¹ ‘Nuove iscrizioni di Capua’, Terza miscellanea greca e romana (Studi pubblicati dall' Istituto Italiano per la Storia Antica, 1971), 298 fGoogle Scholar.

² Antico, tardoantico ed èra costantiniana I (1974), 334–8Google Scholar.

³ ‘Lettura epigrafica e carriere aristocratiche: il caso di Petronio Probo’, Riv. di Fil. III (1983), 170–82Google Scholar, developing ideas already expressed in Helikon 15/16 (1975/1976), 308–11Google Scholar; and MEFRA 95 (1983), 268–72. In an appendix (pp. 178–82) I argue that the reconstruction of Probus' prefectures offered by Mazzarino and Giardina is also mistaken. The long tenure of 368−75 (or even a little later) is his third, not first prefecture. This would make it all the more incredible that he should wish to hold a lower office as well.

⁴ ‘L'inscription de Petronius Probus à Capoue’, Tituli 4 (1982), 547–50Google Scholar.

⁵ To save space I use the convenient ancient aeronyms PPO for praefectus praetorio and PVR for praefectus urbi Romae. And I do not give references for uncontroversial careers where the details can be found easily enough in PLRE.

⁶ And he later reports Gratian's remark that it was Ausonius' seniority as prefect that determined his seniority as consul (ibid. 12).

⁷ In the six surviving consular dating formulae in papyri for the year 371 Probus is invariably styled PPO as well as consul: see Bagnall, R. S. and Worp, K. A., Chronological Systems of Byzantine Egypt (1978), 113Google Scholar.

⁸ Even Chastagnol concedes this.

⁹ See Chastagnol, , Revue Historique 219 (1958), 237–52Google Scholar, and in Recherches sur les structures sociales dans l'antiquité classique: Caen 25–26 avril 1969 (1970), 190–4.

¹⁰ For all the details see now Roda, S., Commento Storico al libro ix dell'epistolario di Q. Aurelio Simmaco (1981), pp. 44 f., 116 fGoogle Scholar.

¹¹ Cicero, , de domo 102Google Scholar; Pan. Lat. XI. (3) 7. 1; Jerome, Ep. 59. 5.

¹² Alföldi, A., A Conflict of Ideas in the Late Roman Empire (1952), 87Google Scholar.

¹³ Cameron, , ‘Anicius Claudius (I. Cret. IV. 322)’, ZPE 57 (1984), 147–8Google Scholar.

¹⁴ (Amm. Marc. XXVII. 11. 2; cf. Claudian, , Pan. Ol. et Prob. 45 fGoogle Scholar., and Novak, D. M., Klio 62 (1980), 482–4Google Scholar.

¹⁵ For the growth of these classes, see Chastagnol, A., La Préfecture urbaine (1960), 433Google Scholar; Jones, A. H. M., Later Roman Empire I (1964), 143Google Scholar.

¹⁶ The sources are collected in PLRE I. 737–8.

¹⁷ I am inferring this from the fact the law was posted at Carthage, but the main point is hardly affected even if Probus was not there in person.

¹⁸ Riv. di Fil. 1983, 178.

¹⁹ Seeck, O., Regesten der Kaiser und Päpste (1919), 10 fGoogle Scholar.

²⁰ For other examples of this formulae, see Seeck, , Regesten, 10Google Scholar.

²¹ See Jones, , Later Roman Empire I, 385–6Google Scholar.

²² Flamant, J., Macrobe et le néo-platonisme latin à la fin du IVe siècle (EPRO 58, 1977), 92Google Scholar, with my review in CP 77 (1982), 378–80.

²³ Seeck, O., Die Briefe des Libanius (1906), 265–6Google Scholar, gives a fuller collection of evidence than PLRE I. 814.

²⁴ All set out in Chastagnol, Fastes 164–9, not counting for this purpose the Greek inscription IG XIV 1019 = CIL VI. 30966 where he is called Lampadius alone.

²⁵ SDHI 44 (1978), 47–54.

²⁶ PLRE I. 208, missing however CJ XI. 75. 2.

²⁷ Martin, R., Palladius, Traité d'agriculture I (livres i et it) (1976), p. viiGoogle Scholar.

²⁸ The relevant passages from the grammarians are conveniently assembled in Doer, B., Die römische Namengebung (1937), 21; 39; 46; 68–9Google Scholar.

²⁹ Gramm. Latini v. 140. 35 Keil.

³⁰ The well-known Gallic family of the Palladii: see PLRE II, nos. 4 and 13–15.

³¹ R. H. Rodger's account in CTC 3 (1976), 195–9, not being aware of the issue, gives less help than it might.

³² See JRS 56 (1966), 25 fGoogle Scholar., developing a suggestion of Mazzarino, , Rend. Ist. Lombardo 71 (1938), 255 fGoogle Scholar.; cf. too CP (1982), 379–80. On the MS tradition of Macrobius see now Barker-Benfield, B. C. in Texts and Transmissions: a survey of the Latin Classics, ed. Reynolds, L. D. (1983), 222–32Google Scholar.

³³ Zur Geschichtsschreibung und Weltanschauung des Ammianus Marcellinus, Klio Beiheft 16 (1923), 4–5Google Scholar; cf. Seeck, O., Regesten, 94Google Scholar.

³⁴ Note that in his Briefe des Libanius (1906), 58, Seeck himself sensibly rejected such identifications with the historian, on the ground that ‘Libanius ihn Marcellinus nennt’. This is not to exclude the possibility that one of the Antiochene Ammiani was related to the historian.

³⁵ See E. Doblhofer's edition, I (1972), 22–3; M. Ferrari, IMU 16 (1973), 1–41; Reeve, M. D., in Texts and Transmissions, 340Google Scholar.

³⁶ So even PLRE I. 491.

³⁷ Doblhofer, I, 24 n. 39.

³⁸ Rather than be swept off our feet by the coincidence that a Claudius is attested in one of the posts Lachanius is known to have held, we should rather pause to reflect on the disturbing fact that only one other holder of this office, normally an annual appointment (see n. 83), is attested in the 90 years between 370 and 459 (see the fasti in PLRE I, p. 1094 and 11, p. 1279)! If Lachanius' career were known to us (like many others) from one inscription rather than one well-known poem, scholars would be less anxious to remedy the lack of confirmation from other sources. Only one of the posts held by Rutilius himself (which must have included at least one junior post in addition to mag. off. and PVR) is independently attested—and that by only one law in a corrupt form.

³⁹ Schulze, G. (W.), Graeca Latina (1901)Google Scholar = Orthographica et Graeca Latina (reprint 1958), 95 f.

⁴⁰ See AJA 89 (1985), 141–5.

⁴¹ See PLRE I. 656.

⁴² In the heading to the Ludus Ausonius as consul balances Drepanius as proconsul; in the heading to Eel., there are no titles on either side.

⁴³ And yet Courcelle, P., Late Latin Writers and their Greek Sources (1969), 226Google Scholar n. 10, emphatically rejected the linguistic arguments brought against the identification by Baehrens, W. A., Hermes 56 (1921), 443–5Google Scholar.

⁴⁴ Further discussed by Cameron, Alan and Schauer, Diane, JRS 72 (1982), 145Google Scholar.

⁴⁵ See the stemma of the family as reconstructed in AJA 1985, 145.

⁴⁶ See Epigrafica 1985.

⁴⁷ Carandini, A., Ricci, A., de Vos, M., Filosofiana: La Villa di Piazza Armerina (1982)Google Scholar; a possibility already taken seriously by Cracco Ruggini, L., ‘La Sicilia tra Roma e Bisanzio’, in Storia della Sicilia III (1980), 68 n. 57Google Scholar. See now below, 66–8.

⁴⁸ Vera, D., Opus 2 (1983), 583–4Google Scholar could have been more emphatic in his rejection of this conjecture.

⁴⁹ Seeck, O., Die Briefe des Libanius (1906), 282–4Google Scholar; PLRE I, 611–12.

⁵⁰ See the texts cited by Doer, B., Die römische Namengebung, 68–73Google Scholar.

⁵¹ Another case (it would appear), though his ‘agnomen’ was not formed in the same way, is the famous calligrapher Furius Dionysius Filocalus (information in Ferrua, A., Epigrammata Damasiana (1942), 21–35Google Scholar).

⁵² ILCV 104, with Hadot, P., Marius Victorinus: Recherches sur sa vie et ses œuvres (1971), 16–17Google Scholar.

⁵³ ILS 5557; ILT 1192; Ann. Epigr. 1955, 52; 1972, 873.

⁵⁴ In fact, as we shall see, Petronius Probus himself.

⁵⁵ See my discussion in ZPE 56 (1984), 167.

⁵⁶ Seeck, , Regesten, 205Google Scholar.

⁵⁷ Riv. di Fil. 1983, 176.

⁵⁸ The stone itself gives the name in the nonsensical form PACIBO. Mommsen was surely correct to suppose that the stonecutter misread his copy. A can often look like R in capital script, and a badly formed O could be misread as CI. In what should be vigentibus in l. 2 the man clearly misread a G as a C.

⁵⁹ If Petronius Claudius the proconsul was in fact the younger brother of Petronius Probus, we might guess that the Claudius came from their mother's family—a connection of no interest to Probus once he had contracted his own match with the principal heiress of the Anicii. He did not pass the name on to any of his three sons, whose names are known in full.

⁶⁰ Arnheim, M. T. W., The Senatorial Aristocracy of the Later Roman Empire (1972), 196–7Google Scholar.

⁶¹ ‘Collegiate Prefectures’, JRS 54 (1964), 78–89Google Scholar = The Roman Economy (1974), at 387–91.

⁶² Essai sur la préfecture du prétoire du Bas-Empire (1933), 109–18.

⁶³ ‘Sulla carriera prefettizia di Sex. Petronius Probus’, Helikon 7 (1967), 414 fGoogle Scholar. = Antico, tardoantico ed èra costantiniana, 328–33.

⁶⁴ ‘Anicianae domus culmen, nobilitatis culmen’, Klio 62 (1980), at 475–80Google Scholar.

⁶⁵ L'Empire chrétien² (1972), 269 n. 2; Giardina, MEFRA 95 (1983), 268–72.

⁶⁶ Die Religionzugehörigkeit der höhen Amtsträger des Römischen Reiches seit Constantins I. Alleinherrschaft bis zum Ende der Theodosian. Dynastie (1978), 297.

⁶⁷ Chastagnol, , Fastes, 164–9Google Scholar.

⁶⁸ PLRE I. 979.

⁶⁹ ‘Die Reichspraefektur des vierten Jahrhunderts’, Rhein. Museum 1914, 25.

⁷⁰ Antico …, 330; see too his full, earlier discussion in Stilicone (1942), 8–22, a masterpiece of misplaced erudition and ingenuity.

⁷¹ On p. xcix of his edition of Symmachus, Seeck postulated a massive double ‘error lapicidae’.

⁷² The original is lost, but in addition to Ferrarino we have the more accurate and evidently independent transcription of Sirmond (see Mommsen's commentary on CIL v. 3344; Mazzarino refers to Dessau's ambiguous comment (on ILS 1266) that the inscription was ‘fortasse non plus semel descripta saec. xv’, but this was surely not intended to imply that Sirmond depended on Ferrarino, for Dessau prints Sirmond's text throughout). Mazzarino misapplies here the criterion of lectio difficilior. The essential point is that virtually all Ferrarino's other divergences from Sirmond are errors, so obviously so that Dessau did not even bother to report them in his edition. E.g. (Sirmond first): 6, Illyrici] Illyici; 7/8, praef. praet. Gal/liar.] prae. et praefec./ Galliar.; 14, eruditissimo] -imum; 18, conss.] coss. Ferrarino also abbreviates (e.g. atq., omnib.) where Sirmond writes out the text in full. The IIII must surely be treated as the merest slip.

⁷³ In his note to Cod. Theod. XI. I. 15 (p. 574).

⁷⁴ Seeck, , Regesten, 228Google Scholar.

⁷⁵ Jones, , Roman Economy, 393Google Scholar.

⁷⁶ See Palanque, , Essai, 115–16Google Scholar.

⁷⁷ Roman Economy, 391, citing Amm. Marc, XXVI 5. 5, ‘Italiam vero cum Africa et Illyrico Mamertinus [regebat]’.

⁷⁸ Mazzarino advanced another objection (Antico …, 328–9). According to Ammianus (xxx. 5. 4), writing of 375, Probus ‘praefecturam praetorio tune primitus nanctus, eamque multis atque utinam probabilibus modis in longum proferre gestiens … plus adulationi quam verecundiae dedit’. According to Mazzarino, Ammianus is saying that Probus' 368–75 prefecture was his first. But why should Ammianus have made this point near the close of Probus' long prefecture in 375 rather than when describing its start in 368 (XXVII. II. I)? In fact we must almost certainly accept Heraeus' insertion of a non before tune primitus. This was a favourite turn of phrase with Ammianus: cf. XVII. II. I, ‘quod non tune primitus accidit’; XXIII. 5.16, ‘non nune primitus (ut maledici mussitant) …’ If so, then the passage could equally well be used as evidence that this was not Probus' first prefecture! More probably, however, Ammianus' point is that although Probus had held his prefecture for a long time by 375, he was none the less still anxious to hang on to it, at what Ammianus saw as the price of his self-respect. Clearly this passage cannot provide any firm evidence either way on the chronology of Probus' prefectures.

⁷⁹ Essai, 117; cf. Jones, , Roman Economy, 388Google Scholar; confirmed by the full discussion of Novak, Klio 62 (1980), 477–8.

⁸⁰ Antico …, 330.

⁸¹ I am assuming (following Jones) that we need no longer consider the possibility of a collegiate prefecture.

⁸² Ep. I. 58 of Symmachus refers to Probus undertaking a second term of office: ‘Sit tibi animus aequus et patiens muneris imperati. Saepe usu venit ut in secundos labores virtus probata reparetur.’ Unless Symmachus merely means ‘again’, this should be a reference to Probus' second prefecture of 366. If so, this would be the earliest datable extant letter of Symmachus (Seeck dated it to 378/9 on the basis of a theory of Probus' prefectures he later abandoned himself; Callu to ‘vers 383’). All Symmachus' letters to Probus should be early, while they were still on speaking terms. If the unnamed ‘civis emeriti’ of Ep. III. 88 is (as often suspected) Probus, then Symmachus was reprimanded from court for not writing the usual letter of condolence on Probus' death.

⁸³ Jones, , Later Roman Empire I, 380–1Google Scholar. Our lists of governors are of course full of gaps, but there are clear indications none the less that both consulars and proconsuls were in principle appointed for a period of one year: see Barbieri (n. 1), 299 n. 2; Mazzarino, Antico …, 304–6; Barnes, T. D., Phoenix 1985Google Scholar.

⁸⁴ Chron. s.a. 372. He certainly spent a lot of time in Illyricum at this period: cf. PLRE I. 738.

⁸⁵ I pass over here the question of the false iteration number II: cf. Palanque, Byz. 9 (1934), 355–9; 706–7; Chastagnol, , Fastes, 177–8Google Scholar.