¹ Garzetti, A., From Tiberius to the Antonines, transl. by Foster, J. R. (1974), 403 fGoogle Scholar.

² Millar, F., The Emperor in the Roman World (1977), 286 ffGoogle Scholar.

³ Pflaum, Proc., 295 f.

⁴ Les carrières procuratoriennes équestres sous le Haut-Empire romain (1960–1961), 3 vols.

⁵ JRS 53 (1963), 194 ff.

⁶ The administrators of Roman Egypt', JRS 65 (1975), 124 ff.

⁷ Proc., Pt. 2 Ch. 4.

⁸ e.g., Proc., 251, refers to ‘le cours normal de son avancement’.

⁹ Proc., 206.

¹⁰ Carr., 395.

¹¹ Carr., nos. 121, 217, 249, 327, 331 bis. The early stages of P. Cominius Clemens' (no. 184) career were so regular that it could be said ‘jusqu'ici pas trace de favoritisme’.

¹² Ad M. Caes. 5. 37.

¹³ Proc., 210 and 295 f.

¹⁴ Fronto and Antonine Rome (1980), 102–3, with notes 52 and 53.

¹⁵ Emperor, passim.

¹⁶ It has been argued that the age of an imperial slave or freedman may have had an influence on the level of office for which he was eligible (Weaver, P. R. C., Familia Caesaris (1972), 224 ff.CrossRefGoogle Scholar). Though the evidence cited shows more of a pattern than any similar evidence for equestrian procurators, it is still less than compelling (G. Burton, ‘Slaves, Freedmen and Monarchy’ (review of Boulvert, G., Esclaves et affranchis impériaux sous le Haut-Empire romain, rôle politique et administratif and Domestique et fonctionnaire sous le Haut-Empire romain: la condition de l'affranchi et de l'esclave du prince), JRS 67 (1977), 162Google Scholar f.).

¹⁷ Millar, Fergus, A Study of Cassius Dio (1964), 104Google Scholar.

¹⁸ ibid., 111.

¹⁹ 52. 20. 1–2.

²⁰ 52. 24–25.

²¹ 52. 24. 1–2 and 52. 25. 5.

²² It may also be worth pointing out that in the codicil of appointment sent by Marcus Aurelius to Q. Domitius Marsianus (AE 1962, 183) there is no hint of a seniority principle. Marsianus' appointment to the ducenariate procuratorship is said to have been due to Marcus' favor, and he is told to maintain the diligentia, innocentia and experientia which he had previously displayed.

²³ A. Sherwin-White, ‘Procurator Augusti’, PBSR 15 (1939), 11 ff. After showing the slow development of the procuratorial service from disparate elements in the first century, Sherwin-White concluded that the system became increasingly ‘mechanical and regular … in the highly developed bureaucracy of the second century’.

²⁴ Pflaum, Proc., 227.

²⁵ Out of the dozens of second- and third-century careers of ex-centurions, only a few give evidence of a complete career which progressed beyond the level of entry (among the primipilares, nos. 167, 225, 330; and of the primipilares bis, nos. 94, 109, 162, 234 in Pflaum's catalogue). Professor E. Birley has pointed out the diverse social origins of primipilares, some of whom would have been of municipal aristocratic origins, while many others rose from the ranks of the legions (Roman Britain and the Roman Army (1953), 118 ff.). To judge by a statement of Cassius Dio (52. 25. 7), the common birth of the latter group was not forgotten and may have influenced their advancement, so it is best to keep them separated from the other careers, as Pflaum did.

²⁶ It would be a mistake to draw too many conclusions from the concept of four discrete subdivisions at the ducenariate level. In the case of C. Iulius Rufus (no. 129) who is known to have been procurator patrimonii and procurator provinciae Raetiae (each attested by a different inscription), Pflaum supposed that he probably held another procuratorship between the two, since the Raetia post belonged to the third echelon. In fact, the only concrete evidence from the second century relevant to the question consists of two complete cursus inscriptions: in one (no. 167) the Raetia post appears as the second ducenariate office, and in the other (no. 156) it appears as the third. The evidence is indecisive, leaving only the weight of argument based on confidence in the system of sub-divisions. But confidence may wane when it is noticed that only a minority (seven) of the seventeen third echelon posts known from sufficiently complete second-century careers were actually held as the third ducenariate post.

²⁷ For comments about variation, Carr., nos. 106 bis, 132; for the seniority principle, nos. 310, 312.

²⁸ Pavis-D'Escurac, H., La préfecture de l'annone: service administratif impérial d'Auguste à Constantin (1976), 392Google Scholar.

²⁹ In the third-century sample, in the case of procurators whose cursus inscriptions mention no militiae, I have counted them as having held no militiae. There are indications that in this period equites became less concerned about giving full details about their military service, and it is possible that they held militiae without including any hint of them on their cursus inscription. If the cursus without militiae were excluded from the sample as uncertain, the coefficient would again be changed little: R = —0·26 (N = 21).

One further calculation may add meaning to the correlation coefficient. The determination coefficient (i.e., the degree to which the length of service in sexagenariate posts is influenced by the number of militiae) is calculated by squaring the correlation coefficient (R). Thus, even in the case of the strongest relationship (R = —0·31 in the third century), the length of service in militiae can at best be said to explain only ten per cent of the variation in the number of sexagenariate posts held. For an explanation of these statistical techniques, Floud, R., An Introduction to Quantitative Methods for Historians (1973), chapter 7Google Scholar.

I have also tested the possibility that the number of militiae influenced the pace of the career over a longer period by calculating the correlation coefficient for the number of militiae against the number of sexagenariate and centenariate posts held before promotion: R = —0·21 for the second century (N = 26) and R = 0·0 for the third century (N = 14). In other words, there was no relationship between the factors during either period.

³⁰ Based on thirty second-century careers.

³¹ Based on twelve careers. (The number of relevant third-century careers is even smaller.)

³² Brunt, art. cit. (n. 6), 137, n. 58, found it impossible to draw conclusions from the limited data available.

³³ Pflaum's comments seem to suggest that he believed in such a hypothesis. E.g., with regard to P. Magnius Rufus Magonianus (no. 236), it is concluded: ‘Les postes se succèdent de telle sorte que chaque catégorie n'est représentée que par un seul emploi, prouvant la longue durée des fonctions, puisque rien ne nous permet de supposer une suite de promotions particulièrement rapide et brillante.’ But the hypothesis about longer tenure of fewer offices is never argued in detail, much less proven.

³⁴ Faustinianus is attested in the ducenariate office in 173, but Pflaum plausibly suggests that he may have gone out to Egypt with his father, the prefect in 170.

³⁵ Dobson, B., ‘The Significance of the Centurion and “Primipilaris” in the Roman Army and Administration’, Aufstieg und Niedergang der römischen Welt II. 1, ed. Temporini, H. (1974), 402Google Scholar. The article includes a discussion of the development of a career structure for centurions.

³⁶ 52. 25. 5.

³⁷ Boissevain, J., ‘When the Saints go marching out: Reflections on the decline of patronage in Malta’, in Patrons and Clients, ed. by Gellner, E. and Waterbury, J. (1977), 88Google Scholar, describes such an evolution of attitudes in modern Malta, where the people increasingly demand services from the government as their rights in accordance with rules, rather than petitioning the services as favours. The lack of expectation about appointment to procuratorships can be contrasted with the expectations which exconsuls held about appointment to the proconsulships of Africa and Asia (Tacitus, Agricola 42).

³⁸ Pflaum's comments about the career of L. Aurelius Nicomedes (no. 163) imply the existence of such attitudes. Carney, T. F., Bureaucracy in Traditional Society (1971), 18Google Scholar, suggests much the same (without argument or evidence).

³⁹ E. Birley, ‘Senators in the Emperor's Service’, PBA 1953, 202.

⁴⁰ Proc., 195 ff.; also Jarrett, M., ‘The African contribution to the Imperial civil service’, Historia 12 (1963), 222,Google Scholar and Dobson, B., ‘Praefectus fabrum in the early Principate’, in Britain and Rome: Essays presented to E. Birley, ed. by Jarrett, M. and Dobson, B. (1965), 77 fGoogle Scholar.

⁴¹ Nos. 121, 145, 187, 209, 264.

⁴² Proc., 206. The circularity of the argument for the abolition of the post under Septimius Severus should perhaps be noted: Pflaum's evidence lies in the fact that no praefectus fabrum appears after the reign of Commodus; on the other hand, the reign of Septimius Severus is given in the catalogue as the terminus ante quem for the prefectures of M. Veserius Iucundianus (no. 209) and M. Porcius Aper (no. 187) because Severus abolished the post. Dobson, art. cit., 78, discards the Severan date for abolition without argument.

⁴³ Though Jarrett makes similar remarks about the special favour enjoyed by African praefecti fabrum (n. 40 above), his African evidence shows the opposite: only seven of sixteen African praefecti became equestrian officers and four reached procuratorships, while half of all African equestrian officers of die same period (nine of eighteen) succeeded to procuratorships. With such a small sample, conclusions cannot be firm, but the sample hardly proves that praefecti were unusually successful.

⁴⁴ e.g., Epist. 4.4, 7.22, 10. 87; Millar, Emperor, 284 ff.

⁴⁵ See below, p. 56 f.

⁴⁶ Pflaum, Proc., 218. The slow ascent of M. Aemilius Bassus (no. 103) through the ranks is attributed to lack of military distinction, while two cases of rapid promotions (nos. 116 and 136) are ascribed to outstanding military service.

⁴⁷ The group referred to appears in Carr., Table A2: nos. 51, 72 and 74 won decorations, but are not known to have reached ducenariate procuratorships.

⁴⁸ M. Vettius Latro (no. 104) and C. Annius Flavianus (no. 202) present similar cases: both won military decorations, but were nevertheless appointed to a total of three sexagenariate and centenariate posts instead of the more usual two.

⁴⁹ Nos. 136, 181 bis and 188 won military decorations and later reached the senate.

⁵⁰ Carr., 336.

⁵¹ Nos. 151, 152,174, 178 bis are not known to have been promoted; nos. 121 and 163 are known to have held one more post. Slightly more than half of all centenariate procurators of the second century are known to have been promoted to ducenariate posts.

⁵² Nos. 66, 73, 117, 140 and 168 are known to have been promoted, and nos. 66 and 117 reached Palatine officia, while the moneta was the last recorded post for no. 206.

⁵³ Nos. 66, 73, 117 and 140 all spent much of their time at the ducenariate level in provincial posts rather than at the emperor's side in Rome (no. 140 had previously been Lucius Verus' ab epistulis Caesaris).

⁵⁴ op. cit., 428 ff.

⁵⁵ Proc., 237 and 252.

⁵⁶ Jarrett, M., ‘An album of the equestrians from North Africa in the emperor's service’, Epigraphische Studien 9 (1972), 206Google Scholar.

⁵⁷ 52. 24. 2: ἐκ τῶν πολλάκις τε ἐστρατευμένων καὶ πολλὰ καὶ άλλα διῳκηκότων.

⁵⁸ 52. 26. 2: ἐπιτηδειότεροι πρὸς πᾶν ἔργον.

⁵⁹ The figures are based on the careers found in Pflaum's catalogue, together with a few additional second-century careers discovered since its publication (Q. Domitius Marsianus, AE 1962, 183; A. Scantius Larcianus, AE 1962, 312; Q. Petronius Novatus, AE 1967, 644).

⁶⁰ Proc., 237.

⁶¹ Dobson, op. cit. (n. 35), 423, lists four procuratorships not held (as far as we know) by primipilares, in part because of lack of expertise in letters. Dobson also outlines four possibilities for procuratorial careers for primipilares; it should be pointed out that only the second possibility, which includes a legionary prefecture, is unique to primipilares.

⁶² Equites promoted from militiae held thirty-one per cent of their lower-level civilian procuratorships in Rome, eleven per cent in Italy and fifty-eight per cent in the provinces (N = 113); the comparable figures for second-century civilians are forty per cent in Rome, eight per cent in Italy and fifty-two per cent in the provinces (N = 50).

⁶³ Pflaum comments that ‘nous ne nous étonnons pas que Julius Julianus, ancien préfet de la flotte pontique, ait été appelé à redevenir un chef naval’ (Carr., 464). Out of thirteen prefects of Italian fleets whose careers are known, only Iulianus and Q. Baienus Blassianus (no. 126) had previously commanded provincial fleets.

⁶⁴ Schulz, F., History of Roman Legal Science (1946), 103 ffGoogle Scholar.

⁶⁵ art. cit. (n. 6), 141.

⁶⁶ Pflaum, Proc., 225 ff., 260 ff. (Pflaum's figures show that this was only a tendency, not a rigid rule.)

⁶⁷ art. cit., 128 f.

⁶⁸ MacMullen, R., Roman Government's Response to Crisis (1976), 54 fGoogle Scholar. talks of equestrian officials being stationed ‘more often than not’ near their homes, and of the reliance on regional expertise. A handful of examples are offered out of the three hundred and fifty careers available in Pflaum's catalogue: for each example it is possible to discover numerous counter-examples of officials who served in different regions during their careers (they are too numerous to list, but it may be noted that in the third century alone the following procurators were moved from one end of the empire to the other: nos. 222, 229, 262, 264, 268, 277, 380, 295, 317). Clearly there was no general attempt at regional specialization.

⁶⁹ A. Scantius Larcianus (AE 1962, 312) was promoted directly from a centurionate to the urban office, while Ti. Claudius Secundinus L. Statius Macedo (no. 109) and an anonymous primipilaris (no. 211) were promoted to it after holding the prefecture of Legio II Traiana. Another primipilaris (no. 182) also held the post. Nos. 106 bis, 142, 168 and 183 were the second-century civilian occupants of the office.

With regard to such urban offices, Pflaum developed the concept of ‘un <arrêt de compensation>’, arguing that those who enjoyed the privilege of holding their offices at Rome paid for the privilege by serving in more posts. Thus, type 1 careers proceeded more rapidly than type III careers (Proc., 245, 246, 287 and especially 252). In fact, among the second-century procurators who were promoted through the ducenariate level (to a Palatine officium or a top ducenariate post) there is no connection at all between the number of urban posts held and the rapidity of movement through this level (the correlation coefficient is precisely 0·0).

⁷⁰ Nos. 89, 113, 139, 162 and 180 held no urban ducenariate offices, and nos. 109, 110, 117, and 134 held one. Presumably it is because of the expectation of financial specialization that Pflaum commented: ‘l'avancement à partir de ce dernier emploi (Lyonnaise et Aquitaine) conduit souvent au secrétariat a rationibus, tandis que le praef. class, praet. Ravennatis est promu ordinairement au commandement de la flotte de Misène, voire au poste d' a censibus equitum Romanorum’ (Carr., 308). In fact, Pflaum's table (Proc., 255 f.) shows three procurators of Lugdunensis (including no. 89) and two naval prefects promoted to be secretary a rationibus during the second century—hardly indicative of two different career paths in accordance with expertise.

⁷¹ See above, p. 54.

⁷² L'annone, 79.

⁷³ 52. 25. 5.

⁷⁴ 52. 19. 4. In 52. 25. 5 Dio talks of the offices as a ‘prize for merit’ (‘athlon aretēs’).

⁷⁵ Paneg. 44.7 f. where Pliny also speaks of a ‘prize’ for virtue.

⁷⁶ AE 1962, 183; Pflaum, , ‘Une lettre de promotion de l'empereur Marc Aurèle’, Banner Jahrb. 171 (1971), 349Google Scholar ff.

⁷⁷ P. Berl. 8334 (= Corp. Pap. Lat. 238).

⁷⁸ Millar, Emperor, 286 ff., emphasizes the emperor's patronal role.

⁷⁹ Seneca, Ep. ad Luc. 19. 3; Philostratus, VS 524, 626. That eloquence and learning were standard characteristics of worthy gentlemen is demonstrated by the fact that most commendationes written by Pliny attest to these qualities (Ep. 2. 9, 3. 2, 4. 15, 6. 6, 10. 4)—even when the post sought is a military tribunate (4. 4, 7. 22).

⁸⁰ Ann. 12. 49; close personal contact must also have been the source of success for Lucius Verus' educator and libertus Nicomedes, who progressed rapidly to an important ducenariate post, despite his servile origin (Pflaum, Carr., no. 163).

⁸¹ Jos., AJ 20. II. 1; see Millar, Emperor, 286 ff. for the importance of patronal petitions.

⁸² Ad Pium 9; the letter also indicates that Fronto had already failed twice on Appian's behalf.

⁸³ Ep. 10. 85, 86a–b. The qualities ascribed to the subordinates included probitas, industria, diligentia, disciplina, integritas, iustitia and humanitas. For a more detailed discussion of these letters, see my Patronage and Bureaucracy in the Early Empire (forthcoming).

⁸⁴ Mor. 814D.

⁸⁵ Pt. 2, Ch. 3.

⁸⁶ See above, p. 50.

⁸⁷ ILS 1191, 2941, 4928 (= CIL VI 1532, 1418, 2132).

⁸⁸ While senatorial and equestrian officials were numbered in the hundreds and their staffs perhaps in the thousands, Chinese bureaucrats were numbered in the hundreds of thousands. Though personal factors affected appointments in the Chinese bureaucracy, there existed elaborate mechanisms for the appointment and promotion of officials in accordance with seniority and merit (as evaluated, for instance, by examinations). The contrast between the Chinese and Roman administrative structures in this respect is marked (see Reishauer, E. and Fairbank, J., East Asia: the Great Tradition (1960), via index)Google Scholar.

⁸⁹ Hirschfeld, Verwaltungsbeamten, 443 f., stressed suffragia, and R. Syme, ‘Pliny the Procurator’, HSCP 73 (1969), 208, wrote of the dominance of personal factors. That most procurators remained amateurs is perhaps most clearly indicated by the fact that the majority probably spent less than ten years in procuratorial service (see Appendix).

⁹⁰ Weber, M., Economy and Society (1968), 958 fGoogle Scholar.

⁹¹ Campbell, B., ‘Who were the “Viri Militares”?’, JRS 65 (1975). 11Google Scholar ff.

⁹² JRS 53 (1963), 196.

⁹³ Pflaum's figures for the number of offices are used throughout, with full knowledge that they are not exactly correct—in part, because of the methodological error of assigning the creation of a procuratorship to the reign of the emperor in which it first appears in the epigraphic record. Given the sporadic appearance of some offices in inscriptions, there is of course no reason to expect them to be recorded in extant inscriptions immediately after their creation. Millar (art. cit., 196) gives the example of the procuratorships of the vicesimae hereditatium.

⁹⁴ Pflaum, , Abrégé des procurateurs équestres (1974), 8 and 23Google Scholar.

⁹⁵ I wish to thank Dr. P. D. A. Garnsey, Miss J. M. Reynolds, Professor F. Millar, Mr. D. Cohen, Dr. R. Duncan-Jones and the Editorial Committee of JRS for their help. Special thanks are due Professor G. Iversen for advice and reassurance concerning statistical methods.