Version classiqueVersion mobile

Les savoirs professionnels des gens de métier

 | 
Nicolas Monteix
, 
Nicolas Tran

Exploring the limits of skilled craftsmanship: The fullonicae of Roman Italy

Miko Flohr

Texte intégral

  • 1 Bakers (pistores): CIL IV, 875 (IX 3, 4-5), 886 (IX 3, 8-9), 7273 (I 8, 6-7); green-grocers (pomar (...)
  • 2 See for this traditional idea e.g. Finley 1973 [1985], p.40.
  • 3 The advises with explicitly mentioned rogatores constitute only about 20% of all texts; only a few (...)

1A remarkable, and perhaps puzzling aspect of Pompeian epigraphy is the degree to which the painted election slogans on the city’s façade feature references to crafts and occupations. There are electoral advises by, to name only a few examples, bakers, green-grocers, goldsmiths, wool-dyers, mule-drivers, and fullers1. Indeed, there are few stronger arguments against the traditional, Finleyan idea that Roman craftsmen were socially marginal people that spent their lives fighting against the disdain of their fellow citizens: here we have an urban community in the heart of the Roman empire, which was closely familiar with the normative preferences of the imperial elite, and yet there are large amounts of supposedly marginal and irrelevant people that actively decided or were encouraged to use their business as an asset in urban politics2. They could easily have left it away, had their occupation been inconvenient to the intended audience: as Henrik Mouritsen has shown, most of the recorded electoral advises were anonymous, and many others only mention the name of the rogator, without further specification of his or her background3. Yet, a not insignificant amount of craftsmen chose to add an explicit reference to their daily work, which suggests that either they themselves, or the people who encouraged them to have these slogans painted, suspected that their craftsmanship would add something to the credibility of their message.

  • 4 For an overview of this evidence see Flohr 2007, p.148. To the fullonicae mentioned in that list, (...)
  • 5 CIL IV, 875 (IX 3, 4-5), 886 (IX 3, 8-9), both related to bakery IX 3, 8-9.
  • 6 CIL IV, 494 (between VII 5, 13 and VII 5, 14); CIL IV, 537 (between VII 5, 17 and VII 5, 18). In b (...)
  • 7 Moeller 1976, p.90.
  • 8 Jongman 1988, p.155-186; Ph. Borgard (et al. 2003, p.22-23) mainly reiterates the claims of W. Moe (...)

2However, what is even more remarkable is that one group of craftsmen clearly stands out compared to the others. Most occupations figure in only one or two slogans, and a few have three. Yet, not less than seven electoral advises can be related to fullones, and these were found on five different locations in the city, whereas none of the other occupations is found on more than two façades. This has nothing to do with the amount of fullonicae in the city: in the excavated area of Pompeii, 13 fulling workshops may be identified, but there were at least twenty-six bakeries4. Yet, for some reason or another, bakers used their occupation in only three slogans, two of which were found on the façade of one and the same bakery5. Similarly, there probably were many caupones in the city, but only two of them have possibly used their occupation to support a candidate for local office6. Why did the fullones of Pompeii stand out in their electoral credibility? The traditional explanation, advocated by Walter Moeller, has been that Pompeii was a centre of textile production and that the fullones, as main agents in the industry, were a socially prominent group whose electoral preferences counted7. This, however, is implausible – there is little reason to presume a Pompeian textile industry of any significance, as W. Moeller conjectured: this has been convincingly shown by Willem Jongman twenty years ago, and despite some recent statements to the contrary by Philippe Borgard, most of W. Jongman’s arguments still stand8.

  • 9 This also has to do with the shape of the ancient loom, which was unable to produce the broadcloth (...)
  • 10 Flohr 2010, p.71; Wild 2002, p.22.
  • 11 Significantly, the only large fullonica without election notices is VI 8, 2.20-21, which also did (...)

3A much more logical interpretation for the enhanced electoral credibility of fullones may be sought and found in analyzing what most of the fullones responsible for the electoral advises actually did in their daily work. Fullers, of course, worked with textiles, and, if we are to believe the Roman sources, with clothes rather than with cloth9. As is widely known, fullones dealt with new as well as with used garments. The aim of their work was not only to remove pollution, but also, and most prominently, to make the clothes comfortable to wear and good looking by raising, shearing and polishing a ‘nap’, a thin, interlaced layer of fibres on top of the textile10. Economically, fullers may be oriented towards private customers or, in the case of new clothes, towards people involved in the clothing trade. In this respect, it is worth noticing that the electoral advices of fullones without exception have been found in relation to fulling workshops which had or were situated in a shop and which thus were – at least partially – orientated towards private customers11. Without excluding the possibility that part of the fulling capacity in these establishments was used for new clothes that still needed to be sold on the market, it is rather likely that these fullonicae and the fullones working in them were known to the public for what they did in maintaining clothes belonging to the personal wardrobe of the individuals who brought them (or had them brought) to the workshop. The public profile and, hence, the credibility, of these fullones thus seems strongly related to the service they delivered to their fellow citizens.

  • 12 These inventories sometimes even specify whether clothes had been fulled or not. Cf. P. Mich. II, (...)
  • 13 Cf. Robaye 1991.

4In that respect, it is worth noting that the contact between a fullo and his customer is much more intense than that between most other craftsmen and their customers. This has a practical and an emotional aspect. On a practical level, it is likely that in many cases a fullo and his client needed to dis cuss, to some detail, the treatment of the garments that were delivered and negotiated what needed to be done for what price and when it needed to be ready. A client, in turn, had to make an active decision to bring the garment to the fullo, which involved leaving home for the specific purpose of going to the fullery, taking the garments with him, and choosing the right direction that would lead to the fullonica of choice. Unlike food and drink outlets, or indeed most other shops, fullonicae could not count on spontaneous sale. Going to the fullo, thus, was a much more conscious process than, for instance, buying fruit and legumes. Secondly, on an emotional level, it must be emphasized that these Pompeian fullones worked with the private property of other people, and that there was a considerable amount of trust involved in the fulling business. Clothes were highly valued personal belongings that were of vital importance to the social performance of their owner. Detailed descriptions of private wardrobes feature on inventories of the possessions of deceased people from Roman Egypt, and several of the curse tablets found in the sacred spring at Roman Bath were directed at the thieves of cloaks and garments12. Handing clothes over to someone else to have them treated was a sensitive issue, as is emphasized by the extensive discussion of the legal responsibilities of fullones in their dealing with another person’s property in the juridical corpora13. It is unlikely that people would easily bring their clothes to a fullo with a bad reputation in the handling them. As a consequence, a fullo working for private customers needed to have his craftsmanship, to some degree, publicly recognized in the community in which he operated, otherwise he would not have any business. This publicly recognized craftsmanship, in turn, could be a valuable asset in local elections: fullones were better known and probably had a more sharply delineated reputation than most bakers, green-grocers and shoemakers. Pompeian fullones were using their occupational title in electoral advises simply because it gave their recommendation more authority.

  • 14 SGDI 2, 1904 (Baunack); Wescher – Foucart 1863, p.289. Baunack dates the inscription between 156 a (...)

5In a certain sense, this puts the fullo at the heart of the discussion about the meaning of craftsmanship and professional knowledge in the world of Roman manufacturing: while individual and group identities based on craftsmanship and professional knowledge obviously were rooted in real artisanal skills, the above example shows how the public recognition of these skills was not only vital to economic success, but also could play a role in social processes not directly related to business. Moreover, it emphasizes that the nature of what you did could matter in this respect. It is not hard to see fullones as highly appreciated artisans who were specialist carriers of the professional wisdom that they had inherited from the previous generation and were to pass on to the next; while there are no direct indications from the Roman period that the craftsmanship of a fullo was thought of in these terms, an interesting illustration of this may be found in a contract inscribed on the walls of the sanctuary of Apollo at Delphi around 150 BC14. This text records how a certain Dromokleidas buys a young slave boy, Sosas, and sends him, for a certain period, to a fuller, Artemidoros, to learn the skills necessary to perform this craft; the contract specifies that Sosas, after having learnt the trade, will have to work as a fuller for the household of Dromokleidas. Fulling, it seems, at least in late Hellenistic Greece could be seen as a craft that needed to be learnt from a specialist. There seems little reason to assume that this was fundamentally different in the Roman period: a decent Roman fullo probably was seen by his fellow citizens as someone who had trained professional skills that enabled him to do things that others could not do. Yet, it is the socioeconomic meaning of those skills – rather than the skills themselves – that is of relevance here.

6In that respect, it may be argued that the material and written evidence for fulling from the Roman world also encourages us to explore the limits of the idea that skilled craftsmanship and professional knowledge are the key to our understanding of the daily world of people involved in artisanal activities. The focus of this paper is on these limits: what may be a useful way to look at the situation of certain fullers not necessarily constitutes the most suitable approach to all people involved in fulling in the Roman world. Indeed, it may be argued that, for understandable reasons, the literary record tends to be biased towards fullones working for private customers: this was what Roman literary authors and their elite audience used to see of fulling in the world in which they lived. Though there are no texts suggesting this, it seems safe to suggest that for fullers who did not usually work for private customers, the social mechanisms related to their profession and their occupational skills may have worked rather differently than for their colleagues. Moreover, the model of the fullo as a skilled artisan defined by his professional knowledge works particularly well in understanding the role of skills and craftsmanship in relatively small-scale workshops that functioned under the direct leadership of one experienced fullo, but it may be less apt to make sense of larger workshops with a large group of workers and more complex and hierarchical divisions of labour. Thus, to be able to understand the degree to which skilled craftsmanship and professional knowledge actually were important aspects of the world of people involved in fulling, it is necessary to take a closer look at the evidence.

Fullones and fullonicae in Roman Italy: the evidence

  • 15 The amount of evidence for fulling from outside Roman Italy is minimal and not easily compatible w (...)
  • 16 Flohr 2006, p.193-194.
  • 17 Flohr – Wilson in press.
  • 18 Flohr 2005, p.59-60.
  • 19 See further Flohr 2010, p.146-159; Flohr 2006.
  • 20 Rinsing is a necessary part of the fulling process, as it is impossible to wear clothes treated wi (...)
  • 21 See Flohr 2010, p.130-134.

7Compared to most other crafts, the activities of fullers in Roman Italy have left a significant body of both material and immaterial traces: immaterial in textual references by Roman authors, and material through epigraphy, iconography, and archaeology15. Key to the interpretation of this evidence is an understanding of the fulling process. As I have argued elsewhere, it is most convenient to think of fulling as a procedure that consisted of three main phases: soaping, rinsing and finishing16. In the first phase, the garments were treated with alkaline chemicals by trampling them, scrubbing them and wringing them out, as has been depicted on the famous paintings from a Pompeian fullonica now in the museum of Naples (fig. 51). The chemicals used in this phase included certain natural clays, commonly known as fuller’s earth (creta fullonica), and ammonia won from aged urine, though the role of the latter probably has been exaggerated by modern scholars17. Soaping was done in so-called fulling stalls, niches surrounded by low walls with a tub fixed in the ground. These fulling stalls play an indispensible role in identifying fullonicae in the archaeological record: without fulling stalls, the identification of fullonicae often is rather uncertain18. After the chemicals had done their work, they needed to be washed out. For this purpose, large fullonicae had complexes of several rinsing basins that were connected to the urban water system (fig. 52). These rinsing complexes were organized in a rather intelligent way: water went in one direction through the complex from the first basin, where the water-supply mouthed, to the last basin, where the drain was. Clothes ran the opposite direction, and thus gradually came in cleaner water. This design not only allowed for the reuse of water, it also streamlined the production process19. At Pompeii, some small fullonicae did not have rinsing facilities, and rinsing probably was done with water from the public fountains on the street20. After rinsing, and, perhaps, drying, it was time to raise the nap and to trim it. This was done by brushing and shearing, and finally, by putting the garment under a textile press, so that the nap was pressed against the textile and ended up more stable (fig. 53). In this last phase, there also was room for some treatments that probably were optional, such as, most prominently, sulphuring and chalking, but it is unclear how commonly these were performed21. With respect to the theme at stake here, it may be noticed that the fact that fulling consisted of several phases means that there were different kinds of ‘professional knowledge’ to be obtained by people involved in the business, and that not all people involved in fulling necessarily obtained the complete range of fulling skills: this had to do with the division of labour and with the degree to which people working in a fullonica actually were involved in all activities. This issue will be further discussed below.

Fig. 51. Painting with fullers at work from fullonica VI 8, 2.20-21 at Pompeii (MANN, Inv. no 9774; photograph N. Monteix).

  • 22 Monumentum Neronis Drusi (CIL VI, 4336); Monumentum Marcellae (CIL VI, 4445); Monumentum Statilior (...)
  • 23 Flohr 2010, p.81-82; contra Dixon 2001, p.9-10.
  • 24 AE 1958, 273; cf. Ferrua 1957, p.609.

8Overlooking the evidence for fulling, it becomes immediately clear that the activity was performed in a variety of contexts. There is epigraphic evidence for fullers working for some large elite households in Rome: in the funerary monuments of the Statilii Taurii, the Volusini and several imperial households, epitaphs of deceased fullones were found22. Rather than being seen as evidence for the involvement of such households in textile production, these inscriptions may be thought to suggest that these families were large and wealthy enough to have their own private fulling service23. There is also epigraphic evidence suggesting there was a fulling workshop associated with one of the macella at Rome, probably that of Livia24. If this workshop really was in the macellum, it is likely that it was situated in a shop, and thus of limited size and capacity, and oriented towards private customers. While it is possible that macellum here is used as a vaguer kind of toponym, referring to some place in the environment of the macellum, the emphasis on the location suggests that the deceased could be associated with it by others – and thus worked in an open, visible, and presumably commercial environment.

  • 25 Flohr 2010, p.83-85.
  • 26 I take the number of fulling stalls as an indicator of scale because other facilities, particularl (...)
  • 27 This fullonica has recently been excavated at Casal Bertone, some two miles outside the Aurelian w (...)

9Yet, the most detailed information about the socioeconomic contexts for fulling is to be found in the material remains of fulling workshops. At this moment, there are 24 identifiable fullonicae in Roman Italy. Thirteen of these are at Pompeii, six at Ostia, three at Rome, one in Herculaneum and one in Florence. The scale and economic orientation of these workshops varied considerably. A large majority of fifteen fullonicae had a shop, but it is worth emphasizing that seven workshops certainly did not have a shop (fig. 54)25. As to the variation in scale, most workshops have two or three fulling stalls, a few have between five and ten, and four fullonicae were of exceptional size, having thirty or more stalls (fig. 55)26. The largest fulling establishment discovered so far counts an amazing amount of 97 stalls, which makes it almost fifty times as large as the median excavated fullonica27. For the present purpose, it seems most useful to subdivide this data set into three groups based on the architectural context in which the workshops were found: fullonicae in tabernae, fullonicae, in Pompeian atrium houses, and fullonicae situated in purpose-built production halls.

Fig. 52. Complex of rinsing basins in fullonica VI 8, 2.20-21 at Pompeii (photograph M. Flohr).

Fig. 53. Cupid brushing textile from the frieze with fulling cupids in oecus q at the Casa dei Vettii at Pompeii (VI 15, 1.27; photograph M. Flohr).

Site

Fullonicae

With ‘shop’

Unclear

No ‘shop’

Pompeii

13

11

-

2

Ostia

6

3

-

3

Rome

4

2

1

1

Herculaneum

1

1

-

-

Florence

1

-

-

1

Fréjus

1

1

-

-

Total

26

15

1

7

Fig. 54. Overview of fullonicae with and without a shop.

  • 28 Flohr 2007, p.135-137; Flohr in press.

10Thirteen of the fullonicae were situated in tabernae, small units organized around a shop. These workshops were, by nature, oriented towards private customers, and were rather small: most had only two or three fulling stalls, though two had four. At Pompeii, most tabernae were directly associated with living space behind or above the shop, but at Ostia, this seems less often to have been the case. The four fullonicae in atrium houses were all situated in Pompeii and Herculaneum, and apart from the Herculaneum workshop, they were considerably larger in size: the three examples at Pompeii have five, six and ten fulling stalls respectively, and each had a complex of three interconnected rinsing basins. Three of these workshops had a shop, but one does not seem to have had a space for interaction with private customers. It may be pointed out that the houses in which these fullonicae were situated were not ‘converted’ into workshops: there is clear evidence that during or even after the construction of these workshops the residential qualities of these complexes were improved28. For example, the presence of piped water was used to embellish the atrium or the peristyle with decorative fountains, and fourth style decorations contemporary to or postdating the fullonica cover the wall of more than one room in each of the Pompeian houses (fig. 56).

Fig. 55. Fulling capacity of the fullonicae of Roman Italy.

  • 29 Flohr in press; NSA 1912, p.286.
  • 30 Ostia: fullonica II, xi, 2, III, ii, 2, and V, vii, 3; cf. De Ruyt 2001; De Ruyt 1996; De Ruyt 199 (...)
  • 31 That is, 251 of all 311 fulling stalls were found in these five workshops. Cf. fig 55.
  • 32 Cf. Flohr 2010, p.90.
  • 33 The fullonica of Casal Bertone was situated just east of Rome, near the point where the major rout (...)
  • 34 It should not be excluded at forehand that the workers in these fullonicae were actually paid for (...)
  • 35 For a more elaborate discussion of this problem see Flohr 2010, p.300-303.

11Moreover, finds of glass pastry beads and other objects suggest that these houses were inhabited by both men and women, and the paintings from one of the fullonicae suggest that children were involved, too29. In other words: the world of these workshops seems to have been one of private households, the core of which probably was constituted by a family of men, women and children. The third group consists of the five exceptionally large fullonicae. Three of these were situated in Ostia, one in Florence, and one in Rome30. What is striking about these workshops is their staggering scale: in number they constitute about a fifth of all fullonicae, but their aggregate capacity (measured in the amount of fulling stalls) constitutes no less than 81% of the excavated fulling capacity of Roman Italy31. Typically, none of these workshops had a shop: they thus seem to have worked for clothing traders rather than private customers, which is also logical given their industrial scale and the risk of garments getting lost or confused32. The location of the establishments in Rome and Ostia suggests that they served the metropolitan clothing market and dealt with imported goods33. Moreover, it is relevant to note that these fulling factories were, essentially, little more than production halls: there is no living space attached, nor is there any space that could be used for administrative purposes or, indeed, storage. Significantly, and remarkably, given the number of people at work, there was not even a toilet. The absence of living space suggests that these workshops were not operated by slaves belonging all to the same household, but by a variety of people living in a variety of places34. This also may be thought to suggest that these people, or some of them, were working for wage35.

12The scholarly significance of all this variety should not be underestimated: the context in which fulling took place, of course, could have consequences for the working lives of the people involved, and could have a profound influence both their occupational identity and their actual professional skills. Whether a worker fulled clothes in a taberna or in a full scale fulling factory made a difference in how he (or she) related to his daily work and profession, and may have considerable impact on the degree to which he saw himself and was seen by one’s fellow citizens as a skilled craftsman: like all identities, an occupational identity is to a considerable extent degree socially constructed in that it is usually negotiated and maintained in the social arena of daily life by the persons involved in the occupation and the people surrounding them. Indeed, an occupational identity is embedded in social interaction: you are not a fullo because you are involved in the craft, but because you and your environment decide, at some point, to attach that label to you on the basis of your involvement in the craft and because of your mutual personal relationships. Not all fullers also were fullones: not everybody who was involved in fulling necessarily thought of himself as a fullo or was perceived as such by the people around him or her. Moreover, besides these potential differences in the public perception and recognition of craftsmanship, the differences between the scale and functioning of workshops may also have led to real differences in the craftsmanship itself: the nature of the division of labour on the shop floor may have a considerable impact on the degree to which a worker in a fullonica actually was able to develop and practise all or most of the professional skills related to his craft.

Fig. 56. Fourth style wall-decorations in fullonica I 6, 7 at Pompeii (photograph M. Flohr).

The daily work environment and the development of occupational identities

  • 36 Cf. Flohr 2009.

13One of the keys to our understanding of the differences between fullers in occupational identity and craftsmanship is our interpretation of the social landscape in which fullers operated. In what kind of environment did they live and work? Several factors may be thought to have played a role in this respect, and they may be broadly divided into two categories: on the one hand, there are factors related to the daily life on the shop floor – that is, to the atmosphere among the staff of one fullonica and the nature of the social ties between the workers of that workshop36. Secondly, there are factors related to the daily interaction with outsiders. As to the first category, we may think of the way in which people actually could communicate during the working day, how they were related to each other outside working hours, how tasks were allocated and to which degree there was a fixed hierarchy within the work force.

14As to the second category of factors, we may think of the way in which people, during the work day interacted with the outside world, through contact with the neighbourhood and interaction with customers coming to bring or collect their garments.

Fig. 57. Plan of fullonica I 6, 7 at Pompeii (drawing M. Flohr).

The communicative landscape of fullonicae

  • 37 No fullonica, for instance, has less than two fulling stalls (see fig. 55).

15Few, if any, fullones seem to have worked completely alone. While there obviously were workshops with a larger and a smaller work force, even the smallest workshops had a production capacity that seems to have been designed for two or three people to work37. People who spend the day working in the same building obviously are likely to communicate. Part of this is necessary for the progress of the production process, but it is of course entirely human to go beyond what is strictly necessary and to develop, at least to some degree, more social forms of interaction – either verbal or non-verbal. Yet, the nature of such interaction is of course dependent on several factors. For example, communicative patterns in smaller groups often tend to be much more integrated than in larger groups: it is easier to have one central conversation in a small room with four people than it is in a large hall with fifty people – especially if they are spread over the entire area. In that respect, fullonicae in tabernae must have had a distinctly more intimate atmosphere than the larger fullonicae in atrium houses, which were split into two separate parts – a shop in front and a work area in the back – between which direct communication was impossible, though the work areas themselves were still rather intimate and integrated (fig. 57). Yet, the most spectacular difference is between the industrial fullonicae of Ostia, Rome and Florence and all the other establishments: these large workshops had a work force of thirty to more than a hundred people. The communicative landscape of such establishments is much more dispersed: workers are also positioned in long rows, which makes verbal communication harder, and non-verbal communication impossible, and the central position of the rinsing complex does not help either: it would have been impossible for fullers to reach someone at the other side of the area without seriously raising the voice or actually going there (fig. 58). In these workshops, many conversations could be going on at the same time, and participation in such conversations was more voluntary than obligatory: people could easily have spent the day relatively unnoticed. This made the atmosphere in these workshops distinctly more dispersed and anonymous than in the small and medium-sized fullonicae in shops and atrium houses.

The nature of ties within the staff network

  • 38 For the exhausting character of fulling see Sen., Ep. 15, 4.

16The differences between the industrial fullonicae and the other fulling workshops can even be made more substantial by analyzing the differences in the social context of these workshops: whereas the fullonicae of Pompeii were all related to domestic space and thus, probably, were staffed from a household, perhaps sometimes with a few outsiders, such a clear and intimate social basis was lacking in the industrial fullonicae of Ostia, Rome and Florence: as mentioned above, the people that made up the staff network are likely to have lived in different places, which means that they generally were not saddled with each other outside working hours. The impact of this is in fact rather important, because people may be thought to develop a different, more personal relation with colleagues with whom they also live than with those with whom they only work. Thus, the workers in the large fullonicae of Ostia, Rome and Florence did not only work in a more anonymous atmosphere, they also shared less outside working hours. As a consequence, it is likely that new workers integrated much less easily in these large and anonymous staff networks than in the small and medium-sized establishments at Pompeii. Besides this, it is relevant to point out that the composition of the work force also differed in terms of age and gender: the household-based establishments of Pompeii included men, women and children, but the industrial fulling factories are likely to have been populated mainly by people hired for their qualities – and as the qualities required resemble each other, it is likely that the people resemble each other too. Many of these people may have been adult males38.

Fig. 58. Plan of fullonica V, vii, 3 at Ostia (drawing M. Flohr).

A hierarchy of tasks

  • 39 For a more elaborate discussion of the social hierarchy of tasks within the fulling process, see F (...)

17A third major difference between the fullonicae of Pompeii and the industrial fullonicae of Ostia, Rome and Florence concerns the nature of differences between workers. Arguably, not all workers in a fullonica had a similar position in the social network made up by the members of the staff: some had a more marginal role, while others were able to develop a more central position. In a certain sense, the position of workers within the staff network may have been related to the division of labour within a fullonica: not all tasks were equally desirable, and workers with better skills – technically or socially – may end up doing the more desirable tasks more often, which, in turn, also may have brought them some prestige or certain other social advantages. It seems that there was a rather clear hierarchy of tasks that is similar to the order of treatments outlined above: it is likely that the lowest position, in terms of desirability, was taken up by the work in the stalls39. This was not only relatively easy, as it required muscular power rather than trained skills, it was also less responsible, as the results could be checked by workers involved in later stages, and, as workers stood with the feet in alkaline chemicals, relatively dirty. The work in the basins was not difficult either, but brought some responsibility with it in the sense that the results of the first phase could be checked, and it was considerably less dirty, as it only involved water instead of chemicals. Finishing, however, was not dirty at all, and one arguably needed concentration and a sharp eye to check the result, also because there were no subsequent phases. Moreover, some finishing tasks also bore a relatively large risk of inflicting damage on the garment through brushing and shearing without caution. Yet, besides this hierarchy intrinsic to the process, it also seems that workers in later phases had a better communicative position: in the large Pompeian fullonicae, basin workers have a central place in the communicative network and were thus much more prominent in the social landscape than the stall workers around them. For example, in the fullonica of Stephanus (I 6, 7), the people in the basins stood in the middle of those working in the stalls, could easily control them, and also controlled the communication with the shop in front of the house, as they could see people approaching the working area, and were the first to be seen by them of the workers in the back yard (fig. 56). Finishing was often done in the shops, which brought finishers informational and communicative advantages over the workers in the work area in the back part of the house: they interacted with customers and thus were the ones to give orders to the others.

Task allocation and social differentiation

18While these status differences are interesting enough in themselves, what makes them of relevance here is the probable differences between fullonicae in the way in which tasks were allocated: in small workshops there may be considerable room for variation: at quiet moments, or in cases of absence or illness, or just naturally, as part of the direct and intimate way in which these workshops were managed. In large workshops, on the other hand, and especially in the industrial establishments, such variation was much less likely to occur: to keep such workshops functioning, it was much more efficient to give each worker standard tasks, so that fewer decisions needed to be taken about who did what. Moreover, the fulling factories of Rome and Ostia specialized in the first two phases of the fulling process – that is: in soaping and rinsing. It seems that the parts of the fulling process that required most skill – the raising and trimming of the nap – were not done in these establishments at all. This not only meant that there was considerably less variation for workers and a lot more monotony, but it also implies that there were fewer chances for workers for personal development in terms of skill mastery: most of the workers in these industrial establishments were hard-working muscles rather than skilled artisans, and they functioned within a probably rather fixed hierarchy which was not easily breached. There simply was no occasion for these people to fully master the craft in all its aspects.

Fulling and the urban community

19The staff networks of the large fullonicae of Ostia, Rome and Florence were thus not so coherent, the atmosphere was rather anonymous, and there were fewer chances for skill development for individual workers. The workshops in tabernae and atrium houses, on the other hand, were integrated, had an intimate atmosphere, and probably a much more fluid task allocation with much higher chances of skill development. Besides that, however, there were also significant differences between the workshops in terms of integration into the wider urban community.

Living and working in a neighbourhood

  • 40 Cf. Laurence 2007, p.102-116. For the identification of this taberna as a fullonica see Flohr 2008 (...)
  • 41 VII 12, 7, VII 12, 9 and VII 12, 13 were all bakeries. Cf. Flohr 2007, p.148.
  • 42 The only exceptions are VI 8, 20-21.2 and VI 16, 3-4.

20Tabernae, obviously, by nature had an open relationship with the urban environment: the shop had a wide opening which not only invited customers in, but also provided the workers inside with a clear view on the street and the shops surrounding it. For example, workers in the fulling stalls in the small commercial fullonica VII 2, 41 at Pompeii could potentially see a great deal of what was happening on the Via degli Augustali, which was in the busy commercial core of the city (fig. 59)40. At the same time, most of the people out there could see the fullers at work and, thus continuously associate him with his occupation. It may be imagined that it was not hard for these fullers to become publicly known as fullones – just like the man next door may have been known as a caupo, and some people on the opposite side of the road may have been referred to as pistores41. This scenario may also be applied to most of the other fullonicae at Pompeii and HerHerculaneum: most of these, too, had shops, and while the workers in the back area were not visible for outsiders, they were integrated into the neighbourhood through their colleagues and housemates at work in the shop42.

Fig. 59. View from fulling stalls in fullonica VII 2, 41 at Pompeii (photograph M. Flohr).

  • 43 At Ostia, fullonicae II, xi, 2 and III, ii, 2 were only accessible through another building, where (...)

21Yet, if a fullonica did not have a shop, workers were not visible for outsiders and could not be publicly associated with the craft they performed. Again, this made the situation of the workers in the large industrial fullonicae fundamentally different from that of people in the smaller establishments. These workshops were surrounded by heavy walls, and most were not even directly accessible from the street, or were situated in a quiet alley or along a dead end road43. Generally, thus, these workshops were almost completely invisible to the outside world, and it is not unlikely that many people in the Roman world did not even know that such workshops existed, or how the circumstances on the shop floor were. People working in these fulling factories could thus not easily be associated by outsiders with their craft and they probably were not publicly known as fullers.

The social impact of having a shop

  • 44 Cf. Flohr 2010, p.65-82.

22Yet, it must be emphasized that the influence of having a shop went beyond sheer visibility: negotiating the maintenance of valuable garments was something distinctly different simply buying everyday commodities. As has been pointed out above, fulling involved a considerable amount of trust and the delivery of garments necessitated some negotiations, which made the contact between fullers and their customers rather intense and personal. Besides this, it may also be worth noting that also the average socio-economic background of the customers of fullones may have been a factor of relevance: demand for fulling was closely related to dress culture, and there is reason to assume that those groups in society that spent most on display dress were also the most frequent visitors of fulling workshops44. Demand for fulling was unequally divided over Roman urban communities, and fullones thus, more than bakers or green-grocers, had access to potentially significant social networks, such as elite households. Even if it was not necessarily always the male head of the elite household himself who came to the fullonica, his clothes – and those of his family – arrived with considerable frequency, and probably with his consent.

23Workers in fullonicae without a shop lacked such contacts. Of course, there will have been business contacts that visited the workshop to bring or collect parties of garments, or just to inspect the production process, but such visits are likely to have been brief and incidental compared to the potentially continuous arrival of private customers in shops. Moreover, given the size of the work force, there is unlikely to have been much interaction between these visitors and the plodding infantry in the stalls: while some frequent visitors may have become, to some degree, known to the workers of the fullery, this was probably not the case the other way around, as it is likely that such people would mainly spend most of their time in the fullonica with the persons that stood at the head of these establishments.

Being a fullo

  • 45 For the most explicit explorations of the cultural tensions related to fulling see Bradley 2002, p (...)
  • 46 There are 18 inscriptions from Roman Italy commemorating deceased fullones, and less than five for (...)
  • 47 For example, pearl-sellers (margaritari) have left 18 commemorative inscriptions; purpurarii 22, a (...)
  • 48 Flohr 2010, p.365-366.

24Based on the above discussion, it is becoming clear that the starting point for developing a strong attachment to the job and a public occupational identity could sharply differ from situation to situation. Yet, it must be mentioned that for those who actually developed a strong occupational identity, there was a ready and rather clearly defined cultural image that could be embraced. There was a clear set of symbols that could be used, and there were several occasions during which the occupational identity could perform an explicit role in public ritual and ceremony. Most of all, epigraphic evidence suggests that there was not something like a social stigma encouraging fullones to hide their occupational background, as has often been suggested by modern commentators45. Instead, it seems like fullones were relatively well-off: compared to most everyday occupations, they seem to have been more frequently commemorated with durable inscriptions46. Indeed, most occupations with more evidence for commemoration are luxury trades that made their money by serving the demands of the Roman elite, such as pearl-sellers, purpurarii, architects and the like47. Collegia of fullones also seem to have been relatively successful in producing epigraphy, especially outside Rome48. The frequent occurrence of fullones in Pompeian electoral advises must also be seen in this light: being a fullo thus was no disgrace. Instead, as argued above, involvement in fulling could present considerable and significant social advantages that may have encouraged rather than discouraged those involved in the trade to embrace the occupational identity.

  • 49 Varr, Men., 86, 4.
  • 50 CIL IV, 4112, 4118 and, most famously, 9131: fullones ululamque cano, non arma virumque.
  • 51 CIL IV, 7963; cf. Moeller 1976, p.89-90.
  • 52 PPM VII, 329-331.
  • 53 E.g. CIL IV, 7163, 9125, 9128.
  • 54 Though not the legal responsibility attached to it, which always rests with the formal owner of th (...)

25As to the symbols related to the professional identity of fullones, one very clear element of the occupational identity of fullones was the ulula, Minerva’s holy owl, which is related to fullones by Varro49. The owl figures a couple of times at Pompeii – both in painting and in epigraphy, and the way in which it is used is very suggestive of the way in which occupational identities were construed both among the fullers themselves and towards the outside world50. There was a fullo who was publicly known as Fabius Ululitremulus (‘Fabius the Owl-fearer’) and who used his full name in electoral advertising, using the symbol related to his trade as an asset in political discourse51. At the same time, there were paintings of a bird-catching ritual in the work area of fullonica VI 14, 21-22 which also involved the ulula. Applied to the wall of the workshop, these paintings were made to be viewed by the workers of the fullonica during the workday, which is likely to have enhanced their personal attachment to their occupation and may have stimulated the development of an occupational identity52. That the owner of a fullonica was not always the only person who could be seen as a fullo is also suggested by several graffiti and election notices related to fulling workshops which do not use the singular fullo, but the plural fullones53. Thus, rather than to the skilled artisan at the head of the workshop, the term ‘fullo’ might just as easily refer to one or more of his workers54.

Fullers, fullones and craftsmanship

  • 55 Especially Moeller 1972; Moeller 1976, p.57-71; for a critique see Jongman 1988, p.182-183.
  • 56 Pompeii: CIL X, 813; Spoleto: CIL XI, 4771.
  • 57 Both faber and centonarius may have been container terms referring to craftsmen in general or craf (...)

26Analyzing the evidence for the construction of occupational identities, it becomes clear that a great deal of it comes from Pompeii. Besides the abovementioned paintings and the graffiti referring to the ulula, there are also the election notices, and there is the famous statue of Eumachia that was erected on behalf of the fullones in the cryptoporticus of the large public building which she had constructed on the forum of the city, and which on the basis of this statue often, wrongly, has been identified as a clothing market55. Of course, most of this abundance of evidence can be directly ascribed to the exceptional circumstances that created the archaeological record at Pompeii: similar types of data simply do not exist in other Italian sites. While this makes it hard to understand the full significance of the Pompeian evidence, it may be suggested that the ‘Pompeian model’ of self-confident artisans with well-developed occupational identities may, in some way or another, be valid for many medium-sized cities of Roman Italy: there were enough fullones around, and they were well-integrated enough in the community to make ‘fullo’ a relevant social label, both for fullers themselves and for their fellow citizens. The relative small scale of most workshops and their strong integration with private households created ideal circumstances for the development of coherent and intimate work groups, which further enhanced identity awareness. In such cities, fullones typically organized themselves in professional organizations: there is a remarkable similarity between the honorific inscription that Pompeian fullones dedicated to Eumachia, and the dedicatory inscription for Minerva produced by the fullones of Spoletium56. In very small cities, with one or perhaps two small fullonicae, such as, for example, Saepinum, things may have been slightly different, and the emphasis in the identity of fullers may have shifted from their specific craft to the fact that they were craftsmen: in such cases, fullones may not even have had their own organization, but may have participated in a collegium fabrum or a collegium centonariorum instead57.

27Yet, there is likely to have been a world of difference between these urban communities and the metropolitan world of Ostia and Rome. The extreme urbanization in Rome and its surroundings led to an urban society on a scale that in the ancient world hitherto had been unknown. The large industrial fullonicae were the result of these developments and present us with a picture that is not easily reconcilable with our traditional ideas about urban manufacturing in preindustrial societies: these workshops are large-scale, rationally organized fulling factories that worked with an extremely large work force in an almost industrial manner. Indeed, the strict division of labour, and the clear separation between working and living were rather uncommon in Europe before the industrial revolution. While we do not have any direct evidence concerning the ideas of the workers in these establishments about their job, the above analysis of the social landscape of these workshops suggests that these people were unlikely to have associated themselves with their work, and that many were probably not very skilled either. These people were perhaps, to some extent, fullers, but by no means must they be seen as fullones.

  • 58 For the bakeries of Ostia see Bakker 1999; for the fish-salting factories of Roman Africa and Spai (...)
  • 59 See esp. Wilson 2008.
  • 60 Rome: besides the abovementioned fullonica in the Macellum Liviae, small fullonicae were discovere (...)

28Arguably, while the fullones of Pompeii can be used to highlight the possibilities of an approach to the world of Roman manufacturing that puts a strong emphasis on the professional knowledge of skilled artisans, it is clear that the situation in the industrial fullonicae of Rome, Ostia and Florence points to the limits of such an approach: the material remains of these workshops leave little room for romantic ideas about skilled artisans and occupational pride. It is also relevant to point out that fulling was not unique in this respect: evidence for other large-scale production facilities point to similar developments in other branches of the economy. The most straightforward example of this may be found in the mills-bakeries of Ostia, but one may also think of the enormous fish-salting facilities along the coastlines of the Maghreb and the Iberic Peninsula58. In a slightly different way, the emergence of water-powered milling installations such as the one at Barbegal, and the rise of the use of moulds in pottery production also suggest that in the imperial era, more and more alternatives to the skilled artisan emerged in the Roman world59. This may simply have been a matter of efficiency and economies of scale: traditional craftsmen may have been skilled and proud of their professional knowledge, but they also may have been relatively inefficient and expensive because they worked on a small scale. This is not to say that skilled artisans disappeared altogether: in Rome and Ostia, too, we find evidence for small-scale, consumer-oriented fullonicae in tabernae60. The point is that alongside the skilled artisan, a parallel universe seems to have emerged that has much more in common with modern industrial mass production than it has with traditional craftsmanship, and to fully understand manufacturing in the Roman world, the differences between these modes of organization should be at the heart of our analysis.

Notes

1 Bakers (pistores): CIL IV, 875 (IX 3, 4-5), 886 (IX 3, 8-9), 7273 (I 8, 6-7); green-grocers (pomari): CIL IV, 149, 180, 183, 202, 206 (all in the southern part of Via di Mercurio), 7261d (I 8, 1-2); goldsmiths (aurifices): CIL IV, 710 (IX 9, 1); wool-dyers (infectores and offectores): CIL IV, 864 (IX 7, 2-3); CIL IV, 7812 (IX 3, 2-3); mule-drivers (muliones): CIL IV, 97 (VI 1, 3-4), 113 (across the road, VI 17, 1-2), and 134 (VI 17, 31-32); fullones: CIL IV, 998 (I 4, 5-6), 2966 (I 4, 26-27), 3476 (VI 14, 21-22), 3529 (VI 15, 3-4), 7164 (I 6, 7), and 9128 (IX 13, 5-6).

2 See for this traditional idea e.g. Finley 1973 [1985], p.40.

3 The advises with explicitly mentioned rogatores constitute only about 20% of all texts; only a few mention occupations. Cf. Mouritsen 1988, p.60-63.

4 For an overview of this evidence see Flohr 2007, p.148. To the fullonicae mentioned in that list, two workshops, of which the identification only recently has been confirmed, need to be added: VI 3, 6 and VII 2, 41. For the latter see Flohr 2008, p.10-12.

5 CIL IV, 875 (IX 3, 4-5), 886 (IX 3, 8-9), both related to bakery IX 3, 8-9.

6 CIL IV, 494 (between VII 5, 13 and VII 5, 14); CIL IV, 537 (between VII 5, 17 and VII 5, 18). In both cases, the reading is uncertain – caupo is a suggestion of the editor and has not been read by the excavators: CIL IV, 494 ‘officially’ reads CAULO, while CIL IV, 537 reads ÆVPO. Two electoral notices certainly mentioning caupones (CIL IV, 336 and 3502) do not seem to be made by persons using the occupation as an explicit argument supporting the message. Whether CIL IV, 629 actually refers to a caupo is unclear as some letters are missing: the editor records ‘[…]COPO’, which could be anything ending with these four letters.

7 Moeller 1976, p.90.

8 Jongman 1988, p.155-186; Ph. Borgard (et al. 2003, p.22-23) mainly reiterates the claims of W. Moeller, but without really addressing W. Jongman’s deconstruction. Particularly his argumentation concerning the function of Moeller’s lanifricariae is weak. For an alternative interpretation of these workshops see Flohr 2007, p.132.

9 This also has to do with the shape of the ancient loom, which was unable to produce the broadcloth, and the nature of roman clothing, which was woven to piece and did not need extensive tailoring. Cf. Flohr 2010, p.66.

10 Flohr 2010, p.71; Wild 2002, p.22.

11 Significantly, the only large fullonica without election notices is VI 8, 2.20-21, which also did not have a shop. The other Pompeian fullonica without a shop (VI 16, 3-4) neither had election notices referring to fullones.

12 These inventories sometimes even specify whether clothes had been fulled or not. Cf. P. Mich. II, 123 (46 AD); SB XVI, 12314 (2nd c. AD); P. Tebt. II, 417 (3rd century AD). For a curse tablets from the sacred spring of Minerva Sulis at Bath see Tomlin 1988, no. 64.

13 Cf. Robaye 1991.

14 SGDI 2, 1904 (Baunack); Wescher – Foucart 1863, p.289. Baunack dates the inscription between 156 and 151 BC.

15 The amount of evidence for fulling from outside Roman Italy is minimal and not easily compatible with the data from the peninsula. Hence, it makes sense to leave this out of the present analysis.

16 Flohr 2006, p.193-194.

17 Flohr – Wilson in press.

18 Flohr 2005, p.59-60.

19 See further Flohr 2010, p.146-159; Flohr 2006.

20 Rinsing is a necessary part of the fulling process, as it is impossible to wear clothes treated with alkaline chemicals that have not been washed out. Most small fullonicae did not have an internal water source. They may have used wooden basins filled with water from the street, but it may also have been economic to just wash clothes out at the nearest fountain with the aid of a bucket.

21 See Flohr 2010, p.130-134.

22 Monumentum Neronis Drusi (CIL VI, 4336); Monumentum Marcellae (CIL VI, 4445); Monumentum Statiliorum Taurorum (CIL VI, 6287-6290); Monumentum Volusinorum (CIL VI, 7281).

23 Flohr 2010, p.81-82; contra Dixon 2001, p.9-10.

24 AE 1958, 273; cf. Ferrua 1957, p.609.

25 Flohr 2010, p.83-85.

26 I take the number of fulling stalls as an indicator of scale because other facilities, particularly rinsing complexes, are less easily measurable and not generally quantifiable. Cf. Flohr 2010, p.86-91.

27 This fullonica has recently been excavated at Casal Bertone, some two miles outside the Aurelian walls at Rome. For a short preliminary report of the excavations see Musco et al. 2008.

28 Flohr 2007, p.135-137; Flohr in press.

29 Flohr in press; NSA 1912, p.286.

30 Ostia: fullonica II, xi, 2, III, ii, 2, and V, vii, 3; cf. De Ruyt 2001; De Ruyt 1996; De Ruyt 1995; Pietrogrande 1976. Florence: Piazza della Signora; De Marinis 1997. Rome: Casal Bertone; Musco et al. 2008.

31 That is, 251 of all 311 fulling stalls were found in these five workshops. Cf. fig 55.

32 Cf. Flohr 2010, p.90.

33 The fullonica of Casal Bertone was situated just east of Rome, near the point where the major routes that connected the metropolis with the mountainous hinterland of the Apennines converged. Cf. Flohr 2010, p.96-99.

34 It should not be excluded at forehand that the workers in these fullonicae were actually paid for what they did, though clear evidence in favour or against this is lacking.

35 For a more elaborate discussion of this problem see Flohr 2010, p.300-303.

36 Cf. Flohr 2009.

37 No fullonica, for instance, has less than two fulling stalls (see fig. 55).

38 For the exhausting character of fulling see Sen., Ep. 15, 4.

39 For a more elaborate discussion of the social hierarchy of tasks within the fulling process, see Flohr 2009.

40 Cf. Laurence 2007, p.102-116. For the identification of this taberna as a fullonica see Flohr 2008, p.10-12.

41 VII 12, 7, VII 12, 9 and VII 12, 13 were all bakeries. Cf. Flohr 2007, p.148.

42 The only exceptions are VI 8, 20-21.2 and VI 16, 3-4.

43 At Ostia, fullonicae II, xi, 2 and III, ii, 2 were only accessible through another building, whereas fullonica V, vii, 3 was situated in a presumably rather quiet road with little traffic. The fullonica at Florence was situated at a dead end road. Cf. Flohr 2010, p.346-347.

44 Cf. Flohr 2010, p.65-82.

45 For the most explicit explorations of the cultural tensions related to fulling see Bradley 2002, p.36-37; Kudlien 2002, p.56-58.

46 There are 18 inscriptions from Roman Italy commemorating deceased fullones, and less than five for probably equally common craftsmen such as tanners, butchers, fishermen and inn-keepers. Cf. Flohr 2010, p.361-362; for an overview of the inscriptions on which this argument is based see http://research.mikoflohr.nl/epigraphy/index.php.

47 For example, pearl-sellers (margaritari) have left 18 commemorative inscriptions; purpurarii 22, and architects 23. Cf. Flohr 2010, p.359, table 5.

48 Flohr 2010, p.365-366.

49 Varr, Men., 86, 4.

50 CIL IV, 4112, 4118 and, most famously, 9131: fullones ululamque cano, non arma virumque.

51 CIL IV, 7963; cf. Moeller 1976, p.89-90.

52 PPM VII, 329-331.

53 E.g. CIL IV, 7163, 9125, 9128.

54 Though not the legal responsibility attached to it, which always rests with the formal owner of the workshop. Cf. Dig., 14, 3, 5, 10 (Ulp.).

55 Especially Moeller 1972; Moeller 1976, p.57-71; for a critique see Jongman 1988, p.182-183.

56 Pompeii: CIL X, 813; Spoleto: CIL XI, 4771.

57 Both faber and centonarius may have been container terms referring to craftsmen in general or craftsmen involved in textile economy respectively. For a recent study on the collegia centonariorum and the broad character of the term centonarius see Liu 2009, esp. p.57-83.

58 For the bakeries of Ostia see Bakker 1999; for the fish-salting factories of Roman Africa and Spain see Wilson 2006b.

59 See esp. Wilson 2008.

60 Rome: besides the abovementioned fullonica in the Macellum Liviae, small fullonicae were discovered on the Celian and Oppian hills. Cf. Astolfiet al. 1990; Pavolini et al. 1993. At Ostia, small fullonicae were identified by Pietrogrande at I, xiii, 3 and II, ix, 1 (Caseggiato della Fullonica). Cf. Pietrogrande 1976, p.9-13, 51-54. To these, a third workshop may be added at II, iv, 5 [12] on the basis of epigraphy referring to a corpus fontanorum (CIL XIV, 4573) and the discovery of three fulling tubs during the excavation. Cf. NSA 1909, p.90. For fullones and fontani see Tran 2007, esp. p.598-601.

Table des illustrations

Légende Fig. 51. Painting with fullers at work from fullonica VI 8, 2.20-21 at Pompeii (MANN, Inv. no 9774; photograph N. Monteix).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-1.jpg
Fichier image/jpeg, 358k
Légende Fig. 52. Complex of rinsing basins in fullonica VI 8, 2.20-21 at Pompeii (photograph M. Flohr).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-2.jpg
Fichier image/jpeg, 153k
Légende Fig. 53. Cupid brushing textile from the frieze with fulling cupids in oecus q at the Casa dei Vettii at Pompeii (VI 15, 1.27; photograph M. Flohr).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-3.jpg
Fichier image/jpeg, 240k
Légende Fig. 55. Fulling capacity of the fullonicae of Roman Italy.
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-4.jpg
Fichier image/jpeg, 70k
Légende Fig. 56. Fourth style wall-decorations in fullonica I 6, 7 at Pompeii (photograph M. Flohr).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-5.jpg
Fichier image/jpeg, 114k
Légende Fig. 57. Plan of fullonica I 6, 7 at Pompeii (drawing M. Flohr).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-6.jpg
Fichier image/jpeg, 33k
Légende Fig. 58. Plan of fullonica V, vii, 3 at Ostia (drawing M. Flohr).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-7.jpg
Fichier image/jpeg, 273k
Légende Fig. 59. View from fulling stalls in fullonica VII 2, 41 at Pompeii (photograph M. Flohr).
URL http://books.openedition.org/pcjb/docannexe/image/5127/img-8.jpg
Fichier image/jpeg, 437k

Auteur

University of Oxford, Oxford Roman Economy Project.

© Publications du Centre Jean Bérard, 2011

Conditions d’utilisation : http://www.openedition.org/6540

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search