1. Seneca Ep. 86.15 ‘ut ait Vergilius noster, qui non quid verissime, sed quid decentissime diceretur aspexit, nee agricolas docere voluit, sed legentes delectare.’

2. ‘Pertinent’: Wilkinson, L.P., ‘The Georgics’ in The Cambridge History of Classical Literature II: Latin Literature, Kenney, E. J., Clausen, W.V. (eds.), (Cambridge, 1982), p. 322Google Scholar. This article, which contains Wilkinson's latest thoughts on the Georgics (others are to be found in his recent translation of the poem for Penguin Classics), is by no means a mere summary of his earlier book The Georgics of Virgil, A Critical Survey, (Cambridge, 1969)Google Scholar. There (p. 15) he referred to Seneca's comment as ‘obvious’. ‘Wise’: Williams, R. D., Virgil, The Eclogues and Georgics² (New York, 1983)Google Scholar, introduction, p.x. (This is the most recent English edition of the Georgics.) For Petrarch's use of the Georgics as a practical handbook: Wilkinson, , op. cit., (1969), p. 290fGoogle Scholar.

3. Williams, , op.cit., introduction, p.xGoogle Scholar. (It is not clear why he chooses to describe the didactic material as ‘intellectual’.)

4. Putnam, M. J., Virgil's Poem of the Earth: Studies in the Georgics (Princeton, 1979), p. 7Google Scholar.

5. Wilkinson, , op. cit. (1982), p. 323 and p. 320Google Scholar.

6. Ibid., p. 320., cf. idem, op. cit., (1969), p. 54. See also Quinn, K., Texts and Contexts, The Roman Writers and Their Audience (London, 1979), p. 134Google Scholar, who assumes that Virgil is describing the world of the small subsistence farmer.

7. Williams, , op. cit., introduction, pp. x–xiGoogle Scholar.

8. See for example the review of recent studies of the Georgics by Gransden, K. W. in JRS 72 (1982), 207–209Google Scholar and especially Griffin, J., ‘Haec Super Arvorum Culta’, CR 31 (1981), 23–37Google Scholar. For the development of the now orthodox literary approach to the Georgics: Otis, Brooks, Virgil, A Study in Civilized Poetry, (Oxford, 1964), p. 145Google Scholar; ‘It is only very recently that some critics – I think especially of Burck, Klingner and Büchner – have refused to treat Virgil with such naivete (sc. reading the Georgics as a poem about agriculture) and have tried to penetrate the deeper meaning of the poem…’ He provides bibliographical details on the history of interpretation of the Georgics in Appendix 6, p. 407. See also Wilkinson (1969), p. 71 ff and p. 314f. Readers of Greece & Rome have been well served with articles on the Georgics: cf. the contributions of Jermyn, L. A. S. in G & R. 18 (1949), 49–69Google Scholar; 20 (1951), 26–37 and 49–59; an early article by Wilkinson, 19 (1950), 19–28; and more recently J. Griffin, 26 (1979), 61–80.

9. The only very approximate dates that we have of the publication of the three agricultural treatises are: first half of the 2nd century B.C. (Cato), second half of the first century B.C. (Varro) and middle of the 1st century A.D. (Columella). Columella (3.3.3) mentions Seneca as the owner of exceptionally productive vineyards near Nomentum. Seneca may have been one of Columella's patrons, although that need not mean that Seneca ever read his or any other available agricultural manual. See: Griffin, M. T., Seneca, A Philosopher in Politics (Oxford, 1976), pp. 290fGoogle Scholar.

10. We know that Aegialus, a freedman, gained a considerable reputation for successful farming near Liternum (Pliny, , N.H. 14.49)Google Scholar.

11. Cf. Food and Agriculture Organization of the United Nations (FAO), Modern Olive Growing (Rome, 1973), pp. 27fGoogle Scholar.

12. Yet it is quite possible that Virgil himself refers to this supposedly new method at Georgics 2.30–1 (and see note ad loc. in Conington's edition). The poet expresses his astonishment that such a replanting system could work — mirabile dictu. One further difficulty in comprehending Seneca is caused by textual corruption (Ep. 86.14), where he is describing what sort of olives are treated in this way. The usual reading of editors appears to mean that the olives are healthy and productive. It would seem senseless then to prune so radically and then transplant.

13. Seneca speaks of keeping the cold and wind from the roots. Cato advises the firming of the soil around the roots at 28.2; 49.1. A serious danger is that otherwise the roots will become waterlogged when it rains: ibid. 61.2.

14. Ancient agricultural advice: Cato 40.1: Varro 1.30; Columella 5.9.6–7, Modern advice: FAO, op. cit., pp. 27f. Modern, yet traditional, practice is also to transplant cuttings (usually into a nursery) immediately after the winter pruning. Virgil perhaps alludes to this by his stress on putator ‘pruner’ (G. 2.28).

15. Varro had clearly stated that in the case of slow-growing trees, such as the olive, it takes too long to grow them from seed and thus advises propagation from cuttings (1.41.6). Virgil makes exactly the same point, but Seneca misunderstands.

16. Cf. Spurr, M. S., ‘The Cultivation of Millet in Roman Italy’, PBSR 51 (1983) 1–15Google Scholar.

17. It would seem also that Seneca is disputing Virgil's statement that beans were a spring crop. Yet, where winters were severe (inland hill and mountain zones and the Po valley), they had to be spring-sown. In warmer areas there was a choice, although it was generally acknowledged that spring-sown yielded less than autumn-sown beans (Columella 2.10.9). Pliny the Elder considered, rightly perhaps, that Virgil was thinking particularly of the Po valley (N.H. 18.120)

18. There are, in fact, only very few real agricultural ‘mistakes’ in the Georgics: he is incorrect to state that repeated ploughing of poor soils causes loss of moisture, as Williams points out in his note on 1.70, following White, K. D., ‘Virgil's knowledge of Arable Farming’ Proceedings of the Virgil Society 7 (1967–1968), 11–22Google Scholar. But it is not usually explained that this was what was generally believed at the time (Columella 2.4.11; Pliny, , N.H. 18.242)Google Scholar, and thus the error can hardly be ascribed just to Virgil. Other mistakes: pear trees can not bear apples, nor plum trees cornels, after grafting (2.33–4; but again a shared error: cf. Varro 1.40.5). He recounts wondrously how mares in Asia Minor are impregnated by wind (3.269–75) but even Columella was prepared to credit similar stories (6.27.7). In Book 4 the queen bee is a rex, and Virgil advocates Bugonia as an effective method of reproducing bees. Yet, again, both were common, if not universal, ancient misconceptions. The female sex of the queen bee was only decisively proved in the mid-18th century: Thomas, K., Man and the Natural World (London, 1983) p. 62nGoogle Scholar.

19. See for example, Kramer, G., ‘The Didactic Tradition in Virgil's Georgics’, in Boyle, A. J. (ed.), Virgil's Ascraean Song. Ramus Essays on the Georgics (Melbourne, 1979), pp. 7–21Google Scholar. (This is one of the new works reviewed by Griffin [n. 8 above], who incidentally approves of Seneca's judgement on Virgil.) See also Otis, Brooks, op. cit., p. 148Google Scholar.

20. Wilkinson, , op. cit. (1969), pp. 67fGoogle Scholar. and Virgil, The Georgics (Harmondsworth, 1982) p. 95Google Scholar.

21. Williams, , op. cit., p. 172Google Scholar.

22. See Rawson, E., ‘The Introduction of Logical Organization in Roman Prose Literature,’ PBSR 46 (1978), p. 14fGoogle Scholar. For a review of previous judgements on the worth of Varro's treatment, whether from the agricultural or compositional point of view: Skydsgaard, J. E., Varro the Scholar, Studies in the First Book of Varro's de re rustica (Copenhagen, 1968), pp. 89 ffGoogle Scholar.

23. Rawson, op. cit., argues that Latin Prose composition on a systematic basis, of the sort now taken for granted, was a gradual development, certainly not complete by Varro's time, although he contributed significantly to its progress. Thus to judge work of that period by later ideals is unjustified.

24. See West, M. L. (ed.), Hesiod Works and Days (Oxford, 1978), pp. 53ff.; 252fGoogle Scholar.

25. Thus, too, when Lucretius says (1.926–950) that he uses poetry to enliven and make palatable the dry reasoning of philosophy, as doctors honey the cup that contains medicine, he does not mean that he will thereby sacrifice the serious didactic content to poetic imagination. Poetic digressions and ‘purple passages’ enhance the directly informative sections and make them more effective.

26. Wilkinson, , op. cit. (1969), pp. 223–5 and (1982), p. 31fGoogle Scholar.

27. For a detailed account of Virgil's adaption of the works of Aratus, Eratosthenes, and Theophrastus: Wilkinson (1969), pp. 234 ff. 242 ff. Some have worried that the rooks which Virgil describes (following Aratus) were uncommon in Italy but see Royds, T. F., The Beasts Birds and Bees of Virgil (Oxford, 1918), pp. 40fGoogle Scholar.

28. Wilkinson, , op. cit. (1969), p. 245Google Scholar: ‘But there is no known predecessor to Virgil's selective and discriminating list, and it may of course represent his own taste. At any rate he has diverged here from Theophrastus.…’ The fullest ancient account of Italian wines is given by Pliny the Elder (N.H. 14.20–76). Varro's comment on Theophrastus is noteworthy: ‘His books are more suited to philosophy students than farmers’ (1.5.2).

29. For their use of sources: Gummerus, H., Der römische Gutsbetreib als wirtschaftlicher Organismus nach den Werken des Caw, Varro und Columella, Klio Beiheft 5 (Leipzig, 1906Google Scholar, reprint edn. Aalen, 1979). For Varro in particular: Skydsgaard, op. cit., White, K. D., ‘Roman Agricultural Writers I: Varro and his Predecessors’, in Aufstieg und Niedergang der römischen Welt 1.4, ed. Temporini, H. (Berlin, 1973), pp. 439–97Google Scholar.

30. The conventional view is that, as a consequence of the growth of slave-staffed villa estates and the eradication of the peasantry in the second century B.C. onwards, cereal cultivation was replaced by the cultivation of vines and olives. See for example: Toynbee, A. J., Hannibal's Legacy, 2 Vols. (London, 1965), 247 ff. 286 ff. and passimGoogle Scholar.

31. Pliny the Elder said that one of the main functions of asses was ploughing (N.H. 8.167). Economical to feed (Columella 7.1.2), the ass would have been inexpensive and its forage not difficult to find even on the least prosperous peasant farm.

32. Cato and Varro both concern themselves with rotation patterns but do not treat all the possibilities envisaged by Virgil. This could indicate Virgil's own experience. Pliny noted Virgil's advice on fallowing but remarked that this was possible only on a large enough farm. Otherwise a rudimentary rotation system should be adopted (N.H. 18.187). No doubt many small holders continued to grow grain year after year on the same land, with resulting low yields.

33. Varro 1.23.5 refers to the making of such a harrow (a ‘drag’ would be a better term, since it is mentioned by Virgil and Columella [2.17.4] as an implement only for levelling ploughed fields), without describing its use. See White, K. D., Agricultural Implements of the Roman World, (Cambridge, 1967), p. 147Google Scholar. Virgil's reference to its function may again be the result of personal knowledge.

34. Spurr, , op. cit., 9Google Scholar.

35. Various other references to fencing sown fields (e.g. 1:270–1; 2.436) show that Virgil was considering a farm with a combination of cereal cultivation and livestock farming. The livestock provides manure for various crop rotations (1.80) and in return will need forage crops and pasture. Fences will have to be constructed as part of the livestock management – to keep the animals in, or out of, fields at different times.

36. Pliny, , N.H. 18.295 and Columella 2.19.2Google Scholar.

37. A convenient summary of the reasons for horse breeding is found in Varro 2.7.15: cavalry, carriages, racing and breeding. Cf. Virgil 3.72–208. They were not employed for ploughing until the Middle Ages: Parain, C., ‘The Evolution of Agricultural Technique,’ in The Cambridge Economic History of Europe Postan, M. M. (ed.), i. 142 ffGoogle Scholar. Virgil's long passage on horses in Book 3 is probably the most obvious proof that his work concerns the agriculture of the upper classes.

38. It has recently been argued that the originator of the agricultural didactic poem, Hesiod, was a wealthy peasant: Millet, P., ‘Hesiod and his World’, PPhS 30 (1984), 84–107Google Scholar, but comparison will show that Virgil's farming was carried out on a much larger scale.

39. E.g. 2.416.

40. E.g. Strabo 5.2.1; Pliny, , N.H. 18.178Google Scholar; Horace, , Odes 3.6.38–9Google Scholar.

41. Flowers: Cato 8.2; Varro 1.16.3; Columella 10.308; It is clear that the dapibus inemptis (4.133) of Virgil's gardener refer only to fruits and vegetables. He must have purchased grain and wine, for which we are told his land was unsuited.

42. E.g. Williams, , op. cit., introduction, p. viiGoogle Scholar.

43. Brunt, P. A., Italian Manpower 225 B.C. – A.D. 14 (Oxford, 1970), p. 329Google Scholar; Hardie, C. G., ‘Virgil’, OCD 2 (Oxford, 1970), p. 1123Google Scholar. If the reading sexaginta veterani is credited, and if the veterans received an average of 35 iugera apiece (Keppie, L., Colonisation and Veteran Settlement in Italy 47 -14 B.C., British School at Rome (London, 1983), p. 90)Google Scholar, then Virgil's father was a rich man. Not all the 2,000-odd iugera would have been included in one farm. Instead the pattern of landholding was to own several estates, of, perhaps, ca. 200 iugera (Varro 1.19.1), often in the same area. An example often cited is Sextus Roscius of Ameria in Umbria, who owned thirteen farms in the Tiber Valley in the vicinity of his home town: Cic. S. Rose. 15;20. A poem sometimes ascribed to Virgil, Catalepton 8, speaks of his father missing Mantua and Cremona after his expropriation. Perhaps his estates were located in the territories of both towns.

44. Wilkinson, , op. cit. (1982), p. 22fGoogle Scholar. It can be noted here that Wilkinson's attempt to back up the conventional notions with the statement that in the Po valley and the Naples area the small holder still flourished (n. 6 above) is, as it stands, misleading. Archaeological surveys over the last fifteen years have shown it is likely that small farms remained numerous throughout many areas of Italy. And we know too from archaeology that there were plenty of agricultural villas in the Po valley and Campania. The point of departure for such evidence remains the important article: Frederiksen, M. W., ‘The Contribution of Archaeology to the Agrarian Problem in the Gracchan Period', Dialoghi di Archeologia 4–5 (1970–1971), 330–357Google Scholar.

45. Williams, G., Tradition and Originality in Roman Poetry (Oxford, 1968), pp. 294f., 303 ffGoogle Scholar.

46. Varro 1.17 for slaves, hired labourers and (probably) tenant farmers. Cf. Columella 1.7. References in the Georgics to labour include: 1.210 exercete viri which will refer to ploughmen, either slave or hired; 1.259 ff. the sort of tasks that can be done by the various members of the workforce (cf. alii, 1.264), which resemble the prose writers' admonitions to keep the labourers always at work, even if the weather is too bad to go out: e.g. Cato, 2.3; 39.2. Work can be done after dark (Cato 37.3, Columella 11.2.90) and before light (Varro 1.36) in winter. Besides not working, another danger was that the agricultural slave might run away, even from the most vigilant owner's estate (Cato 2.2.) and this is referred to by Virgil at 1.287. 1.291 f. perhaps refer to a tenant couple (and the Hesiodic nudus ara, sere nudus, 1.299, suits such a picture), whereas the reapers of 1.316 were probably a mixture of slaves and hired labourers (cf. Varro 1.17.3). Slave girls to carry out the spinning (1.390) came under the supervision of the slave bailiffs female partner, as Columella later made clear (12.3.6). The planting of vines was (and is) a very hard job, because the earth must be dug to a depth of up to four feet, depending on the nature of the terrain (Columella 3.13.8). It was probably usual to hire labourers for this operation, since Columella refers to a conductor, a ‘contractor’ (ibid. 3.13.12). Thus Virgil's reference to the establishment of a vineyard with ‘much expense’ (multa mercede, 2.62) should be taken literally. Yet the ‘digger’ referred to at 2.264 is probably a slave, as is the uinitor (2.417). A vinedresser was a skilled and valued slave, but nfossor also could be skilled (thus the formula optimus fossor, Columella 11.1.12). The division of labour on a slave-staffed villa estate was quite marked (see especially Columella ibid.).

47. Williams, , op. cit., introduction p. x–xiGoogle Scholar.

48. Caesar, , B.C. 1.18.4;1.24.2Google Scholar.

49. Toynbee, op. cit., passim.

50. Brunt, , op. cit., pp. 269–77Google Scholar.

51. On alternative sources of food see the chapter ‘Wild and Cultivated Plants’ in Frayn, J. M., Subsistence Farming in Roman Italy (London, 1979), pp. 57–72Google Scholar. Virgil mentions hunting: 1.271, 307–9; and refers to collection of berries, and sometimes acorns, as food: 1.305–6; 1.148; 1.159.

52. Children's work could include: vine trimming, cutting back ferns, hen keeping (Columella 11.2144; 2.2. 13; 8.2.7.) and shepherding (Varro 2.10.1)

53. Appian, , B.C. 5.19;56 (raids in the south of Italy)Google Scholar. After Octavian recovered Sardinia he ‘strengthened the coast of Italy with many garrisons’, ibid. 5.80. Even if, as Cicero said in another context, merely the threat of hostile attack caused men to flee their fields (De Imp. Pomp. 16), that would mean only a very temporary setback for agriculture.

54. The figures comes from Res Gestae (25.1) and may well be exaggerated, since Augustus represents the war against Sextus Pompey as a war entirely against slaves and pirates.

55. Brunt, , op. cit., pp. 330–343fGoogle Scholar. Brunt believes that the Perusine War was more widespread than is argued here (cf. p. 29O.f.). But the only other hard evidence is the sacking of Sentinum and the capitulation of Nursia, small towns in comparison with Perusia (Dio, 48.13).

56. For a typical comment about the failure of veterans as farmers: Wilkinson, , op. cit. (1982), p. 320Google Scholar. On contemporary laments see, for example, Williams, , op. cit., p. 95Google Scholar on Eclogue 1.70–2; Keppie, , op. cit., p. 101Google Scholar.

57. To explain Varro's enthusiasm some commentators consider that his praise of Italy belongs to a much earlier draft of the work, since they believe in an agricultural crisis in the thirties B.C. See Heurgon, J., Varron, Economie Rurale: Livre Premier (Paris, 1978), p. 105Google Scholar. Editors of Virgil will point out that the laus Italiae was a well-established traditional literary feature which had nothing to do with reality. A recent acute study shows, perhaps rightly, that the passage contains some pessimistic elements in the form of moral admonition against violence and the pursuit of luxury: Thomas, R. F., Landscapes and People in Roman Poetry. The Ethnographical Tradition. PCPhS Supp. Vol. 7 (1982), pp. 39 ffGoogle Scholar. But such moralising is also part of the agricultural manual tradition (see n. 65 below).

58. Williams, , op. cit., p. 156Google Scholar. Cf. the long-standing unchallenged words of Page, T.E., Virgil, Bucolics and Georgics,(Leicester, 1974; first edition 1898), p. 242Google Scholar, … ‘The words admirably connect the whole lament for the ruin of Italy with the subject of the Georgics’.

59. Williams, , op. cit., p. 156 (on line 1.509)Google Scholar.

60. E.g. Jones, A. H. M., Augustus (London, 1977), p. 142fGoogle ScholarPubMed.

61. Wilkinson, , op. cit. (1982), p. 320Google Scholar.

62. Horace, , Odes 4.15.5Google Scholar: fruges et agris rettulit uberes; cf. Veil, . Pat., 2.89.4Google Scholar. The Ara Pads was dedicated in 13 B.C. (the year in which it is quite conceivable that Horace wrote 4.15), although not ‘opened’ until 9 B.C. For an interpretation of the monument as a piece of Augustan propaganda: Weinstock, S., ‘Pax Augusta and the Ara Pads’, JRS 50 (1960), 44–58Google Scholar;cf. Toynbee, J. M. C., ‘The Ara Pacis Reconsidered’ JRS 51 (1961), 153–6Google Scholar. For Horace as propagandist: Brink, C. O., Horace on Poetry, Epistles Book II. The Letters to Augustus and Florus, (Cambridge, 1982), pp. 523 ffGoogle Scholar.

63. On Augustus’ settlement of veterans see: Brunt (1971), pp. 326 ff and Keppie, op. cit., pp. 115 ff., for works of public utility, including roads. The campaign against brigandage: Suet, . Aug. 32.1Google Scholar.

64. Suet, . Aug 18.2Google Scholar. After continual difficulties with the grain supply to Rome, Augustus reached a settlement which suited grain merchants, the Roman poor, and arable farmers (aratores). In the context, the ‘farmers’ were surely Italian and local to Rome, and it could have been that the cultivation of cereals in the environs of the city was declining because of growing imports. Although the precise details are unclear, it does seem that, in this local and particular case only, Augustus intervened directly in Italian agriculture: Suet, . Aug. 42Google Scholar.

65. Cato, praef.: farming is better than all other pursuits, as our ancestors rightly judged. Varro, 1.13.6–7: modestly-sized agricultural villas are better (as our ancestors thought) than luxurious and parasitic suburban villas; 1.69.3: violence is endemic in the city of Rome; 2 praef: luxury and idleness in towns; Rome relies on imports; farmers have left their land for the easy life at Rome; 3.1.1–5: the country is sacred to the gods and country life more noble and traditional. All these moral topics are dealt with also by Pliny, , N.H. 1.1–22Google Scholar, and in the long preface of Columella.

* I should like to thank Malcolm Davies and Elizabeth Rawson for their valuable criticism of an earlier draft of this article.