¹ I am grateful to Drs. Beryl Rawson, P.A. Gallivan and R. Develin for helpful suggestions and criticism. The remaining obscurities are my own. References to inscriptions are to CIL unless otherwise stated.

The following special abbreviations are used:

Boulvert, EAI = Boulvert, G.Esclaves et Affranchis impériaux sous le Haut-Empire romain : rôle politique et administratif (Naples 1970)Google Scholar

Boulvert, DF = Boulvert, G.Domestique et Fonctionnaire sous le Haut-Empire romain: la condition de l’Affranchi et de l’Esclave du Prince (Paris 1974)CrossRefGoogle Scholar

Chantraine, FS = Chantraine, H.Freigelassene und Sklaven im Dienst der rümischen Kaiser: Studien zu threr Nomenkldtur (Wiesbaden 1967)Google Scholar

Eck, SOI = Eck, W.Die Staatliche Organisation Italiens in der höhen Kaiserzeit (Munich 1979)Google Scholar

Hirschfeld, Verwalt.² = Hirschfeld, O.Die Kaiserlichen Verwaltungsbeamten bis auf Diocletian (2nd ed.) (Berlin 1905, repr. 1963)Google Scholar

Pflaum, CP = Pflaum, H.-G.Les Carrières procuratoriennes équestres sous le Haut-Empire romain (3 vols.) (Paris 1960–1)Google Scholar

Smallwood = Smallwood, E.M.Documents illustrating the Principates of Gaius, Claudius and Nero (Cambridge 1967)Google Scholar

Weaver, Fam.Caes. = Weaver, P.R.C.Familia Caesaris: a Social Study of the Emperor’s Freedmen and Slaves (Cambridge 1972).CrossRefGoogle Scholar

² For all its admirable thoroughness and the valuable insight it provides, a certain lack of feeling for chronology seems to me one of the weaknesses of Millar’s, Fergus monumental work, The Emperor in the Roman World (London 1977)Google Scholar, though I would certainly not go so far in criticism (or ‘bias’) as Hopkins, Keith in his review, JRS 68(1978), 178–86.Google Scholar

³ This view, which cannot be argued further here, is at variance with that espoused, for example, by Pflaum, H.G. in his fundamental works on the equestrian procurators, esp. Les Procurateurs équestres sous le Haut-Empire romain (Paris 1950),Google Scholar Les Carrières procuratoriennes équestres sous le Haut–Empire romain (Paris 1960–1), and in numerous subsequent articles. On the work of G. Boulvert, see further below. An admirable formulation of this position is by Brunt, P.A. ‘The Administrators of Roman Egypt’, JRS 65 (1975), 124–47, at 141 f.Google Scholar

⁴ For his earlier volume, Esclaves et Affranchis impériaux sous le Haut-Empire romain (Naples 1970), see the reviews by Chantraine, Gnomon 48 (1976), 477–82,Google Scholar and by Weaver, Labeo 18 (1972), 218–26.Google Scholar It follows the succession of publications in this field during the last decade and a half (see Gnomon 41 [1969], 173 ff., 421 ff.; 48 [1976], 482 ff.) which have transformed, or rather virtually created, the field itself and drawn attention to its intrinsic and crucial importance for the study of the administration and society under the whole period of the early empire. As with EAI, this latest work (DF) is based on massive use and citation of all the relevant sources. It contains a full discussion of the modern literature from Mommsen and Hirschfeld onwards, including an amount of specialist legal material.

⁵ As is well brought out by Hopkins, Keith in the section on political eunuchism in his recent book Conquerors and Slaves (Cambridge 1978).Google Scholar

⁶ See, in general and for examples, Fam. Caes. 269 f. and further, Section VIII below.

⁷ For the acquisition by purchase of slaves for the Imperial familia, see Appendix below.

⁸ Cf. Chantraine, FS 297.

⁹ DF 15 n. 34.

¹⁰ See JRS 558(1968), 110ff.; Fem. Caes. 149 ff. and further Section VI below.

¹¹ In any case, the fiscus castrensis can only serve as a terminus post quem, i.e. after the date of its foundation, which could be Flavian rather than Claudian.

¹² See Fam. Caes. 46–7; Chantraine, FS 264–8, esp. 268.

¹³ DF 121 n. 63; cf. EAI 83 n. 476; 177 n. 593; 182 n. 640; 246 n. 320–1. Also Smallwood, No. 185.

¹⁴ Cf. Chantraine, FS 26–30, 32–4; Fam. Caes. 22.

¹⁵ DF 123 n. 73a; cf. Chantraine, FS 20 n. 20; 21 n. 27.

¹⁶ DF 121 n. 63.

¹⁷ Some examples: 10. 6324 = ILS 1734: ‘Ti Claudius Aug. 1. Quir. Alcibiades, qu[i fuit?] praegustator et a c[u]biculo Neronis’; 14. 3644 = ILS 1942: ‘C. Iulius Aug. 1. Sam[ius], accensus Divi Claudi et Neronis Augusti’; 6. 3970: ‘Coetus Herodianus, praegustator Divi Augusti’; 6. 1887: ‘Fortunatus Aug. 1. …ab epistulis, accensus patron(o) Divo Aug. Vespasiano, lictor curiatius…’. Cf. also 8.12657 = ILS 1744: ‘Valentinus ex numero cubiculariorum Aug(usti)’.

¹⁸ E.g. 3. 536 = ILS 1575: ‘Theoprepes Aug. lib., proc. domini n(ostri) M. Aur. Severi Alexandra Pii Fel(icis) Aug(usti)’; 6. 8498 = ILS 1738: ‘M. Aurelius Augg. lib. Prosenes, a cubiculo Aug., proc. thesaurorum, proc. patrimoni, …’; and for private procurators of members of the Imperial family other than the emperor: 6. 8500 = 11. 1753 = ILS 1490: ‘Domitius Lemnus, proc. Germanici Caesaris’; 10. 1738: ‘Cn. Domitius Chrysanthus, proc. Domitiae Domitianf. On the use of the title ‘proc. Aug(usti)’ by freedmen, see Fam. Caes. 268; and (most recently) Chantraine, Historia 24 (1975), 613 ff.Google Scholar

¹⁹ Cf. Pflaum, , Proc. Eq. 10 ff.;Google ScholarCP passim.

²⁰ Degrassi, (Doxa 2 [1949], 103)Google Scholar commented that the plural ‘a paedagogis’ shows that there may have existed more than one Imperial paedagogium for the training of young Imperial slaves in Rome in the period in question. This should not be pressed, but if true, the appointment of a procurator in charge of a system of training institutions, rather than a single establishment, could point to a later rather than earlier date.

²¹ For further details, see Fam. Caes. 225.

²² DF 16 n. 49.

²³ Tac. Ann. 14. 65. 1; Dio 62. 14. 3; Suet. Nero 35. 5; cf. Oost, S.I. ‘The Career of M. Antonius Pallas’, AJPh 79 (1958), 113–39, at 138.Google Scholar

²⁴ Cf. JRS 58 (1968), 110 ff. and further in Section VI below.

²⁵ Fam. Caes. 233 n. l; cf. 231.

²⁶ Syria 34 (1957), 278–84 at 281 = IGLS 5 (1959) 1998 = AE 1958, 236 = SEG 17 (1960) 755.

²⁷ Pavis d’Escurac, H.La Préfecture de L’Annone (Paris 1976), p. 329;Google ScholarBradley, K.R. ‘Claudius Athenodorus’, Historia 27 (1978), 336–9;Google Scholar against Millar, F.JRS 53 (1963), 199.Google Scholar

²⁸ As suggested by Pflaum, CP 964–5 (No. 49 bis).Google Scholar

²⁹ Bradley, op. cit. 339.

³⁰ Pflaum, CP 964–5;Google Scholar Pavis d’Escurac, 329 f., 387 f.

³¹ He could not have been freed even earlier, by Pallas, as suggested by Griffin, M.T.JRS 52 (1962), 105,Google Scholar as his name would then have been simply ‘M. Antonius Pallantis 1 (ibertus) Carpus’.

³² See Fam. Caes. 97 ff., esp, 102 f.

³³ Fam. Caes. 212 ff.

³⁴ Cf. 6. 18358: Fl(avia) Aug. lib. Helpis Caenidiana, who is married to Callistus Aug. lib. Hyginianus and has a daughter Ulpia Calliste.

³⁵ The term ‘verna’, while it need not indicate slave status, is generally associated with slaves and freedmen and their familia, whether lib. liberti or freeborn, and is often used of those who die at an early age. See Starr, C.G.CPh 37 (1942), 314;Google Scholar and the 112 examples given, exempli gratia, by Chantraine, FS 171 n. 137.

³⁶ As by Boulvert, DF 296 n. 182 who argues that enrolment in the tribe Palatina is sufficient indication of legitimate birth; cf. ibid. n. 183, for examples from Quirina, including ILS 1535. In the case of 10. 6666 (Antium): ‘Eros Aug. 1. procurator Caenid(i)anus cum M. Antonio Quii. Candido filio suo’, it is difficult to accept that the freeborn son, Candidus, derives his nomen from his father Eros rather than from his mother. She is not mentioned (as being irrelevant) in an inscription dedicated to tribuni of an Imperial domestic collegium at Antium. Eros, a former slave of Antonia Caenis (who was in turn a freedwoman of Antonia Minor) would have had to pass back into the familia of Antonia Minor, have been freed by her and subsequently, as Aug(ustae) l(ibertus), have found his way into an Imperial procuratorship. I do not follow Chantraine (FS 305–6) who suggests ‘Aug(usti) Kibertus)’ would then have to be understood in a wider sense (‘in weiterem Sinne’), whatever that could mean.

³⁷ See Fam. Caes. 37 ff.; Chantraine, FS 101 ff.

³⁸ It is not proposed here to multiply references to modern authorities, beginning with Momigliano, A.D.Claudius: The Emperor and his Achievement (Oxford 1934, repr. 1961),Google ScholarScramuzza, V.The Emperor Claudius (Cambridge, Mass. 1940),Google Scholar through to Boulvert, EAI 91 ff., 333 ff. Full bibliographies are given in each of these works. The literature on Claudius is enormous.

³⁹ Momigliano, 46.

⁴⁰ EAI 341–73.

⁴¹ Also included by Dessau among the ‘Tituli domus Augustae condicionis libertinae et servilis’.

⁴² Cf. Weaver, Historia 14 (1965), 511 n.21;Google ScholarBoulvert, EAI 105 n. 70.Google Scholar

⁴³ Cf. Chantraine 165 f., 168 f. who plausibly conjectures that the Greek form of the status indication influenced this order.

⁴⁴ This inscription from Tusculum, which runs: dis manibus/Ciaudiae Primigeniae/ lib. Aug. benemerenti/Mamilia Albana/filiae/suae. Vixit annis XV/mensib(us) IIII, exhibits other unusual features. The mother, Mamilia Albana, if freeborn (as she presumably is), cannot have given birth to a slave-born daughter, nor indeed, if she were a freedwoman of a Mamilius or Mamilia at the time of the birth. Even if the mother were a slave then, the name of her daughter cannot be as stated unless she were sold or transferred to the ownership of the emperor while still a baby or very young. Claudia Primigenia is not only unusual in her name, but also as a woman with status indication in the Familia Caesaris, and as a freedwoman, into the bargain, manumitted at the very early age of 15 years. The only hypotheses I can suggest to account for this nomenclature are either (1) that Claudia Primigenia is not the daughter (‘filia’) of Mamilia Albana at all, but her step-daughter or foster-child and that she was born of an Imperial slave mother; or (2) that Primigenia was born of a freeborn mother cohabiting with an Imperial slave, who was her father, and that she then suffered the reduction to slave–status that was the penalty that could be imposed for children of such unions under the Senatus Consultum Claudianum of A.D. 52. (Gaius 1.84; Tacitus, Ann 12.53.1. For a discussion of the (controversial) S.C. Claudianum and its effects on status of children and wives of Imperial slaves, see Fam. Caes. 162 ff.; and for a discussion of the role of women in the Imperial familia and the distribution there of females/males with status-indication, ibid. 170 ff., esp. 173 with Table IX.) In any event 14.2690 is a remarkable inscription.

⁴⁵ See Fam. Caes. 76 ff.; and the exhaustive treatment of Chantraine, 281 ff.

⁴⁶ See Fam. Caes. 87 ff., esp. 89; also CQ 14 (1964), 311–15; Chantraine 128 ff.

⁴⁷ These personal cognomina, as used by e.g. L. Aurelius Marcianus Aug. lib. (9. 5828), are to be clearly distinguished from those second cognomina (or agnomina) in -ianui, used in addition to the normal cognomen and which reflect the name of the previous owner of the slave before he was transferred to the familia of the manumitting owner (see Section III above) e.g. C. Iulius Divi Aug. 1. Delphus Maecenatianus, whose personal cognomen was Delphus and whose former master was Maecenas.

⁴⁸ For further details and references, see Weaver, CQ 14 (1964), 314;Google Scholar Chantraine, FS 128 f.

⁴⁹ For a recent discussion of the significance of ‘Aug(usti)’, ‘Aug(ustorum)’ in the titles of Imperial freedmen, see Chantraine, H. ‘;Zur Nomenklatur und Funktion-sangabe kaiserlicher Freigelassener’, Historia 24 (1975), 603–16, esp. at 613 ff.,Google Scholar where it is argued that the discrepancy in the use of ‘Aug.’ in occupational titles by equestrians and Imperial freedmen represents not a ‘distinction of principle’, as I had previously argued (Fam. Caes. 278, cf. Historia 14 [1965] 465 n. 29), but rather a difference in practice, to be viewed in the context of other aspects of nomenclature. The earliest case is 6. 9015 = 29847a = ILS 8120: ‘Ti. Claudius Aug. lib. Eutychus proc. Augustor(um)’ which raises difficulties over the interpretation of ‘Augustorum’ in the nomenclature of Imperial freedmen. In status nomenclature I have argued (Fam. Caes. 58 ff., cf. Historia 13 [1964] 188–9) for joint manumission or dedication of inscription in a period of joint rule, while conceding that in occupational nomenclature ‘successive’ rather than ‘joint’ emperors can be implied. Chantraine (FS 224–63, esp. at 243 ff.; 24 [1975], 603 ff.), on the other hand, argues strongly for ‘Augustorum’, in periods other than joint rule, as referring always to ‘successive’ emperors, both in status and occupational nomenclature. Without wishing to pursue the argument further here, in the example mentioned above (ILS 8120), it still seems to me significant that there is a discrepancy between the status nomenclature and the occupational title of Ti. Claudius Eutychus, who is ‘Aug(ustt) lib(ertus)’, but ‘proc(urator Augustor(um)’, with the singular used in the one case and the plural in the other, despite the fact that Eutychus would have been manumitted earlier than his being appointed procurator and thus been a freedman under more emperors than he may have been a procurator.

⁵⁰ Domitius Lemnus was fonnerly dated to the reign of Domitian (e.g. Dessau, ad loc.; Hirschfeld, Verwalt.² 40 n. 3) and thus included by McCrum, and Woodhead, in their Select Documents of the Principates of the Flavian Emperors (Cambridge 1961) No. 201;Google Scholar but he really belongs under Nero, and so appears in Smallwood, as No. 179. For the arguments, based on personal nomenclature and occupational titles, see Weaver, Historia 14 (1965), 509–12;Google Scholar Chantraine, FS 65 n. 15.

⁵¹ For the rationale behind what on the surface appears to be a quite extraordinary cumulation of posts, one in Rome and the other in Ephesus, see the discussion in Pflaum, CP p. 269 f.Google Scholar Pflaum, plausibly; suggests that Eudaemon first held the post proc. hereditatium when Hadrian was in Rome during the period 125–129, and then accompanied the emperor on his travels to the East, where during his stay in Ephesus he also was in charge of the Imperial finances in the province of Asia. After this Eudaemon accompanied the emperor to Antioch, 129–130, where he held his next post of proc: Syrlae.

⁵² His wife was Ulpia Euhodia, which implies a date no earlier than Trajan. His inscription runs: Ulpiae Euhodiae/coniugi optimae/T. Flavius Aug. lib./Delphicus/ tabularius a ratio[n(ibus),[p] roc. ration(um) thesaurorum,/hereditatium,/fisci Alexandrini. This wording has led some to infer a cumulation of all three posts, but erroneously. Cf. Boulvert, DF 122 n. 69, 131, 136, 140.

⁵³ There is also ‘P. AeliusHikrus Augg. lib. qui proc(uravit) Alexandriae ad ra(tionem) patrimonii’, presumably in Egypt, but his inscription (14. 2504 = ILS 1491) comes from Tusculum, near Rome. His date could be anywhere from c. 140 to 169.

⁵⁴ Hirschfeld, Verwalt.² 41 n. 1;Google Scholar Boulvert, EAI 104 n. 66.

⁵⁵ See the discussion by Chantraine, FS 259 f.

⁵⁶ Cf. Fam. Caes. 241 ff. Some 150 tabularii are attested in the Familia Cae saris. From the Julio-Claudian period the overall proportion of those that could be still of slave status drops from just over 10% to under 5%.

⁵⁷ For a similar consistency in the age-at-deafh figures for dispensatorei, see Fam. Caes. 225 f.

⁵⁸ 6. 8467: ‘T. Fl(avius) Aug. lib. Epagathus, tabular(ius) viae Salariae’, who died aged 48. His wife was Ael(ia) Plusias, hence probably at least second century. The others are:

6. 8466 = ILS 1606: ‘M. Ulpius Aug. lib. Eutychus, tabul(arius) viae Appiae’. Died aged 40 years under Traian or Hadrian.

6. 8476 = ILS 1544: ‘Donatus Augustorum (ser.), tabularius fisci frumentar(ii)’. Died aged 29 years. His father was P. Aelius Aug. lib. Donatus and his mother Aelia Caenis. A date under M. Aurelius is indicated.

6. 8484: ‘T. Flavius Aug.l. Celadus, tabularius marmorum Lunensium’. Died aged 37 years.

6. 8488 = ILS 1607: ‘Salvius Aug. lib., tabularius aquarum’. Died aged 29 years 10 m. No date indicated.

6. 37744: ‘P. Aelius Aug. lib. Epaphrod[itus], tabularius rationis castrens[is]’. Died aged 33, probably under Hadrian; wife: Flavia Margaris.

14. 4482–3: ‘Ingenuus Aug. lib., tabularius portus Aug(usti)’. Died aged 29 years 10 m. His wife was Flavia Crispina and his brother Flavius Faustus — hence Flavian-Trajan in date.

To these should be added 6. 9072: ‘Catervarius Augg. lib. ex tabular(iis)’ whose wife, Lucida Augg. vern(a) died aged 33 years, after 15 years of marriage. Hence, he himself is likely to have been at least 35, perhaps over 40, as his retirement from the post (ex tabular [iis]) has already occurred.

⁵⁹ 8.12595, 12596, 12882, 24702; 2. 3235 = ILS 1555.

⁶⁰ See Fam. Caes. 246 ff.

⁶¹ Quite the contrary. Pallas showed every sign of increased influence, status and wealth after Agrippina’s marriage to Claudius, reaching the high point of his career in 52 with the honour of ornamenta praetoria and the offer of HS 15 million (Tac. Ann. 12. 53. 1 f.), which he could afford to decline. See further, on Pallas, Oost, S.I.AJPh 79 (1958), 113–39,Google Scholar esp. 126 ff. with 128 n. 37.

⁶² At best in this period one can admit the distinction between the public state treasury, the aerarium, and the congeries of fisci making up the emperor’s financial administration, which came to be called the fiscus, over which the a rationibus exercised controlling interest, if not accountability. But a further distinction within the Imperial fiscal administration between the a rationibus dealing with the emperor’s public revenues and the procurator a patrimonio, presiding over his private financial interests, seems to me prima facie unlikely for the Julio-Claudian period, even if such a distinction could be satisfactorily drawn. (For a different view, see Boulvert, EAI 96 ff., esp. 103 n. 64 bis.) From the well-known description by Statius in Silvae 3. 3. 86–105 of the duties of the a rationibus, little can be gained about the relationship between the patrimonium and the central financial bureau even in the time of the early Flavians and less in the time of Claudius under Pallas, and in the reign of Nero under Phaon. (For Phaon, Fam. Caes. 259, 289; Boulvert, EAI 97 n. 37.) Some of the items referred to — gold from the Iberian mines (89), corn from Africa (91–2) — can be assumed to derive from properties belonging to the patrimonium; others from Imperial provinces (90–1); others again from neither, e.g. ivory from India (94–5). As Boulvert points out (EAI 104 n. 67), Statius is not distinguishing particular categories of materials by reference to their point of origin, except in a very general sense. He is rather seeking to convey the empire–wide scope of the revenues to be accounted for by the central financial accounts office, the a rationibus. Cf. on lines 99 ff., Eck, SOI 75 f.

⁶³ Boulvert, EAI 104 η. 66, 105 n. 75.

⁶⁴ On the omission of status indication by women in the Imperial familia see Fam. Caes. 173 ff.

⁶⁵ If Mia Prima were an Imperial freedwoman, she could only have been manumitted by Gaius or, conceivably, Augustus — because of the praenomen borne by her son, C. Iulius Felix, which would have to derive from that of the manumittor, unless he was bom after his mother’s manumission. The emperor Tiberius is ruled out for this reason. So too, for practical purposes is Augustus, on grounds of the long time interval since A.D. 14.

⁶⁶ See Fam. Caes. 102 f., 112 ff.

⁶⁷ For full discussion, see Fam. Caes. 112–36.

⁶⁸ For details and discussion, see Fam. Caes. 97–111.

⁶⁹ In 6. 8507, where he is simply Primigenius Aug. 1., tabularius patrimoni, he is also named ‘tutor’. It is perhaps suggestive that the initial name on the funerary inscription following on the same stone is C. Flavius C. f. Quir. Firmianus, who died aged 5 years and 4 months. Clearly freeborn and possibly (but by no means certainly) the ward of Primigenius.

⁷⁰ Boulvert, EAI 105 n. 75: ‘une inscription…qui n’est donc certainement pas postérieure au règne de Claude’.

⁷¹ On the hereditates, see esp. Hirschfeld, Verwalt.²110 ff.Google Scholar

⁷² Boulvert, EAI 137 f.

⁷³ Already noted above, pp. 82 f.

⁷⁴ As by Boulvert, EAI 137 n. 299.

⁷⁵ See the discussion in JRS 58 (1968), 110 ff., and also Fam. Caes. 40, 149 ff.

⁷⁶ The nomenclature of this inscription is discussed in detail in JRS 58 (1968), 111 f.

⁷⁷ EAI 93 n. 9, 97 n. 38, 193 n. 699; DF 121 n. 63,124 n. 74,167 n. 352. Boulvert follows the prevailing view that Bassus’ nomen must be ‘Claudius’. However, Chantraine, FS 116, leaves open the possibility of earlier marriage to explain the nomen of his son. This is much the most likely explanation. Cf. Fam. Caes. 254 ff.

⁷⁸ See Fam. Caes. 21 f.; Chantraine, FS 15 ff.

⁷⁹ Fam. Caes. 22; Chantraine, FS 34 and 331 no. 276, where he instances NS 1922, 413 n. 25: Cerdo Ti. Claudi Germanici Aug. 1.

⁸⁰ See refs. in Chantraine, FS 21 n. 24. The reference to 6. 8473 is a tabeltarius, who is sub–clerical.

⁸¹ Discussed briefly in Fam. Caes. 34 with n. 1. The following is a slight modification of the age limits there given. In lines 7–8, the possible reading ‘fil(io)/b(ene) m(erenti)’ is to be rejected, as this would make Urbanus the father of Chrysogonus Lesbianus; the latter is, however, a ‘vicarianus’, a former slave of a slave–born Lesbius, (cf. Chantraine, FS 321, No. 205) subsequently admitted into the Familia Caesaris. This would be most improbable if his father was already in the Imperial familia.

⁸² On adiutores as junior clerical in status, see Fam. Caes. 236 ff., and on adiutores tabulariorum and adiutores a commentariis, ibid. 239 f. This latter group are characteristic of the provincial centres and, to a more limited extent, of minor departments in Rome (as here), but not of the large Palatine bureaux.

⁸³ 6. 8449 = ILS 1552 (see Section VIII below); 6. 8422; cf. Weaver, PCPhS 10 (1964), 91.Google Scholar

⁸⁴ See Fam. Caes. 54 ff.; Chantraine, FS 193 ff.

⁸⁵ In the slave nomenclature of privati, ‘n(ostri)’ makes an earlier appearance and there are Neronian examples connected with the Imperial familia: e.g. 6. 11242–3: Agathopus Actes η. ser.; cf. the dedication from the Thracian Chersonese ‘balneum populo et familiai Caesaris n(ostri)’ dated to A.D. 55.

⁸⁶ Weaver, CQ 15 (1965), 323 ff.;CrossRefGoogle ScholarFam. Caes. 35 f.; Chantraine, FS 77–89;Google Scholar Boulvert, EAI 95 f. with n. 29; DF 41 ff. These works give references to earlier discussions (and solutions) of the problems, going back to Hirschfeld and before, which it is unnecessary to record here.

⁸⁷ It is true that Seneca, Apocolocyntosis 15, describes Claudius’ final fate with the words: ‘Caesar ilium [i.e. Claudius] Aeaco donat; is Menandro liberto suo tradidit, ut a cognitionibus esset’. This has been taken by Hirschfeld (Verwalt.² 329 n. 5), Besnier, (Les Affranchis impériaux à Rome de 41 à 54 après J.C. [Paris 1947–8], 150),Google Scholar Boulvert (EAI 96), and others, to indicate the existence of such a post in the Imperial chancellery, if not a fully-fledged bureau a cognitionibus, under Claudius, who is assumed to have created it. Mommsen (Staatsr.³ 2. 926 n. 1) disagreed. The difficulty is, however, that Seneca assigns Claudius to be a cognitionibus of a freedman, thereby implying the existence of such posts in the households of privati as well as the Imperial familia, no doubt ever since Augustus. (Cf. Besnier, op. cit. 153.) A similar difficulty in regard to the ab epistulis, a libellis and a rationibus is raised in the charges brought against the Torquati Silani in A.D. 64 and 65 (Tac. Ann. 15. 35; 16. 8), where it is clear that these ‘nomina summae curae’ are merely the pretext for judicial murder, not the substance of the charges.

⁸⁸ Not 60 years, as stated erroneously in Fam. Caes. 35.

⁸⁹ DF 41–3.

⁹⁰ A distinction which he maintains (DF 42 with n. 250), and reproves Chantraine for ignoring (FS 78); differently expressed, however, in EAI, although equally hesitantly: ‘son nom est sans doute tiré de celui d’Aelius le père d’Aelia Paetina’ (ib. 95).

⁹¹ Boulvert, DF 42, rightly rejects the hypothesis of E. Cuq, in his Etudes d’épigraphie juridique (1881), that Theodotus was, as a slave, part of the dowry of Paetina to Claudius, and subsequently manumitted by Claudius — hence his status indication ‘Aug(usti) lib(ertus)’ — with the consent of Paetina, the nomen ‘Aelius’ being retained as a mark of such consent. Other fanciful solutions have been proposed, too many to be discussed further here.

⁹² As Atticus conferred on his freedman Dionysius the praenomen of his friend M. Cicero (Cic. Att. 4. 15. 1).

⁹³ The basic discussion of this question and all the relevant cases is now Chantraine, FS 77–100, esp. 89 ff. Boulvert (DF 43 n. 258), in support of his case that certain irregularities are possible, calls attention to Suetonius, Ner. 32. 2: ‘e libertorum defunctorum bonis…qui sine probabili causa eo nomine essent, quo fuissent uUae’ familiae quas ipse [Nero] contingeret’. This does not mean that certain freed-men of Nero without plausible reason took the nomina of families related to the Imperial family, i.e. that they changed their nomen. They need not, and could not. Imperial freedmen of the period would have such nomina anyway; i.e., besides ‘Claudii’, many would be ‘lulii’ (e.g. ‘Ti. Iulius Aug. 1.…, the father of Claudius Etruscus), ‘Antonii’ (e.g. M. Antonius Pallas), or ‘Domitii’ (e.g. L. Domitius Lemnus). These must be the nomina ‘quo fuissent ullae familiae quas ipse contingeret’; (for a similar use of ‘contingere’ by Suetonius in the sense of ‘being related to’, cf. Galb. 2. 1 : ‘Neroni Galba successif nullo gradu contingens Caesarum domum.’). Nor are meant here freedmen with these same nomina but belonging to families unconnected with the Imperial house. The numbers of these would have been legion. Chantraine (FS 81 f.) connects this passage with Pliny Eld. NH 12. 12 and the ease of Thessalicus, the eunuch freedman of M. Claudius Marcellus Aserninus, who after the death of his patron irregularly attempted to claim status as freedman of the emperor, potentiae causa. He suggests that freed-men with Imperial nomina such as ‘Iulius’, ‘Claudius’, etc. were making use of the Imperial status indication ‘Aug. lib.’ to which they were not entitled, attempting to cash in on the status it would confer. He compares also Tacitus, Hist. 2. 92. 3, wherein some freedmen of exiled patrons had attempted to transfer themselves to the Imperial familia and become ‘ipsis dominis potentiores’.

This may well be right. But the passage should also be seen in its immediate context: Nero, desperately short of money, resorts to desperate measures — ‘calumniis rapinisque intendit animum’. The pluperfect tense of ‘fuissent’ should refer to previous nomina in the Imperial familia, i.e. ‘Iulius’, ‘Livius’, ‘Antonius’, ‘Domitius’, etc. Nero, in predatory and irrational mood, is penalizing the estates of those freedmen who had been manumitted by earlier emperors or members of the Imperial family as well as of those who had irregularly claimed membership of the Imperial familia (presumably without the knowledge or consent of their real patrons), and possibly even of those freedmen who simply had such nomina as ‘Iulius’, ‘Antonius’, etc. The executors of such freedmen, whatever status they had laid claim to, would have to prove this claim (‘probabili causa’) — presumably no easy matter under these circumstances — or suffer confiscation of five-sixths of the estate.

⁹⁴ Cf. Chantraine, FS 83.

⁹⁵ See the references collected by Chantraine, FS 88 n. 112.

⁹⁶ See further, CQ 14 (1964), 314 f.

⁹⁷ For full lists of names of wives of Imperial freedmen and slaves, arranged in chronological order, see Fam. Caes. 307–13 (Appendix III).

⁹⁸ Cf. CQ 15 (1965), 323 ff.; Fam. Caes. 35 f.; Chantraine, FS 77 ff., who also discusses a third case of the same conflicting nomina, and exceptional use of Imperial status indication, NS 1917, 7: Ti. Claudio Aug./lib. Eutrapelo/patri piissi/mo et dulcis/simo T. Aelius/Aug. lib. Paris/filius b.m.f. At least in this case the father has the earlier Imperial nomen and the son the later one, not the other way round. Eutrapelus is in fact (as NS 1917, 6 reveals) an Augusti liberti libertus, the freed-man of an Imperial freedman, Ti. Claudius Aug. 1. Paris and the irregularity consists in his use of the direct Imperial status indication ‘Aug. lib.’ where one might have expected at least ‘Aug. l(ib). l(ft>).’; e.g. 6. 5909: C. Iulius Augusti l(iberti) l(ibertus) Priamus; cf. 6. 20002, 32450; 14. 2302; AE 1953, 24 = Gordon, Alb. 1. 90: M. Livius Augustae liberti libertus Tanais, dated to A.D. 45(?). Freedmen of freedmen could survive the original manumission of their patron by very long periods — as is the case with Eutrapelus here.

⁹⁹ On the whole question of the S.C. Claudianum, see Fam. Caes. 145 ff., esp. 162 ff.; cf. Chantraine, FS 86 f.; Boulvert, DF 309 ff.

¹⁰⁰ FS 86 f.

¹⁰¹ Which should be added to the list given in Fam. Caes. 75; cf. Chantraine, FS 150 n.48.

¹⁰² For detail and list, see Fam. Caes. 145 ff.; 156 ff. Cf. Boulvert, DF 309 ff.

¹⁰³ See Crook, J.A.CR 17 (1967), 7 f.Google Scholar

¹⁰⁴ For the full argument on this difficult anomaly, see Fam. Caes. 162 ff.

¹⁰⁵ Momigliano, Claudius 46; for a Claudian date, but duly cautious about generalizing for every province from the one example of Achaea, Boulvert, EAI 133 ff.

¹⁰⁶ On the vicesima hereditatium, see now esp. Eck, W.SOI 125–45;Google Scholar and on its administration, 129 ff.

¹⁰⁷ 6. 5554 = ILS 1547: ‘Τ. Aelius Aug. lib. Agathopus, proc. XX heredit(atium)’. The fragmentary Greek inscription, CIG 2980 (Ephesus), claimed by Boulvert (EAI 135 f. with n. 291) to prove the existence of a freedman procurator there from the time of Claudius, is to be assigned rather to an equestrian career from the late second century. See Eck, W.Chiron 5 (1975), 387;Google Scholar Pflaum, CP No. 193: M. Aurelius Mindius Matidianus Pollio.

¹⁰⁸ 6. 31032 = ILS 1418; Pflaum, CP No. 54.

¹⁰⁹ AE 1973, 485, first published by Z. Tasliklioglu in 1971; cf. Moretti, L.RFIC 102 (1974), 454–8;Google ScholarEck, W.Chiron 5 (1975), 365–92 (with full discussion of the equestrian career).Google Scholar

¹¹⁰ See Eck, SOI 129 ff.

¹¹¹ On the other hand, while a Flavian date cannot be ruled out for 6. 8475 = ILS 1542: d.m./T. Flavio/Apollonio/ a libellis f(isci) f(rumentar(ii)/Secundus/ Caesaris/nostri ser./Crescentianus/dispfensator) XX/hereditat(ium), it can just as easily be Trajanic, but not later. Cf. Boulvert, EAI 258 n. 423; Eck, SOI 131 n. 91 thinks the date is probably Flavian.

¹¹² SOI 116f., 131.

¹¹³ Cf. Moretti, L.RFIC 102 (1974), 454ff.,Google Scholarcriticized by Eck, Chiron 5 (1975), 391f.Google Scholar and SOI 131 n. 90.

¹¹⁴ See Section I above; for the age data on freedman procurators with references, see Fam. Caes. 225 f.; 269 f.

¹¹⁵ Fam. Caes. 270.

¹¹⁶ 116For full details of his extraordinary career, see Fam. Caes. 284 ff.

¹¹⁷ Cf. Smallwood, No. 182 = 6. 32775 = 33131 = ILS 2816, Ti. Iulius Aug. lib. Xanthus, manumitted by Tiberius, but surviving to at least the reign of Nero to become subpraefectus classis Alexandriae.

¹¹⁸ AE 1968, 489 = Knibbe, D.JOAI 47 (1964–5) Beibl. col. 25–8.Google Scholar

¹¹⁹ AE 1972, 574 = Knibbe, D.JOAI 49 (1968–71), Beibl. col. 19 ff.Google Scholar

¹²⁰ In the early second century A.D.C. Iulius Photinus Celer, with his nomen ‘Iulius’ and his two cognomina, cannot be an Imperial freedman. Nor is it clear how he could be an equestrian adhttor of an Imperial freedman procurator. On equestrian adiutores, see Pflaum, CP 1264 s.v. ‘adiutor’; cf. Fam. Caes. 231 ff.