² For a convenient summary of the (inconclusive) debate as to whether Julian was born in 331 or 332 see Demandt, A., Die Spätantike (Handbuch der Altertumswissenschaft, Abt. 3 Teil 6) (München, 1989), p. 94, n. 2.Google Scholar

³ So Penella, R. J., Greek Philosophers and Sophists in the Fourth Century a.d. Studies in Eunapius of Sardis (Leeds, 1990), pp. 2–3Google Scholar. Goulet, R., ‘Sur la chronologie de la vie et des oeuvres d'Eunape de Sardes’, JHS 100 (1980), 60–72, prefers 349.CrossRefGoogle Scholar

⁴ See Petit, P., ‘Les Sénateurs de Constantinople dans l'oeuvre de Libanius’, L'Antiquité Classique 26 (1957), 347–82CrossRefGoogle Scholar. Libanius' famous invective against the courtiers of Constantine and Constantius is in Or. 42.23–4; cf. Or. 1.76.

⁵ Fl. Dionysius (PLRE I, pp. 259–60, n. 11) had become acquainted with Libanius' family while he was consularis Syriae (Or. 1.36; cf. Eus. V.C. 4.42.3) sometime between 329 and 335.

⁶ PLRE 1, pp. 611–12 s. v. Musonianus. The acquaintance, attested since the early fifties (Lib. Ep. 580 = Fr. 3), might well go back to the time when Strategius came to Antioch as comes of Constantine in 325 (Eus. V.C. 3.39.3; 62.1). See further Seeck, O., Die Briefe des Libanius zeitlich geordnet (Texte und Untersuchungen zur Geschichte der altchristlichen Literatur N. F. 15; Leipzig, 1906), pp. 282–4, n. IV.Google Scholar

⁷ Or. 1.31; 39–44; Ep. 209.

⁸ PLRE I, pp. 818–19, n. 1. The identification of Seleucus the son of Ablabius and father of the deaconess Olympias (PLRE I, pp. 642–3, n. 2) with Seleucus the courtier of Julian, contested by Schouler, B., ‘Hommages de Libanios aux femmes de son temps’, Pallas 32 (1985), 123–48CrossRefGoogle Scholar, at 146 n. 68, gets further confirmation from a letter of John Chrysostom to Olympias saying that she came to Christianity ‘ex asebous oikou’ (Ep. 8, 5c).

⁹ PLRE I, pp. 2–3, n. 4.

¹⁰ PLRE I, pp. 846–7, n. 2.

¹¹ Or. 12.31; 18.9–11, 206.

¹² AM 21.10.8 = Jul. E.L.F. 21; Jul. Or. 7.22, 227c–228d; Caes. 315d, 328d–329d, 336a–c. For discussion see Vogt, J., ‘Kaiser Julian über seinen Oheim Constantin den Großen’, Historia 4 (1955), 339–52Google Scholar, reprinted in: idem, Orbis (Freiburg, 1960), pp 289–304Google Scholar. But Vogt's treatment of Libanius and Zosimus (at 351–2 or 302–4 respectively) is inadequate.

¹³ The crucial point is that Julian in the first panegyric on Constantius (Or. 1.21, 26b) dates the death of Constans (January 350) to the sixth year after the battle.

¹⁴ See Portmann, W., ‘Die 59. Rede des Libanios und das Datum der Schlacht bei Singara’, ByzZ 87 (1989), 1–18.Google Scholar

¹⁵ Pace Callu, J.-P., ‘Un Miroir des princes: le “Basilikos” libanien de 348’, Gerion 5 (1987), 133–52Google Scholar, at 135–6; Portmann (loc. cit.), 1, 12–13.

¹⁶ Barnes, T. D., Athanasius and Constantius. Theology and Politics in the Constantinian Empire (Cambridge, MA, 1993), pp. 315–16Google Scholar, n. 49 suggests the praetorian prefect Fl. Philippus (PLRE I, pp. 696–7, n. 7) pointing to the fact that Libanius praises the emperors for replacing their praetorian prefects regularly (Or. 59.164). I find it hard to believe that Libanius would deliberately have offended the man who commissioned the speech.

¹⁷ A governor of Bithynia named Pompeianus (PLRE I, p. 712, n. 3) staged a rhetorical contest between Himerius and Libanius: Lib. Ep. 742; Decl. 46; Him. Or. 53.

¹⁸ While vicarius Ponticae, Fl. Philagrius (PLRE I, p. 694, n. 5) invited Libanius to deliver a speech: Or. 1.70–72.

¹⁹ See Kraft, K., ‘Die Taten der Kaiser Constans und Constantius’, JNG 9 (1958), 141–86Google Scholar, esp. 182–3; accepted by Callu, loc. cit., 137–8. The maiorina coinage depicts (1) the battle of Singara during which the Persian heir to the throne was killed (cf. Or. 59.98–120), (2) Constans' crossing of the channel (cf. Or. 59.137–41), (3) the transfer of a Frankish tribe to Roman soil (cf. Or. 59.127–34), and (4) the transfer of a city in Adiabene to Thrace (cf. Or. 59.83–7).

²⁰ Eusebius uses the expression for Constantine (V.C. 1.1.1; 1.56.1; 3.55.4; 4.68.2), for God (V.C. 3.43.4; 4.29.4) and for Diocletian (V.C. 1.14.4)!

²¹ The observations of Petit, P., ‘Libanius et la Vita Constantini’, Historia 1 (1950), 562–80Google Scholar have not in my view been invalidated by the critique of Moreau, J., ‘Zum Problem der Vita Constantini’, Historia 4 (1955), 234–45Google Scholar, reprinted in: idem, Scripta minora (Heidelberg, 1964), pp. 124–34Google Scholar who argued that Libanius followed the lost history of Praxagoras.

²² See Kloft, H., ‘Zur Vita Constantini I 14’, Historia 19 (1970), 509–14.Google Scholar

²³ For a conclusive demonstration that the first war of Constantine against Licinius is ignored in the Vita Constantini (1.49–2.18) see Vogt, J., ‘Die Vita Constantini des Eusebius über den Konflikt zwischen Konstantin und Licinius’, Historia 2 (1954), 463–71.Google Scholar

²⁴ theos: Or. 59.50, 125. ho kreitton: Or. 59.48 bis, 58, 72, 142.

²⁵ See Baynes, N. H., ‘Rome and Armenia in the Fourth Century’, EHR 25 (1910), 625–43CrossRefGoogle Scholar, reprinted with revisions in: idem, Byzantine Studies and Other Essays (London, 1960), pp. 186–208Google Scholar: Ensslin, W., ‘Zu dem vermuteten Perserfeldzug des rex Hannibalianus’, Klio N. F. 11 (1936), 102–10.Google Scholar

²⁶ See Barnes, T. D., ‘Constantine and the Christians of Persia’, JRS 75 (1985), 126–36.Google Scholar

²⁷ See Warmington, B. H., ‘Ammianus Marcellinus and the Lies of Metrodorus’, CQ n. s. 31 (1981), 464–8CrossRefGoogle Scholar; Matthews, J., The Roman Empire of Ammianus (London, 1989), pp. 135–6, 448.Google Scholar

²⁸ See Lightfoot, C. S., The Eastern Frontier of the Roman Empire with Special Reference to the Reign of Constantius II (D. Phil., Oxford, 1981), p. 37, n. 116.Google Scholar

²⁹ Dig 39.4.11 (Paulus); 48.4.4 (Scaevola); Expositio 22.

³⁰ See Hoffmann, D., Das spätromische Bewegungsheer und die Notitia Dignitatum (Epigraphische Studien 7; Düsseldorf 1969/1970), Vol. 1, pp. 265–77Google Scholar with notes 599–727 in Vol. 2, pp. 110–17.

³¹ For a good survey of emperors' responses to criticism in circus or theatre (not, however, using the evidence from Lib. Or. 19 and 20) see Alan, Cameron, Circus Factions. Blues and Greens at Rome and Byzantium (Oxford, 1976), pp. 157–92.Google Scholar

³² Förster's conjecture (ekeinon for ekeinou) is rightly rejected by Norman. There is no need and hence no justification to change the transmitted text.

³³ PLRE I, p. 407, n. 1.

³⁴ Ensslin, W., ‘Dalmatius censor, der Halbbruder Konstantins I’, RhM 78 (1929), 199–212Google Scholar; PLRE I, pp. 240–1, n. 6. But CTh 12.17.1 is unlikely to have been addressed to Constantine's half-brother since members of the imperial family are not otherwise attested as recipients of laws.

³⁵ PLRE I, p. 226, n. 7.

³⁶ Auson. Prof. 17.9–12; 18.9–13.

³⁷ Lib. Or. 14.29–30 = Jul. E.L.F. 20.

³⁸ For 314 see Habicht, C., ‘Zur Geschichte des Kaisers Konstantin’, Hermes 56 (1958), 360–78Google Scholar; for 316: Kienast, D., ‘Das bellum Cibalense und die Morde des Licinius’, in: Wissemann, M. (ed.), Roma renascens. Beiträge zur Spätantike und Rezeptionsgeschichte. Ilona Opelt von Freunden und Schülern zum 9.7.1988 in Verehrung gewidmet (Frankfurt/Main etc., 1988), pp. 149–71.Google Scholar

³⁹ Straub, J., ‘Konstantins Verzicht auf den Gang zum Kapitol’, Historia 4 (1955), 297–313Google Scholar, reprinted in: idem. Regeneratio imperii. Aufsätze über Roms Kaisertum und Reich im Spiegel der heidnischen und christlichen Publizistik (Darmstadt, 1972), pp. 100–18.Google Scholar

⁴⁰ Paschoud, F., ‘Zosime 2, 29 et la version païenne de la conversion de Constantin’, Historia 20 (1971), 334–53Google Scholar, reprinted in: idem, Cinq études zur Zosime (Paris, 1975), pp. 24–62.Google Scholar

⁴¹ I accept the arguments of Liebeschuetz, J. H. W. G., Antioch. City and Imperial Administration in the Later Roman Empire (Oxford, 1972), pp. 270–6Google Scholar as modified in CR n. s. 29 (1979), 30 that Or. 48 was composed earlier than Or. 49, probably in 384/5.

⁴² Petit, P., Libanius et la vie municipale à Antioche au IVe siècle après J.-C. (Institut Français d'Archéologie de Beyrouth. Bibliothèque archéologique et historique 62; Paris, 1955), p. 54.CrossRefGoogle Scholar

⁴³ PLRE I, pp. 478–9, n. 35.

⁴⁴ Or. 1.74–80, esp. 75–6, 215, 279; 30.37; Ep. 441.4; 633.2.

⁴⁵ For the food supply of Constantinople see Dagron, G., Naissance d'une capitale. Constantinople et ses institutions de 330 à 451 (Bibliothèque Byzantine. Études 7: Paris, 1974), pp. 530–41.Google Scholar

⁴⁶ Lib. Ep. 143, Or. 17.19; cf. Or. 18.206; AM 25.4.23–4, 24.3.4 (speech of Julian); cf. 24.4.9.

⁴⁷ Licinius' economical use of funds is implicitly acknowledged by Lactantius (M.P. 46.12: ‘tenax in largiendo’). Eusebius merely reproduces the usual stereotypes about the behaviour of tyrants (H.E. 10.8.13; V.C. 1.55.1–2: size of fiscal lots exaggerated, men long dead included in the tax registers); cf. Anon. Val. 22: many rich men executed in order to confiscate their properties.

⁴⁸ CTh 13.10.2. The attribution to Licinius was proposed by Grégoire, H., ‘About Licinius' Fiscal and Religious Policy’, Byzantion 13 (1938), 551–60Google Scholar. The attribution to Maximinus Daia, proposed by Mitchell, S., ‘Maximinus and the Christians in a.d. 312: A New Latin Inscription’, JRS 78 (1988), 105–24Google Scholar, at 122–3, seems less likely. In order to ascribe the law to Maximinus, not only the inscriptio but also the subscriptio has to be changed, since Maximinus did not recognize the third consulate of Licinius, making himself consul for the third time in 313.

⁴⁹ For Tatianus see PLRE I, pp. 876–8, n. 5. Petit, P., ‘Sur la date du Pro Templis’, Byzantion 21 (1951), 285–309Google Scholar has in my view failed to prove that Libanius sent the speech while the praetorian prefect Cynegius (PLRE I, pp. 235–6, n. 3), harshly criticized in §46, was still in office (384–8).

⁵⁰ Eus. L.C. 8.1–4; V.C. 54, 57; Jul. Or. 7.22, 228b; Anon. De reb. bell. 2.1; Lib. Or. 30.6, 37; 62.8; Zos. 5.24.6; Joh. Mal. p. 324 Bonn.

⁵¹ Disseminated in secret as long as sons of Constantine reigned, it first shows up in Julian's Caesares (336a/b). Sozomenus deemed it necessary to refute it at length: H.E. 1.5.

⁵² A similar periodization of Constantine's reign is to be found in Aur. Vict. 41.12.

⁵³ Local circulation is indicated by the fact that the speech culminates in an invective against a local enemy of Libanius: Or. 62.63–73. Terminus post quern is the suppression of the revolt of Procopius (Or. 62.58–60). Since the sketch of the emperors' attitudes to Greek rhetoric (Or. 62.8–18) ends with Julian, the speech was almost certainly composed before Valens' death.

⁵⁴ In view of these explicit statements it is puzzling to find that in his Autobiography Libanius praises a citizen of Heraclea Pontica whom he had met in 340 for having ‘consorted more with gods than with men on earth’ although a sacrilegious legislator inflicted the death penalty on anyone who dared to do so (Or. 1.27). Two (not mutually exclusive) explanations might be suggested. Firstly, Libanius refers to divination (cf. Or. 19.20) rather than to sacrifices which did not become liable to the death penalty before 356/7 (CTh 16.10.6). Secondly, since the statement covers the whole life-time of the man, it cannot be precisely located in time. The emperor in question could, therefore, well be Constantius.

For a different view see Barnes, T. D., ‘Christians and Pagans in the Reign of Constantius’, Entretiens Hardt 34 (1987), 301–37, at 330, n. 135.Google Scholar

⁵⁵ It was not, however, shared by Eunapius who accused Constantine of having pulled down the most celebrated temples: V.S. 461.

⁵⁶ Barnes, T. D., Constantine and Eusebius (Cambridge, MA, 1981), p. 377, n. 11.Google Scholar

⁵⁷ Eus. V.C. 2.45; 2.23, cf. CTh 16.10.2.

⁵⁸ According to Eusebius (V.C. 3.56) the famous sanctuary (for the site and its history see Robert, L., ‘De Cilicie à Messine et à Plymouth avec deux inscriptions grecques errantes’, Journal des Savants [1973], 162–211Google Scholar, reprinted in: idem, Opera Minora Selecta, vol. 7 [Amsterdam, 1990], pp. 225–75Google Scholar) was razed to the ground by soldiers acting on Constantine's orders. But when Julian in 362 passed through Tarsus he was approached by ‘the priest of Asclepius’ and granted his plea that the local bishop should restore the columns which he had taken away from the temple (Zon. 13.12). In 355 an altar dedicated to ‘Asclepius of Aegeae’ was set up in Epidaurus (IG IV² 438). Libanius attributes the demolition to Constantius: Or. 30.38–9; cf. Ep. 695, 1342.

⁵⁹ Joh. Mal. p. 324 Bonn with the discussion in Dagron, op. cit., pp. 373–7 (but Joh. Lyd. De Mens. 4.2 is poor evidence for the performance of pagan rites during the consecration).

Mango, C., Le Développement urbain de Constantinople (IVe–VIIe siècle) (Travaux et mémoires. Monographes 2: Paris, 1990 ²), pp. 34–6Google Scholar points out that the few churches which can plausibly be attributed to Constantine can hardly have given Constantinople a thoroughly Christian outlook as Eusebius would have us believe (V.C. 3.48).

⁶⁰ The traditional interpretation of Constantine's Letter to the Eastern Provincials has recently been challenged by Barnes, T. D., ‘Constantine's Prohibition of Pagan Sacrifice’, AJPh 105 (1984), 69–72Google Scholar) who argues that document means the exact opposite of what it has always been taken to mean: in his view the Letter is not an edict of toleration, but an edict of persecution. For a convincing refutation of this view see Errington, R. M., ‘Constantine and the Pagans’, GRBSt 29 (1988), 309–18, at 311–12.Google Scholar