The Sale of Wine

In this chapter, I wish briefly to outline a study of the means by which, during the Roman Empire, wine produced on an estate ended up on consumers’ tables or in their cellars. The evidence available lends itself to seeing the question from two points of view, the producer’s and the consumer’s.1


The choice facing an Egyptian producer in the second century AD can be seen from an Oxyrhynchus papyrus (P. Oxy. LIX. 3989).2 An ex-gymnasiarch from Oxyrhynchus, sending instructions to someone who is looking after his estates, adds this at the end: ‘You did well to sell the wine to traders rather than retail it in small amounts’ (каАш? enoinoa? tov oivov i^nopiKW? nwArfoa? ка1 KOTvAloa?).

It should be noted that the only choices appear to have been selling wine by the cotyla (a quarter of a litre)—that is to say, retailing in very small quantities, as in a bar—and selling to a trader. I have no evidence suggesting that a producer might sell two or three amphorae directly to a consumer who wanted to buy some at the cellar door.

First published in E. Lo Cascio (ed.), Mercati permanent e mercati periodici nel mondo romano: atti degli Incontri capresi di storia dell’economia antica (Capri, 13-15 ottobre 1997) (Bari: Edipuglia, 2000), 223-35.

  • 1 Similar arguments have been presented recently, not limited to wine, by L. de Ligt (1993: 164-5), and J.-J. Aubert (1994: 244-76). N. Morley (2000: 213 ff.) also touched on this subject at the same conference. The bulk of the evidence about wine traders was gathered by Schuster (1958) and Schlippschuh (1974).
  • 2 I am very grateful to Fabienne Burkhalter for drawing my attention to this papyrus.

The Sale of Wine KotvX%€iv

The term KorvXioag can be explained, at least in part, by the advice of Varro, Rust. I. 2, 23:

si ager secundum viam et opportunus viatoribus locus, aedificandae tabernae deversoriae, quae tamen, quamvis sint fructuosae, nihilo magis sunt agri culturae partes (‘If a farm lies along a road and the site is convenient for travellers, an inn might be built; however profitable it might be, still it would form no part of agriculture’). (trans. W. Hooper, rev. H. B. Ash, adapted)

What he is describing are the activities of both hotelier and innkeeper. We can see some evidence of this in the tabernae Caediciae spoken of by Festus,[1] who indicates that they are on the edge of the Appian Way and that they bear the names of their owner. They have long been associated with the Vicus Caedicii, which, according to Pliny (XIV. 62), was in the middle of the Ager Falernus. The Oxyrhynchus wine-grower could have sold his wine by the cotyla in such places.

Raymond Billiard (1913: 183), drawing on Breton’s book (1855) on Pompeii, which mentions an oenopolium in the House of Pansa, speaks of this as ‘an essential annex to the owner’s vineyards’, where ‘a dispensator, a slave whose special job this was, retailed fruit, vegetables, and wine from his master’s estates to local housewives’. Billiard compares this activity with the way wine was sold by Florentine nobles in the nineteenth century, ‘poured by the jugful in a little room on the ground floor of their magnificent palaces’. He gives no other examples of any direct contact between producers and consumers. It was from this passage that Roger Dion, in a fine example of circularity, drew his conclusion that we can find ‘in Greco-Roman Antiquity the origin of the custom which was constant in France until the reign of Louis XV’, selling at the half-door and by the jugful, and which ‘the greatest nobles engaged in with neither scruple nor shame’, though it did not amount to ‘being an innkeeper’.[2]

The truth is that the shops surrounding the fine House of Pansa, as it is called, were rented out, at least in ad 79.5 There is nothing to show that they were ever used for selling the landlord’s wine; and A. Wallace-Hadrill (1994), who has recently analysed the cheek-by-jowl siting of tabernae and aristocratic houses, never goes that far. This can remain an open question, in the light of Vitruvius (VI. 5, 2): Qui autem fructibus rusticis serviunt, in eorum vestibulis stabula, tabernae... sunt facienda (‘But those who depend upon country produce must have stalls for cattle and tabernae in the forecourt... ’) (trans. F. Granger, adapted). The type of houses of which it could be said that they

fructibus rusticis serviunt remains unclear—though it probably does not include the House of the Faun in Pompeii, despite its tabernae opening onto the street and communicating with the atrium.[3]


Evidence on the various ways of getting the vintage from the grower to the trader is much more abundant. We can define three of them.

1. The producer makes and looks after his wine; the trader comes to the cellar door to make the purchase and take delivery after the vinalia in April or later. In the laws governing wine, with their rules about tasting, measures, and responsibility during the keeping period, there is no shortage of instances of this; and the jurists[4] see this procedure as the most widespread method of sale. Buyers took the wine away in skins (Cato, 154) or amphorae. If the amphorae were Dressel 1s, much evidence demonstrates that the pozzolana stoppers bore the names of traders, who brought their stamps with them on their boats;[5] and, while on board, they even fashioned cork stoppers to be overlaid by the pozzolana.[6] It was these traders who then set about decanting the wine into amphorae.

An example that, though different, still seems to confirm this can be seen in the building in Oplontis that has been called the villa of Crassius Tertius or Oplontis B. It is a great courtyard with a portico, 22.5 metres x 10 metres, with a series of numbered cellae opening onto it—in short a horreum. In the cellae, remains of agricultural products have been found, clover, hay, lucerne, and, in one of them, 7m3 of pomegranates. At four points in the courtyard itself piles of empty amphorae have been found, more than 600 in all, with two ovens for melting pitch. Among the amphorae, there are not only local varieties but Greek ones as well, from the Ager Falernus or Surrentum, and even one that had previously been reused to carry liquamen. On the white coating of the portico wall, there is a red inscription: N.NEGOT, which must be some sort of allusion either to a negotiator] or to negot[iatores]. The whole impression is one of premises let out to traders as space for storage and work, where vinarii, among others, had reusable amphorae pitched for the next vintage. They were a category of relatively modest traders, who needed no great number of new amphorae and were presumably active in regional distribution of the produce from some of the villas in the Pompeii area.[7]

The wine-grower himself could also undertake the decanting into amphorae and the sale of wine in these containers: there are plenty of examples of workshops attached to villas; and the Digest (XVIII. 6, 1, 1) makes quite clear that there were two possibilities, selling in bulk or in amphorae: Et ante mensuram periculo liberatur si non ad mensuram vendidit sed forte amphoras vel etiam singula dolia (‘And even before it is measured, [the vendor] will be released from responsibility if he did not happen to sell it by measure, but sold it in amphorae or by dolia’) (trans. S. P. Scott, adapted).

2. The producer sells the standing crop. Three texts outlining this method are traditionally quoted: the leges of Cato (147-8), distinguishing the lex vini pendentis from the lex vino in doliis; the mention by Pliny of the sale of heaps of grapes from the vineyard of Remmius Palaemon: emptori addicta pendente vindemia (XIV. 50); and Pliny’s letter on the remissiones (VIII. 2).

With the third of these sources, there are two difficulties, identified in an influential article by De Neeve (1990) on the estates of Pliny the Younger: on the one hand, at IX. 16 and 20, Pliny is dealing with his vintage, whereas it is generally accepted that this was undertaken by whoever had bought the crop pendente vindemia; and, on the other, he has sold to several traders, and one wonders how they went about dividing the tasks and making use of the villa’s equipment. De Neeve's solution is a complicated one and has not met with much agreement. In my view, there is a simpler solution: Pliny’s words are vendideram vindemias certatim negotiatoribus ementibus. Nowhere in the letter does it say the crop was sold standing, despite which that is unanimously accepted to be the case, especially since Heitland's statement three-quarters of a century ago that Pliny had sold ‘evidently the hanging crop’.[8] Editors in the nineteenth century were less rash, both G. E. Gierig in 1800 and M. Doring in 1843 harbouring the same doubts as I do.

Somebody who buys a standing crop is wagering on two things, the first of which is the future price of the wine. This, however, is almost negligible compared with the other risk: the quantity and quality of the vintage, which will remain fraught with uncertainty until the wine is made.[9] There are other letters by Pliny that give evidence of how momentous this uncertainty could be.[10] Yet, in VIII. 2, Pliny is referring only to the price: invitabat pretium, et quod tunc, et quod fore videbatur. Had he sold the standing crop, he would undoubtedly have mentioned the second of the two wagers. The truth is that Pliny auctioned off his new wine, in batches, which means that traders sampled it, then made offers as to price and quantity; some of them have paid, some have not paid, and no one has yet taken delivery. Reading this meaning from vindemias is entirely plausible: Vergil, for instance, uses it to refer to must: spumat plenis vindemia labris (‘the vintage foams in the brimming vats’) (Georg., II. 6, trans. H. R. Fairclough).[11] So this letter by Pliny should be used instead as evidence for the first of the three ways mentioned above. Sales of standing crops are still well attested by the other two sources quoted.

As for how such sales were managed, a trader would buy the grapes still on the vines at some time before harvesting began. He was then responsible, either personally or through a sub-contractor, for doing the vintaging, though he used the equipment on the estate (treading vat, press, and lacus). Two different possibilities are attested for what might happen next: in Cato, the wine was made on the estate and remained in store until being removed (at the latest by the Kalends of October, in other words, the following harvest); while Digest, XIX. 1, 25, implies that, after treading, the must was taken away to ferment elsewhere.[12] This second source is an interesting one, for it presupposes that there were wine-making premises and storehouses belonging to traders, unattached to the vineyards and the pressing establishments, and hence that there were autonomous trading growers who had at their disposal solid infrastructure.

We get a reasonably similar picture from Digest, XVIII. 6,2. At XVIII. 6,1,4, there is a mention of the problem of a sale of wine in the dolia for which no date has been set for it to be taken away. In that case, the wine must be removed before the wine-grower needs his dolia again for the next vintage. If not, recourse might be had to the famous and tragic measurement by basket, which meant pouring the wine away. This is the eventuality mentioned by Cato. In the following paragraph, however, a clarification by Gaius is mentioned:

Hoc ita verum est si is est venditor cui sine nova vindemia non sunt ista vasa necessaria; si vero mercator est, qui emere vina et vendere solet, is dies spectandus est quo ex commodo venditoris tolli possint. (The above is also true provided that the vendor is someone who needs these vessels only for the new vintage. But if he is a merchant who is in the habit of purchasing and selling wine, he must watch for the moment when the wine can conveniently be removed without inconveniencing the seller.) (trans. S. P. Scott, adapted)

Here we have a trader who bought wine in bulk, always sold it on in bulk, presumably to other traders or to retailers; so, if he was to buy new wines and conduct other transactions, he needed his dolia. However, he did not sell retail, but rather sold to someone who bought in a single batch the wine contained in several dolia, if not the whole lot. This gives a slightly different image of a trader from that of the buyer of a standing crop, who was a mere go-between, though he too owned heavy equipment and was very different from the small retailer and even from those in the horreum belonging to Crassius Tertius at Pompeii.16

It would be nice if archaeology had turned up examples of these independent storage cellars. Rosa Plana-Mallart and Michel Christol have argued that one good example of this is the major workshop producing amphorae and tiles at Llafranc (Catalonia): three ovens, an area for drying tiles, and a dumping ground for amphorae have been identified. A name stamped on amphorae and tiles, P. Vsulenus Veiento, is no doubt that of a member of a great family of Narbo, who, if we are to accept the authors’ reading of the inscription CIL XII. 4426, was more particularly the father of a duumvir and flamen belonging to the last years of the reign of Augustus. Near the pottery workshop stood a building in which were found the base of a dolium in situ and many fragments of other dolia.17 The sites being excavated would appear to be the urban area of a small port. The authors are adamant that Usulenus Veiento was a trader. And, in that case, he could easily have been a buyer of standing crops of wine in the local area (where small properties seem to have played a leading part), who aged the wine in his cellar near the harbour, and a maker ofthe amphorae in which he exported the wine he produced. We could possibly see a similar function in the building excavated at number 16, boulevard de la Republique in Aix-en-Provence: a large shed much longer than broad (more than 41 metres x 9.5), which held dolia, up against the city walls. No wine-pressing facility has been discovered.18

However, these hypotheses are nowhere near proven, and it may well be that these discoveries belong to quite standard production facilities. The hypothesis that there was a large villa at Llafranc has been advanced; and the existence there of an urban settlement has not been established.19 Nor does [13] [14] [15] [16]

it seem impossible that Usulenus Veiento was merely the owner of a large pottery workshop. It would not have been out of the question to have a vineyard in Aix near the ramparts; and it appears plausible that the storehouse had an agricultural function, especially since the authors say it was no doubt transformed at some later stage into a cowshed. The area of these two excavations was not extensive enough to rule out the existence of a wine press near the storehouses, neither of which was completely unearthed. So we must wait for a reliable archaeological model of such buildings, which are known to have existed but whose distribution remains uncertain.

3. This third category is that of the producers who undertake the transport of their own wine from the estate. Their existence is indisputably attested in texts from the Digest that list equipment that is useful for deportare, exportare, evehere, or vectare agricultural produce: amphorae, casks, skins, draught animals, carts, and boats.[17] Varro (I. 16) makes the point that the estate needs to be established in an area quo fructus nostros exportare expediat,[18] and Columella (I, 2, 3) stresses the importance of having waterways close by, quo deportari fructus etper quod merces invehi possint (cf. Cato 1, 3).

These two texts show consistency in the vocabulary they use: nowhere is there a suggestion of sales, but only of conveyance. It could be argued that what they are dealing with is supplying the landowner’s town house, which was most definitely done,[19] and very probably they are talking about part of the transport that this entailed. Let us not read too much into that last comment, though, for clearly Varro at least means selling in the nearby town. Perhaps this is again evidence of contact with the consumer (possibly in the context of nundinae), whether in a mode akin to that of the KOTvXiZeiv (wholesaling in small quantities), or even selling in larger amounts, in amphorae, for example (though there is no evidence for this). It may also refer to selling to mercatores or to local caupones. It is difficult to say whether producers went looking for buyers much farther afield (boats, on rivers as on the sea coast, were useful even over short distances), and, if so, who were the buyers. Here we start floundering among a multiplicity of hypotheses, which it would be pointless to develop further.23

By way of conclusion, let me underline the diversity of the possibilities that I have canvassed. In the light of these, it would still, as usual, be useful to know what was the dominant paradigm and what was less frequent; and, also as usual, the sources we have do not run to that. Which model is held to be standard is influenced by personal impressions. My own impression is that the screen that trading interposed between producers and consumers was pretty solid and pretty diverse, and that the producing activity was as a rule quite separate from any level of trade that went beyond local sales. Wine arriving at Ostia, I feel, was generally conveyed there by members of the corpus impor- tantium et negotiantium vinariorum24 or even, if it had come from the east coast, by the navicularii maris Hadriatici of Ostia and Rome,25 or in the third century by the negotiantes vini supernates et Ariminenses of Rome,26 and not by producers who owned ships and made amphorae. However, there are gaps of uncertainty, breaches through which any scholars convinced that great Italian landowners were directly involved in trading their own agricultural produce may step at their ease. Bruce Frier comments that ‘none of the jurists’ hypothetical cases concerns sale to ultimate consumers’,27 to which one could reply that the Digest explicitly quotes a mere three instances of wine being bought by a mercator or a negotiator;28 at other places, it speaks only of emptor. In such matters, it is not easy to prove anything. The case of Remmius Palaemon, however, who owned officinae promercalium vestium,29 who increased spectacularly the productivity of his estate, showing he had a knack for investing and a concern for profit-making, and then was content to sell his wine as a standing crop, strikes me as a rather eloquent example of a well- internalized and all but insurmountable disjunction between agricultural production and the business of being a trader.

  • [1] Lindsay, 39. 2 Dion (1959), quoting Legrand d’Aussy (1782).
  • [2] 5 CIL IV. 138.
  • [3] [Nowadays, I would be less cautious; see p. 15.]
  • [4] Yaron (1959); Frier (1983: 258).
  • [5] Stamps have been retrieved from the wreck at Cape Negret (Ibiza) and from the Gulf of Fos;see Hesnard and Gianfrotta (1989).
  • [6] Joncheray (1994: 18). This wreck has yielded up some cork stoppers completely shaped,others barely started, and intact sheets of cork.
  • [7] See Lagi De Caro (1985: 552-4). I also draw, with much gratitude, on a seminar presentedin Marseilles (December 1992) by Dr Lagi De Caro.
  • [8] Heitland (1921: 322). The idea made its first appearance in the Venice edition of 1519 byG. M. Cattaneo.
  • [9] Billiard (1913: 183) is mistaken in saying that the texts by Cato do not refer to a bulk saleand that the final transaction depends on the later calculation of the amount of wine. Forevidence against that, see Dig. XVIII. 1, 8, and 39.
  • [10] IX. 16, 1; 20, 2; 28, 2.
  • [11] See also the text by Seneca quoted at the end of this chapter. As for the use of the plural, it isconstant in Pliny, who never uses vindemia in the singular.
  • [12] Qui pendentem vindemiam emit...sive lectam uvam calcare sive mustum evehereprohibeatur...
  • [13] [Pliny gives further evidence of cellars for wine storing and ageing belonging to traders(XXIII. 40); these were the traders who invented smoke-ageing in their cellars: mangones ita inapothecis excogitavere].
  • [14] Christol and Plana-Mallart (1997, 1998).
  • [15] Nibodeau, Nin, and Richarte (1989).
  • [16] Llinas i Pol and Sagerra i Aradilla (1993: 106): ‘The dimensions of the site would appear torule out a villa, unless what we have here is one of great proportions, along with manyouthouses . . . The layout of any urban settlement is completely uncertain and there is noevidence of streets or notable buildings.’ Even Christol and R. Plana-Mallart (1998) hint thatthey cannot rule out production, when they mention ‘structures from the first century ad withproducing and storing functions’.
  • [17] Dig. XXXIII. 7, 12, 1 and perhaps XLVII. 2, 21, 5. Also Cicero (Verr. 2.v.46). To my mind,Dig. XIV. 3, 4 makes it appear far from certain that the goods conveyed by institores ad homineshonestos are agricultural products, the meaning of the passage being actually ambiguous; seeAubert (1994: 7, n. 26). On this problem, see Di Porto (1984b: esp. 3238-41); de Ligt (1993:164-5). I need not stress my own preference for much less assertive language than Di Porto uses;his article suffers from its reliance on an outmoded interpretation of the stamps of Sestius andL. Titius on the amphorae and stoppers of amphorae from the Grand Congloue wreck.
  • [18] The text of Varro 1, 2, 14 (vilicus... appellatus a villa, quod ab eo in eam convehunturfructus et evehuntur, cum veneunt) is more ambiguous: the symmetry between bringing (con-vehere) the crops to the villa and carrying them out (evehere) at the time of sale may be taken tomean merely that they were being brought out of the storehouses or at the most taken outside thegates of the estate.
  • [19] Whittaker (1985: 58-9). Serving wine of one’s own making is an elegant touch: Fronto, eloquentia, I. 1 (Naber, 115) and see Veyne (1979a).
< Prev   CONTENTS   Source   Next >