paul

In the Jewish War Josephus represents Agrippa II as delivering a speech in Jerusalem (dramatic date, May-June A.D. 66) designed to discourage the Jewish insurgents from embarking on a full-scale war against Rome (BJ 2.345-401). A large section of the speech (358-87) is devoted to a review of Roman power and resources. If such a speech was given at the time and place indicated, Josephus who was present in Jerusalem may well have heard it, and certainly when Agrippa, according to Jospehus, was sent a copy of the War, he replied testifying to its accuracy (Vit. 362; 364-7), but there really is no reason not to believe that the speech as reported was written by Josephus himself in accordance with the conventions of Graeco-Roman historiography; the idea, for example, that the Romans have achieved their empire with God’s help (2.390) can be found also outside the speech at 1.390 and 3.293. Two passages of this speech are of interest for the present discussion: in the first (383) it is claimed that the yearly produce (karpoi) of North Africa maintains the people of Rome for eight months, while in the second (386) Egypt is said to send grain (sitos) to Rome for four months.

These data have often been combined with the statement of the fourth century Epitome de Caesaribus, 1.6: huius (sc. Augusti) tempore ex Aegypto urbi annua ducenties centena milia frumenti inferebantur, resulting in a supposed annual import of grain to Rome in the first century of 60,000,000 modii. This way of treating the ancient sources had been magisterially rejected by Beloch in 1886 as “unmethodisch”¹ and Rostovtzeff showed that even if accepted the combination was difficult to interpret in the light of other information.² But the combination continued to be employed by some authors,³ though more recently there has been a tendency for scholars not only to disjoin the evidence of Josephus and the Epitome but even in some cases to suppose that the evidence of one or both sources is unreliable.⁴ It seems an appropriate time to take a closer look at the evidence of these two sources, first and briefly the Epitome, secondly and at greater length, Josephus’ report of Agrippa II’s speech, particularly since there is a tendency in many cases to seize upon the raw data provided by the sources without weighing the reliability of the contexts in which the data appear.

In his study of the Epitome, Schlumberger candidly concedes that the figure of 20,000,000 modii claimed as the import from Egypt to Rome in the time of Augustus “finds no confirmatory parallel in the whole ancient tradition,” yet he also maintains that “no reason can be found why it should have been falsified.”⁵ In support of this contention, in addition to citing the authority of Gardthausen and Johnson (above, n. 3), he ventures the suggestion that the figure originated with Augustus’ breviarium totius imperii (Suet. Aug. 101) or “a similar set of statistics from the Augustan period”, and that perhaps it was transmitted to the model of the Epitome and other similar works by Livy. These notions fail to convince: in the first place though the breviarium totius imperii (on which see further, below) is said to contain an account of vectigalia, all grain imported from Egypt did not fall into that category; secondly, the contents of the breviarium became publicly known only on Augustus’ death in A.D. 14 and Livy’s history ended in 9 B.C.; thirdly, there is no evidence for another Augustan set of statistics and though there are later attempts to record and publish the rationes imperii, “there is no sign later of any comparable effort either to collect or to publish full accounts of the resources of the empire.”⁶

Schlumberger, however, in his chapter “Epitome und Sueton”, has also successfully argued the case that the Epitome is heavily dependent on a fourth century Vorlage which in turn derives from Suetonius. Since Suetonius’ life of Augustus contains no reference to the supposed amount of 20,000,000 modii this datum must derive either from the author of the Epitome or more probably from his Vorlage. Since 20,000,000 is a suspiciously round figure and no one has yet accounted for the emergence of a genuine figure of that nature in a work of the fourth century, perhaps it should be seen as a rhetorical addition to the record, either as a protest against the diversion of Egyptian grain to Constantinople, with consequent difficulties for Rome,⁷ or as, to quote The Mikado, “merely corroborative detail, intended to give artistic verisimilitude to an otherwise bald and unconvincing narrative.” Schlumberger indeed argues that the Epitome preserves other rare and valuable pieces of information from the first century; (1) the name of the victor over Saturninus who conspired against Domitian, namely Norbanus Lappius (Epit. 11.10), and (2) the praenomen of Julius Vindex, namely Gaius (Epit. 5.6).⁸ Unfortunately it has been shown conclusively that there were two separate persons, Norbanus and Lappius,⁹ and secondly, even if Dio-Xiphilinus 63.22.1² is the only other extant source to give the praenomen Gaius, it is bound to derive from a Latin source, probably Tacitus in a lost portion of the end of Annals, and in any case it would be a likely guess that a Gallic noble family with the name Julius had been enfranchised by Caesar;¹⁰ and in PIR² of the Julii whose praenomina are preserved, Gaius is overwhelmingly the most common one. It is still to be demonstrated that the Epitome (which incidentally omits “Vindex”, a more significant element of the name) is as valuable a source for the first century as Schlumberger maintains, and hence there is no reason to regard its figure of 20,000,000 modii as particularly authoritative.

When we turn to Josephus’ we must note in the first place that the aim of Agrippa’s speech is to deter the people from war (2.343) and it is in his interest to magnify Rome’s power and resources. But also, Josephus’ statistics are often unreliable:¹¹ muria at 2.361 might be the rhetorical equivalent of “a large number” but we should recognize right away that this is not the style of an official breviarium and if such a document does underlie the speech, as is often maintained, it has been overlaid by Josephan additions; Asia (apparently the province since Bithynia, Cappadocia etc. are separately mentioned, 2.368) is credited with 500 cities (2.366), a number repeated, in a rhetorical context, only by Philostratus, VS 2.1.4 (548), in the time of Nerva; Pliny’s figure for the populi of Asia was 282.¹² Modern calculations yield lower totals still, e.g. 73 cities of Asia are said to have coined under Augustus and Tiberius.¹³ In the case of Gaul also there are disparities between Josephus and other sources: he speaks of 305 tribes (2.372) while Appian writes of the conquest of 400 tribes (Celt. fr. 1.6; cf. BC 2.150.627) and Plutarch of 300 (Caes. 15); also of almost more than 1200 cities (2.373), whereas both Plutarch and Appian refer to the conquest of more than 800 cities by Caesar.¹⁴ At 2.377 Josephus represents Agrippa as referring to the presence of 8 legions in Germany; the correct total for A.D. 66 was 7 and X Gemina was not moved to Germany until late A.D. 69 or A.D. 70.¹⁵ Another possible anachronism, noted by Smallwood, may be contained in the reference to the sending of hostages and the offer of submission by the Parthians (2.379; cf. 389). This seems to refer to the mission of Tiridates which did make its appearance in Rome in A.D. 66 but perhaps not early enough to have been referred to by Agrippa at the dramatic date of the speech.¹⁶ As noted by Ritterling, there is no reference in the speech to Raetia, Noricum and Moesia; more surprisingly there is no reference to Syria, the legions of which constituted the most obvious threat to the Judaean insurgents; Ritterling suggests that the situation in the East could be presupposed as known to Agrippa’s audience,¹⁷ but it would hardly be rhetorically effective to pass over the most pressing danger. There is almost no reference in the speech to auxiliary forces. These errors, omissions, exaggerations and possible anachronisms must be held to weaken the general reliability of the speech and to predispose the careful reader to distrust the data on the grain-supply from Africa and Egypt.

But before looking at the evidence for the grain-supply in more detail, it will be useful to examine what Josephus reports in this speech about the population of Egypt (2.385). He claims, on the basis of the poll-tax returns, that the population, excluding Alexandria, was 7,500,000. Now, as pointed out by A.C. Johnson, the poll tax was paid only by those between the ages of 14 and 65 and its amount varied widely from one community to another, so it was a very uncertain basis for calculating the size of the population.¹⁸ But why would Josephus choose to calculate the size of the population in this way, rather than refer to the census-returns (especially if, as often supposed, he were citing from an official document)? Johnson indeed assumes that Josephus also had access to the census data but if so, why would he need to refer to the poll-tax returns? Is it possible that the reference to the poll-tax has been introduced to lend a specious authority to a figure derived from other sources? According to Diodorus the population of Egypt was “of old” about 7 million and in his own day (following the reading of all mss. except M, which has a blank) no less than 3 million (1.31.8). For the figure of 7 million Diodorus is probably following Hecataeus of Abdera,¹⁹ and Josephus’ account may also go back to this author who wrote on the Jews as well as on Egypt (Jos. Ap. 1.183-204) and was interested in questions of population. Josephus’ figure should perhaps not be cited so readily as it frequently is as independent evidence for conditions in Egypt in his own day.

In the references in Agrippa’s speech to the grain-supply from Africa (“the third part of the inhabited world”) and Egypt (BJ 2.383 and 386 respectively) an immediate problem that arises is that if these two sources together account for Rome’s total yearly supply of grain (for eight months and four months respectively) there is no room for supplies from other quarters, e.g. Sicily, Sardinia, Spain and Gaul, not to speak of Italy itself, which we know to have continued to produce grain.²⁰ Perhaps this is one reason why Mommsen assigned to Africa the provision of only one-third of the annual amount of Rome’s grain-supply, the other reason being that BJ 2.382-3 could be read as including Egypt in Africa, so that one-third for Egypt should be deducted from the African total.²¹ In fact it does not seem likely that Egypt would be counted twice over, though there appears to be some confusion in Josphus’ account between the Roman province of Africa, where the legion was stationed, and the whole extent of Roman conquests in North Africa. Of course, it is not disputed that in the first century of our era Africa and Egypt were the major suppliers of grain to Rome (cf. e.g. Tac. Ann. 12.43.4) but clearly there is something wrong with the proportions cited in Agrippa’s speech. Must we then entirely abandon these data? Perhaps not. Reasons have been cited above for doubting whether the information in Agirppa’s speech can derive from an official breviarium totius imperii whether of Augustan or other date, but it is notable that with the exception of that for Germany the provincial legionary totals are exact. These totals, however, must have been a lively topic of discussion in Flavian circles in A.D. 69 and until the new régime was established (by which time, incidentally, the total of legions of Germany was eight) and Josephus could hardly fail to have had access to them. In addition, Vespasian in A.D. 69 was urgently aware of the importance of Egypt and its grain-supply (BJ 4.605-606) and when he advanced there to secure the province, he was accompanied by Josephus. When Titus (and Josephus) returned to resume the Judaean campaign he was joined by Ti. Julius Alexander, former prefect of Egypt, who remained with the Roman forces as praefectus castrorum (BJ 2.309; 5.45-46; 6.237). Ti. Alexander is described as “the most tried of all his (sc. Titus’) friends for loyalty and sagacity.” A native of Alexandria, Jewish by upbringing though he later renounced his faith (AJ 20.100), he was the son of the Alabarch and nephew of Philo, the Jewish philosopher.²² If anyone was likely to be well informed about Egypt and its relations with Rome, it would be Ti. Alexander and Josephus had abundant opportunity to consult him, not only in the Roman camp in Judaea, but also later at Rome where Ti. Alexander may have been appointed praefectus praetorio, perhaps even as colleague of Titus. There is a good chance, therefore, that with access to such a source, Josephus’ information about the amount of grain supplied by Egypt to Rome was exact. But as far as we know he had no informant of similar calibre concerning circumstances in Africa, though Vespasian had intended to secure the grain-supplies there as well (Tac. Hist. 3.48). Though there are no reliable figures about African grain production or about the amount of African grain sent to Rome, there is some indication that the expansion of the African grain-supply could only occur after adequate roads and harbours were constructed and after the creation of the large imperial estates there in the reign of Nero.²³ It also seems likely that the activities of Tacfarinas between A.D. 17 and 24 hawd some effect on the African grain-supply, though Tacitus’ narrative is scanty (Ann. 2.52; 3.20-21; 74; 4.23-25). Hence, though there is no doubt that Africa was a major supplier of grain to Rome in the first century of our era, the amount of exported grain credited to Africa in Agrippa’s speech is more likely to be exaggerated because Josephus ignores supplies coming from sources other than Africa and Egypt, African grain-production was probably still developing in the middle of the first century and as far as we know Josephus did not have access to the kind of information about Africa that he could have about Egypt.

Footnotes

1 Julius Beloch, Die Bevölkerung der griechisch-römischen Welt (Leipzig 1886) 411; cf. 31 and n. 2. Controversy on the question, however, has continued until the present.

2 M. Rostovtzeff, RE 7 (1910) 126-187 s.v. frumentum, esp. 136; cf. 133; also doubted for other reasons by U. Kahrstedt in L. Friedländer, Darstellungen aus der Sittengeschichte Roms, 9.10. Aufl. (Leipzig 1921) 4.18, n. 1, though his interpretation of Plut. Caes. 55 is mistaken.

3 Cf. V. Gardthausen, Augustus und seine Zeit 1.1 (1891) 457; 2.1 (1891) 244 n. 35; Tenney Frank, Economic Survey of Ancient Rome (= ESAR) 3 (Roman Africa — Haywood) 42-43; 5 (Rome and Italy of the Empire — Frank) 219; the wording in 2 (Roman Egypt — Johnson) 481-482 and in 5.140 is more circumspect.

4 Cf. e.g. H. Pavis d’Escurac, La préfecture de l’annone (B.E.F.A.R. 226) (Rome 1976) 170-171; K. Hopkins, Conquerors and Slaves (Cambridge 1978) 97-98; G. Rickman, The Corn Supply of Ancient Rome (Oxford 1980) 118; Appendix 4, 231-235; P. Garnsey in Trade in the Ancient Economy edd. P. Garnsey, K. Hopkins, C.R. Whittaker (Berkeley/Los Angeles 1983) 119; 201, n. 5.

5 J. Schlumberger, Die Epitome de Caesaribus (Vestigia, Bd. 18) (Munich 1974) 19 and n. 7.

6 Suet. Cal. 16.1; F. Millar, The Emperor in the Roman World (Ithaca, NY 1977) 267-268.

7 Cf. Rickman (above, n. 4) 198-199; H.P. Kohns, Versorgungskrisen und Hungerrevolten im spätantiken Rom (Antiquitas, Reihe 1, Band 6) (Bonn 1961) 41; 215-216; Claudian, Bell.Gild. 49-65.

8 Schlumberger (above, n. 5) 72, n. 42.

9 PIR² L 84; J. Assa, Akten des 4. internationalen Kongresses für griechische und lateinische Epigrafik (Vienna 1964) 31-39; B.W. Jones, Domitian and the Senatorial Order (Philadelphia 1979) 30-31 and n. 8.

10 See PIR² I 628.

11 See e.g. H. Drüner, Untersuchungen über Josephus (Diss. Marburg 1896) 36-42; H. Peter, Die geschichtliche Literatur über die römische Kaiserzeit (Leipzig 1897) 1.399.

12 NH 5.150; A.H.M. Jones, Cities of the Eastern Roman Provinces 2nd ed. (Oxford 1971) 506-508, argues that Pliny’s information is based on official documents published by Augustus and Agrippa.

13 D. Magie, Roman Rule in Asia Minor (Princeton 1950) 1.472; 2.1335, n. 16. Even in the second and early third centuries only 168 cities in Asia coined and only 153 are known to have had the organization of a polis, Magie, 2.1502, n. 25.

14 Cf. C. Jullian, Histoire de la Gaule (Repr. Brussels 1964) 1.180 and n. 1; 2.214-215, n. 1. Jullian’s total of tribes is about 500.

15 Ritterling RE 12.1 (1924) 1261-1262; 12.2 (1925) 1680 s.v. legio.

16 Dio 63.1-7; Suet. Nero 13; Josephus, Jewish War tr. G.A. Williamson, revised with notes by E.M. Smallwood (Penguin Books 1981) 160, n. 36.

17 Ritterling (above, n. 15) 1262.

18 ESAR 2 (Roman Egypt) 245.

19 See A. Burton, Diodorus Siculus, Book I: A commentary (Leyden 1972) 8. Most commentators accept Dindorf’s emendation of triakosion (sc. muriadon, i.e. 3,000,000) to heptakosion; see. e.g. J.K. Davies, CAH 7.1² 264 and n. 34.

20 Evidence in Rickman (above, n. 4) 101-108; 112-113.

21 T. Mommsen, Römische Geschichte 5¹⁰ (1927) 651-652 = The Provinces of the Roman Empire tr. W.P. Dickson, 2 (1909) 337. Mommsen gives no reason for his figures.

22 See PIR² I 139.

23 See Kahrstedt (above, n. 2) and Rickman (above, n. 4) 111; cf. F. De Martino, Wirtschaftsgeschichte des alten Rom, übers. v. B. Galsterer (Munich 1985) 500.