© Journal of Roman Archaeology 26 (2013)

The textile economy of PompeiiMiko Flohr

A fiercely debated aspect of Pompeii’s history is the nature of its textile economy. The large-scale excavations of the late 19th and early 20th c. led to the discovery of unusually large numbers of workshops, artefacts, iconographic representations and texts that point to the existence of a variety of textile crafts in the town, which stirred the imagination of several scholars. For most of the 20th c., Pompeii was seen as a centre of export-oriented textile production where wool was turned into finished textiles to be sold on the imperial market. The idea that Pompeii was a Roman version of a mediaeval ‘textile town’ was most elaborately discussed by W. Moeller in the 1960s and 1970s, but it would be attacked by primitivist scholars including M. I. Finley and, especially, his then-student W. Jongman.1 While Finley and Jongman were able to expose some critical weaknesses in the arguments of those who championed the idea that Pompeii had an export-oriented textile industry, they failed to develop an alternative. As a consequence, there is no realistic framework through which one might make sense of Pompeii’s textile economy.

The present article aims to present a new perspective. It begins with a brief summary of the 20th-c. debate and sketches how developments of the last decade have changed the field. Next, it re-assesses the evidence for textile crafts and trade at Pompeii. The evi-dence will then be used to rethink the functioning of Pompeii’s textile economy of the third quarter of the 1st c. A.D., and to reconstruct the investment strategies of the entrepre-neurs involved in it. Finally, I will consider the economic background against which these strategies evolved, and how Pompeii’s textile economy may have functioned within local, regional and supra-regional networks of trade and exchange.

The history of Pompeii as a ‘textile town’

Evidence for aspects of the textile economy has been noted since the early excavations. The first fullonica (VI 8, 20-21.2) was excavated in 1825-26 and immediately recognized for what it was. The building of Eumachia on the E side of the Forum, excavated in 1820, was (wrongly) interpreted as a clothing market as early as the 1860s.2 However, the idea that Pompeii might have been a place where textiles were produced on a large scale does not seem to date back much further than the early 20th c., and must be seen as a result of the large-scale clearance that began in the 1860s and continued until the start of the First World War. The bulk of the evidence that has played a rôle in discussions about Pompeii’s textile economy was brought to light in that half-century.

From Tenney Frank to Jongman

The first scholar to coin the idea that Pompeii was a textile town was Tenney Frank. In 1918 he published an article on its economic life in which he claimed that the town was

1 Moeller 1976; Finley 1985; Jongman 1988.2 For the excavation and identification of fullonica VI 8, 20-21.2 see, e.g., Bechi 1827, Förster 1828,

207. Early identifications of the Building of Eumachia as a clothing market include Dyer 1869, 123; see also Mau 1892 (Moeller 1976, 64 is wrong to ascribe the idea to Mau 1908, 112-13). See also Breton 1855, 130, who sees it as the home of the collegium of the fullones. Fiorelli 1875, 260 considered it a big fullery.

M. Flohr54

an “important center in the clothing trade”.3 Frank based his theory on the “inordinate number of fulleries” that could be found in the town, and on the then-widespread idea that the Building of Eumachia was a cloth hall, which, as he argued, through its dimen-sions emphasized the scale of the clothing trade. His brief argument was reprinted in the books he wrote later, from where it was picked up by others, including, most prominently, M. Rostovtzeff.4

Yet it was not until shortly after the Second World War that Frank’s ideas were used as a starting point for a more thorough exploration of the Pompeian textile industry, by W. Moeller. Moeller’s argument was slightly more nuanced and considerably more elabo-rate than Frank’s. Besides the sheer quantity of evidence, it was for Moeller also the fact that all stages in the production chain of woollen textiles were identifiable: he made a careful (though partially erroneous) analysis of the identifiable textile workshops, ascrib-ing each type to a specific phase in the process — from the washing of raw wool to the polishing of finished garments. Moreover, he argued that the importance of the textile industry was not only attested through the large Building of Eumachia, but also through the dominant rôle in local politics of, particularly, fullers, a picture that he based on M. Della Corte’s interpretation of electoral slogans on the façades of houses and shops.5

Moeller’s work, published more than 15 years after it was written, landed in a schol-arly environment that just began to be dominated by Finley’s primitivist ‘take’ on the ancient economy.6 Finley emphasized the primarily local orientation of ancient urban economies; his model did not really have a place for urban economies based on export- oriented production.7 Moeller’s work completely neglected the primitivist model, and unsurprisingly it met with a critical reception.8 The most serious criticism came from Jongman, who devoted a chapter in his book on the economy and society of Pompeii to thoroughly deconstructing the Frank/Moeller model.9 Jongman argued, rightly, that Moeller had been overly optimistic in his interpretation of the epigraphic evidence, and that there was very little that warranted Moeller’s claim that the Pompeian town council was dominated by individuals from the textile business. Jongman was also able to show, convincingly, that Moeller had grossly overestimated both the amount and the scale of the textile-workshops. Particularly problematic was the fact that Moeller presented virtually no reliable evidence for the most important and time-consuming phases in textile produc-tion, those of spinning and weaving.10 Jongman therefore concluded that the evidence rather suggests a scenario in which Pompeii had a small-scale industry serving the local market.11 While Jongman’s criticism of Moeller is generally well-founded, it is important to see it in its scholarly and ideological context: in trying to argue that Pompeii was an

3 Cf. Frank 1918, 234-35. 4 See Frank 1927, 260-64; id. 1940, 262. Also Rostovtzeff 1957, 577-79.5 Moeller 1976, 91-93; Della Corte 1954.6 Moeller completed his thesis on the Pompeian woollen industry in 1962, and published it,

almost unchanged, in 1976.7 Finley 1985.8 See, e.g., the review (1980) by R. P. Duncan-Jones. Finley himself, in the second edition of The

ancient economy (1985, 194), used Moeller as a key example of the, in his eyes, simplistic and impressionistic methodologies of some of the critics of his model.

9 Jongman 1988, 155-86.10 Ibid. 162-65.11 Ibid. 184.

The textile economy of Pompeii 55

archetypical Finleyesque ‘consumer city’, Jongman had to get rid of any scenario in which Pompeii derived substantial income from export-oriented manufacturing.12

Recent years: an archaeological revolution

In practical terms, Jongman’s work more or less settled the debate for much of the 1990s.13 Yet the last decades of the 20th c. also saw significant changes in Pompeian archae-ology, which eventually would have a profound effect on the study of the town’s economic life, including that of its textile economy. From the 1980s onwards there was a notable increase in the number of (foreign) archaeologists and a change of focus towards docu-menting and analyzing the material remains of larger complexes or even building blocks in their entirety.14 From the remains of workshops encountered by these projects a wealth of new evidence became available in published form.15 Moreover, from the turn of the 21st c. increasing numbers of scholars were interested in the archaeology of Pompeii’s urban economy, leading to a number of projects devoted to studying the material remains of shops and workshops.16

Particularly relevant here is the work of two French scholars, Ph. Borgard and N. Mon-teix. In the early 2000s Borgard systematically studied the dyeing workshops of Pompeii as part of a larger project run by the Centre Jean Bérard, and did considerable work on a number of other textile establishments.17 While Borgard was primarily interested in the technological aspects and in understanding how these workshops functioned, he also dis-cussed the economic background of the textile workshops. The quantity and quality of the Pompeian evidence led him to revive Moeller’s idea that Pompeii in A.D. 79 was a textile town.18 However, he added a new dimension by suggesting that the situation in 79 was the result of developments in the course of the 1st c. A.D. which transformed Pompeii from a centre of wool production to a centre of textile production in the very last years of its existence.19

The monograph recently published by Monteix focuses primarily on the shops and workshops of Herculaneum, but his discussion of textile crafts includes the Pompeian

12 Ibid. 201-3. It has to be pointed out that some of the reviews of Jongman’s monograph (e.g., Banaji 1989; Purcell 1990) were highly critical of his approach in general, though they accepted his deconstruction of the Frank/Moeller model; see also the more recent criticism by Wilson 2002, 235-36.

13 See, e.g., Laurence 1994, 57-61.14 Key pioneering projects in this respect were V. Strocka’s Haüser in Pompeji project (e.g., Seiler

1992) and R. Ling’s project at Insula I 10 (id. 1997). After 1997, there was a boom in projects (e.g., Dickmann and Pirson 2002; Coarelli and Pesando 2006; Amoroso 2007) combining the study of standing remains with excavations beneath the floor level of A.D. 79.

15 E.g., the fulling workshops of VI 16, 3.4 and VI 16, 6 were published by Seiler 1992; a wool-related workshop in the House of the Postumii (VIII 4, 4) was studied and published by Dick-mann and Pirson 2001.

16 E.g., Robinson 2005; Ellis 2004a and 2004b; Ellis and Devore 2009 and 2010; Ellis et al. 2011; Bor-gard et al. 2003; Flohr 2007a, 2008 and 2011a. Also Poehler et al. 2011, 5-7.

17 Borgard 2002; Borgard and Puybaret 2003; Borgard et al. 2003; Borgard and Puybaret 2004.18 Borgard et al. 2003, 23-24, Borgard and Puybaret 2004, 58-59.19 Borgard and Puybaret 2004, 58-59. Borgard’s argument is based on his impression that what

he interprets as wool-washing workshops seem to be older than the fulleries and the dyeing workshops, which he all dates to the last years of the town. It has to be noticed, however, that Borgard has not consistently dated these workshops; moreover, his interpretation of these workshops is controversial. See below p. 58 ff.

M. Flohr56

evidence too.20 Particularly relevant is that Monteix is the first to collect and analyze in a serious manner the full body of evidence for spinning and weaving from the two Vesuvian towns, and that he points to the existence of at least a couple of places where this evidence appears in contexts suggesting market-oriented production.21 Comparing the evidence from the two towns, he suggests, cautiously, that the Pompeian textile economy seems to have operated on a much larger scale than that of Herculaneum, which might indicate sur-plus production; this, then, would support Borgard’s revised version of the Frank/Moeller model.22

Shortcomings of the debate

While Borgard and Monteix make much better use of the archaeological evidence than Moeller and Jongman had ever done, in the end their (admittedly cautious) reinstating of Pompeii as a ‘textile town’ fails to convince. This has to do not only with the fact that the evidence does not unequivocally support such a scenario; a key problem is also that they, like Moeller and Jongman, operate within a paradigm laid out by Frank almost a century ago. That paradigm in many ways reflects the industrializing western world that helped shape it, but from an early 21st-c. perspective it has a number of shortcomings that can-not be overcome by simply looking harder at the evidence. There are three key problems:1. There is a strong methodological optimism: using the Frank paradigm implies assum-

ing that our evidence can tell us something about the scale of Pompeii’s textile export in absolute terms. This is problematic for two reasons. First, as we do not know the size of the local textile market, we cannot possibly understand which proportion of the town’s output must be seen as surplus.23 Second, there is reason to be pessimistic about our insight into the output of Pompeii’s textile workshops: even if we would be able to understand the maximum capacity of each individual workshop, we would still have no clue about the degree to which this maximum capacity was normally used.24 Whether Pompeii had a textile industry that produced surplus for supra-local markets is a question that the material remains of identifiable textile workshops cannot answer unequivocally on their own.25

2. There is a strong tendency towards what one could call ‘conceptual reductionism’. The focus of the debate is exclusively on market-oriented production and export, whereas it is obvious that Roman textile economies were much more complex than that: manufac-turing and trade are subsets of a larger whole that also includes retail, maintenance, and re-use — issues that too often are taken for granted, and have played no meaningful rôle in past debates about Pompeii’s textile economy. It is the larger whole that needs to be

20 Monteix 2011, 169-217. A key strength of Monteix’s work lies in the fact that it publishes an enormous amount of barely accessible evidence assembled from the archives (and not only for Herculaneum but also for Pompeii).

21 Ibid. 175-85.22 Ibid. 216-17. Monteix rightly stresses that it is necessary to put the evidence from Pompeii and

Herculaneum in a much wider, comparative perspective.23 In this respect, knowing the size of the population is insufficient, for one would also have to

have an understanding of average textile consumption per capita.24 For a comparable argument see Flohr 2013, 74-79.25 Particularly problematic is the fact that there is no site with which Pompeii in this respect can

plausibly be compared. Monteix’s comparison with Herculaneum is a good first step, but a key problem is that it is impossible to see whether the differences between them are due to the two towns being hugely different in size (as is generally assumed) or are indeed related to Pompeii having a rather well-developed and export-oriented textile economy. Cf. Monteix 2011, 217.

The textile economy of Pompeii 57

addressed, not the subset. We must shift the debate from discussing ‘textile production’ towards discussing the textile economy as a whole, in an integrated manner.

3. The paradigm suffers from geographical isolationism: past accounts have taken Pom-peii out of its regional environment and analyzed the town as if it existed in a vacuum. There is no way to understand Pompeii or its textile economy without some awareness of the nature of the many local and regional networks in which it played a rôle, and which had a profound influence on the socio-economic processes and strategies that shaped the archaeological record. However complicated it is to understand the complex dynamism of the Bay of Naples in the 1st c. A.D. and however vague the evidence is, this needs to be a much more central part of the debate.

In focusing on the economic context of the evidence, dealing with the Pompeian textile economy in the broad sense of the word, and discussing the geographic context of its tex-tile market, the present article aims to address these shortcomings. Yet first it is necessary to examine the actual evidence.

The Pompeian textile economy and the evidence for it

The basic question as to what truly counts as evidence for Pompeii’s textile economy is less straightforward than it may seem. Besides some scattered pieces of evidence, two com-mon workshop types are securely identifiable: dyeing workshops and fulling workshops (fullonicae). A third, probably used for felt-making, can be linked to the textile economy because it is found in direct association with fulleries and dyeing workshops. The problem lies with a fourth group of workshops, which in the past have been referred to as officinae lanifricariae and have been associated with the preparation of raw wool. Their interpreta-tion is crucial for our understanding of the Pompeian textile economy, especially because they have been identified in large numbers. Hence, it makes sense to deal with them first.

Thinking harder about the “lanifricariae”

The “lanifricariae” are the second most numerous workshop type in Pompeii, with 15 securely identified.26 Of all other workshop types in the town, only bakeries have been identified in greater numbers.27 They are characterized by a long but rather low work-ing bench incorporating one or two furnaces over which a large but shallow lead bowl is installed (fig. 1). This bowl, which generally spans the entire width of the working bench, has a vent at its back; the working surface of the bench is covered with waterproof plaster. In addition, these workshops tend to have one or two small basins, in some of which traces of a white sediment can be observed.28

The idea that these workshops had something to do with wool or textiles goes back to the middle of the 19th c., when G. Fiorelli, amongst others, expressed the idea that

26 These include workshops I 3, 10; I 3, 15; I 3, 16.18; VII 2, 27; VII 3, 24-25; VII 4, 39-40; VII 9, 41; VII 9, 43; VII 9, 44; VII 10, 5.8.13; VII 11, 2-3; VII 11, 5; VII 12, 21.17; VII 12, 22-25; VII 12, 30; VII 4, 35. Compared to the workshops listed by Monteix 2011, 170, I am adding I 3, 10 and rejecting the interpretation of I 3, 30; IX 3, 13; and IX 3, 14, which lack evidence for the installations char-acteristic of this workshop type. The interpretation of VII 13, 21 is uncertain, as the furnace is of a different shape and the basin is lacking. For I 8, 19, VI 16, 3-4, VII 14, 5.17-18 and VIII 4, 4, see the next section. At 2007b, 148, I followed Moeller 1976, 30-35, and failed to include I 3, 10 and VII 2, 27.

27 There are 24 bakeries with mills in Pompeii, and another 8 without; cf. Flohr 2007b, 148.28 Such traces are still visible in workshop VII 9, 43.

M. Flohr58

they served as fulleries.29 Their name, however, was invented by Moeller, who derived it from a graffito referring to a lanifricarius that was found in the environment of one of these shops.30 Rightly distinguishing these workshops from the town’s fullonicae, Moeller thought that they were involved instead in wool-scouring: in his view, the “lanifricariae” played a rôle in the first phase of his textile industry — as workshops where the wool, directly after shearing, was cleared of lanolin and other natural pollutants. This is a conten-tious claim, and it was attacked by Jongman, who rightly pointed out that the term officina lanifricaria must be seen as a scholarly fiction, and argued that there is no evidence that these workshops had anything to do with wool production at all.31

While it has been generally accepted that these workshops were not known as lani-fricariae, Jongman’s rejection of the idea that they served for wool-scouring has not been universally accepted.32 Many scholars still assume at least some link between these work-shops and wool production; they include Borgard and Monteix, who have recently defended Moeller’s identification,33 yet rather than epigraphy, it is now the archaeology of these workshops and its relation to literary evidence that is brought forward to sup-port their interpretation. Crucially, several scholars have pointed to the similarity between the work installations in these workshops and an alleged description of wool-scouring by Pliny the Elder (NH 29.35)34 — a problematic claim: first, Pliny’s text does not refer to

29 For the cases of VII 12, 17.21 and VII 12, 22-25, see Fiorelli 1875, 285-88. Probably these are the ‘fulleries’ referred to by Frank 1918, 235.

30 CIL IV 1190; Moeller 1976, 33.31 Jongman 1988, 167-68.32 Cf. Borgard and Puybaret 2004, 52, Monteix 2011, 170.33 Borgard and Puybaret 2004, 51-52, Monteix 2011, 170-71. See also Robinson 2005, 91. Laurence

(1994, 60-61), despite being critical of Moeller’s model, also assumes these workshops were involved in wool-production.

34 Cf. Borgard and Puybaret 2004, 51; Robinson 2005, 91.

Fig. 1. Pompeii workshop VII 12, 21.17, two furnaces with shallow lead bowl.

The textile economy of Pompeii 59

wool-preparation, but describes a way to obtain oesypum, the grease adhering to the wool that had a variety of medicinal and cosmetic uses. To separate the oesypum from the wool, the wool had to be cooked in bronze over a low fire. Nowhere does Pliny imply that this was also the common procedure for degreasing wool destined for textile production, nor does his description of the facilities necessarily point to workshops with a layout resem-bling that of the “lanifricariae”, including the basins and the working benches.35 While, theoretically, installations like those found in “lanifricariae” could perhaps be used for wool-washing, other uses are not excluded.

Indeed, the few indications we have point away from wool and textiles. Three argu-ments may be highlighted. In the first place, in several of these workshops excavators found quantities of animal bones.36 The excavators of workshop I 3, 15 even noticed that the bones all came from the upper legs of animals (the report does not specify the species).37 This is incompatible with wool-scouring, and strongly suggests a relation with meat con-sumption: upper legs typically belong to the most meat-rich parts of both cows and pigs. Perhaps the “lanifricariae” were involved in bone-working after cooking the last remains offlesh from the bones; alternatively, they may have been involved in producing some kind of cooked meat product. Secondly, we may emphasize the spread of these workshops. Unlike securely-identifiable textile workshops, which are spread randomly over the urban area, the “lanifricariae” are concentrated in two clusters (fig. 2): in Regio VI, east of the forum, and in Regiones I and VIII, close to the theatre quarter. This clustering in the town’s cen-tre suggests that they were performing a task somehow related to places that were much frequented by people. While few of these workshops had a shop, it is possible that their location was shaped by concentrations of demand; they may have sold their produce on the

35 Contra Robinson 2005, 91.36 These include I 3, 15; I 3, 16; and, possibly, VII 9, 44. Cf. Monteix 2011, 170-71. 37 See Galella et al. 1869, 297-98 and 302-3; Cf. Flohr 2012, 70.

Fig. 2. Pompeii with locations of identifiable textile workshops (dots) and officinae lanifricariae (stars).

M. Flohr60

street or in market stalls.38 In any case, unless one would assume that the Macellum served as a wool market (for which there is no evidence), it is hard to see what would have trig-gered such a concentration if these workshops were devoted to wool preparation. The fact that the spatial concentration of these workshops distinguishes them from identifiable tex-tile workshops cannot be neglected. Thirdly, their spatial contextualization sets them apart from textile workshops. Contrary to dyeing workshops and fullonicae, most “lanifricariae” are isolated from their direct environment, and few have open connections with public space: they were hidden behind thick walls with narrow doors and small windows.39 In houses, the workrooms are rigidly separated from residential rooms.40 Possibly this has to do with the fact that they were working on dead animals, but that these workshops seem to have a controversial position in their environment is hard to relate to cooking wool in order to separate it from lanolin: if fulling and dyeing are not controversial, why would wool-preparation, which worked with the same basic substances, be problematic? It seems safe to suggest, then, that, to our present knowledge, “lanifricariae” had nothing to do with wool or textiles.

Dyeing workshops

The identification of the town’s dyeing workshops is secured by epigraphic evidence: electoral slogans on the façades of two of these workshops refer to infectores and offectores,41 craftsmen who are related to the dyeing of textiles and raw wool by several Latin authors. The workshops are recognizable by high furnaces over which large and deep lead caul-drons have been installed (fig. 3). Some workshops also have one or more basins. Six dyeing workshops have been fully excavated; a seventh (not yet brought to light) is identifiable by

38 On the shops associated with lanifricariae, see Borgard and Puybaret 2004, 50. 39 Cf. Flohr 2007b, 136; 2012, 70.40 E.g., in the case of houses VII 10, 5.8.13, VII 12, 17-21, and VII 12, 22-24. 41 Infectores: CIL IV 7812 (on the façade of IX 7, 2); offectores: CIL IV 864 (on the façade of IX 3, 1-2).

Fig. 3. Pompeii, dyeing workshops V 1, 4 and V 1, 5 with their characteristic furnaces.

The textile economy of Pompeii 61

a cauldron standing on the threshold of its entrance.42 The technology of these workshops is much better understood through the work of Borgard, who used experimental archae-ology to understand their functioning, which enabled him to reconstruct and use parts of workshop V 1, 4.43

Borgard discerns two types of dyeing workshops. Four have a hierarchically ordered ensemble of cauldrons, one larger cauldron being used for a preparatory treatment and a set of 4 smaller cauldrons for the dyeing.44 The other two have no such hierarchy and appear less well-organized, while the cauldrons are also smaller.45 Borgard argues that the first group was involved in dyeing raw wool, whereas the second group was involved in re-dyeing used textiles.46 Essentially, this idea is based on literary sources distinguishing between the two activities: the former was done by infectores, the other by offectores.47 A strong argument in favour of Borgard’s distinction is the fact that the epigraphic evidence matches it: the electoral slogan of the infectores is related to a workshop probably devoted to dyeing raw wool (IX 7, 2; fig. 4), whereas that of the offectores is related to a workshop belonging to the second category (IX 3, 1-2). Following Borgard’s interpretation, we note that only the four workshops of the first group can possibly be seen as (indirect) evidence for textile manufacturing, while the second group by definition belongs to the realm of maintenance and re-use.

42 Fully excavated are I 8, 19; V 1 4; V 1, 5; VII 2, 11; VII 14, 5.17-18; and IX 3, 1-2. Not fully exca-vated is IX 7, 2.

43 The experiment was presented to a scholarly audience at the Purpureae Vestes conference in 2008, but has not yet been published. A preceding experiment was published in Puybaret, Bor-gard and Zérubia 2008.

44 These include workshops I 8, 19; V 1, 4; V 1, 5; VII 14, 5.17-18. What we know of the workshop suggests that IX 7, 2 also belongs to this group.

45 These include workshops VII 2, 11 and IX 3, 1-2.46 Borgard and Puybaret 2004, 52-57.47 See especially Festus 112.6. Cf. Borgard and Puybaret 2004, 52.

Fig. 4. Pompeii, dyeing workshop IX 3, 1-2 with three cauldrons.

M. Flohr62

The dyeing workshops vary in capacity. The standard size of the production ensem-ble in the first group seems to be 4 or 5 cauldrons (which comes down to 3 or 4 colors). Workshop VII 14, 5.17-18 is the biggest dyeing workshop of this group: it has two sets of cauldrons (one of 5, one of 4). Workshop I 8, 19 has 4 cauldrons and so does workshop V 1 5, but workshop V 1 4 has 5. The interpretation of these last two workshops is prob-lematic: as they were situated directly next to one another in two shops belonging to the same property, it is hard to understand how independently they operated. Is it possible that they should be seen as one workshop divided over two adjacent spaces? Of the two workshops of the second group, VII 2, 11 has also 9 cauldrons, divided in sets of 2 and 3, whereas workshop IX 3, 1-2 has one set of 3 cauldrons and thus a relatively modest pro-duction capacity.

Fullonicae

As is the case with the dyeing workshops, fullonicae are linked to the textual record through electoral slogans on the façades of some workshops, which refer to fullones. Full-ing aims at improving the physical (and visual) qualities of woollen textiles, and consists of three main phases: soaping, rinsing and polishing; it was generally performed on fully finished clothes, both new and used ones.48 Typical installations found in fullonicae are the “fulling stalls” consisting of tubs embedded in the floor surrounded by low walls covered by waterproof plaster (fig. 5); in these tubs, workers treated clothes with chemicals by tram-pling them underfoot; they could use the two surrounding walls to support their arms.

Thirteen fullonicae may be identified at Pompeii, of which 12 have been excavated.49 Within this data-set there is a fundamental distinction between 4 medium-sized work-shops which, alongside the fulling stalls, had a complex of 2 or 3 rinsing basins supplied by

48 Flohr 2013, 57-64.49 Excavated fulling workshops include I 4, 7; I 6, 7; I 10, 6; V 1, 2; VI 3, 6; VI 8, 20-21.2; VI 14, 21-22;

VI 15, 3; VI 16, 3-4; VI 16, 6; VII 2, 41; and IX 6, a.1. Large quantities of epigraphic evidence make it likely that there was a thirteenth, unexcavated fullonica at IX 13, 5-7. Cf. CIL IV 9125, 9128 and

Fig. 5. Pompeii, fullonica V 1, 2 with three fulling stalls and waterproof workfloor in front (photograph taken during cleaning, 2008).

The textile economy of Pompeii 63

the town’s water system (fig. 6), and 8 small workshops which do not appear to have had internal rinsing facilities.50 The fullonicae with a rinsing complex also had a higher number of fulling stalls (between 5 and 10, as opposed to 2 or 3), which implies that they were built to operate on a larger scale.51 Yet both types are directly associated with fullones through epigraphic evidence, and there is no indication that the process carried out in them dif-fered in any meaningful way. Thus, while earlier scholars thought of the large fullonicae as cloth-finishing factories and of the small ones as laundries, such a distinction is not visible in the archaeological record.

Unlike dyeing workshops, fullonicae have been identified reliably in a number of other places in Italy, most prominently Ostia and Rome, and a comparison shows significant differences which are crucial to understand the economic rôle of fulling at Pompeii. Com-pared to the larger establishments at Rome and Ostia, the Pompeian fulleries are small: measured in the number of fulling stalls, the largest fullonicae of Rome and Ostia each are at least as large as all 12 excavated Pompeian fullonicae together.52 In terms of its fulling capacity, Pompeii is ‘small fry’. To this must be added a fundamental difference in the eco-nomic orientation of these workshops. Unlike at Rome and Ostia, a large majority of the fullonicae seem to have had a shop.53 Thus they seem to have worked at least in part for a

9131. Ten of these have recently been thoroughly studied in the field: Flohr 2007a (I 4, 7; VI 15, 3; IX 6, a.1); id. 2008 (VI 8, 20-21.2; VI 14, 21-22; VII 2, 41) id. 2011a (I 4, 7; I 10, 6; V 1, 2; VI 3, 6; VI 15, 3; VI 16, 3-4).

50 The large drain mouth on the street in front of entrance IX 13, 5 suggests that the unexcavated fullery in this complex belonged to the larger group of medium-sized establishments.

51 The exception is VI 16, 3.4, which had only 3 fulling stalls next to its rinsing complex.52 Flohr 2013, 77.53 On the identification of shops, see also Flohr 2007b, 133; Cf. Borgard and Puybaret 2004, 50;

Laurence 1994, 100.

Fig. 6. Pompeii, fullonica VI 8, 20-21.2, with two basins of the rinsing complex (photograph taken during cleaning, 2007).

M. Flohr64

clientele of private customers: the only two fulling workshops that did not have a shop are the large fullonica VI 8, 20-21.2 and the medium-sized VI 16, 3-4. The consumer-oriented nature of fulling in Pompeii also contrasts with the dyeing workshops in town, of which only IX 3, 1-2 had a shop.54

Felt-making workshops

The third workshop type that can be directly related to the town’s textile economy is not easily understood: there are no direct ties with textual or iconographic evidence, and the installations found in them are not sufficiently specific to allow scholars to interpret them unequivocally. Yet their relation with wool-processing is beyond reasonable doubt, as three are directly related to other textile workshops: two are connected to a dyeing workshop (I 8, 19 and VII 14, 5.17-18), and one is in a work area also containing a fullonica (VI 16, 3-4). A fourth workshop of this type is found in the House of the Postumii (VIII 4, 4). The workshops are characterized by one or two small lead cauldrons fixed over a furnace, a basin, and long, low working benches with a travertine working surface (fig. 7). There was space for workers to stand on both sides of these benches, and around the working benches many drainage provisions are to be found.

While Moeller did not recognize these workshops as a separate category, Borgard has suggested that these workshops were, essentially, more advanced versions of the lanifri-cariae and that they served for wool-washing.55 The key argument behind Borgard’s inter-pretation was that the equipment in these workshops resembled that of the lanifricariae, so

54 The situation with V 1, 4 and V 1, 5 is complicated: although they are situated in tabernae and thus probably also worked with private customers, they do not have as much space to interact with such customers as did the fullonicae; the fact that they are so close together and may have operated as a whole might suggest a business model aimed at professionals rather than only at end-users.

55 Borgard and Puybaret 2004, 57-58.

Fig. 7. Pompeii, VI 16, 3.4, with remains of felt-making workshop.

The textile economy of Pompeii 65

if the identification of the lanifricariae as wool-scouring plants is to be rejected, nothing is left to support the idea that these workshops served for wool-scouring. A more attractive alternative was recently suggested by J.-A. Dickmann, that these workshops were devoted to felt-making.56 Traditionally, felting is done with the aid of a heated mixture of water and detergents, which would explain the presence here of heated cauldrons.57 The hot water mixture is poured over the fibres, which are then kneaded and scrubbed until they felt. This needs to be done on a smooth and even surface that does not absorb the water and chemicals used, and does not interfere with the fibres; for this purpose the travertine working benches are ideal. The liquid waste generated would explain the drainage facili-ties around the working benches. The basins in these workshops probably were used to rinse the felt subsequently.

As noted by Dickmann, the equipment in these workshops can be reconciled with a depiction of felt-making on the façade of taberna IX 7, 7 (fig. 8). It shows workers that are referred to as “quactiliari” (felt-makers) by an electoral slogan on the same façade.58 The painting shows 8 workers standing around two working tables that are not functionally different from the travertine working benches; the cauldron depicted is comparable to the cauldrons found in these workshops, in that it is small and closed;59 further, in the paint-ings the workers stand in an arrangement that is also suggested by the layout of these 4 workshops: two on each side of the working table.60 There are no differences in capacity between these 4 workshops: all have two working benches of roughly similar size, and provided space for 4 or 8 workers. The paintings and texts near shop IX 7, 7 suggest that a fifth workshop lies hidden in the unexcavated building behind this façade. Of these 5 workshops, two were directly related to a shop, but the three that were related to other textile workshops were not.61 This suggests that these workshops had a slightly different business model, one that relied less directly on spontaneous sales to local customers.

56 Cf. Dickmann 2011, 60-62; id. 2013.57 On the technology of felt-making, see Moeller 1976.58 CIL IV 7838.59 Based on the paintings on the façade of IX 7, 7, Moeller identified two felt-making workshops

in Pompeii (I 12, 4 and IX 3, 16), but the material remains are too badly preserved to be sure and other interpretations are possible. The limited size of the rooms argues against Moeller’s inter-pretation: there is barely space for work installations like the ones depicted on the paintings. Cf. Moeller 1971; id. 1976, 51-54.

60 The main difference is related to the discharge of liquids: while the painted working tables have a rim and seem to discharge in the cauldron, the travertine benches drained on the floor. See also Dickmann 2013.

61 VIII 4, 4 and IX 7, 7 had a shop; I 8, 19, VI 16, 3.4 and VII 14, 5.17-18 did not.

Fig. 8. Pompeii workshop IX 7, 7, painting on façade showing felt-makers at work.

M. Flohr66

Spinning and weaving

As far as archaeologically recognizable workshops are concerned, the evidence is restricted to dyeing workshops, fullonicae, and felt-making workshops. The evidence for spinning and weaving is much more fragmentary and harder to make sense of from an economic point of view. The problem is not that weaving and spinning are unidentifiable in the material record: spindles, spindle-whorls and loom-weights, which have been found in many places in the town, particularly domestic contexts, provide clear indications for these processes.62 The proliferation of these items suggests that it was not uncommon for households to have some access to the basic equipment needed for turning raw wool into textiles, but, as Monteix rightly noted,63 loom-weights in many cases have been found in small numbers — which suggests re-use rather than the presence of a loom: looms can safely be identified only when large numbers of loom-weights are found together. The number of confidently identifiable looms in the town is rather limited: Monteix, using a minimum number of 20 loom-weights as a criterion, counts only 16.64

As several scholars have argued, looms do not, in themselves, constitute evidence for market-oriented production: it is widely accepted that weaving that took place within pri-vate households could serve its internal needs. For the looms found in domestic contexts it is thus impossible to distinguish, archaeologically, between household-oriented weaving and market-oriented weaving.65 Of the 16 identifiable looms, thirteen are found in hous-es, and only 3 were found in places that may suggest a ‘commercial’ context,66 yet even in those cases the evidence is ambiguous: while the one-room taberna VII 16, 19 and the two-room taberna I 6, 10 perhaps may be seen as weaving shops, the evidence says little about the economic context in which they operated: did they work on demand for private customers, or regularly for professionals? In the case of taberna IX 2, 5, the position of the loom in a back room (itself with a possibly domestic function) does not exclude a scenario in which it was used for household purposes.

This leaves us with very little material evidence upon which to build, and the epigraph-ic record is equally dissatisfying. There are no painted electoral slogans; and while there is a graffito mentioning a textor, suggesting that the occupation was not unknown to Pom-peians, it does not refer to his work.67 A graffito from the peristyle of House VI 13, 6.8-9 seems to record the names and productivity of 10 (female) spinners, but it does not provide the contextual information necessary to appreciate the economic context in which this took place.68 While Moeller, following M. Della Corte, identified a number of places where graf-fiti indicated “large-scale spinning and weaving”, his identifications have all been shown to be wrong by Jongman and, recently, Monteix.69

The relative silence of the evidence makes it hard to grasp the economic dynamics of spinning and weaving in Pompeii. It is likely that professional, market-oriented spinning

62 Cf. Allison 2004, 146-48.63 Monteix 2011, 180.64 Ibid. 183-84.65 See, e.g., Allison 2004, 148; Monteix 2011, 185-86.66 Tabernae I 6, 10, VII 16, 19, and IX 2, 5; cf. Monteix 2011, 186.67 CIL IV 8259.68 CIL IV 1507. Cf. Monteix 2011, 175-76.69 Moeller 1976, 39-41, listed I 10, 8; V 3, 10; VI 13, 6.8-9; VII 4, 57; IX 7, 20; and IX 12, 1-5, as textri-

nae; but see Jongman 1988, 162-65; Monteix 2011, 183-85.

The textile economy of Pompeii 67

and weaving took place, but the evidence does not tell us to which degree they played a defining rôle in the town’s textile economy. Still, the evidence for weaving implies a rather different investment compared to the other branches of the urban textile economy: the 16 buildings with a loom also seem to have contained only one loom. Commercial weaving, if it existed, was thus less concentrated in formal ‘workshops’ and took place in produc-tion units of an even smaller scale: individual weavers working at home or in a small taberna, perhaps with support from one or more spinners. This, combined with the rela-tive paucity of epigraphic and material evidence when compared to other textile crafts, suggests that professional spinning and weaving did not play a central rôle in the textile economy. In any case, the burden of proof at this point for those wishing to argue for sub-stantial commercial weaving is considerably greater than for those inclined to argue for more household-oriented scenarios.

Trade and retail

A final category of evidence is related to trade and retail. Due to the traditional emphasis on production, it has not received much attention from scholars, apart from the traditional idea that the Building of Eumachia was some kind of textile market (as this idea has been destroyed by an array of scholars70 it does not need to concern us here). For trade as well as retail, the evidence is scarce and vague. Important are the paintings from the Praedia of Julia Felix, which show at least two clothing traders negotiating a sale with a customer in a context that has, for good reasons, been identified as Pompeii’s forum (fig. 9);71 prob-ably the scene is meant to show itinerant traders who operated on a regional level selling their wares to Pompeian consumers. Then there are the paintings on the façade of felt- making workshop IX 7, 7 (fig. 10, cf. fig. 8). While the felt-making scenes are to the left of the entrance, to the right is a shop scene: a woman behind a counter, and in front of her a table on which the wares to be sold (probably hats) are on display. The combination of the two pictures on either side of the entrance suggests that the owners of this workshop wanted to convey a message of locally oriented production. Lastly, a graffito on one of the columns of House VII 2, 16 mentions a certain Vecilius Verecundus, who was a vestiarius, but the economic context in which he operated is unclear.72

While this evidence in itself contains little information about the realities of trade and retail in Pompeii’s textile economy, it does suggest that at least part of the local public

70 See especially Jongman 1988, 178-84; cf. Moeller 1976, 65-72. For a discussion of the Eumachia building see Fentress 2005, 225-29.

71 Cf., e.g., Holleran 2012, 200-1.72 CIL IV 3130.

Fig. 9. Pompeii, II 2, 2 (Praedia of Julia Felix), paintings with scenes from the Forum.

M. Flohr68

image of the textile economy was connected to its interaction with consumers. This is also true for the public image of the fullones. The fullones are by far the best represented occupa-tional group in the electoral epigraphy: many workshops show evidence for fullones using their occupational status in electoral campaigning. The fact that fullones could do so must be seen as related to the consumer-oriented character of their workshops: it was through the frequent interaction with customers about the treatment of their clothes that their occu-pational status, and their electoral credibility, was constructed.73 Thus it is not unlikely that many Pompeians associated local textile crafts primarily with consumption, and not so much with production. To some extent this image must have been rooted in reality.

Characterizing Pompeii’s textile economy

The question is how we can use this rich and varied body of evidence to understand the nature and performance of Pompeii’s textile economy as a whole. In this respect three issues are relevant. First, it is useful to rethink what we can actually say about the manu-facturing output generated in the town. Second, how do the different textile crafts relate to each other? To what extent was the textile economy integrated into a coherent chain of production? Third, we can raise the issue of the nature and context of investment in workshops: who were those investing in the town’s textile economy, and what were their investment strategies?

Manufacturing and its output

Quantification of the level of output at Pompeii is impossible, but there are several rea-sons to suspect that the rôle of market-oriented manufacturing was much smaller than Frank, Moeller and Borgard have assumed. One point concerns the absence of evidence for wool preparation. Thus any raw wool entering the town had already been washed and, probably, carded. This does not in itself argue against large-scale textile manufacturing, but it does point to the fact that, for textile production on a larger scale, the town would have depended on imports from farther away. For wool coming from flocks grazing in the vicinity of Pompeii, it makes sense to centralize the washing and carding in urban work-shops, but for wool arriving after a trading process it is more useful to wash and card the

73 For a more elaborate argument, cf. Flohr 2011a, 96-100; id. 2013, 337-38.

Fig. 10. Pompeii, workshop IX 7, 7, painting on façade showing the selling of products.

The textile economy of Pompeii 69

wool close to the source since that reduces weight and volume, and also makes it easier for intermediate traders to check its quality. That at least some wool was washed close to its point of origin is suggested by inscriptions referring to wool-washers that are found only in Cisalpine Gaul, an area also known for its production of raw wool.74 While some wool probably entered Pompeii from its own hinterland, the evidence for agriculture in Pompeii’s environs suggests a strong emphasis on viticulture and cereals.75 None of the villae there has returned evidence for sheep-rearing, and the flourishing Vesuvian wine economy implies that there was not much land left free for grazing. As Jongman already suggested, textile manufacturing in Pompeii was not fostered by large quantities of wool coming from the immediate environs.76

A second issue is that a significant proportion of the dyeing capacity probably did not support textile manufacturing but felt-making. Of the three excavated dyeing work-shops focusing on raw wool, two were related to a felt-making workshop, where in all probability a considerable proportion of the felt was made with wool dyed in the same establishment. It is likely, especially in the case of the large workshop VII 14, 5.17-18, that these workshops also dyed wool to be spun and woven into textiles, but to which extent is unclear. In the case of workshop I 8, 19, which had half the dyeing capacity of workshop VII 14, 5.17-18 but a felt-making workshop of similar size, it is possible that most of the dyed wool was turned into felt, and that the surplus available for spinning and weaving was limited. Thus, while the dyeing workshops leave open the possibility that some textile manufacturing took place in the town, their total capacity does not argue in favour of pro-duction on a significant scale.

Finally, the evidence for fulling also is not necessarily directly related to local textile pro-duction. First, most of the fulling workshops were both fairly small and oriented towards private customers. Measured in fulling stalls, this amounts to some 43% of the town’s exca-vated fulling capacity (19 out of 45 fulling stalls), while of the remaining 57% more than half were serving private consumer demand at least partially: fullonicae I 6, 7 and VI 14, 21-22, while large enough to handle professional demand, had a shop and so also served private customers; the two workshops without a shop stand for 22% of Pompeii’s exca-vated fulling capacity. As a consequence, a considerable part of the town’s known fulling capacity served the local consumer market; these fullonicae either dealt with used clothes that had to be recovered, or with home-woven products that had to be finished.

As far as fulling for a ‘professional’ clientèle is concerned, one may question the idea that this should necessarily be related to textiles that were produced in Pompeii and destined for an export market. Following mediaeval examples, scholars have thought of fulling as the last phase in a chain of textile production. Evidence suggests that this was not generally the case in the Roman world.77 The large-scale fullonicae of Ostia and Rome seem to have been related to imported clothes that were fulled, by traders, after transport, before being

74 AE 1987, 443 (Altinum), refers to lanari purgatores. CIL XI 1031 (Brixia), AE 1927, 100 (Brixia), and AE 1946, 210 (Regium Lepidum) all refer to lanari carminatores, which does not strictly refer to washing but does seem to refer to some preparatory wool-working. The same is true for the lanarii pectinarii mentioned in CIL V 4501 (Brixia) and AE 1946, 210. On wool-production in Cis-alpine Gaul, see Noé 1974; Vicari 1994.

75 See, e.g., Moormann 2007; Jongman 1988, 97-154.76 Cf. Jongman 1988, 159-61.77 Cf. Flohr 2013, 79-84.

M. Flohr70

sold on the metropolitan market.78 This also may have been true for the larger Pompeian fullonicae: they may have dealt with imports rather than with locally produced textiles.

The above points, taken together, suggest that we should hesitate to see Pompeian tex-tile manufacturing as export-oriented: raw materials had to be imported, the urban dyeing capacity was limited, and fulling primarily served consumer demand. This picture is cor-roborated if we turn to the way in which textile crafts relate to each other.

Integration of the textile economy

It may be argued that Pompeii’s textile economy does not directly present itself as a smoothly running ‘production chain’ from raw wool to finished product; the evidence suggests that it was fragmented rather than integrated. A cluster of workshops, the dye-ing and felt-making workshops, dealt with raw materials, and another set of workshops (the fullonicae) dealt with the finishing product, with very little in between. Crucially, no evidence for spinning and weaving has been found in direct relation to any of the other production processes attested.

As far as weaving in domestic contexts is concerned, this leaves room for two scenarios, which are not mutually exclusive. In the first, consumer-driven, scenario, spinning and weaving were oriented towards household needs: people bought their wool after it had been dyed, turned it into a textile, and then, if necessary, brought it to their fullo to have it polished before beginning to use it. The second scenario was more entrepreneur-driven: spinning and weaving were market-oriented and were directed by traders/entrepreneurs, who bought up the wool, had it dyed, and gave it to spinners and weavers working for them, who turned it into textiles and returned it to the trader, who then had it fulled, if desirable, and sold it on the market. Such a ‘putting-out-system’ cannot be excluded for Pompeii, as it would be invisible in the archaeological record. However, there is no evi-dence from elsewhere in the Roman world that such systems were common.79

Moreover, as argued above, the aggregate dyeing capacity available to support such a system would have been limited. Basically it would have consisted of dyeing workshop VII 14, 5.17-18 and the twin dyeing workshops V 1, 4 and V I, 5.80 It is possible that some of the wool dyed in these establishments was given to professional weavers. The situation with the latter two workshops, which belonged to the Casa del Toro (V 1, 3.6-9) is espe-cially interesting as the small fullonica V 1, 2 was in the same property (fig. 11); the owners could have the returned garments fulled on the spot if they wished so. Yet a putting-out- system, if one existed in this property, operated on a very small scale: fullonica V 1, 2 was unable to deal with large batches of clothes, and both the fullonica and the dyeing work-shops were situated in tabernae, which suggests they also dealt with private customers — one cannot exclude the possibility of people bringing their own wool to have it dyed there. In both cases, an eventual putting-out system is unlikely to have contributed to the generation of meaningful volumes of export trade.

While the evidence does not allow us reliably to fix a ratio between ‘consumer-driven’ and ‘entrepreneur-driven’ textile manufacturing, it seems to make sense to assign to the

78 Cf. ibid. 84-87.79 For a recent discussion on putting-out-systems in textile production see Dross-Krüpe 2011,

183-89.80 Possibly IX 7, 2 should be added to this, but the fact that it was situated in a shop suggests that

it also aimed at private customers.

The textile economy of Pompeii 71

former a much larger rôle than past accounts have done: a 50 : 50 ratio probably overes-timates the scale of market-oriented weaving. In combination with the limited amount of evidence for weaving found in non-domestic contexts, this would mean that textile manu-facturing at Pompeii was a system in which ‘the market’ played a relatively limited rôle. To some extent, the dyeing workshops and, especially, the fullonicae should be seen as performing services in support of individuals spinning and weaving their own textiles at home. However, this is not the whole story, as becomes clear if one takes a closer look at investment patterns.

The nature and context of investment

As far as Pompeii’s textile economy is concerned, three categories of investors can be discerned in the archaeological record. A first category had a substantial and broad eco-nomic portfolio in which the textile economy played only a limited rôle. Typically, their property included a range of tabernae of which only one ended up being a textile workshop. Most of these workshops are small fullonicae, including fullonica I 4, 7 which belonged to the Casa del Citarista (I 4, 5.25), fullonica I 10, 6 related to the House of the Menander (I 10, 4), fullonica VI 3, 6 connected to house VI 3, 7, and fullonica VII 2, 41 which was part of the House of Popidius Priscus (VII 2, 20). In addition, there is the felt-making workshop in the House of the Postumii (VIII 4, 4). In none of these cases can we expect the workshop to have dominated the proprietor’s income portfolio. Ideally, this is how élites organ-ized their urban commercial activities in Italy: a variety of low-investment, small-scale shops and workshops allowed investors to spread their risks and enjoy a stable in-

Fig. 11. Pompeii, V 1, 3.6-9, with the two dyeing workshops in shop 4-5 and fullonica in shop 2.

M. Flohr72

come.81 It may be added that, with one exception, the sizes of the complexes to which these shops were related belong to the top 10% of the town’s housing stock (fig. 12, grey): ranked by size and room number, they score, on average, 95.05 on a scale of 0 to 100.82

A second category of investors also built small workshops, but as part of a much small-er economic portfolio (fig. 12, white). There are three examples: fullonica VI 15, 3 was the only shop belonging to an atrium house of modest size (VI 15, 2); the small dyeing work-shop IX 3, 1-2 was set in an equally modest house, and fullonica IX 6, a.1 was situated in one of three shops possibly belonging to the medium-sized atrium house IX 6, 3. These houses rank much lower in the urban hierarchy, scoring 71.52 on average, which means that they belong to the bottom half of the town’s housing stock.83 This was a pattern of small-scale investment comparable to the first group but on a more modest level; it is one that is likely to have been equally, if not more, common in the towns of Italy. These workshops fit into the model of a consumer-driven service economy as suggested by the preceding sections.

81 Cf. Robinson 2005, 101-2; Wallace-Hadrill 1994, 134-41. On the House of the Postumii and its investment portfolio, see Dickmann and Pirson 2001.

82 This ranking was based on a database of 1026 independent structures in Pompeii that have been fully excavated. For example, House I 4, 5.25 scores 1023/1026 = 99.71% on number of rooms and 1020/1026 = 99.42% on surface area, leading to an average of 99.56% (which is the highest average in the town). The other scores are: I 10, 4: 98.88% (overall ranking: 9th); VII 2, 20: 97.22% (overall 25th); VIII 4.4: 94.79% (overall 44th); VI 3, 7: 84.8% (overall 151th). A ‘struc-ture’ is defined as a set of rooms that are internally connected to each other; thus independ-ent tabernae are counted as separate structures from the house to which they, architectonically, belonged.

83 The lowest 50% of the hierarchy in the database consists of independent tabernae, so anything below 75% ranks among the smaller houses: IX 3, 1-2: 70.71% (overall 298th); IX 6, 3: 71.1% (overall 295th); VI 15, 2.26: 72.76% (overall 279th).

Fig. 12. Relative size of complexes related to textile workshops.

The textile economy of Pompeii 73

The third group of investors, however, seems to have followed a different strategy (fig. 12, black). They concentrated most of their economic efforts on the textile economy, and reserved significant space in their properties to build large-scale production facilities. This group includes the fullonicae of houses I 6, 7, VI 8, 20-21.2, and VI 14, 21-22, the dyeing/felt-making workshops belonging to houses I 8, 1 and VII 14, 5.17-18, and the dyeing workshop of house VII 2, 11. Further, there are the workshops belonging to houses V 1, 3.6-9 (one or two dyeing workshops and a fullonica) and VII 16, 7.38 (two fullonicae and one felt-making workshop). The level of investment in this group is clearly above average: it represents 50% of all complexes, but 78% of the production capacity (Table 1).84 Construction of these workshops generally involved large-scale investment and major adaptations to the house, and their scale also meant that they could not be staffed from a small nuclear family; in order to run them, extra labour (either hired or bought) would have been necessary.85

TABLE 1RELATIVE CAPACITY OF THE WORKSHOPS OF EACH OF THE THREE GROUPS

Group No. Dyeing cauldrons Fulling stalls Felt-making sets Average % of expectedI 5 (31%) - 9 (21%) 1 (25%) 15% 48%II 3 (19%) 3 (9%) 5 (11%) - 7% 36%III 8 (50%) 31 (91%) 30 (68%) 3 (75%) 78% 156%

16 34 44 4

The houses in which these workshops were embedded were slightly smaller than those of the first group (87.23 on average), though much larger than the second.86 Rather than belonging to the core of the urban élite, these people mostly belonged to sub-élite groups. While some of these houses are also connected to other commercial spaces, it is likely that the textile economy contributed significantly to the household’s income. The produc-tion capacity of the workshops of this third group, and the investment involved in their construction, suggests that they were built by people expecting a relatively constant and substantial flow of orders. In other words, these workshops were not built to deal with pri-vate consumer demand, but also or uniquely worked with the larger batches of products and materials coming in and going out through the professional circuit. This is also seen in their remains: while with one exception all of the workshops belonging to the first two groups had a shop, half of the workshops in this group did not. Their scale, as well as their economic orientation, suggests some sort of involvement in supra-local networks of trade and exchange.

Pompeii thus shows a combination of two investment patterns: small-scale, locally- oriented investment but, alongside it, larger-scale investment that was related to the town’s rôle in its wider economic environment. This makes it even more relevant to consider the regional economic context within which the town functioned.

84 As to the fullonicae, all rinsing complexes of Pompeii — which represent large-scale investment because of their size and their connection to the water-supply — are found in the fullonicae belonging to this group (I 6, 7; VI 8, 20-21.2; VI 14, 21-22; and VI 16, 3.4).

85 The degree to which houses were adapted varied from case to case: cf. Flohr 2011b and 2012.86 I 6, 7: 76.12% (241st); I 8, 1-3: 79.78% (201st overall); VII 2, 11: 80.26% (195th overall); VI 14, 21-22:

84.75% (152th overall); VII 14, 5.17-18: 91.76% (81st overall); VI 16, 7.38: 92.98% (62th overall); VI 8, 20-21.2: 95.18% (41th overall); V 1, 3.6-9: 97.03% (26th overall).

M. Flohr74

Pompeii and the regional economy of dress

Pompeii was not an isolated place. As the largest urban centre in the S part of the Bay of Naples, conveniently located on a low hill overlooking the Sarno estuary, it was well- connected by roads to Stabia and Surrentum towards the southwest, the coastal zone below Mt. Vesuvius to the north, and the Sarno plain and ager Nucerinus to the east.87 While there is debate about the precise location of its harbour, it is likely that the town was integrated into the regional maritime network dominated by Puteoli.88 Because of its central location, Pompeii had the possibility of playing a key rôle in networks of exchange and trade in both the S part of the Bay of Naples and within the Sarno valley, which will then have affect-ed the strategies embraced by those involved in the textile economy. We need to look at specific properties of the regional environment since any account of the Pompeian textile economy depends on assumptions which we should try to reconstruct, even if the picture remains largely theoretical because the evidence is highly fragmentary.

The Sarno plain and the ager Nucerinus (fig. 13)

All trade and exchange between the Sarno plain and the wider Mediterranean world passed through Pompeian territory, and a considerable proportion of it (both exports and imports) is likely to have been transshipped in the town’s direct environment. While there is no evidence from the Roman period for sheep-rearing or transhumance in the (upper) Sarno valley, it is possible that some of the agricultural produce from this region consisted of wool, and if it was exported Pompeii would have been the first major transshipment point and perhaps also a destination for some of it. On the other hand, like the town’s

87 According to the Tabula Peutingeriana, a road connected Pompeii, Stabiae and Surrentum. For the ager Nucerinus, see Plin., NH 3.60, where it is described as if it was situated between Pompeii and the town of Nuceria.

88 For recent discussions about Pompeii’s harbour, see Curti 2008; Stefani and Di Maio 2003.

Fig. 13. Pompeii in its regional environment.

The textile economy of Pompeii 75

immediate environs, most of the land between Vesuvius, Nola and Nuceria comprised a well-watered plain that is likely to have been intensively cultivated. Sheep-rearing would not have been the most profitable strategy, and, with the exception of the higher slopes of Vesuvius and the narrow valley of San Severino beyond Nuceria, the landscape was prob-ably dominated by vineyards, horticulture and grain.89 This also suggests that there was no major export flow of raw wool passing through Pompeii. Yet there is more to trade than export. As Strabo (5.4.8) emphasizes, Pompeii’s harbour and the Sarno plain also played a key rôle in the supply network of Nuceria and Nola. Imported wool and textiles destined for those towns are likely to have passed through Pompeii’s harbour, which will have creat-ed opportunities for Pompeians willing to invest in the textile economy. Indeed, for vestiarii selling wool or clothes in Nuceria and Nola, Pompeii would have been a convenient base. If such vestiarii generally had their imported textiles fulled before they sold them, Pompeii would have been the logical place to do so, just as Ostia was for traders operating at Rome.90 This extra demand for fulling may have played a part in the economic strategies that led to Pompeii’s three larger fullonicae. Likewise, regional demand for dyed wool or felt may have made it attractive for entrepreneurs to invest in felt-making and dyeing workshops.

The southern bay of Naples and Puteoli

It is thus possible that the Sarno plain and the ager Nucerinus, particularly Nuceria and Nola, had some impact on Pompeii’s textile economy, yet more important perhaps was the S coast of the Bay of Naples, and not only because of the numbers of people living there but also because of who they were and how they spent their lives and money. The concentration of luxury villas is well-known.91 Some belonged to equestrian or senatorial families or had ties to the imperial house.92 The culture of otium required a suitable outfit, and those who wished to participate had to spend on high-quality clothes.93 Élites consti-tuted a natural market for imported clothes.94 From Pompeii traders could easily reach the villas on the Sorrentine peninsula and along the coast below Mt. Vesuvius; it was the best base for vestiarii operating on a regional scale. Pompeii’s place in the distribution network of the southern Bay of Naples created special opportunities for investment within the tex-tile economy.

Pompeii was also tied in to networks of supply through the harbour of Puteoli, in the 1st c. A.D. home to a cosmopolitan trading community and still Italy’s main harbour for long-distance trade to both the E and W Mediterranean.95 Other than one inscription (2nd c. A.D.) referring to a negotiator sagarius, there is little direct evidence for a textile trade at Puteoli, but the maritime traffic to its harbour is likely to have generated a steady arrival of textiles.96 This proximity of Pompeii to the pan-Mediterranean supply system of Rome is likely to have had a positive effect on the textile market: traders were well-connected

89 The network of transhumance, which covered parts of Campania, was oriented towards the Adriatic coast, and did not connect to the Sarno valley. Cf. Waldherr 2002, 435.

90 See above, p. 69.91 On these luxury villas, see Pesando and Guidobaldi 2006, 279-86 and 415-27.92 See especially D’Arms 1970, 73-115 and 171-233.93 Cf. Flohr 2013, 64-68.94 A similar point was made by De Ligt 1993, 99.95 For a recent study of the economic history of Puteoli, see Jaschke 2007; also D’Arms 1974 and

Purcell 1984.96 For the negotiator sagarius, see CIL X 1872. For the date, cf. D’Arms 1974, 121 n.136 (‘Antonine

period’).

M. Flohr76

and had easy access to both information and to goods and products that could be sold or further processed. This lowered transaction costs in the textile economy and gave traders and consumers in the Bay of Naples, and especially at Pompeii, an advantageous position compared to those operating elsewhere in Italy.

Conclusion

Pompeii was not a ‘textile town’. Frank and Moeller were mistaken in believing that the unique quantity and variety of evidence for textile crafts point to large-scale export- oriented manufacturing. Recent attempts to revive their model fail to address both the character of Pompeii’s archaeological record and the economic realities of the world in which the town operated. However, neither is the opposite true. Jongman’s minimal-ist view of a small-scale, locally-oriented textile economy is as unconvincing as was the Frank/Moeller model because it underestimated the degree to which Pompeii was inter-woven with its regional environment. Pompeii was not an isolated city; rather, Pompeii’s textile economy must be seen as a mixed, import-driven consumer economy that, while satisfying local consumer demand, played a central rôle in regional distribution networks, both around the southern Bay of Naples and in the Sarno plain. Pompeii’s geographical position enhanced investment opportunities for textile entrepreneurs to a level that was probably rare elsewhere. The amount and scale of investment in fulling and dyeing work-shops visible in Pompeii does not seem to be common in comparable urban centres in, e.g., the Apennine region or S Italy.

Yet while textile crafts and trade at Pompeii were characterized by their strong integra-tion into regional and supra-regional economic networks, urban textile manufacturing was penetrated by market exchange only to a relatively limited extent. A substantial, though unknown, proportion of spinning and weaving took place in domestic contexts as part of regular or seasonal household activities. This household-oriented basis of textile manu-facturing was probably something that Pompeii had in common with many Italian urban centres of the 1st c. A.D. This was the traditional model of self-sufficiency, which pervaded (female) élite ideology,97 and it continued to play a rôle in satisfying textile demand in many towns, including Pompeii. Yet with the economic integration of the Mediterranean, a more market-oriented textile economy emerged too, especially in the vicinity of large or wealthy centres of demand. It is within this supra-local textile economy that Pompeii could develop as a centre in its own region’s distribution system, and it is in light of this model that the town’s larger textile workshops should be interpreted. m.flohr@hum.leidenuniv.nl Doelensteeg 16, Leiden

AcknowledgementsThis article results from research conducted at the Oxford Roman Economy Project, generously funded by Baron Lorne Thyssen. I thank the Soprintendenza Archeologica di Pompei for granting access to the site. Nicolas Monteix, Luuk de Ligt and Andrew Wilson provided useful and welcome criticism, comments and suggestions.

BibliographyAllison, P. M. 2004. Pompeian households. An analysis of the material culture (Los Angeles, CA).Amoroso, A. 2007. L’Insula VII, 10 di Pompei. Analisi stratigrafica e proposte di ricostruzione (Rome).

97 On the issue of lanam facere as a symbol of female virtue, see, e.g., Larsson Lovén 2007; Cottica 2007.

The textile economy of Pompeii 77

Banaji, J. 1989. Review of Jongman 1985, in JRS 79, 229-31.Bechi, G. 1827. “Pianta, e spaccato della fullonica di Pompei e delle case delle fontane,” Reale Museo

Borbonico 4, 48-49.Borgard, Ph. 2002. “A propos des teintureries de Pompéi: l’exemple de l’officina infectoria V 1, 4,” in

J.-C. Béal and J. C. Goyon (edd.), Les artisans dans la ville antique (Paris) 55-67. Borgard, Ph. and M.-P. Puybaret 2003. “Approche archéologique du travail de la laine au Ier siècle

après J.-C.,” in G. Cresci Marrone and M. Tirelli (edd.), Produzioni, merci e commerci in Altino pre-romana e romana (Rome) 299-318.

Borgard, Ph. and M.-P. Puybaret 2004. “Le travail de la laine au début de l’empire: l’apport du modèle pompéien. Quels artisans? quels équipements? quelles techniques?,” in C. Alfaro, J.-P. Wild and B. Costa (edd.), Purpureae vestes I. Actas I symposium int. sobre textiles y tintes del Mediterráneo en época romana (Valencia) 47-59.

Borgard, Ph., J.-P. Brun, M. Leguilloux and M. Tuffreau-Libre 2003. “Le produzioni artigianali a Pompei. Ricerche condotte dal Centre Jean Bérard,” RStPomp 14, 9-29.

Breton, E. 1855. Pompeia (Paris).Camodeca, G. 1999. Tabulae Pompeianae Sulpiciorum (TPSulp.): edizione critica dell’archivio puteolano dei

Sulpicii (Rome).Coarelli, F. and F. Pesando 2006. Rileggere Pompei I. L’insula 10 della Regio VI (Rome).Cottica, D. 2007. “Spinning in the Roman world: from everyday craft to metaphor of destiny,” in Gil-

lis and Nosch 2007, 220-28.Curti, E. 2008. “Il tempio di Venere fisica e il porto di Pompei,” in P. G. Guzzo and M.-P. Guidobaldi

(edd.), Nuove ricerche archeologiche nell’area vesuviana (Rome) 47-60.D’Arms, J. H. 1970. Romans on the Bay of Naples. A social and cultural study of the villas and their owners

from 150 B.C. to A.D. 400 (Cambridge, MA).D’Arms, J. H. 1974. “Puteoli in the second century of the Roman Empire: a social and economic

study,” JRS 64, 104-24.Della Corte, M. 1954. Case ed abitanti di Pompei (Rome).De Ligt, L. 1993. Fairs and markets in the Roman Empire. Economic and social aspects of periodic trade in a

pre-industrial society (Amsterdam).Dickmann, J.-A. 2011. “Space and social relations in the Roman West,” in B. Rawson (ed.), A compan-

ion to families in the Greek and Roman worlds (Chichester) 53-72.Dickmann, J.-A. 2013. “A ‘private’ felter’s workshop in the Casa dei Postumii at Pompeii,” in M. Gleba

and J. Pasztokai-Szeöke (edd.), Making textiles in pre-Roman and Roman times: people, places, identi-ties (Oxford).

Dickmann, J.-A. and F. Pirson 2001. “Die Casa dei Postumii in Pompeji und ihre Insula. Bericht über die 4. Kampagne 2000,” RömMitt 108, 329-51.

Dickmann, J.-A. and F. Pirson 2002. “Die Casa dei Postumii in Pompeji und ihre Insula. Fünfter Vor-bericht,” RömMitt 109, 243-316.

Dross-Krüpe, K. 2011. Wolle–Weber–Wirtschaft: Die Textilproduktion der römischen Kaiserzeit im Spiegel der papyrologischen Überlieferung (Wiesbaden).

Duncan-Jones, R. 1980. Review of Moeller 1976, in CR 29.1, 190-91.Dyer, T. H. 1869. Pompeii. Its history, buildings, and antiquities (New York).Ellis, S. 2004a. “The distribution of bars at Pompeii: archaeological, spatial and viewshed analyses,”

JRA 17, 371-84.Ellis, S. 2004b. “The Pompeian bar: archaeology and the role of food and drink,” Food and History 2.1,

41-58.Ellis, S. and G. Devore 2009. “The fourth season of excavations at VIII.7.1-15 and the Porta Stabia at

Pompeii: preliminary report,” FOLD&R 146.Ellis, S. and G. Devore 2010. “The fifth season of excavations at VIII.7.1-15 and the Porta Stabia at

Pompeii: preliminary report,” FOLD&R 202.Ellis, S., A. Emmerson, A. Pavlick and K. Dicus 2011. “The 2010 field season at I.1.1-10, Pompeii: pre-

liminary report on the excavations,” FOLD&R 220.Fentress, L. 2005. “Catastae, chalcidica and cryptae in Early Imperial Italy,” JRA 18, 220-34.Finley, M. I. 1985. The ancient economy (2nd edn., London).Fiorelli, G. 1875. Descrizione di Pompei (Napoli).Flohr, M. 2007a. “Cleaning the laundries. report of the 2006 season,” RStPomp 18, 131-36.Flohr, M. 2007b. “Nec quicquam ingenuum habere potest officina? Spatial contexts of urban production

at Pompeii, AD 79,” BABesch 82, 129-48.Flohr, M. 2008. “Cleaning the laundries, II. Report of the 2007 campaign,” FOLD&R 111.

M. Flohr78

Flohr, M. 2011a. “Exploring the limits of skilled craftsmanship: the fullonicae of Roman Italy,” in N. Monteix and N. Tran (edd.), Les savoirs professionels des gens de métier romains (Rome) 87-100.

Flohr, M. 2011b. “Reconsidering the atrium house. Domestic fullonicae at Pompeii,” in Poehler, Flohr and Cole 2011, 88-102.

Flohr, M. 2012. “Working and living under one roof: workshops in Pompeian atrium houses,” in A. Anguissola (ed.), Privata luxuria: towards an archaeology of intimacy (München) 51-72.

Flohr, M. 2013. The world of the fullo. Work, economy and society in Roman Italy (Oxford).Förster, F. C. 1828. “Die neuesten Ausgrabungen in Pompeji, Herkulanum und Stabiä,” Berliner

Kunst-Blatt 1.7, 206-12.Frank, T. 1918. “The economic life of an ancient city,” CPh 13, 225-40.Frank, T. 1927. An economic history of Rome (New York).Frank, T. 1940. An economic survey of ancient Rome. V. Rome and Italy of the Empire (Baltimore, MD).Galella, A., A. Fraja and D. Scognamiglio 1869. “Relazione officiale dei lavori eseguiti dal 1 maggio a

31 dicembre 1869,” Giornale dei Scavi di Pompei 11, 295-310.Gillis, C. and M.-L. Nosch (edd.) 2007. Ancient textiles. Production, craft and society (Oxford). Holleran, C. 2012. Shopping in ancient Rome: the retail trade in the Late Republic and the Principate

(Oxford).Jaschke, K. 2007. Die Wirtschafts- und Sozialgeschichte des antiken Puteoli (Rahden).Jongman, W. 1988. The economy and society of Pompeii (Amsterdam).Laurence, R. 1994. Roman Pompeii — space and society (London).Ling, R. 1997. The Insula of the Menander at Pompeii, vol. I. The structures (Oxford).Larsson Lovén, L. 2007. “Wool work as a gender symbol in ancient Rome. Roman textiles and ancient

sources,” in Gillis and Nosch 2007, 229-36.Mau, A. 1892. “Osservazioni sull’edificio di Eumachia in Pompei,” RömMitt 7, 119-30.Mau, A. 1908. Pompeji in Leben und Kunst (Leipzig).Moeller, W. O. 1962. The woolen industry at Pompeii (Ph.D. diss., Univ. of Maryland).Moeller, W. O. 1971. “The felt shops of Pompeii,” AJA 75, 188-89.Moeller, W. O. 1976. The wool trade of ancient Pompeii (Leiden).Monteix, N. 2011. Les lieux de métier: boutiques et ateliers d’Herculanum (Collection du Centre Jean

Bérard 34).Moormann, E. 2007. “Villas surrounding Pompeii and Herculaneum,” in J. Dobbins and P. Foss

(edd.), The world of Pompeii (London) 435-54.Noé, E. 1974. “La produzione tessile nella Gallia Cisalpina in età romana,” RendLombAcadSciLett 108,

918-32.Pesando, F. and M.-P. Guidobaldi 2006. Pompei Oplontis Ercolano Stabiae (Rome).Poehler, E., M. Flohr and K. Cole (edd.) 2011. Pompeii: art, industry, infrastructure (Oxford).Poehler, E., M. Flohr and K. Cole 2011. “Introduction,” in iid. (edd.) 2011, 1-9.Purcell, N. 1984. “Puteoli,” in M. Frederiksen and N. Purcell (edd.), Campania (British School at

Rome) 319-58.Purcell, N. 1990. Review of Jongman 1988, in CR 40, 111-16.Puybaret, M.-P., Ph. Borgard and R. Zérubia 2009. “Teindre comme à Pompéi. Approche expérimen-

tale,” in C. Alfaro Giner and L. Caroli (edd.), Purpureae vestes II. Vestidos textiles y tintos (Valencia) 185-93.

Robinson, D. 2005. “Re-thinking the social organisation of trade and industry in first century AD Pompeii,” in A. MacMahon and J. Price (edd.), Roman working lives and urban living (Oxford) 88-105.

Rostovtzeff, M. 1957. The social and economic history of the Roman Empire (Oxford).Seiler, F. 1992. Casa degli Amorini dorati (VI 16, 7.38) (Munich).Stefani, G. and G. Di Maio 2003. “Considerazioni sulla linea di costa del 79 d.C. e sul porto dell’antica

Pompei,” RStPomp 14, 141-95.Vicari, F. 1994. “Economia della Cispadana romana: la produzione tessile,” RivStorAntichità 24,

239-60.Waldherr, G. 2002. “Das System der calles (Herdenwanderweg) im römischen Italien. Entstehung

und infrastrukturelle Bedeutung,” in E. Olshausen and H. Sonnabend (edd.), Zu Wasser und zu Land. Verkehrswege in der antiken Welt (Stuttgart) 429-44.

Wallace-Hadrill, A. 1994. Houses and society in Pompeii and Herculaneum (Princeton, NJ).Wilson, A. 2002. “Urban production in the Roman world. the view from North Africa,” PBSR 70,

231-74.