A graffito from La Graufesenque and \'samia vasa\'

A graffito from La Graufesenque and ‘samia vasa’ By Anthony King The recent excavations at La Graufesenque have uncovered a number of new graffiti. One of these (A25b) is considered here for the light it sheds on the Latin name for samian ware or terra sigillata.1 The potteries that produced samian ware at La Graufesenque have long been known for the graffiti, now numbering about a hundred, which were scratched onto the plates made in the kilns there. The graffiti have been used to elucidate the organization and the possible methods of production at the potteries, although the cryptic and fragmentary nature of most of them has made detailed discussion difficult. The common formula of the texts consists of a list of potters, each coupled with a name of a particular vessel and a number representing the amount of that pottery produced within a time-span that is not now ascertainable. A25b, and a small portion of a similar text, A26b, stand out from amongst these, however, as they appear to give a monthly tally of tasks carried out by various slaves. The names of these slaves also appear on other graffiti of the more usual sort and it can be reasonably assumed that A25b is also connected with the pottery industry. In fact, certain words in it bear this out, as the text and translation which follow demonstrate. ] . ateliae puerorum ex xi aucust[is ]in x k septebres se]cundus agileius dies xiiiis ar[gilam ]dierum xxx iii ad capuries xi[ ca]Iistus o]nesimus ad sabros ii ad crau[cinam ]ad craucinam iii it onesimus[ mat]eriem erigenda i ] dierum xxx ]ae iii calistus ad samiandum [ ]… eos viciedos iii mercatu a[d em ].s materi erigedam [ ]argilam iii di[es ] ]s ada[

{140}

1

(Tally) of the slaves of Atelia from 11 (Kalends) August (22 July)[ ] to 10 Kalends September (23 August). Secundus and Agileius 14½ days at the clay. ]out of 30 days, 3 at Capuries, 11?[ Calistus Onesimus to Sabros 3 days, to Craucinam[ ]to Craucinam 3 days and Onesimus[ ]cutting wood 1 day ]out of 30 days

Number A25b in R. Marichal, Revue des Etudes Anciennes Ixxvi (1970), 266-92, where a photograph is published. It is also in ibid., Comptes Rend. de l'Acad. des Inscr. (1971), 188-212. It was brought to my attention by Paul Middleton during a seminar at the Institute of Archaeology, London in May 1978.

] ? 3 days. Calistus ad samiandum [ ]… eos Viciedos 3 days at [ ] market ].us cutting wood [ ]at the clay 3 days[ ].us at/to ?[ At least five people are listed, all carrying out a variety of activities which can be divided into the visiting of other localities or the preparation and making of pottery. It is the latter group which is of interest here. The dates suggest that potting activity went on in the summer, as might be expected, when conditions were good enough to avoid the use of drying kilns. Argilam probably means preparing the clay for throwing. Clay-digging is unlikely to have been the activity represented for it would have been dug out the previous winter, when the ground was soft and so that the extracted clay could have been suitably weathered. Cutting wood is presumably an activity associated with the firing of the kilns, although this connection can only be inferred. This leaves the word samiandum which, as its context would indicate, probably represents some aspect of potting activity. 2 Further elucidation of the meaning of this word can only be attained by examining its context in other passages, and it is to them that this study will now turn. Verbal usages of the stem sami- or sam- are rare, occurring only in two late texts which are as follows:3 arma tersa sint, ferramenta samiata, calciamenta fortia Let their weapons be clean, their tools polished, their boots sound. Historia Augusta, Aurelian 7 ... loticas vel catafractas, contos et casides frequenter tergere, curare, sameare. (It is the responsibility of the decurion … to ensure that his men) often clean, care for and polish their breast-plates or their armour, and their pikes and helmets. Vegetius 2.14 Both are used in contexts that imply the cleaning of armour, and as such would to be of no help in defining samiandum in the graffito. However, these passages will be seen later to have a vital bearing on the meaning of the word. As a word connected with pottery samiandum has many early and well-known precedents. The stem occurs as an adjective three times in Plautus, qualifying vas (Bacchides 202), vasis (Captivi 291) and poterio (Stichus 694), in Cornificius in a use where pottery is implied (Rhe. ad Herr. 4.51/64), in Lucilius with testam (Sat. 7.21) and with catino (Sat. 13.5), twice in Cicero with capudines (de Rep. 6.2.2) and vasa (pro Mur. 75), in Tibullus with testae (2.3.47), in Valerius Maximus with vasa implied (7.5.1) and twice in Pliny, firstly with an implied fictile connection (NH 35.160) and later with testa (NH 35.165). Late uses occur in Serenus Sammonicus with olla (671), in Tertullian with vasa (Apol. 25.13), in the Vulgate; testa de samiis terrae (Isaiah 45.9) and in Ausonius with luto in a sentence about pottery (Epig. 2). 4 {141} 2

The grammar of the phrase ad samiandum is faulty. See Marichal, CRAI op. cit. (note 1), 198, for a discussion of this point. 3 From the verb comes the noun samiator mentioned in Diocletian's price-edict (7.33-7) and in the Corpus Glossariorum Latinorum 2.117.22; 2.178.8; 2.223.12 and 3.25.57. 4 Tertullian advers. Marcionem 3, 5, 3 uses samias in the phrase ... (non tamen) ut de glebis credas te umquam placentas et samias coacturum... Facciolati and Forcellini in their Lexicon under samia suggest that it is a type of cake, similar to the meaning of placentas, but considering the more general meaning of samias in connection with pottery, it would seem more sensible to ascribe it the latter meaning here also. Plautus (Menaechmi 179) has a transferred use of samiae in connection with a fragile door.

Two other late Roman writers made the first attempts to derive the word. Nonnius ascribes it two meanings; samium est testeum (398.26) and samium rursum acutum (398.31) in support of which he quotes the two passages from Lucilius and the passage from Cicero's pro Murena mentioned above. Neither meaning is strictly accurate, for samium is only one specific kind of testeum, despite its almost cliched usage in the Republican writers, and acutum in fact refers to the implied meaning of testam … samiam rather than simply samiam in the passage from Lucilius about castration (Sat. 7.21). Isidorus is more circumspect. He states that putant ab oppido Sarno Graeciae habere nomen (Etym. 20.4.6. See also ibid. 20.4.3 and 14.6.31). This geographical derivation is undoubtedly right, and there was possibly a time during the Republic when pottery from Samos, or pottery that the Romans thought was from Samos, was in use in Italy. By the time of Plautus, however, the geographical meaning of the adjective when applied to the pottery had been lost, and it would appear that from the late Republic onwards the word had a wider meaning. In all probability a distinctive quality of the original pottery had been copied elsewhere and the name continued to be applied to the copies, in much the same way as 'china' is now accepted as a general term for fine earthenware. A clue to the nature of that distinguishing feature is contained in the graffito discussed at the beginning of this paper, to which we will now return. The verb samiare, as we have seen, means to polish. This must be due to the widespread availability of some sort of polishing stone or powder referred to as terra samia which, as Galen and Pliny indicate, was originally quarried in Samos, but may also have come from other deposits.5 The ancient sources make clear that this was the material used for polishing gold and other metals. It was probably kaolinite to judge from its description and its medical properties. 6 If we bear this in mind, the word samiandum in the graffito takes on a new significance. It must refer to some potting activity that gives the vessels a polished appearance. Evidence from samian ware kiln sites does not support any notion that the pots were actually polished with terra samia, but of course the possibility of finding some in future excavations cannot be ruled out. Thus the word in the context of the graffito must have a transferred meaning relative to the feature of the pot that appears to be polished, namely the slip or gloss. Examination of methods of applying the slip on samian ware would suggest that it acquired its polished appearance during the preparation of the clay base in the slip, and is dependent on the fineness and concentration of the clay particles in the slip and the addition of a suitable deflocculant to the slip-suspension in order to ensure that fineness and concentration. A samian slip has to be shiny before the vessel is fired for it to have the familiar gloss in its fired state, and the shininess is most probably due to the alignment in the same direction relative to the surface of {142} the vessel of the clay particles in the slip and this will only occur when a very fine slip has been prepared. It is probably not dependent on the presence of illite or montmorillionite as samian-type slips have been produced experimentally from non-illitic clays. 7 Thus, samiandum probably refers to the preparation of the slip from the clay. 5

Pliny NH 28.194; 31.117; 35.191 and 36.152. Galen Methodus Medendi 4; de Simplic. Medic. Temp. ac Facult. 9; de Succedaneis s.v. Samia (C. G. Kuhn (ed.), Claudii Galeni Opera Omnia x (Hildesheim, 1965), 298-9; 12, 178 ff. and 19, 727). Isidorus Etym. 20.4.6 mentions that the cretam for samian pottery possibly came from near Rome. This may be a confused reference to a quarry for samian earth. Other relevant references are Isidorus Etym. 16.1.7 and 16.4.13, Corpus Glossariorum Latinorum 5.330.39; 5.391.48 and 5.610.17, Dioscorides 5.153-4, Oribasius 13 s.v. Γή, Aetius 1.2.6, Marcellus Empiricus 8.5, Paulus Aeg. 7.3 s.v. Γή, and Turnebus Advers. 27.21 (on CIL VI.10242 or VI.25857, cf. also VI.25856). The nomen samiarius on these inscriptions may be occupational in origin, for in the Corpus Glossariorum (2.223.12) and in Joh. Lydus (de magistr. 1.46) this word is given an occupational meaning. 6 K. C. Bailey, The Elder Pliny's Chapters on Chemical Subjects (London, 1932), part 2, 240-1 and 264. Kaolinite is used today to give the shine on glossy paper. Kaolin deposits on Samos are referred to in Pauly-Wissowa's Real-Encyclopädie s.v. Samos cols. 2183-4. 7 J. D. Bestwick and T. A. Smith, 'The surface finish of samian ware', Science and Arch. xii (1974), 21-31; A. Winter, 'Terra Sigillata und andere Glanztontechniken', Keramische Zeitschrift viii, No. 3 (1956), 110-3; ibid., 'Terra Sigillata und "Firnis",' Saalburg Jahrbuch xiv (1955), 74-9; ibid., Die Antike Glanztonkeramik. Praktische

Polished slip or gloss was first successfully used on Greek red-figured ware, and it may be to this that the Latins first applied the term samia vasa. However, if the hypothesis developed here is correct, this is not because of the mistaken impression in Italy that all Greek pottery came from Samos, but is due to the impression – also, of course, mistaken – that the gloss was due to the use on the pottery of terra samia to bring up a shine. The word later became an epithet for glossy pottery and hence in poetry for fine pottery in general. It passed into common parlance enough for the potters at La Graufesenque to scratch on a discarded plate that they were making the pottery glossy by samiandum, when in fact they were simply preparing the slip. One difficulty remains. Pliny's passage on vasa samia (NH 35.160) is usually translated with the meaning that Samian ware is still praised for use at sacrifices but that Arretine, Sorrentan, Po Valley, Spanish and Pergamene pottery is also used. This would imply that the vasa samia in question are from Samos itself, or possibly an earlier type of pottery in Italy like Campanian blackgloss ware, and that despite the polished appearance of the other wares where they are attested archaeologically they do not merit the adjective samius, and therefore that samius does not mean polished at all. In fact, this problem can be circumvented by retranslating the passage as follows: maior pars hominum terrenis utitur vasis. samia etiam nunc in esculentis laudantur. retinent hanc nobilitatem et arretium in italia et calicum tantum surrentum, hasta, pollentia, in hispania saguntum, in asia pergamum. 'Most people use earthenware vessels.8 Samian vessels are even now praised for use at meals. Arretium in Italy, and equally Surrentum, Hasta and Pollentia, and Saguntum in Hispania, and Pergamum in Asia have retained this noble use for their vessels.' In this way samia is a generic term for the pottery from the geographical locations in the following sentence, which is a list of production centres for that ware. The existence of an Arretine potter called L Tettius Samia and of Arretine workshop slaves called Samus are relevant in this connection, for they might indicate that samia was a term in use at Arretium for the pottery made there, and which also became a nickname for one of the potters. 9 The slave names may also have had the same derivation but in these cases a geographical origin cannot be ruled out. Thus, there is some independent evidence that the Versuche (Mainz, 1978). Marichal (in REA op. cit., note 1) cites evidence from the Banassac potteries that the slip was brushed, possibly to polish it, but the position of the brush-marks in the bottom of the bowls makes it more likely that the marks are a result of mopping up excess slip. Brush-marks would be much more frequent on the vessels found as kiln waste if this was the only way of producing a polished appearance. 8 The reference to sacrifices (in sacris) at the beginning of 35.160 does not apply to this section of the text, but only to sentences in the preceding sections (35.158-9). In fact, Pliny's line of thought changes in section 160, so that, if we were to divide the original text into paragraphs as in a modem piece of literature, a suitable beginning to the paragraph on pottery would be the line halfway through section 160, maior pars hominum terrenis utitur vasis, which would form the opening sentence of his consideration of pottery vases that closes with the phrase ut Regium et Cumae in section 165. Therefore, the text and translation has deleted in sacris (at sacrifices) and translated in esculentis as 'at meals' or possibly 'as tableware', rather than ‘at food offerings’. 9 A. Oxé and H. Comfort, Corpus Vasorum Arretinorum (Bonn, 1968), refs. 1967-74 and commentary for L. Tettius Samia, and refs. 476 and 1441 for the slave names. It must be pointed out that in B. Hartley, 'Some wandering potters', p. 252 and note 9, in J. Dore and K. Greene (eds.), Roman Pottery Studies in Britain and Beyond (Oxford, 1977), evidence is noted from stamps of potters at La Graufesenque and Lezoux that early workers at those centres thought of themselves as making (vasa) aretina. It is probable, however, that they are simply copying similar stamps from the original workshops (cited in Oxé and Comfort op. cit. this note) in an attempt to pass the pots off as Arretine products. Arretine vessels seem to have been sufficiently distinctive to have been noticed separately by several literary sources (W. Hilgers, Lateinische Gefassnamen (Dusseldorf, 1969), 22, note 70) and, presumably, by the populace in general. Arretinus or an abbreviation stamped on the pottery would probably have been a good selling point.

usual translation of the passage is mistaken and can be reworked to fit in with the suggested origin and use of the term samius. {143} In conclusion, we have seen that the words samiandum and also samius have a definite connection with the pottery industry of La Graufesenque, and an explanation has been put forward for the nature of that connection. The last point to raise is its relevance to the modern terms for the pottery – terra sigillata or 'samian ware'. If, as seems likely, samius was a general term applied to glossy pottery, it would seem in order to use and even prefer, the modern 'samian ware'. This in fact was the case in Britain and Germany during the nineteenth century. Terra sigillata was also used in France at this time, and in the wake of subsequent preference by the Germans for this term and attempts to change English usage away from 'samian ware', 10 the former description has enjoyed a vogue amongst Roman archaeologists. In fact, terra sigillata is an abuse of the original meaning of the phrase's two constituents. Terra means earth, not pottery, and sigillata means 'with little figures' and thus should only be applied to decorated ware. 11 This reminder of the frequent mishandling of Latin words in modern contexts reinforces the case for using the term 'samian ware'; a phrase which provides suitably general terms of reference for the pottery in question and which derives from a phrase probably actually used in the Roman period for red- and black-gloss pottery.

Institute of Archaeology, London.

10

F. O. Waagé, 'Vasa Samia', Antiquity xi (1937), 46-55. His case for reserving 'Samian ware' (with a capital letter) for pottery from Samos is quite reasonable. However, he also dismisses 'samian ware' as a general term without sufficient justification. Cf. also A. H. Pitt-Rivers, 'The Roman red-glazed pottery', in Excavations in Bokerly Dyke and Wansdyke (London, 1892), 294-6. 11 In fact terra sigillata refers to the practice of stamping tablets of medicinal earth which was prevalent in the Middle Ages (H. Brunsting, 'Terra Sigillata', Westerheem, xxi (1972), 252-68; J. W. Mellor, 'Terra Sigillata not Samian ware', Trans. Ceramic Soc. xxiii (1924), 176-7), and derives ultimately from the tablets of Lemnian earth stamped with the figure of Artemis or her symbol, a goat (Galen, de Simplic. Medic. Temp. ac Facult. 9, chap. 2. C. Fredrich, 'Lemnos', Mitt. des deutschen archäol. Instituts: Athen. Abteilung xxxi (1900), 72-4. L. de Launay, 'Notes sur Lemnos', Revue Archéologique ser. 3, xxvii (1895), 318-23). Early antiquaries probably thought that pieces of decorated samian ware were these tablets of terra sigillata and, accordingly, the term was transferred to the pottery. Cf. also W. Hilgers op. cit. (note 8), 22 and 26. Brunsting's article, cited above, contains a full consideration of the use of the term terra sigi!lata as applied to samian pottery by nineteenthcentury archaeologists.

© Anthony King 1980, 2011 & 2015 Published by Prof. A. C. King, FSA, FRAI, Department of Archaeology, University of Winchester, Winchester, Hampshire SO22 4NR, UK. Anthony King hereby asserts his moral right always to be identified as author of this work in accordance with the provisions of the UK Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced in any form whatsoever or by any method whatsoever without prior permission of the author. Quotations may however be taken for academic purposes. This paper is a revised version, incorporating a corrigendum, of an article originally published in 1980 as: King, A C 1980. A graffito from La Graufesenque and ‘samia vasa’. Britannia, 11, 139-143. Published online: 09 November 2011 doi:10.2307/525677 Link to this article: http://journals.cambridge.org/abstract_S0068113X00014306 The corrigendum was originally published in 1981 as: King, A C 1981. A graffito from La Graufesenque: corrigendum. Britannia, 12, 311. Published online: 09 November 2011 doi:10.2307/526266 Link to this article: http://journals.cambridge.org/abstract_S0068113X00015087 The original page numbers of the 1980 article are indicated in the revised paper as {140}, etc. Copies are available for download from the University of Winchester Data Repository, and from academia.edu (Anthony King’s profile page). Anthony King is Professor of Roman Archaeology at the University of Winchester. He has directed excavations on many Roman sites in Britain and Italy, as well as being a leading specialist in the analysis and interpretation of animal bone assemblages from Roman sites. He is author of Roman Gaul and Germany (1990), the chapter on Mammals in Jashemski and Meyer’s The Natural History of Pompeii (2002) and Coins and Samian Ware (2013).