Hostname: page-component-586b7cd67f-2brh9 Total loading time: 0 Render date: 2024-11-30T23:48:20.537Z Has data issue: false hasContentIssue false

Taxatio and Pollicitatio in Roman Africa*

Published online by Cambridge University Press:  24 September 2012

Peter Garnsey
Affiliation:
University of California, Berkeley

Extract

Numerous inscriptions from different parts of the Empire, and particularly from North Africa, testify to the munificence of the urban aristocracy. By munificence I mean the spending of wealthy individuals on, for example, monuments or projects of construction of one sort or another for their cities, or handouts of money, food or other commodities to their fellow-citizens. The significance of munificence from an economic viewpoint can be readily appreciated: it would not be an exaggeration to say that the prosperity of the cities rested in large part on the generosity of their leading citizens. In this paper, however, I will be concerned with some of the political and institutional implications of munificence. If the cities were financially dependent on their aristocracies, then the possession of wealth and the willingness to spend would clearly hold the key to both the acquisition and the retention of power. At the same time, we might expect some form of public control to have been exerted over aristocratic spending. In this connection, it may be significant that much of the expenditure of which we have record was incurred by individuals when they assumed magistracies or priesthoods or entered the local council. When would the wealthy have more readily submitted to financial levies than in the context of election victories? On the other hand, would it have been necessary to exact contributions from successful politicians, who would perhaps have shown their gratitude—and self-esteem—without any prompting through some form of public expenditure ? If any were inclined to hesitate, would they not have responded when reminded of the liberality of their predecessors ? In fact, it is not difficult to show that the dictates of the law, the weight of custom, and personal considerations and motives are all relevant; to determine the relative importance of the three factors is a more formidable task.

Type
Research Article
Copyright
Copyright ©Peter Garnsey 1971. Exclusive Licence to Publish: The Society for the Promotion of Roman Studies

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 The best and most complete account of the expenditures of the urban aristocracy in North Africa and Italy is that of Duncan-Jones, R. P., PBSR 17 (1962), 47 ff.Google Scholar; 18 (1963), 159 ff.; 20 (1965), 189 ff. See bibliography there for other references. This paper draws upon African evidence alone and its conclusions may not be relevant to other parts of the Empire.

The following abbreviations are employed in addition to those in standard use: Beschaouch = Beschaouch, A., Karthago 13 (1968), 125223Google Scholar; C = Corpus Inscriptionum Latinarum, vol. VIII; Duncan-Jones = Duncan-Jones, R. P., Papers of the British School at Rome 17 (1962), 47115CrossRefGoogle Scholar; Leglay = M. Leglay, Akte des IV. Internatiovalen Kovgresses für griechische und lateinische Epigraphik, 1962 (1964), 224–33; Veyne = Veyne, P., Karthago 9 (1958), 91109Google Scholar; Villers = Villers, R., Revue historique de droit français et étranger 18 (1939), 132Google Scholar.

2 See Veyne, 93–6; Leglay, 228; and Duncan-Jones, 66, on the distinction between summa legitima and pollicitatio. The account given here in summary is not accepted by Beschaouch, 154ff.; p. 127 below.

3 The sum exacted varied according to the city and the office. See the table in Duncan-Jones, 103–4. On the summa legitima, ibid. 65 ff., with bibliography; and my forthcoming article in Historia.

4 On Trajan's regulation see p. 120 below; for pollicitatio in general see the accounts of Villers and Veyne. The promise was usually made immediately after entry into office (Veyne, 93, n. 3) and was recorded in the acta of the council (ibid. 96). Originally the making of a promise was spontaneous; that it became morally obligatory and therefore quasi-universal by the second century in Africa (as Veyne asserts, 94–5 ) cannot be shown.

5 On summa legitima, see refs. in n. 3 above.

6 See n. 42 below.

7 Phrases used include ‘ampliata pecunia’, ‘adiecta (amplius) pecunia’, and ‘multiplicata pecunia’. For ‘pollicitatio’ see e.g. ILAfr. 222; ILAlg. 1, 951; cf. C 2353 (‘pollicitator’).

8 At the same time, it may be admitted that the fee is referred to relatively often in Africa.

9 For adiectio entailing pollicitatio see, e.g., C 18234 (Lambaesis); 18649 (Diana). In both towns, as it happens, promises in honour of the same offices (the flaminate in each case) are recorded: see 18214; 4588. Comparisons of this kind between inscriptions involving the same office are possible in respect of the following towns (among others): Sitifis (aedileship), Sutunurca (flaminate), Madauros (flaminate), Hr. Kudiat Setieh (magister), Verecunda (flaminate), Diana (flaminate, duovirate), Lambaesis (flaminate), Thamugadi (augurate, duovirate), Cirta (aedileship, triumvirate). The last five towns mentioned furnish the best evidence (Diana and Thamugadi in respect of the duovirate in both cases). Comparisons involving different offices may be relevant in, e.g., Verecunda (pontificate), Thamugadi (quinquennalis), Cirta (pontificate).

10 Two inscriptions are from Cuicul, in that part of Africa Proconsularis which became the province of Numidia in A.D. 197–8. Six from four towns come from Proconsularis Zeugitana and two from two towns from Proconsularis Byzacena (these are two areas of Africa Proconsularis as divided by Diocletian).

11 The meaning of ‘taxatio’ has been touched upon or discussed by several scholars. See C p. 1241 (Schmidt); DS, ‘honoraria summa’, p. 237, col. 1 (Cagnat); RE v A, col. 73–5, s.v. ‘taxatio’ (Kaser); Leglay, 229; Beschaouch, 155–7.

12 Pliny, , h.n. 13, 92Google Scholar (table); 37, 49 (amber); 9, 120 (banquet); 7, 56 (rarities); Seneca, de Ben. 3, 10 (ingratitude). Note that Cleopatra's bet with Antony is presented by Pliny as a sponsio. This looks forward to the use of ‘taxatio’ by legal writers (see below).

13 Pliny, , h.n. 35, 136Google Scholar (exchange-rate); Suet., Div. Aug. 41 (qualification); Gaius 38 (auction-trials).

14 Gaius, , Inst. 4, 4852Google Scholar.

15 For ‘dumtaxat’, cf. Lex Urs. 70 (= FIRA 2 i, p. 182). Here the duovirs are informed that they may spend up to 2,000 sesterces each of public money on games: ‘ex pecunia publica in sing(ulos) Ilvir(os) d(um) t(axat) HS ∞ ∞ sumere consumere liceto’. The 2,000 sesterces which they had to provide themselves for the same purpose was, in contrast, a minimum: ‘unusquisque eorum de sua pecunia ne minus HS ∞ ∞ consumito’. ‘Dumtaxat,’ however, may mean either ‘no more than’, ‘up to,’ or ‘not less than’, ‘at least.’ For the second sense in a legal context see Dig. 50, 16, 202.

16 See Dig. 12, 3, esp. 4, 2 and 5, 1; Buckland, Text-Book of Roman Law 3 (1963), 659. Ulpian, (Dig. 6, 1, 68Google Scholar) states that where a man who was ordered to restore could not do so and had seen to it (through trickery, dolus) that he could not, he was to be condemned ‘quantum adversarius in litem sine ulla taxatione in infinitum iuraverit’. That is to say, only the plaintiff's good faith placed limitations on his valuation.

17 Pro Tullio 7. See, however, Gaius, , Inst. 3, 224Google Scholar, where ‘aestimare’ is used of the plaintiff's assessment.

18 cf. Apul, ., Met. I, 24–5Google Scholar.

19 Dig. 50, 16, 192; cf. ibid. 202.

20 Suet., Div. Aug. 41.

21 Du Cange, Gloss, s.v. ‘taxa’, ‘taxare’, ‘taxatio’; Ernout, Meillet, Diet, etym. 3 s.v. ‘taxa’, ‘taxo.’

22 See, e.g., Cagnat, art. cit.

23 Cf. AE 1914, 43 (3,000 HS; statue); 1916, 14 (4,000 HS; statue); 1914, 236 (4,400 HS; statue); 1914, 44 (6,000 HS; statue).

24 It was apparently not unknown for an official to make donations to a city of which he was patron. L. Harmand, Le patronat sur les collectivités publiques (1957). 398–9, gives three instances, One of which is relevant to North Africa. See C 2661 = ILS 5788 (Severinius Apronianus). Our D. Fonteius Frontinianus (L. Stertinius Rufus) was probably patron of Cuicul. See AE 1925, 23–4, where ‘patrono col.’ has been restored. For inscriptions relating to this man see B. E. Thomasson, Die Statthalter der römischen Provinzen Nordafrikas (1960), 11, 178; cf. PIR 2 F 472. At least some of the legati Augusti stationed at Lambaesis contributed to the adornment of the city. See espec. C 2630 (L. Matuccius Fuscinus); cf. AE 1920, 21. It is not certain that they were patrons of the town.

25 It is also the formula favoured in the town of Cuicul itself where the amount of the adiectio is not specified. See C 8300; 8318–9; 20148; AE 1908, 242; 1913, 154; 1914, 236; 1914, 237; 1916, 12 and 16; BAC 1911, p. 115; 1919, p. 97. I have found only one exception, AE 1916, 35–6 (‘multiplicata pecunia’). On the other hand, to my knowledge ‘ampliata pecunia sua’ occurs nowhere else. This might lead one to take ‘ampliata’ as neuter plural in agreement with the ‘things constructed“. ‘Pecunia sua’ as opposed to the regular ‘sua pecunia’ (or ‘s.p.’) is rare, but does occur in C 8318–9 from Cuicul. (The only other example I have found is in ILAlg. I, 181 from Calama.) I take it that ‘ampliata pecunia sua’ is shorthand for ‘ampliata pecunia pecunia sua’ (or the order of the last two words might be reversed). Cf. C 18234 (Lambaesis): ‘ampliata pecunia praeter legitimam s(ua) p(ecunia)’; see also ILAlg. 1, 950 (Hr. Kudiat Setieh), where a specific adiectio is mentioned: ‘et amplius adiectis a se SS ∞ ∞ n(ummum) sua [pecunia]’.

26 C 2353 = ILS 5467. Cited in Villers, 22.

27 Dig. 50, 12, 14; Villers, 23 ff.; Veyne, 96 ff.

28 The statement that Frontinianus was patron of Thamugadi, rests on Eph. Epig. VII, 349Google Scholar, dated A.D. 162, which may refer to a legate who was patron of the colony. Frontinianus was patron of Diana (e.g. C 4599), Verecunda (C 4232) and Cuicul (AE 1925, 23–4).

29 Pliny as governor of Bithynia consulted Trajan about a half-finished theatre at Nicaea: ‘Huic theatro ex privatorum pollicitationibus multa debentur’, see Ep. 10, 39. Pliny may have discovered this in the course of his investigations into the finances of the cities, or the matter may have been brought to his notice by the authorities of the city. Dio of Prusa threatened to call in an earlier governor of the same province in order to force those who had promised to subscribe to his scheme to beautify the city to carry out their promises: see Or. 47, 19. Veyne cited this passage, and considered it possible that the town of Thamugadi asked Frontinianus to intervene (97).

30 Leglay, 228.

31 There is no way of deciding this matter from the inscriptions. Only a small proportion of the gift inscriptions that were set up have survived. Some do not record promises. Not all of those which do mention a sum or indicate unambiguously what a sum represents. The evidence is as follows: in a high proportion of the inscriptions recording promises, the sum mentioned in the promise is cited with the comment that it was surpassed (e.g., ‘ampliata pecunia’). Where ‘ampliata pecunia’ or an equivalent does not occur, it is not always easy to tell whether a sum that is mentioned represents the final amount expended on a project or the original estimate of the cost of a project. Cf. AE 1914, 43: ‘statuam quam ob honorem pont(ificatus) promisit ex HS III mil(ibus) super legitimam posuit’; and 44: ‘statuam quam ob honorem pontif(icatus) super legitimam ex HS VI mil(ibus) promisit posuit’. Both inscriptions are from Cuicul. Cf. also C 18241 (Lambaesis) with C 4235 (Verecunda). Where the figure cited is not a round one, it is a fair assumption that it represents the final sum expended. See, e.g., Boeswillwald, etc. Timgad (1905), p. 318 (Thamugadi); C 14296 (Thubba). Where use is made of the formula ‘ex HS (…) quae … promisit’, it is equally clear that the sum is the estimate arrived at in the making of the promise. See, e.g., C 8466 (Sitifis); 14370 (Avedda); ILAlg. II, 34 (Rusicade). In cases where the wording is vague or ambiguous it is possible that a sum cited represents the final cost of a project, and that the donor chose to suppres s the original estimate, perhaps because it was not surpassed or surpassed only narrowly. But this can only be a conjecture. Some other inscriptions are quite uninformative—no sum is recorded at all. See ILAlg. 11, 559, 560, 562, 675 (Cirta); etc. Note that in Beschaouch no. 2 (Mustis) the wording suggests that the promise left the project unspecified.

32 See n. 23. Besides, an exedra seems to have been a rare gift—I have not found another example in the price lists compiled by Duncan-Jones—and it may have been difficult to arrive at even an approximate estimate of the cost.

33 There is one further problem. If there was no agreed figure, how can the son claim to have increased it ‘pecunia sua’? The son is perhaps claiming that he spent more money on the structures than the minimum required to put them up somehow. The columns had to be of marble, it would seem, but otherwise the quantity—surveyor presumably had a free hand to choose the cheapest materials possible. But Honoratus went further than this out of generosity—or to salvage his family's pride. Dr. Duncan-Jones has suggested to me another solution: the son had perhaps announced a level of outlay before the exedra was begun which was exceeded in the event.

34 On nos. 2 and 4 see Duncan-Jones, 66–7, n. 53, against Schmidt, C p. 1241; and p. 127 below.

35 If, on the other hand, we were to say that Honoratianus had promised not just an opus (a statue), but an opus at 10,000 sesterces, we would be using a formula which (with variants) occurs in the vast majority of those inscriptions which mention pledges. The idea might be expressed as follows: ‘opus quod (or statuam quam) ex HS X mil(ibus) n(ummum) promiserat, … fecit (or posuit)’. Only occasionally is it directly stated that a sum has been promised. See Beschaouch no. 2 (Mustis): ‘cum HS X (milia) in opus munificentiae promisisset’; cf. ILAlg. I, 10 + BAC 1938–40, p. 135 (Hippo Regius); ILAlg. 1, 181 (Calama); 11, 501 (Cirta); AE 1914, 237 (Cuicul); ILAfr. 390 (Carthage). See also BAC 1893, p. 157 n. 27: ‘promissis HS I (mille)’; cf. C 2341 (both Thamugadi); 26527 (Thugga).

36 This practice has been noted briefly by Villers, art. cit. 18 and Duncan-Jones, 69.

37 See C 14370 (Avedda); cf. 12058 (Muzuc); 12006 (Sarra). Another formula which carries the same implications runs ‘(opus) additis (or adiectis) ad legitimam summam HS (…. ) ex HS (…. ) pollicitus fuerat … posuit.’ See C 4193; cf. 4187; 4194 (Verecunda); 23107 (Hr. Sidi Navi). In some instances where the summa legitima is put to an opus no mention is made of a promise. See C 14791 (Hr. Debbik): ‘… statuam ex HS IIII mil(ibus) n(ummum) legitimis ampliata pecunia posuit’; 885 (Medeli); 25702 (Thuburnica); ILAfr. 300 (Sutunurca). Cf. the formula ‘inlatis legitimis HS (….) … opus … posuit’, which occurs in C 25468 (Munchar); also AE 1946, 234 (Themetra). But see n. 41 below.

38 In the first six instances cited below the sum promised or spent is known or stated to have been less than or equal to the summa legitima: AE 1914, 43 (Cuicul); C 4577 (no ref. to promise); 4588 (Diana); 2711; 18234 (no ref. to promise; Lambaesis); ILTun. 714 (Thuburbo Maius). (The same applies in the case of C 4594 + 18649 (no ref. to promise; Diana) and AE 1914, 237 (Cuicul), on which see n. 41 beginning). Cf. AE 1908, 242; 1914, 44 and 236; 1916, 12 and 16; BAC 1911, p. 115–6; 1919, p. 97; C 8300 (Cuicul); L. Leschi, Études d'épigraphie (1957), p. 274 (Diana); C 2344 + 17812; Boeswillwald, op. cit. p. 318; AE 1901, 191; 1941 49 (Thamugadi). (An exception, as internal evidence shows, is C 12058 (Muzuc).) Another formula implying the separateness of the summa legitima and the payment to the opus runs’ (opus quod ob honorem) amplius ad summam honorariam pollicitus est, ex HS (…. ) posuit.’ See ILAlg. 1, 1236 (Thubursicu Numidarum); 10 (Hippo Regius). Cf. BAC 1893, p. 157, n. 27 (Thamugadi); AE 1914, 40 (Lambaesis; no ref. to promise). ILAlg. 1, 181 and 185 (Calama) are ambiguous.

39 See Duncan-Jones, 108, n. 96. This inscription is omitted by Beschaouch.

40 I take ‘legitimae summae’ as dative of the end desired (or as dative of purpose), not as partitive genitive. By this interpretation, Honoratianus paid the whole of his summa legitima and not only a part of it into the treasury. Cf. C 12058 (Muzuc), where the two summae honorariae of 1,600 HS each (cf. Duncan-Jones, 103, n. 347) when doubled make a total of 6,400 HS. See also C 19122 (Sigus); 26255 (Uchi Maius); IRTrip. 43, cf. 116 (Sabratha).

41 This would have been explicit if the inscription had run: ‘… praeter HS V mil(ia) n(ummum) quae ex legitima summa ob eundem honorem aerario intulit …’ cf. ILAlg. 11, 569; 675 (Cirta); C 4594 + 18649 (Diana); AE 1914, 237 (Cuicul); ILAlg. 11, 42–3 (Rusicade). (Only in the last two cases is a promise recorded.) This formula, or variants, points clearly to a separation of the summa legitima from the in opus payment in the cases concerned. Where ‘super (or praeter) legitimam summam’ occurs by itself we may perhaps understand ‘quam rei publicae intulit’; for refs. see n. 38 above. Cf. the use of ‘(et) amplius (eo)’ in AE 1941, 46 (Thamugadi); ILAlg. 1, 1223 (Thubursicu Numidarum; no ref. to promise); 3007 (Theveste); C 12370 (Thuburbo Maius; no ref. to summa legitima or promise). As for the phrase ‘inlatis reipublicae (aerario) legitimis’, it is sometimes set off against a statement of expenditures ‘sua pecunia’, and this implies that the summa legitima was not contributed to the opus. See, e.g., ILAlg. 1, 2130 (Madauros); 11, 501 and 675 (Cirta); ILAfr. 451 (Bulla Regia); cf. AE 1914, 237 (Cuicul). (Only in the last of these inscriptions is a promise mentioned.) Perhaps the same deduction can be made with respect to C 858 and 12382 (Giufi), where the word ‘prius indicates that payments for the summa legitima and the opus were made at distinct times. The word order seems suggestive in the case of C 4579 (Diana); 2341 and 17838 (Thamugadi). The following seem more ambiguous: C7079 (Cirta); 1577 + 15572 (Mustis); 17837; 17864 (Thamugadi); 22693 (Gigthis); 24640; ILAfr. 390 (Carthage); ILAlg. 1, 2151 (Madauros). Only in AE 1946, 234 (Themetra) is it stated outright that a statutory payment was spent on an opus: ‘statuam inlatis in earn HS DCCC n(ummum) quos honor(e) sufetat(us) deb(ebat) posuit …’ The same implication is present in C 25468 (Munchar).

42 See C 4583 (Diana); C 7097–8; AE 1918, 44; ILAlg. 11, 471; 675 (Cirta); ILAlg. 1, 951 (Hr. Kudiat Setieh); 1236 (Thubursicu Numidarum); 11, 34 (Rusicade); ILAfr. 222 (Abbir Cella); etc. Most promises must have been honoured by the promiser after his term of office was over. In the following cases it was honoured by relations or heirs: C 2353 (Thamugadi); 4193; 4197 (Verecunda); 12058; 12067 (Muzuc); 14370 (Avedda); 15202 (Thignica); 1577 + 15572; 15576 (Mustis); 19121 (Sigus); 20144 + Leglay; BAC 1911, p. 116; AE 1949, 40 (Cuicul); 23107 (Hr. Sidi Navi); 24003 (Sutunurca); 26498; ILAfr. 561 (Thugga); ILAlg. 1, 2035 (Madauros); 11, 559 (Cirta); Karthago 9 (1958), 92 (Vina); etc.

43 This is an inference from the absence of any reference to a summa legitima.

44 It cannot be proved that the final, as opposed to projected, expenditures of the three diverged. But the likelihood that they corresponded exactly is surely remote.

45 Beschaouch, no. 2 (A.D. 117); C 15576 (A.D. 164–5). In C 1578 (A.D. 222–235?), the figure of 10,000 seems to represent an adiectio rather than the taxatio. But the text is doubtful.

46 Beschaouch, no. 2. The honorand, Placidus, like Laetus, pledged 10,000 sesterces for the flaminate.

47 The suggestion is not that, for example, flamens always promised 10,000 sesterces, no more, no less, but that in normal circumstances they promised at least 10,000.

48 Closely parallel to the African inscriptions are, e.g., CIL XII, 697 (Arelate); ILS 5765 (Turri, Sardinia); SEG xx, 95–6 (Cestrus, Cilicia), cf. IGR III, 422 (Ariassos, Pisidia).

49 Cuicul with over twenty promises is the clearest case.

50 Dig. 50, 12, 14 and p. 120 above.

51 See n. 44 above. In one case the final expenditure is known. Kappianus (C 15576) spent 40,000 sesterces on a temple. This sum is not likely to have been matched by any of the other four flamens: the ‘ampliata pecunia’ formula and its equivalents do not hide a large adiectio. But both the project and three-quarters of the amount had been bequeathed to Kappianus by his brother. It can be presumed that Kappianus, if left to his own devices, would have given rather less than 40,000 sesterces.

52 For example, sums promised in honour of the flaminate at Verecunda include 4,000 (three times), 2,400 and 9,000. The promise of 9,000 and two of the promises of 4,000 are recorded in inscriptions of succeeding years. See C 4196–7 (A.D. 212); 4202 (A.D. 213).

53 Cf. ILS 6957 (= CIL II, 4514) of Barcino, where a legacy specifies an upper limit, indicated by ‘usque at’, to be spent on games, and a precise amount, indicated by ‘ex’, to be used for oil for public distribution. It should be noted that ‘ex’ in the African inscriptions occurs in conjunction with sums representing both final costs and preliminary estimates of projects.

54 M Beschaouch no. 2.

55 C p. 1241.

56 Cagnat, art. cit. Leglay also translates ‘taxatio’ as ‘tax’.

57 Beschaouch, 157. Leglay allows that pollicitatio and summa legitima might be identical where the fixed summa legitima was waived and the individual invited to calculate his own. For Beschaouch it seems that every summa legitima was arrived at by the individual on his own calculation.

58 One example will suffice. C. Iulius Secundinus, flamen of Verecunda, put up a statue for which he had promised 9,000 sesterces, in addition to the payment of a summa legitima of 2,000 sesterces. (He also gave sportulae.) See C 4202 + 18494.

59 See nos. 7 and 8 above.

60 See n. 40 above.

61 The phrase ‘praeter legitimam pollicitationemve’’ in AE 1901, 191 (Thamugadi) is almost a parallel, but asyndeton is avoided.

62 The usage was popular in Cuicul, but it occurs in inscriptions from at least ten other towns.