to proclaim it also in Sparta. They also could claim that their action was not warlike. See A. Andrewes in A. W. Gomme. H.C.T. ad loc.

¹¹ Thuc. 5.50.

¹² Thuc. 1.112. It was proclaimed as a Sacred War (the Second). When Delphi regained independence it gave strong support to Sparta (Thuc. 1.118.3).

¹³ Thuc. 1.121 and 143.

¹⁴ Xen. Hell. 4.5.1.

¹⁵ In 398 b.c. King Agis of Sparta went to Olympia to offer sacrifice; he had previously been banned from this, but this time no-one prevented him. In the settlement with Elis which followed, Sparta did not deprive her of the presidency of the Games, although Pisa was claiming it (Xen. Hell. 3.2.26 and 31). So Sparta's piety stood up well to strain.

¹⁶ Xen. Hell. 7.4.28–32.

¹⁷ Diod. 16.23.

¹⁸ Dem. 19.25.

¹⁹ Paus. 3.10.3.

²⁰ This reputation is endorsed, among contemporary historians, explicitly by Herodotus and Xenophon, the believers, and implicitly by Thucydides, the sceptic (cf. 5.54). As a conscientious historian he could not omit oracles and omens when men acted under their influence. See also Plato, Alc. 2.148d–149c and Cicero, , De Div. 1.95Google Scholar.

²¹ King Demaratus was deposed at the order of the Pythia even though he was popular (Hdt. 6.65–7; Paus. 3.4.4). Kings Agesilaus and Pleistoanax were affected by its pronouncements (Xen. Hell. 8.3.1; Thuc. 5.16). A revealing story about the different attitudes of Sparta and Athens concerns the murder of the Persian envoys before 490 b.c. The Spartans regretted it and sent two noblemen as scapegoats to appease the wrath of Talthybius. The Athenians betray no such scruples.

²² Op. cit. in n. 4.

²³ Thuc. 7.42.

²⁴ Xen. Anab. 6.1.24.

²⁵ Xen. Anab. 6.4.19–25.

²⁶ Greek generals were empowered to take decisions, if need be, contrary to the omens (Pritchett, op. cit. in n. 3, 48–9). We hear of a Spartan commander, Anaxibios, who did so and suffered defeat (Xen. Hell. 4.8.36). If a Spartan could do this we can be sure that other Greeks did. Of course, we are unlikely to be told of this in cases where generals ignored omens and got away with it. Generals might also ‘edit’ omens as the Athenian general Thrasyllus did before the battle of Arginusae (Diod. 13.97.5–7). The Spartan King Agesipolis ‘interpreted’ a portent in 388 b.c.

It is possible that the Ten Thousand, in their isolation and peril, and with the pious Xenophon among them, were very susceptible like the Athenians at Syracuse.

²⁷ Hdt. 9.10.

²⁸ Paus. 3.5.8.

²⁹ Rationalisations of this delay are common and denounced by Popp (op. cit. n. 4, 49) and Pritchett (op. cit. in n. 3, 78–9). Burn, A. R., Persia and the Greeks (London, 1962)Google Scholar, professes the same view in general (pp. 262, 508) but deserts his principle in this case.

³⁰ But it should be noted that the Syracusans delayed their pursuit of the retreating Athenians in order to celebrate a feast (Thuc. 7.73.2) and the Corinthians also delayed their fleet for the Games (Thuc. 8.9.1). But in neither case was action urgent.

³¹ Xen. Hell. 4.7.2.

³² The withdrawal of Spartan troops from Messenia under truce during wartime in order to celebrate the Hyacinthia (Paus. 4.19.4) cannot be confidently treated as historical, though it may reflect a real event and certainly reflects Greek views of the Spartan pattern of behaviour. Cf. the real event of 390 b.c. (Xen. Hell. 4.5.11), when, however, no truce was made.

³³ Hdt. 6.106 and 120.

³⁴ It would, of course, be quite circular, and therefore illegitimate, to use the Spartan delay as an argument for the existence of such wars. It is necessary to provide good independent evidence for them and this has not been done. The Helot War of c. 490 b.c., in which some believed, has been dealt with by Wade-Gery, H. T. in Studies in Ancient Greek Society and Institutions (Oxford, 1966), 289–302Google Scholar: cf. also GHI, p. 47, no. 22. The coin evidence used by Wallace, J. F. in JHS 74 (1954), 32–5CrossRefGoogle Scholar cannot be dated with sufficient accuracy to prove his case. It is true that Sparta received no support from her Arcadian allies in her (belated) Marathon expedition but nor did she from Corinth, Elis and her other allies, by whom no revolt is alleged.

³⁵ Hdt. 7.206.

³⁶ Hdt. 8.26.

³⁷ Hdt. 7.207, 8.40 and 74.

³⁸ If Sparta had decided that more troops were needed at Thermopylae, and increased her contingent, she would no doubt have pressed her allies to do the same, arguing that the need of Greece, and their oaths, should over-rule the normal truce, which was in any case between Greeks and had no relevance to a foreign threat. If the Persians had already been threatening the Isthmus nothing would have been said about the Olympic truce. The Eleians in 364 b.c. were clearly thought to be right in breaking the truce against violators, since Xenophon credits them with divine inspiration (cf. n. 16) and even the Arcadians restored their presidency of the Games (Hell. 7.4.34–5).

³⁹ Op. cit. in n. 3 above, vol. 1, p. 120.

⁴⁰ Hdt. 9.7.

⁴¹ Hdt. 9.8–10.

⁴² Hdt. 9.28. The departure by night can hardly have been meant to surprise the Argives (Popp, op. cit. 112–13) since the dates of the Festival would be known. It must have been a dramatic gesture.

⁴³ When news of the occupation of Pylos in 425 b.c. reached Sparta we are told that owing to a festival they delayed action – but their main army was already away in Attica, so it is not likely to have been a major feast (Thuc. 4.5).

⁴⁴ The garrison was rotated between Sparta's allies (Thuc. 7.27) but the King was probably there all the time as Thucydides implies in this passage, and he would need a small force of Spartiates, possibly the King's Three Hundred (for which see Hdt. 7.205), to provide stiffening and control.

⁴⁵ Xen. Hell. 5.29.

⁴⁶ Xen. Hell. 4.5.11–17. Different details are given in other accounts but none record a smaller disaster.

⁴⁷ Xen. Hell. 6.4.16.

⁴⁸ On the whole question of Roman attitudes to warfare, see Harris, W. V., War and Imperialism in Republican Rome (Oxford, 1979)Google Scholar.

⁴⁹ See, e.g., Bonniec, H. Le, ‘Aspects religieux de la guerre à Rome’, in Brisson, J.-P., ed., Problèmes de la guerre à Rome (Paris, 1969), 101–3Google Scholar.

⁵⁰ Macrob. Sat. 1.16.16.

⁵¹ Macrob. Sat. 1.16.21, 24. Cf. in general on all these days Michels, A. K., The Calendar of the Roman Republic (Princeton, 1967), 62–6Google Scholar; Nock, A. D., ‘The Roman army and the Roman religious year’, Harvard Theological Review 45 (1952), 190–2CrossRefGoogle Scholar. Nock suggests that the Volcanalia, in early Autumn, should be added to the list.

⁵² CIL I² (1893), 296. Most of these days are dies comitiales. It is possible that the Fasti Antiates mark 6 October as nefastus because this was a black day (see below, n. 58). If so, this was a mistake, since the other fasti do not agree. Cf. Michels, op. cit. 66, 133.

⁵³ Plut. Camillus 19.7–8. Michels, op. cit. 25 notes that this day is marked in the Fasti Antiates Maiores. The ban on all military activity in March implied in Suet. Otho 8 and Tac. Hist. 1.89 probably refers only to a taboo specific to the Salii; cf. Balsdon, J. P. V. D., ‘The Salii and campaigning in March and October’, CR 16 (1966), 146–7Google Scholar. See below, p. 164.

⁵⁴ Dion. Hal. 4.49. See the brief account in Fowler, W. W., The Roman Festivals of the Period of the Republic (Oxford, 1899), 95–7Google Scholar; Garlan, Y., transl. Lloyd, Janet, War in the Ancient World (London, 1975), 50–1Google Scholar.

⁵⁵ Cf. Alföldi, A., Early Rome and the Latins (Ann Arbor, 1963), 29–34Google Scholar. Alföldi's denial of early Roman control of the rituals need not be accepted.

⁵⁶ Note the complaints against Flaminius for his failure to carry out the Feriae Latinae before taking over his command in 217 b.c. (Livy 21.63.5); cf. also Livy 25.12.1–2 about 212 b.c.

⁵⁷ Alföldi, op. cit. 32; CIL 1² (1893), 58.

⁵⁸ Plut. Lucullus 27.7.

⁵⁹ Appian, Mithrid. 12.85 gives a different account, but Eckhardt, K., ‘Die armenischen Feldzüge des Lukullus’ Klio 10 (1910), 96–111Google Scholar, prefers Plutarch's version, which he believes to be derived from Sallust.

⁶⁰ Quoted by Macrob. Sat. 1.16.26; Aul. Gell. Noct. Att. 5.17.3–5.

⁶¹ Macrob. Sat. 1.16.26;. When discussing Cannae at Sat. 1.16.26, Macrobius does not suggest that the date was unpropitious because it was a dies postriduanus but because it fell four days before the Nones.

⁶² Livy 22.10.6 records the explicit mention of such days in the words of the people's vow taken in 217 b.c.

⁶³ Macrob. Sat. 1.16.22–7; cf. Varro, , De L.L. 6.29Google Scholar, and Livy's own comments in Livy 22.10.6.

⁶⁴ Caven, B., The Punic Wars (London, 1980), 136Google Scholar.

⁶⁵ Livy 22.45ff.; Polybius 3.112.

⁶⁶ Appian, , Hannib. 18Google Scholar.

⁶⁷ Florus 1.12.15–22 may refer to the unpropitious day when he states that dux, terra, caelum, dies, tota verum natura led to the destruction of the Roman army, but dies here probably refers to the bad weather.

⁶⁸ Frontinus, , Strat. 4.7.30Google Scholar; cf. Livy 37.37ff. The day concerned was probably the dies postriduanus 2 May. The battle of Magnesia took place towards the end of December (Julian date); cf. Walbank, F. W., Philip V of Macedon (Cambridge, 1940), 332Google Scholar. This would correspond to the end of April or beginning of May according to the current Roman calendar.

⁶⁹ Cf. Livy 22.20.6 for a deliberate vow that performance of religious rites will not be invalidated if they occur on a black day.

⁷⁰ Macrob. Sat. 1.16.20.

⁷¹ Macrob. Sat. 1.16.7.

⁷² For recent discussions, see Szemler, G. J., The Priests of the Roman Republic: A Study of Interactions between Priesthoods and Magistracies (Brussels, 1972)Google Scholar; Wardman, A., Religion and Statecraft among the Romans (London, 1982), esp. 17–20Google Scholar.

⁷³ See the list of restrictions in Marquardt, J., Römische Staatsverwaltung ² iii.328ff.Google Scholar

⁷⁴ Szemler, op. cit. 166–7; Broughton, T. R. S., MRR i.289Google Scholar.

⁷⁵ Livy 39.45.4.

⁷⁶ Appian, B.C. 1.65. Cf. Szemler, op. cit. 171; Broughton, , MRR ii.52Google Scholar.

⁷⁷ OCD ² s.v. Flamines. Cf. PWRE vi.2486–90 on the gradual relaxation of restrictions.

⁷⁸ Livy 24.8.10; 37.51.1–7; Cic. Phil. 11.8, 18. In the two latter cases the opposition was led by the current pontifex maximus but the political motives behind his action are obscure.

⁷⁹ Harris, op. cit. 37, suggests that Crassus objected to a command in a province too nearly pacified to leave him an opportunity for glory.

⁸⁰ Livy 41.15.9.

⁸¹ Livy 41.15.10; note that Livy records in the same passage an identical oath taken by another praetor, M. Cornelius.

⁸² Livy 42.32.1–3.

⁸³ The speaker in Tac. Ann. 3.58 assumes, probably wrongly, that a failure to carry out statutory sacrifices, rather than any specific taboo, is the only reason for stopping priests from leaving Rome.

⁸⁴ Polybius 21.13.10–14; cf. Scullard, H. H., Scipio Africanus: Soldier and Politician (London, 1970), 205Google Scholar. The account in Livy 37.33.6–7 is a confused version of Polybius; cf. Briscoe, ad loc., who also comments on the date.

⁸⁵ Val. Max. 1.1.9 gives an anecdote to this effect about an otherwise unknown L. Furius Bibaculus, who was praetor and a Salian priest probably in the second century b.c. For imperial examples, see PWRE s.v. ‘Salii’

⁸⁶ Broughton, , MRR i.436Google Scholar. He celebrated a triumph against the will of the Senate by invoking the protection of his daughter, who was a vestal. It seems unlikely that such a man will have broken the taboos of a salius, especially since he took great pride in his expertise at the Salian dances (Macrob. Sat. 3.14.14.).

⁸⁷ For a demonstration that Africanus was not a magistrate, see Scullard, H. H., Roman Politics 220–150 B.C. (Oxford, 1951), 284–5 (Appendix XVI)Google Scholar.

⁸⁸ Helgeland, J., ‘Christians and the Roman army from Marcus Aurelius to Constantine’, ANRW II 23.1 (1979), 724–834Google Scholar denies the influence of pacifist arguments on pre-Constantinian Christians; L. J. Swift, ‘War and the Christian conscience’, ibid. pp. 835–68 asserts it. Both articles have large bibliographies.

⁸⁹ Helgeland, op. cit. 816.

⁹⁰ Orosius 7.37.2. Orosius may exaggerate the degree of the sacrilege in his desire to find a religious justification for the later sack of Rome: admitting that at Pollentia pugnantes vicimus, he asserts that because Easter had been violated victores victi sumus.

⁹¹ It may be significant that Claudian and Prudentius, who wrote full accounts immediately after the battle, do not mention or try to excuse the date. Cf. Claudian, , De Bello Getico 616ff.Google Scholar; Prudentius, , Contra Orationem Symmachi 2.696ff.Google Scholar; Prudentius even asserts (2.745) that hic Christus nobis Deus adfuit et mera virtus. But perhaps they wanted to avoid embarrassing their patrons, Honorius and Stilicho.

⁹² Agatharchides, ap. Jos. c.Ap. 1.209–11; Strabo, , Geog. 16.2.40Google Scholar; Seneca, ap. Augustine, , De Civ. Dei 6.11Google Scholar; Frontinus, , Strat. 2.1.17Google Scholar; Plut. De Superstit. 8, pp. 169. Cf. Vaux, R. De, Ancient Israel: its Life and Institutions (London, 1961), 258–67Google Scholar.

⁹³ I Samuel 23.9f.; 30.7f. Cf. Hengel, M., Die Zeloten ² (Leiden and Cologne, 1976), 277–95Google Scholar.

⁹⁴ The last holy wars occurred in the reign of David according to De Vaux, op. cit. 265; in the time of Saul according to Encyclopedia Judaica viii.347.

⁹⁵ Jeremiah 21.5.

⁹⁶ Hengel, op. cit. 289–93; cf., e.g., the destruction of the altars in Philistine territory described in I Maccabees 5.68. Against Hengel's arguments for the revolt against Rome, see below, note 155.

⁹⁷ De Vaux, op. cit. 262.

⁹⁸ I Samuel 14.9–12.

⁹⁹ I Samuel 23.10–12. God speaks from the ephod through the Urim and Thummim; cf. Numbers 27.21; Deuteronomy 33.8. The way this oracle functioned is obscure, but it was in regular use during the early monarchy; cf. I Samuel 28.6.

¹⁰⁰ Ahab in the ninth century b.c. asked the advice of 400 prophets before going to war (I Kings 22.5–6). Their prophecy was a deliberate ploy by God to entice him to destruction (I Kings 22.20–23).

¹⁰¹ Contrast the prophet who volunteered correct information to Ahab (I Kings 20.13–14, 22, 28) with Elisha's statement to Jehoram that he could refuse to prophesy if he so wished (II Kings 3.13–14).

¹⁰² Jos. c.Ap. 1.41. However, the Hasmonaean John Hyrcanus was so closely in touch with τ⋯ δαιμόνιον that he was never ignorant about the future (Jos. B.J. 1.69).

¹⁰³ Hecataeus, ap. Jos. c.Ap. 1.201–4. As the seer took the auspices, a Jewish archer shot the bird he was observing, arguing that if the bird had been gifted with divination it would not have let itself be killed.

¹⁰⁴ Goodman, M. D., State and Society in Roman Galilee, A.D. 132–212 (Totowa, 1983), 208Google Scholar.

¹⁰⁵ Josephus, A.J. 3.218, claims that the oracle ceased only two hundred years before his time, but the Mishnah, Sotah 9.12, dates its end to the last of the first prophets and no consultation of the oracle is recorded after the Exile (586 b.c.).

¹⁰⁶ Joshua 10.10–14.

¹⁰⁷ Since such inspiration permitted insight into the divine will without the benefit of portents, these latter are naturally rarely recorded, but cf. Amos 4.6ff.

¹⁰⁸ Cf. the portents recorded by Jos. B.J. 6.284–315, with a parallel account, presumably from the same source, in Tac. Hist. 5.13. See the detailed discussion of both passages in Fornaro, P., Flavio Giuseppe, Tacito e L'Impero (Turin, 1980)Google Scholar. On ‘false’ prophets, see Jos. B.J. 6.285–6.

¹⁰⁹ Leviticus 25.4–5; Nehemiah 10.32; Jos. A.J. 14.202.

¹¹⁰ 1 QM 2.8–9; Vermes, G., The Dead Sea Scrolls in English (Harmondsworth, 1962), 126Google Scholar. Cf. Yadin, Y., The Scroll of the War of the Sons of Light against the Sons of Darkness (Oxford, 1962), 20Google Scholar.

¹¹¹ Jos. A.J. 13.230–5; B.J. 1.59–60.

¹¹² North, R., ‘Maccabean Sabbath Years’, Biblica 34 (1953), 501–15Google Scholar. Wacholder, B. Z. (Hebrew Union College Annual 44 (1973), 153ff.)Google Scholar is the most recent author to attempt to reconstruct the cycle of Sabbatical Years in later antiquity. He concludes (pp. 166ff.) that Josephus is impossibly confused about the Sabbatical Year which fell around 37 b.c. when Herod captured Jerusalem (A.J. 14.475; cf. also R. Marcus, Loeb edition, ad loc.). None the less, if his reconstruction is correct the popular rebellion against the Roman census in the autumn of a.d. 6 took place at the beginning of a Sabbatical Year (Jos. A.J. 18.1–6, 26). Against his dates, however, see Bar-Kochva, B., The Battles of the Hasmonaeans: The Times of Judas Maccabaeus (Jerusalem, 1980), 258 (in Hebrew)Google Scholar.

¹¹³ Goodman, op. cit. (n. 104). 102–3.

¹¹⁴ Deuteronomy 16.1–16.

¹¹⁵ Jos. B.J. 2.515–16. Those left behind were massacred by Cestius Gallus. On the extent to which the pilgrimage was observed, see Jeremias, J., Jerusalem in the Time of Jesus (London, 1969), 58–84Google Scholar.

¹¹⁶ Leviticus 23.7–8, 21, 28–32, 35–6. Observation of such rest days outside Jerusalem is attested at Elephantine as early as the fifth century b.c.; cf. Segal, J. B., The Hebrew Passover from the Earliest Times to A.D. 70 (London, 1963), 224Google Scholar.

¹¹⁷ Jos. A.J. 13.252.

¹¹⁸ Jos. A.J. 14.487. Philo, , De Spec. Leg. 1.186Google Scholar testifies to the zeal with which the Day of Atonement was universally observed.

¹¹⁹ Jos. B.J. 2.515, 517.

¹²⁰ Jos. A.J. 13.252. See Rajak, T., ‘Roman intervention in a Seleucid siege of Jerusalem?’, GRBS 22 (1981), 72Google Scholar; Pucci, M., ‘Jewish-Parthian relations in Josephus’, The Jerusalem Cathedra 2 (1983), 13–16Google Scholar.

¹²¹ Smallwood, E. M., The Jews under Roman Rule (Leiden, 1976), 146, 163, 166, 264Google Scholar.

¹²² See the authors quoted above, n. 92.

¹²³ Jos. A.J. 12.277.

¹²⁴ Jos. A.J. 14.63–4.

¹²⁵ Jos. A.J. 14.63; B.J. 1.146.

¹²⁶ For the clearest expression of this view, see Herr, M. D., ‘The problem of war on the Sabbath in the Second Temple and the Talmudic periods’, Tarbiz 36 (1961), 248ff. (in Hebrew)Google Scholar.

¹²⁷ Op. cit. (n. 112), 331–42. The arguments in the rest of this paragraph are his.

¹²⁸ On mercenaries, see especially Kasher, A., The Jews in Hellenistic and Roman Egypt (Tel Aviv, 1978), 46–53 (in Hebrew)Google Scholar; on the building of the walls, cf. Nehemiah 4.3, 12, 15–17; 6.15; Bar-Kochva, op. cit. 333, 429.

¹²⁹ Jos. c.Ap. 1.209–11; for the narrative of events, see A.J. 12.5–6.

¹³⁰ I Maccabees 2.29–37; II Maccabees 6.11.

¹³¹ II Maccabees 5.25–6; 8.25–8; 15.1–5; cf. the detailed discussion by Bar-Kochva, op. cit. 336–41.

¹³² Bar-Kochva, op. cit. 336–7.

¹³³ Nickelsburg, G. W. E., Jewish Literature between the Bible and the Mishnah (London 1981), 78–9Google Scholar.

¹³⁴ Jubilees 50.12–13.

¹³⁵ Nickelsburg, op. cit. 79; De Vaux, op. cit. (n. 92), 483.

¹³⁶ Jos. B.J. 2.391–4.

¹³⁷ Stern, M., Greek and Latin Authors on Jews and Judaism (Jerusalem, 1974), i.511Google Scholar.

¹³⁸ Jos. A.J. 18.322–3. It could perhaps be argued that Asinaeus was ignorant of Jewish custom in Palestine.

¹³⁹ Jos. B.J. 4.99; Titus was Josephus' patron and will have known if the story was true.

¹⁴⁰ Frontinus, , Strat. 2.1.17Google Scholar.

¹⁴¹ Jos. Vita 161; B.J. 2.634.

¹⁴² Frontinus' evidence is rarely taken seriously (see, e.g. Stern, op. cit. (n. 137), i.510–11) on the grounds that (1) Titus, not Vespasian, conquered Jerusalem and (2) other sources make no allusion to Sabbath observance as an obstacle to Jewish success in the revolt of a.d. 66–73. But Frontinus does not mention the siege of Jerusalem and the incident to which he refers may have been earlier in the war; the silence of the other sources is explicable if the Jews did fight, but fought badly (see below), especially since Frontinus does not himself say that they did not fight, only that they lost; and Frontinus, writing when and where he did, ought to have known.

¹⁴³ Herr, op. cit. (n. 126), 248, 254, with reference to Tosefta, Erubin 4 (3). 6 (Lieberman text).

¹⁴⁴ See Goodman, op. cit. (n. 104), 9, for the proper use of the Tosefta for social history.

¹⁴⁵ Jos. A.J. 12.276; cf. I Maccabees 2.40.

¹⁴⁶ Exodus 31.14; 35.2; Numbers 15.32.6.

¹⁴⁷ See, e.g. Stone, M. E., Scriptures, Sects and Visions (Oxford, 1982)Google Scholar.

¹⁴⁸ Neusner, J., Rabbinic Traditions about the Pharisees before 70, 3 parts (Leiden, 1970)Google Scholar.

¹⁴⁹ Jos. c.Ap. 1.212, where the issue is presented starkly.

¹⁵⁰ Jos. B.J. 1.146.

¹⁵¹ Sifre to Deuteronomy 203–4 (ed. Finkelstein, )Google Scholar; Tosefta, Erubin 4 (3), 7 (Lieberman text).

¹⁵² Jos. Vita 12; note that Josephus' own definition at A.J. 14.63–4 contradicts Shammai's. The suggestion by Graetz, H., Geschichte der Juden III 2⁵ (1905), 799Google Scholar, that Jewish fighters in the revolt against Rome relied on Shammai's dictum to clear their consciences is pure surmise.

¹⁵³ Juster, J., Les Juifs dans l'Empire romain (Paris, 1914), ii.265–79Google Scholar argues that Jewish soldiers were not rare in the Roman period, but see the more cautious approach of Sevenster, J. N., The Roots of Pagan Anti-Semitism in the Ancient World (Leiden, 1975), 154–6CrossRefGoogle Scholar. Observance of the Sabbath was the reason given by Dolabella for permitting exemptions for Jews from Roman military service (Jos. A.J. 14.226), but Applebaum, S., ‘Jews in the Roman army’, in Roman Frontier Studies, 1967: The Proceedings of the Seventh International Congress held at Tel Aviv (Tel Aviv, 1971), 181Google Scholar, stresses Jewish objections to the ruler cult and other pagan religious acts required of Roman soldiers.

¹⁵⁴ Jos. B.J. 2.450, 456, 517; both priests (B.J. 2.409–10, 453) and the wider Jerusalem mob (B.J. 2.517) broke the Sabbath in this way.

¹⁵⁵ The view proposed by Hengel, op. cit. (n. 93), 293–6, that the breaking of the Sabbath taboo shows the Jewish Revolt to have been a holy war, founders on the considerable general evidence that the revolt was not inspired by religion. For some secular reasons for the rebellion, see Goodman, M. D., ‘The First Jewish Revolt: social conflict and the problem of debt’, Journal of Jewish Studies 33 (1982), 417–27CrossRefGoogle Scholar.

¹⁵⁶ Jos. B.J. 7.260.

¹⁵⁷ Johns, A. F., ‘The military strategy of Sabbath attacks on the Jews’, Vetus Testamentum 13 (1963), 482–6CrossRefGoogle Scholar.