#lang scribble/base
@(require "../Bibliography/cite.rkt"
          scriblib/footnote
          "../Images/image-support.rkt")

@define-cite-footnote[footnote
                      make-footnote
                      cite-footnote]

@title{Commercial Production of Textiles}

@author{Morgan Lemmer-Webber}

@section{Introduction/framing}

In spite of the domestic associations between women and textiles outlined above, commercial production of textiles in the Roman empire is typically framed in scholarly sources as a predominantly male endeavor. And yet there's a dichotomy between some sections of textile commercialization being well documented and some sections being almost entirely overlooked in both the archaeological evidence and in the scholarly discourse.   

Scholars often focus on @index['("Dye Shop")]{dyeing} and @index['("fullery/fullonica")]{fulling} as indicators of commercial production of textiles at a given site. In part this is due to the dearth of evidence for commercial spinning and weaving workshops.@cite-footnote[Flohr2016 24] The evidence that does exist for commercial production is inherently biased against women based on three main factors: archaeological survival bias, the tools and materials required for various stages of textile production, and the way that the Romans viewed the work of women and slaves.

The problem of survival bias manifests in the very nature of the materials used in the early stages of textile production. This erases the kind of clear evidence that exists for other stages within the archaeological record, such as @index['("Dye Shop")]{dyeing} and @index['("fullery/fullonica")]{fulling}.

The patriarchal nature of Roman society passes on its biases to contemporary scholarship, leading to the under-representation of women's roles in textile production. The Roman ideal of women was restricted to locally domestic tasks as opposed to tasks that were visible within the public sphere. Due to this, activities that probably constituted real economic activity by women within the household is diverted into the moral realm of "traditional wife", leading scholars to underestimate its importance. Looking at Roman commercial textile production from the hindsight of a post-industrial society, we as scholars are tempted to think of commercial activity as something that happens in organized industrial centers. The more likely economic model for the early stages of textile production (preparation of the wool, spinning, and weaving) is closer to that of a cottage industry than a modern warehouse. If we take away the assumption that commercial production happens outside the home, then the path for women's involvement is revealed.

@section{Grounding for a @as-index{cottage industry}}

I have already used the term @index['("cottage industry")]{cottage industry} several times in this dissertation; perhaps this term could use some stronger definition and grounding. Cottage industry is a common enough term in the literature, and refers to work done on small scale, particularly domestic, production levels, which then fed back into a larger system.@cite-footnote[Hafter1985 74] This already acknowledges not just subsistence production but a wider economy, of which domestic production is integrated.@cite-footnote[Boeke1942] Cottage industry is often used interchangeably with the "Putting-out system," both of which involve labor being performed within the household and a third-party (whether this be a manufacturer, merchant, wholesaler, or @italic{@as-index{lanarius}}) paying for that labor then selling the final product into a broader market. However, the "putting-out" system prioritizes the manufacturer who has access to raw materials and tools which they put-out to laborers, who are then paid by the piece. This system lowers their overall cost by removing operating costs of a central production center and maintenance of their workers. Cottage industry more frequently refers to the system from the point of view of the laborers, who either have access to raw materials or purchase raw materials, use their own tools to produce the product, then sell the product by piece to the manufacturer.  Both of these are in contrast to the handicraft model where the laborer would sell the final product to the consumer directly in a local market.@cite-footnote[Boeke1953 100]

While all of these terms initially indicated production past subsistence, confluence and vocabulary drift has somewhat muddied their distinction. In use, authors have used the term cottage industry to indicate household production for small scale, local economies despite the initial intention of the term including the possibility for either local trade or within a larger trade network.@footnote{ See @cite[Prentice1983], 18-24 for a discussion of various interpretations of cottage industries.}  In reference to eighteenth century production, the term 'proto-industry' was favored to indicate production within homes of goods intended for larger-scale markets.@cite-footnote[Gullickson1981 179] While this terminology is applicable on the verge of the industrial revolution, it seems less relevant to the ancient economy. I find no reason to believe that this is a binary choice between industry production and domestic production or production for a local economy or a global trade network.@footnote{ Historically, cottage industries have existed alongside centralized production centers elsewhere. For example, see: @cite[Hareven2003 56].} A realistic economic model might combine all of these factors (particularly in early stages of textile production involving cleaning, spinning, etc).

@;Access to tools raw materials:

@;-Homestead model: Self-sufficient farms produce raw-materials at a higher than subsistence level. Laborers produce textiles on the homestead at higher than subsistence level in order to sell into a larger market.

@;-Domestic production to meet the needs of the household is augmented when external materials are available to sell into a larger market.

@;-Laborers purchase raw materials (either outright or on credit), produce textiles, and sell them into a larger market (thus getting a return on investment of the materials and profit)

@;-Manufacturer owns the tools and materials, leases the tools out to laborers, allots portions of raw materials, then pays laborers by the piece for the textiles they produce in their homes

@;Sometimes even within the household, the head of the household 'owns' the tools or materials that others within the household use it (women and servants)

@;Patterns of Sale:

@;-Domestic subsistence: household production to fulfill household needs (this is not a cottage industry because textiles are not produced for the intention of sale)

@;-Handicrafts: household production sold into a local market. In this model, the laborers who produce the goods know and interact directly with the customer. This could be for money, barter, or gifts of social obligation

@;-Municipal market: household production sold into a larger municipal market. In this model the laborers may know the customer, but likely do not. However there is still a sense of shared collaborative interest (think state-level market as opposed to either the town market or a global trade network)

@;-Global market: household production sold to a merchant/wholesaler/etc who then sells the goods into a larger trade network. The laborers do not know the customer, and the product may even pass between multiple merchants before it reaches the final customer.

@; These rules can apply to either the import or export of materials

@; The chances of using currency (as a shared system of value) rather than barter increases with the number of steps between the producers and the consumers.

@; Expanding outward into larger trade networks gives a greater sense of security because it offers market diversity

@;Part-time, occasional work vs. specialized work

@;-Many of the above models are 'part time' because textile production is only one part of the larger sets of domestic or agricultural duties

@;-In cases where cottage industry textile production is a generational endeavor, only the head of the household is considered 'full time' or even employed, even though other members of the family (wives, children, grandparents) contribute to the work. These other members of the family may contribute to the textile work in addition to other household duties.

@;-Textile work can happen intermittently in the space between when material and time are available
@;     -seasonal production (textile work is done over the winter when there is less agricultural work)
@;     -Childcare and other domestic duties
@;     -Education (in the case of children contributing)
@;     -Prostitution

@; context switching is difficult in and of itself, so the part-time nature of the work doesn't inherantly devale uthe level of skill required

Since physical evidence of textile production is scarce in the general case (survival bias is a regular theme in this dissertation), perhaps we can at least lean on whether or not we have evidence of other cottage industries in the Roman world. In @italic{Olive Production and the Roman Economy: The Case for Intensive Growth in the Roman Empire}, Robert Bruce Hitchner provides an example within the world of olive oil.  First, Hitchner lays out that we see evidence in many rural areas of localized presses for large-scale olive oil production:

@nested[#:style 'inset]{
  [...] between the late 1st and 4th centuries there was substantial growth in the number of of rural agricultural settlements devoted to olive production, and that this growth was not restricted to areas under olive cultivation but extended to areas previously underdeveloped agriculturally.  In an area of 1500km in the djebel [mountain or hill] to the west of @italic{Lepcis Magna}, for example, Matttingly estimates that there were more than 750 presses established in the Roman period, approximately 1 press for every 2 km.  Over 350 presses are known in an area of 1,500 km, in the Sbeitla-Kasserine-Thepte region of central Tunisia. In the Guadalquivir valley in Spain as many as 161 of the 1,500 recorded Roman period rural sites show evidence of pressing facilities, and the actual number of presses in the valley "could have been well in excess of 1000".

  In Africa and Tripolitania, where environment and post-Antique histoircal developments have contributed to a high level of site preservation, a significant number of sites show evidence of having had multiple presses (17 in one example, more often 3-5).  [...] the unpretentious character of most of the associated structures is an unequivocal indication of the intensely industrial character of oil production at these sites, suggestive of an intent to produce large volumes of surplus oil on a regular basis.  Indeed, from the standpoint of economic attitudes and responses, the construction of these oilery sites reflects considerable capital investment in the future potential of the mass oil market noted above.  By any standard, this must be considered a marked exception to the supposed heavy rent-seeking mentality of Roman elites.@cite-footnote[Hitchner2012 "75-76"]}

This demonstrates that rural areas were pulled into a wider industrial production system, and also might make a reasonable case that rather than a purely parasitic approach by Roman elites, substantial investment to shape the area may have been made which benefited both local rural residents and the rent-seeking-or-not elites both.  But a purely primitivist approach could still make a case at this point that this may still be interpreted as agricultural locals who could have otherwise survived off of purely subsistence being exploited by those privileged enough to participate in the market economy.  And a large portion of that analysis, particularly in terms of power imbalances, rings true.  However, what follows indicates that stopping here would miss a significant part of the story:

@nested[#:style 'inset]{
  Almost certainly related to this phenomenon is the proliferation of small farms with one or two presses often in close proximity to oileries, and frequently on agricultural marginal lands (piedmont and mountain zones).  That is, the decision to construct a large stone, lever press, particularly when much more modest means for extracting oil for subsistence needs were available, implies that surplus oil production was the ultimate objective of the small farm occupants.  Although the capital for these presses is likely, in many instances, to have come from the owners of the nearly oileries interested in the oleocultural development of marginal lands in or around their estates, we may also see in these arrangements an effort by the farms' occupants, whether independent small-holders, free tenants or even slaves, to better their lot.@cite-footnote[Hitchner2012 "76"]}

What we read from this is that small farms, while undoubtedly exploited by the elites of Roman society, when given the chance, would willingly take steps to feed into and benefit from the market structure itself.  There is no denial that the larger presses of industry within town existed, and yet still we see sophisticated presses serving needs beyond subsistence within smaller domestic farms.  This is what we mean by cottage industry: industrial production and domestic production can exist side by side.  While unequal in power distribution, there is nonetheless economic integration.

This is not a dissertation about olive oil, so why the long digression on this topic?  The point here is to point out a generalization.  A common theme of primitivist approaches is that rural communities are merely exploited by a market structure but otherwise not participants in it.  A common focus of modernist research is on the higher level and elite side of the industrial economy.  Yet if smaller farms willingly installed advance presses to "better their lot" within wider economic participation, then we have reason to believe that smaller domestic centers participated and benefited in selling back to the wider economy and exports.

If a cottage industry existed in the Roman empire for olive oil, it would be difficult to believe this would be necessarily @italic{exclusive} to olive oil. One could even argue that the barrier for entry to small farms participating in the export economy of olive oil was higher than the barrier for entry for individual households to participate in the larger commercial textile economy since the cost of @index['("spindle/whorl")]{spindles} was reasonably low.@footnote{ @cite[DiocletianEdict] XXIII.1 5}  Of course this is not proof enough on its own that the same applies to textiles, so we now turn our eye towards that endeavor: examining evidence@footnote{And, in keeping in the theme of this dissertation, absence of evidence where we must ask where our biases filled in the gaps.} for commercial and domestic production both, as well as the potential role of women in each. I will first outline the existing evidence for commercial production of textiles as it is frequently discussed in economic history, with a focus on centralized production centers, then cycle back to re-examine the archaeological evidence from domestic contexts within the context of the cottage industry. 


@section{Archaeological evidence}

@;Latin term for dye shop?

It is difficult to reliably identify centralized textile production centers archaeologically. @index['("fullery/fullonica")]{Fulleries} (@italic{fullonicae} and @index['("Dye Shop")]{dyehouses} both require specialized equipment make them easier to identify, though even in these cases the archaeological evidence can be ambiguous. Most houses were not equipped to accommodate these stages of textile production, so these facilities were used by multiple households. However, in the evidence from Pompeii, many @index['("fullery")]{fulleries} or @index['("Dye Shop")]{dyeshops} are connected to houses with separate street access for the shop, much like bakeries.@footnote{ For a discussion of @italic{fullonicae} connected to atrium houses, see @cite[Flohr2011]} This indicates that these commercial endeavors were potentially still linked to household textile production. 

@; NDC: e.g. Caecilius Jucundus

@;footnote{I still have to go back and do a more thorough check for archaeological evidence of fulleries and dyeshops in the case study sites}

@index['("fullery/fullonica")]{Fulling} is the final stage of Roman textile production. The cloth is exposed to cleansing agents (the options could vary but include the ammonia from human urine, soapwort, and fullers' earth), agitated by being tread under food, rinsed, and combed to tease out some of the fibers.@cite-footnote[Wild1970 83] After the fulled fabric dries it shrinks down in size tightening the weave. In essence, the combination of heat and friction effectively felts a layer on top of the cloth. The final step is to finish the cloth by sheering off any excess fibers to create a smooth surface. @index['("fullery")]{Fulling workshops} are identifiable by tubs embedded into the floor with low waterproofed walls separating each tub into stalls.@cite-footnote[Flohr2013 62] Some @index['("fullery")]{@italic{fullonicae}} from Pompeii are equipped with basins that are linked to the town's water supply for rinsing the fabric.@cite-footnote[Flohr2013 62] While this is the final stage in the production process, fulling also served a role in the refurbishing of used fabrics either for use by the existing owner or for preparation for the resale market. Miko Flohr has identified twenty-two @index['("fullery")]{@italic{fullonicae}} from Pompeii, Ostia, Rome, Herculaneum, and Florence.@cite-footnote[Flohr2013Fullo 26] In Timgad, Andrew Wilson has identified twenty-two workshops that were likely @index['("fullery")]{fulleries}, however could also have been used for cold-water dyeing.@cite-footnote[Wilson2004 237]

@;**Evidence of fulleries in Ephesus and Trier (expand)

The @index['("Dye Shop")]{dyeing} of textiles could be done at nearly any stage of textile production, raw wool, spun thread, woven fabrics, or completed fabrics.@footnote{Diocletian's Price Edict lists prices for both unprocessed silk and wool as well as spun silk and wool dyed purple; Diocletian, @italic{Edictum de pretiis rerum venalium} XXIV.1, if this was done at these stages for purple dye, was presumably done at various stages in other colors as well.} Furthermore, used fabrics were often re-dyed to refresh their appearance. @index['("Dye Shop")]{Dyeing shops} can be identified archaeologically by deep lead cauldrons with furnace installations below to heat the chemicals.@cite-footnote[Flohr2013 60] In some of these shops, the cauldrons are arranged in a hierarchy of sizes to accommodate different stages in the dyeing process. Other shops have a less regimented arrangement of cauldrons. 

In Pompeii, two other variations of workshop space are often considered in relation to textile production. Like dyeing and fulling workshops, these two types both involve workbenches with built in furnaces and drainage. In the first type, often identified as @italic{lanifricariae,}@footnote{The term @italic{lanifricariae} was coined by Moeller based off of a reference to a @italic{lanifricarius} in a graffito near one of these shops, it is not an ancient term.@cite[Moeller1976] This identification has been backed up archaeologically based on comparisons of the workbenches to Pliny's descriptions of a method for collecting grease from the wool, @italic{NH} 29.35, @cite[BorgardPuybaret2004]. This identification is contested by @cite[Jongman1988 167] and @cite[Flohr2013 59] due to insufficient and occasionally conflicting evidence} the table top is coated with waterproof plaster and includes a shallow lead basin that spans the width of the surface.@cite-footnote[Flohr2013 57] These shops have been often been interpreted as spaces for cleaning wool.@cite-footnote[Moeller1976 33] The second type has a travertine work surface and incorporates small lead cauldrons over the furnaces. These could have been a variation on the @italic{lanifricariae}, or used for felt making.@cite-footnote[Flohr2013 65] The debate over the identifications between @index['("Dye Shop")]{dye shops}, @index['("fullery/fullonica")]{fulleries}, and these ambiguous variations of buildings with vats and furnaces at Pompeii indicates that while these work-spaces leave far more of an archaeological footprint than spinning or weaving spaces, they are by no means easily recognizable. Outside of Pompeii and Ostia, these structures are fairly rare and much harder to systematically analyze.@footnote{ Wild, for example, mentions the complete lack of dye-works in the northern provinces, whether this is due to regional variations in practices, survival bias, or that sites such as Pompeii and Ostia were in fact production centers is unknown, @cite[Wild1970 81].}

@;**Expand list of scholars
Many scholars lament the absence of textile tools that can be identified for commercial production.@footnote{@cite[Flohr2013 66], @cite[Flohr2016 24]} Unlike @index['("fullery/fullonica")]{fulleries} and @index['("Dye Shop")]{dye-shops}, there are no unique features or equipment that would distinguish a building as a weaving warehouse or a space for spinning. While textile tools such as @index['("spindle/whorl")]{spindles} and @index['("loom weight")]{loom weights} have been found in a myriad of locations, there have been few locations that contain high enough concentration of these tools to suggest a warehouse. In a rare example of a centralized production center, roughly 200 @index['("loom weight")]{loom weights} were discovered in a building at the ancient site near modern Viale Tiziano near ponte Milvio.@cite-footnote[Lipkin2012 43] In contrast, only three sets of loom weights large enough to constitute a functional set were found in potentially commercial contexts at Pompeii and even these only represent a single loom per structure.@footnote{@italic{Tabernae} I 6, 10, VII 16, 19, and IX 2, 5, @cite[Monteix2010 186], @cite[Flohr2013 66]} This apparent dearth of evidence has translated into a relative silence on the early stages of textile production in a commercial context.

@; NDC: also weaving may have been done on the second stories of houses which generally don't survive.

@; NDC: I think there's a lot one can do with negative evidence, if you're careful. What other facilities would a commercial production center need? Does it suggest that such centers are undiscovered, and therefore rare, and therefore large? Or that they are unrecognized, perhaps small and unpretentious? Is it possible that cloth in Pompeii was ctually woven by women/slaves in houses on non-warp-weighted looms and then sold? Just wondering...

@;section{Epigraphic evidence}

@;As discussed in the chapter on ceremonial associations with textile production, funerary inscriptions of job titles provide solid evidence for specific roles within the commercial process. Funerary inscriptions provide a minimal amount of demographic information as well since they often combine names, job titles, and occasionally portraits. As I've already covered this information elsewhere, I will not go into much detail here other than to note that both women's and men's roles appear in commemorative inscriptions though the gender parity is not necessarily conclusive.

Many of the tools used for the early stages of textile production were made out of perishable materials and therefore do not survive. At a typical Roman site, the evidence that we can expect for spinning and weaving consist of @index['("spindle/whorl")]{spindle whorls} or @index['("loom weight")]{loom weights} made out of stone, clay, or other non-perishable materials. However, a review of the tools from Karanis, where the arid weather conditions were more conducive to preservation, show a large number of @index['("spindle/whorl")]{spindle whorls} made out of wood. Even without direct evidence, the abundance of wood available in Italy and Northern Europe in comparison to Egypt suggests that there were likely more @index['("spindle/whorl")]{spindles} in use at other sites than the non-perishable specimens that survive. In fact, it may even be more likely that a commercial workshop for @index['("quasillaria/spinner")]{spinners} would use wooden spindles as they are more cost-efficient than the glass, bone, ivory, or stone whorls and more durable than ceramic whorls that are often found in domestic contexts. As of yet, we have no definitive archaeological evidence of the Roman two-beam loom since it was an entirely wooden construction; however the representations of the loom type from the temple of Minerva in the Forum Transitorium @figures["ForumTransitorium"] and the Hypogeum of the Aurelii @figures["Aurelii"] as well as the low quantities of @index['("loom weight")]{loom weights} suggest that the two-beam loom was the dominant type in use during the Roman Empire.@cite-footnote[Wild1976]

In addition to the perishable nature of these tools, spinning and weaving have fewer restrictions on where they can be performed. Since @index['("spindle/whorl")]{drop spindles} are portable, spinning can be done anywhere. Once a loom is dressed (prepared for weaving and strung with warp threads) it must stay in place until the project is completed and therefore requires a controlled space with sufficient light where it can remain for the duration of use. However, the wooden frame of a loom is easy to dismantle and move when not in use, therefore making it semi-portable. As stated above, the furnaces and workstations for @index['("fullery/fullonica")]{fulling} and @index['("Dye Shop")]{dyeing} require a more permanent location making the use of centralized production centers inevitable. This is particularly pertinent in urban environments where space within residential homes was limited thus having @index['("fullery/fullonica")]{fulleries} and @index['("Dye Shop")]{dye} shops that served multiple households or commercial endeavors was far more efficient.

@section{Epigraphic Evidence}

While the funerary inscriptions honoring domestic textile skills discussed above appeared exclusively on women's graves, epitaphs listing professional textile-related titles include both men and women. Unlike the formulaic epigraphs lauding domestic production of textiles, those with job titles tend to be shorter and include the deceased's name, their job title, and potentially their freed/slave status and their employer or owner's name. Because of this, there is less to parse in the inscriptions themselves and I will present the job titles themselves and include a list of relevant inscription numbers from the @italic{Corpus Inscriptionum Latinarum} for each in the footnotes. It is important to note that these professional titles span both domestic and commercial production. The majority of the funerary inscriptions that list job titles were for slaves or freedmen/women, many of whom served within a @italic{domus} while others worked in centralized production centers. 

The only job title that was exclusively associated with women was that of @index['("quasillaria/spinner")]{@italic{quasillaria}}, or spinner.@footnote{@italic{CIL} 6:9495, 9849a, 9840. @cite[Treggiari1976] 82, @cite[LarssonLoven1998] 75} While this task accounted for a large portion of the labor hours invested textile production, it could be done from almost anywhere and was likely often done from within the home even if the product of spun-wool was then sold to more commercial manufactures. The extent of textile production that occurred within each @italic{domus} varied greatly and would have been dependent on size, status, and number of servants. The monument of the Statilii, a columbarium in Rome, reflects a broad range of textile jobs within the slaves and freedmen of the Statillii family including: eight @index['("quasillariae")]{@italic{quasillariae}}, two @italic{@index['("textor/weaver")]{textores}} and one @italic{@as-index{textrix}} (weavers), a @italic{@as-index{lanipendius}} (textile supervisor), a @italic{@index['("sarcinator/sarcinatrice/sarcinatrix")]{sarcinator}} and three @italic{sarcinatrices} (tailors).@cite-footnote[Dixon2000 12] This level of differentiation suggests either a large self-sufficient household providing for itself, a household that produced textiles for the market from the @italic{domus}, an advanced cottage industry, or a centralized production center operated by the Statilii family.@footnote{@italic{CIL} 6:6339-6346 (MS), @cite[Treggiari1976 82] @cite[Hasegawa2005 31]} The @index['("quasillaria/spinner")]{spinners} attested by title in epigraphs realistically represent only a small portion of slaves and servants who would have spun wool as part of their domestic service. It is probable that most households had fewer servants doing a broader array of task instead of a large enough labor force to have dedicated servants for each specialized task.

The job of weaver, @italic{@index['("textor/weaver")]{textor}/@as-index{textrix,}}@cite-footnote[LarssonLoven1998 75] could be undertaken by a woman or a man either within the home or in centralized weaving houses. Two of the four weavers whose ashes were incorporated in the columbarium of the Statilli were women.@footnote{@italic{CIL} 6360-6362, @cite[Treggiari1976] 82.} The @italic{@as-index{lanipendius}/@as-index{lanipenda}} was the supervisor of textile production within a @italic{domus} who measured out the daily amount of wool to spin.@cite-footnote[LarssonLoven1998 75] Within many households, this was the only textile-related job title commemorated in the epigraphic evidence.@cite-footnote[Treggiari1976 82] Whether this role was that of a man or woman was apparently dependent on the social class of the household. Imperial residences and those of the wealthiest aristocrats employed male @index['("lanipendius")]{@italic{lanipendii}} as a status symbol.@footnote{@italic{CIL} 3976-3977 (ML), 6300 (MS), 37755, 8870, 9495} Female @italic{@index['("lanipenda")]{lanipendae}} were employed by families that were prosperous but less ostentatious.@footnote{@italic{CIL} 9496, 9497, 9498, 34273, 37721} The @italic{@index['("lanipendius" "lanipenda")]{lanipendius/a}} worked under the supervision of the @italic{matrona}, and in more modest households, this role would have been filled by the lady of the house herself.@cite-footnote[Treggiari1976 83]

@;Kleijwegt: Highlighted the word "on" in sentence starting "Two of the four weavers on the monument of the Statilii were women" and put a note saying "Whose ashes were incorporated in the columbarium of the Statilli" ... I'm not sure if he wants me to incorporate that into the sentence, or if he's asking me whose ashes...

@;Kleijwegt, in response to lanipendius/a: Perhaps you should consider incorporating what it is that makes scholars post this claim rather than just the claim; I don’t know what the evidence is that should convince me and therefore I am not convinced ... How does Treggiari establish that they did this as a status symbol?  Is this not something that is up for revision?  After all, Treggiari said this almost fifty years ago

The epigraphic evidence suggests that other roles in the Roman textile market were almost entirely occupied by men. The task of @index['("Dye Shop")]{dyeing} was carried out by the  @italic{@as-index{tinctor}},@footnote{@italic{CIL} 6:9936} @italic{@as-index{infector}},@footnote{@italic{CIL} 4:7812; 5:997; 6:33861} or @italic{@as-index{offector}}@footnote{@italic{CIL} 4:864} respectively depending on whether they were dyeing fleece, new material, or re-purposed cloth. These titles only appear in epigraphs for men.@cite-footnote[LarssonLoven1998 74] We have only a loose idea of the duties of the @italic{@as-index{lanarius}}@footnote{@italic{CIL} 6:9498} and his counterparts the @italic{@as-index{linarius}}@footnote{@italic{CIL} 5:1041, 3217; 6:7468; 11:3209, 6228} and @italic{@as-index{purpurarius}} who were involved in some managerial role in the production or sale of wool, linen, and purple fabrics respectively.@cite-footnote[LarssonLoven1998 74] These managerial roles are almost exclusively associated with men, the only exceptions being two female @italic{@index['("purpuraria")]{purpurariae}},@footnote{@italic{CIL} 6:9846, @italic{CIL} 637820, @cite[Dixon2001]} one female @italic{@index['("linaria")]{lintearia}},@footnote{ @italic{Lintearia} is a variation of the word @italic{linaria}, @italic{CIL 2.4318a}} and one female @italic{@index['("linaria")]{linaria}} who is commemorated alongside a @italic{@as-index{linarius.}}@footnote{@italic{CIL} 5:5923. @cite[LarssonLoven1998 75]}

@; Kleijwegt: lintiarii [the alternative spelling is lintearius; there is a Fulvia from Tarraco who is a lintearia; CIL 2.4318a = CIL 22.14.1284]

After the cloth has been produced, dyed, and distributed, the final stage is clothing production. Tailors can be divided into two groups. @italic{@index['("vestiarius" "vestiaria")]{Vestiarii/vestiariae}} were tailors working commercially out of shops.@cite-footnote[Treggiari1976 84] In this profession, the majority of epigraphs are for male @italic{@index['("vestiarius")]{vestarii}}@footnote{@italic{CIL} 6:2825, 4044, 4196, 4476, 9962-70; 4:3130; 5:324, 3460, 7379-80; 9:1712; 10:3959-60, 3963; 11:868-69, 6839. @cite[LarssonLoven1998 76]} with a few outlying cases of female @italic{@index['("vestiaria")]{vestariae.}}@footnote{@italic{CIL} 6:8557, 9961, @cite[Treggiari1976] 85; @cite[LarssonLoven1998 76]} @italic{@as-index{Vestifices/vestifci/vestificae}} were part of a domestic staff that produced clothing for the household.@footnote{@italic{CIL} 5206, 9980, 9744, @cite[Treggiari1976 84]} This role could be filled by either female or male servants. The related role of @italic{@as-index{sarcinator/sarcinatrice/sarcinatrix,}} who mended existing clothes within a household, could be filled by men or women but was more frequently performed by women.@footnote{@italic{CIL} 6:4028 (ML), 9038, 8903, 3988, 4029 (ML), 5357, 4467, 9039, 4434, 4468, 9037, 6349-6451, @cite[Treggiari1976], 85} Many of these same textile-related professions that we've seen in funerary inscriptions appear in Plautus' @italic{Aulularia} (@italic{@as-index{lanarius}, @index['("linarius")]{lintones}, @index['("fullery/fullonica")]{fullones}, @index['("sarcinator/sarcinatrice/sarcinatrix")]{sarcinatores}, @index['("textor/weaver")]{textores,} @index['("infector")]{infectores}}) as part of a long list of the excessive ways that wives spend money.@footnote{ Plautus, @italic{Aul} 505-522} Since this is a literary text and refering to hypothetical merchants who could be of either gender, all of the job titles are in the masculine form.

From this epigraphic evidence, it is evident that certain jobs within textile production were separated by gender. With the exception of some outliers, this split falls roughly along the division between public and private spheres. Jobs that could be performed in the house could be taken by women or men, those that required the worker to work in specialized production centers or required contact with larger trade or commercial networks were performed by men. Within the @italic{domus,} female servants predominated though male servants could be indicators of social status.@footnote{Though the relatively small number of inscriptions that Treggiari is basing this interpretation could cloud the issue. It is also possible that male and female slaves had similar domestic tasks that they performed in their daily lives but their commemorations were chosen to fit gender stereotypes. @cite[Treggiari1976 83]} It is perhaps also worthwhile to note that not all servants or slaves were commemorated and a large portion of the textile workforce within the household likely had multiple roles to fill rather than a designated position.

@; *Kleijwegt: I am becoming more and more upset by this qualification, which, I know, is not yours.  Is it not feasible that the Romans simply used slaves, male and female, wherever they were required?; I am thinking here of the parallel case of babysitters, who could be male or female

The disproportionate number of men and women commemorated with job titles is likely also tied to the way that commercial efforts within a household were viewed. For example, the funerary inscription of Gaius Cafurnius Antiochus and his wife Veturia Deutera, both freed slaves, provides something of a puzzle.@footnote{ @italic{CIL} 6.9489} Gaius is identified as a @italic{@as-index{lanarius}} but no title is provided for his wife. This absence of evidence, however, does not necessarily mean that she did not work. She clearly would have worked as a slave, but that labor was less likely to be highlighted in her commemoration. As the wife of a @italic{@as-index{lanarius}} is entirely probable that she played some role in his textile business. As freedmen they emulated higher social circles in which women were not employed, therefore they might list his job title alone, leaving her own contribution implied.@cite-footnote[LarssonLoven2013-production "119-120"]

@; NDC: What does this tell you about the process of textile production, though?

@;As discussed in the previous section, women are more likely to be identified by their domestic merits than their professional roles based on Roman gender ideologies.@cite-footnote[LarssonLoven2007 124] Therefore, even within a craft that is so culturally associated with women as textile production, professional titles appear in higher quantities of epigraphs for men than women. In the case of a freedwoman who worked as a spinner, it is more For many domestic servants, tasks like spinning and weaving were likely part of a larger set of duties and therefore their 


Political graffiti often reference textile-related professional associations or job titles. This source has been most thoroughly explored in Pompeii, where the proximity of these graffiti to suitable structures has been applied in identifying the functions of several workshops. In the cases of the @index['("fullery/fullonica")]{fulleries} and @index['("Dye Shop")]{dye} workshops discussed above, this method has helped identify different types of workshops. Epigraphic evidence from Pompeii suggests some level of specialization between shops that specialized in dyeing raw material, @italic{@index['("infector")]{infectores}}, and those that focused on secondary dyeing of used textiles, @italic{@index['("offector")]{offectores}}. This identification is based on an electoral slogan for @italic{@index['("infector")]{infectores}}, IX 7, 2 associated with a workshop with varying sizes of cauldrons for presumably separate stages of @index['("Dye Shop")]{dyeing}; and another for @italic{@index['("offector")]{offectores}} with no such distinctions between the sizes of their cauldrons. @cite-footnote[Flohr2013]  

@; Kleijwegt: Scholars who have commented on this [infectores/offectores say that such a distinction was highly exceptional.  I don’t think I agree with this, nor should you, but it needs to be mentioned.

This method, however, can be rather imprecise. The building of Eumachia in Pompeii poses an interesting case-study. An inscription at the entrance identifies that Eumachia, a public priestess, funded the building and dedicated it to Concordia Augusat and Pietas,@footnote{@italic{CIL} X 810} two personifications of the Empress Livia associated with the Imperial cult.@cite-footnote[LydingWill1979 38] While the location, off of the Forum of Pompeii, and its size, roughly sixty-seven by fourty meters, suggest that the building had a public function, the structure itself does not indicate the building's function.@cite-footnote[DAmbra2012 401] Some connection to the textile industry is suggested by a portrait statue of Eumachia herself dedicated by the @index['("fullery/fullonica")]{fullers}.@footnote{@italic{CIL} X 812} This inscription, and the presence of cisterns, vats, and basins which were recorded in early descriptions of the building which are no longer present, led to the identification of the building as a @index['("fullery/fullonica")]{fullery}.@cite-footnote[Moeller1972 323] Due to the buildings incongruity with other @index['("fullery/fullonica")]{fulleries}, Mau discounted these vats and basins as belonging to the repairmen restoring the building after the 62 C.E. earthquake and instead proposed that the colonnade of the building served as a textile market.@cite-footnote[Mau1892 119] Given the lack of archaeological evidence, it could also have served as a weaving warehouse.@cite-footnote[Jongman1988 178] Moeller suggests instead that the building served as a meeting place for members of the wool trade to conduct business, including but not limited to the @index['("fullery/fullonica")]{fullers}.@cite-footnote[Moeller1972 325]

Other graffiti can be informative without giving us a clear indication of the purpose of the building. A list of ten women's names appear in a graffito from the house of Eudoxus in Pompeii, accompanied by an account of the quantities of wool to be spun for warp threads, @italic{stamen,} and two types of weft threads, @italic{trama} and @italic{subtemen}.@cite-footnote[Moeller1969 566] In the same portico are the names of seven male @index['("textor/weaver")]{weavers.} Such examples blur the lines between what would be considered commercial production vs. domestic. It is possible, as della Corte suggests, that this structure was formally adapted into a textile workshop. It is likewise possible that the names inscribed on the portico were those of slaves and servants who worked within the structure the @italic{domus} but had a focus on textile production. With seventeen textile craftsmen attested in the same space, does the distinction between the two even remain relevant?

@; NDC: RE: house of Eudoxius: are these quantities large? This sounds really interesting. what more can you say? Are the quantities domestic or commercial?

Other inscriptions referencing textile workers are entirely separated from the context of a production center. In the instance of the @index['("textor/weaver")]{linen-weavers} guild from Ephesus, for example, the inscription marks a preferential seating area for members of the professional association. It informs us that such a guild existed in Ephesus and that membership in the guild had a certain social status associated with it, but it tells us little about the production methods or locations. Another epigraph from Ephesus links @index['("fullery/fullonica")]{fullers} to the goddess Diana.@footnote{@italic{SEG} 34,1124; @cite[Kleijwegt2002 106]}



@;TODO: Expand this section


@section{Administrative Evidence for Commercial Production}

Since the commercial textiles market had an impact on the public sphere, we end up with a lot more administrative evidence for this category of textile work than we've seen in domestic or ceremonial contexts. This loosely defined category of evidence includes the bureaucratic records of the empire including imperial decrees, 

The @italic{Notitia Dignitatum} is an administrative document which lists civic appointments that oversee specific functions in the provinces. Though the date of the manuscript itself is unknown, the information documents various administrative roles dating from the Tetrarchy through ca. 430 CE.@cite-footnote[Sinnigen1963 806] The text identifies two textile-related administrative positions for Trier, the @italic{Procurator gynaecii Triberorum, Belgicae primae}@footnote{@italic{Notitia Dignitatum Oc} XI 58} and the @italic{Praepositus barbaricariorum siue argentariorum Triberorum.}@footnote{@italic{Notitia Dignitatum Oc} XI 77} As discussed above, the precise job description for both of these titles is unknown. The @italic{gynaecaeum} was likely a weaving house of some sort.@cite-footnote[Wild1976-Gynaecea] The @italic{barbaricarii} were textile workers that specialized in luxury textiles for ceremonial purposes that incorporated silver and gold into their production.@cite-footnote[Sinnigen1963 807]
@; Are there examples of barbaricarii other than Trier?
@; Find more examples of textile related jobs in the Notitia Dignitatum. 


@; NDC: so these aren't typical weaving facilities but have to be more tightly regulated. Are the gyn. necessarily commercial in the sense of public?



The climate of Egypt preserves evidence in the form of informal letters or formal contracts which can give us more detailed information about the bureaucratic end of the textile industry. An apprenticeship contracts from Karanis reflects the education of a domestic servant of Aurelius Ision by a professional @index['("textrix")]{weaver}.@footnote{P. Mich. Inv. 5191} The terms of this contract indicate that the slave girl and all of her maintenance costs -- such as food and clothing -- are transferred from her master to her teacher for the period of the contract, after which she will presumably return and apply her new skills either within her master's home or toward her master's commercial interests. As such, this serves as an example of both commercial and domestic production. In a similar contract from Oxyrhynchus a father, Pausiris, son of Ammionios, apprentices his son, also named Pausiris, to a master @index['("textor/weaver")]{weaver}, Epinikos, son of Theon.@footnote{P. Mich. 81} This contract reflects a young man's education for his future career. Papyrilogical evidence also documents labor contracts. In one such contract from Karanis, the laborer, Aurelia Taesis, daughter of Asklepiades and Sarapous, indentures herself to Aurelia Thaisarion, daughter of Komon to work off her father's debt of eighteen thousand silver drachmai.@footnote{P. Mich. Inv. 2819}

@; NDC: but do any of these specify where the person will work?

Wax-tablets from the House of Caecilius Iucundus in Pompeii archive the transactions of an active @index['("fullery/fullonica")]{fullery} owned by the city.@cite-footnote[Flohr2013Fullo 18] In regards to @index['("Dye Shop")]{dyeing}, Strabo discusses how the hot springs in Hierapolis were ideal for dyeing wool, affording textiles dyed with roots a similar quality to those died with higher end materials such as crocus.@footnote{ Strabo, @italic{Geogr.} 13.4.14}

@;Evidence of the purchase and exchange of cloth goods likewise survives. A receipt of garments outlines a variety of clothing items tunics, cloths, shirts, and an embroidered panel.@footnote{P. Mich. Inv. 1050} Letters of correspondence can also give a less formal indication of the textile market. A letter from Karanis requests white cottons specifically from Rome.@footnote{P. Mich. Inv. 5638, item 1} Another letter from Karanis reflects an exchange of textile goods in which the a young man acknowledges receipt of a short cloak from his father, Claudius Terentianus, and indicates that in return he sent a bag containing two mantles, two capes, two linen towels, two sacs, and a linen covering.@footnote{P. Mich. Inv. 5390} In another letter from the same private collection, Claudius Terentianus writes to his father Claudius Tiberianus to acknowledge receipt of a cloak, tunic, and girdled clothes, but requests clothing more suitable to his military service including a beaver skin cloak, a girdled tunic, and trousers.@footnote{P. Mich. Inv. 5391}
@; NDC: Where is cotton grown. Relevence of these passages to the question at hand?

@;Legal documents on a more mundane scale than the lofty imperial edicts could also document textiles. Marriage contracts also document specific textile products in the form of the woman's dowry. Examples from Karanis vary from vague references to clothing@footnote{P. Mich. Inv. 4703} to more specific examples with types and values of clothing. One such contract from Karanis includes a white tunic valued at twenty drachmai in addition to forty silver drachmai.@footnote{P. Mich. Inv. 6551} The dowry for a far wealthier woman from Philadelphia included two tunics, two light mantles, a Syrian cloak valued at 430 Augustan drachmai, and a striped garment in addition to larger assets in the form of jewlery, real estate, and a slave.@footnote{P. Mich. Inv. 508} In a loan contract from Antinoopolis, Flavius Christodoros pledges "a garment made in Egypt after the Tarsic fashion and an undergarment-shirt" as collateral towards his loan from Aurelia Euphemia.@footnote{P. Mich. Inv. 478} 

@subsection{Diocletian's Price Edict} 

@;- Passages from books and poems

Diocletian's Price Edict of 301 CE established the maximum prices for consumer goods, raw materials, labor and shipping costs for a broad range of industries in response to inflation.@cite-footnote[Kropff2016]} Since the Edict is a list of products and prices, each line item includes a brief explanation of the product or labor in question, the unit of measurement it is evaluated by, and the price in @italic{denarii communes}. Due to this format, it is limited in what it can tell us but is useful in establishing the values and roles of textiles and textile manufacturing more broadly in the commercial economy.

In framing this information, it is important to remember that this is by no means an exhaustive list of products and services and due to its function in capping prices, many of the items included are extreme luxuries. Out of the over 1,200 entries in the price edict a single pound of "purple dyed silk" ties with a male lion for the highest priced item at 150,000 denarii .@footnote{@cite[DiocletianEdict] XXIV 1 for Purple dyed silk, XXXIV 1 for the lion} The high cost of this luxury textile is caused by the high cost of purple dye and the fact that silk was imported from China.@cite-footnote[Wild1976 169] Textiles manage to range from some of the lowest price points listed to the highest, partially because they range the whole gamut from the low-grade wool (XXV.1 5 at Den 25) and fabrics used for the clothing of slaves up to the sumptuous purple silks for the imperial court (XXIV.1 1 at Den 150,000).

@bold{Table 7.1:} Textile tools from Diocletian's Price Edict
@tabular[(list (list @bold{Tool} @bold{Material} @bold{Price range in denarius} @bold{Chapter})
               (list "Shuttle" "Boxwood" "14" "XXIII.1 1")
               (list "Shuttle" "Wood" "30" "XXIII.1 2")
               (list "Weavers comb" "Boxwood" "12" "XXIII.1 3")
               (list "Weavers comb" "Wood" "14" "XXIII.1 4")
               (list "Spindle" "Boxwood" "12" "XXIII.1 5")
                (list "Spindle" "Wood" "15" "XXIII.1 6")
               (list "Sewing needle, very fine" "Ivory" "4" "XVI.1")
               (list "Sewing Needle, second quality" "Ivory" "2" "XVI.1"))]

@; NDC: significance of this?

The tools in the edict lists are categorized by material (Table 7.1). The wooden tools give a lower-end price for tools made of boxwood, and a higher-end price for other wood tools. The wooden textile implements include a weaving shuttle (XXIII.1 1-2 at Den 14 for boxwood, 30 for wood), a weavers comb (XXIII.1 3-4 at Den 12 for boxwood, 14 for wood), and a @index['("spindle/whorl")]{spindle with a wooden whorl} (XXIII.1 5-6 at Den 12 for boxwood, 15 for wood). In the ivory section of the edict there are two sewing needles, one of very fine quality, the other of second quality (XVI.1 Den 4-2). It is important again to note that this is a list of maximum prices, for example we know that lower-cost needles exist such as the bronze and bone needles from Trier and Ephesus @figures["TrierNeedles" "ThurB109" "ThurB49"].

@bold{Table 7.2:} Wages for textile workers from Diocletian's Price Edict
@tabular[(list (list @bold{Descriptor} @bold{Unit} @bold{Price in denarius} @bold{Chapter})
               (list "Silk worker" "Per day" "25-40" "XX.1 9-11")
               (list "Woman Weaver of tunicas" "Per day" "12-16" "XX.1 12-13")
               (list "Wool Weaver" "Per 1lb" "15-40" "XXI.1 1-4")
               (list "Linen Weaver" "Per day" "20-40" "XXI.1 5-6")
               (list "Fuller" "Per Garment" "20-600" "XXI.1 1-26"))]
@; separate the high/low wages

@; NDC: much more significant: who's paying these people a daily wage, how does that work? what is the arangement here? Many specify 'with maintenance"

@; NDC: is gender specified in Latin? is it clear that these are all masculine or is the word conventional?

The cost of labor for textile production varies depending on the material, quality of work, and demographic (Table 7.2). @index['("textor/weaver")]{Wool weavers} are compensated based on the quality of material between coarse or third quality wool (XXI.1 4 Den 15), second quality wool (XXI.1 3), and those working with wool from Tarentum, Laodiceia, or Altinum (XXI.1 2 Den 30). Weavers working with 'sea wool', or thread spun from byssus, earned higher for the delicate material (XXI.1 1 Den 40).@footnote{This extremely fine fiber was collected from secretions from @italic{Pinna nobilis} mollusks @cite[Laufer1915 104]} Compensation for linen was based on the quality of the work rather than the materials with second quality work (XXI.1 6 Den 20) and first quality work (XXI.1 5 Den 40). Compensation for silk weaving was based on purity and complexity with the categories of part silk (XX.1 9 Den 25), pure silk (XX.1 10 Den 25), and pure silk checkered (XX.1 11). Weaving in wool and byssus was compensated per pound whereas weaving in linen and silk is measured per day of work.

@;@footnote{Check if this is actually pounds, or if that was a more accessible translation for this publication @cite[DiocletianEdict]}

@; Who is paying? Is this really evidence for the kind of modern system we tend to imagine? Other possibilities? Don't fall into the modernist trap.

The most relevant wage distinction for the purposes of this study is the difference in cost between the @index['("textrix")]{woman weaver} of tunicas of soft cloth (XX.1 12-13 Den 12-16) from the other @index['("textor/weaver")]{weavers}. While not explicitly stated as silk in the description, these titles listed in the chapter for silk workers. That means that a @index['("textrix")]{female weaver} working with silk or soft cloth would make roughly the same amount as a @index['("textor/weaver")]{male weaver} working in the lowest quality of coarse wool. Since the edict sets maximum costs by quality, this is the best indicator for wage inequality between male and female textile workers. The edict does not include a price for spinning in general, but does give a price for those spinning purple silk (XXXIV.1 14-16 Den 60-116/unica) and purple wool (XXXIV.1 17 Den 24/unica).@footnote{ An unica is a Roman ounce and is 27.28 grams, see: @cite[Lauffer1971 54].} Since the edict is giving maximum costs and the wages for labor increase with the cost of the materials in other instances, it is fair to infer that the cost of labor for spinning the various qualities of wool would have been significantly less.


@;bold{Table 7.3:} Fulling costs of "new hooded cloaks" by region of production from Diocletian's Price Edict
@;tabular[(list (list @bold{Region} @bold{Price in denarius} @bold{Chapter})
@;               (list "Nervii" "600" "XXII.1 21")
@;               (list "Laodiceia" "175" "XXII.1 22")
@;               (list "Ripensi" "300" "XXII.1 23")
@;               (list "Noricum" "200" "XXII.1 24")
@;               (list "Other" "100" "XXII.1 25"))]

@;The cost of fulling varies quite wildly because it is determined per garment depending on size, quality of materials, and age of the garment. On the low end, fulling an undecorated coarse wool shirt would cost Den 20 (XXII.1 3) whereas fulling a new hooded cloak of wool from the Nervii would cost upwards of Den 600 (XXII.1 21). The list indicates that the regional production center where a garment was made has a bearing on the price of fulling the garment as well (Table 7.3). If we look at "new hooded cloaks" from Nervii (XXII.1 21 Den 600), Laodiceia (XXII.1 22), Ripensi (XXII.1 23), and Noricum (XXII.1 24), we see a wide range in the fulling costs, presumably due to the quality of material and craftsmanship for each of those production centers.

@bold{Table 7.4:} Raw fibers and materials from Diocletian's Price Edict
@tabular[(list (list @bold{Fiber} @bold{Unit} @bold{Price in denarius} @bold{Chapter})
               (list "White unprocessed silk" "Per 1lb" "12,000" "XXIII.1 1-2")
               (list "Unprocessed silk, dyed purple" "Per 1lb" "150,000" "XXIV.1")
               (list "Wool dyed purple" "Per 1lb" "300-50,000" "XXIV.1 2-12")
               (list "For Spinning purple silk" "Per unica" "60-116" "XXIV.1 14-15")
               (list "For Spinning purple wool" "Per unica" "24" "XXIV.1 16")
               (list "Wool" "Per 1lb" "25-175" "XXV.1 1-5")
               (list "Sea wool/byssus" "Per 1lb" "150" "XXV.1 6")
               (list "Rabbit's hair, unsorted" "Per 1lb" "100" "XXV.1 7")
               (list "Combed, unspun flax" "Per 1lb" "16-24" "XXVI.1 1-3")
               (list "Spun linen yarn" "Per 1lb" "72-1,200" "XXVI.1 4-12")
               (list "Linen unit of fabric" "Per 1 web" "200-11,000" "XXVI-XXVIII")
               (list "Unit of purple fabric" "Per 1 web" "2,500-36,000" "XXIX 30-48")
               (list "Spun gold" "Per 1lb" "72,000" "XXX.1 2"))]

@; NDC: This is interesting but your dissertation is not about all aspects of textiles; keep the focus@

As far as textiles themselves go, the Price Edict reflects every stage from raw materials through woven fabric as well as completed garments (Table 7.4). Unspun fibers are listed for silk (XXIII.1 1-2 Den 12,000/lb), wool (XXV.1 1-5 Den 25-175/lb), byssus (XXV.1 6 Den 150/lb), rabbit's hair (XXV.1 7 Den 100/lb), and flax (XXVI.1 1-3 Den 16-24/lb). The only spun yarns listed are linen (XXVI.1 4-12 Den 72-1,200/lb) and gold (XXX.1 2 Den 72,000/lb). These raw and spun fibers are sold in one pound units.@footnote{A Roman pound is 327.45 grams, Lauffer 1971, 54-55} No dyes themselves were listed in the price edict, but book XXIV.1 contains purple products including the aforementioned purple silk (XXIV.1 1 Den 150,000/lb), wool dyed in various shades and origins of purple (XXIV.1 2-12 Den 300-50,000/lb)

@;Roman clothing was woven to the size and shape specifications for the specific garment rather than woven as larger rectangles and cut to a pattern. In fitting with this, the linen fabric listed in the price edict were measured per web. This unit of measurement refers to the fact that Roman garments were woven in one piece in a standard size for a standard use. The prices for fabric, therefore, ranged quite drastically based on the size, function, and quality of the intended final product (XXVI-XXVIII Den 200-11,000). A woman's @italic{dalmatica,} a variant on a tunic, of the highest quality was capped at Den 11,000 per web (XXVI.1 34); the men's @italic{dalmatica} of the same quality was priced at Den 10,000 per web (XXVI.1 39). The same article of clothing of the third quality for common people or slaves was merely Den 600 for a woman's and Den 500 for a man's @italic{dalmatica}. A face cloth, which by nature is a much smaller piece of cloth, ranged from Den 3,250 per web for the highest quality without colored bands (XXVI.1 99), eighteen tiers of quality lower on the list is a face cloth of the third quality of coarser linen for common people or slaves at Den 200 per web (XXVI.1 119). 

Independently, each of these lists only gives us a constrained set of information, but combined they can offer greater insights into the Roman textile economy. If we cross-reference between the cost of tools, cost of labor, cost of materials, and cost of final products it is clear that materials make up the primary portion of the real cost.@cite-footnote[Jongman2000 191] Though spinning gets the least representation of the textile jobs represented in the edict, it is perhaps the best example for a full review because a single @index['("spindle/whorl")]{spindle} is the only tool necessary to complete the task. The initial investment in tools for spinning is reasonably low at 12 denarii for a boxwood spindle and 15 for a higher quality wood. The cost of a pound of unprocessed white silk is 12,000 denarii. A pound of that same silk dyed purple is 150,000 denarii (XXIII.1, XXIV.1 1).@footnote{inferring a cost for the dye and labor of @index['("Dye Shop")]{dyeing} that silk to at 138,000 denarii} The wage for spinning pure purple silk is 116 denarii per unica, which converts to 1/12 lb. The wage for a silk weaver using pure silk is 25 denarii per day.@footnote{**calculate amount of time it takes to spin/weave.} Sadly, all of the woven fabric prices that survive on the edict are for linen products, so that is as far as we can take the analysis on a single object.

@; NDC: Cost of purple dye? How much do you need to dye a pound of silk? If you're getting into the economics of this, do it fully. But perhaps purple silk isn't the material to start with. Nobody is going to buy a spindle and start spinning silk ... 

The breakdown of this analysis is that the barrier of entry for the tools and skillset to carry out this work would be relatively negligible, a solid @index['("spindle/whorl")]{spindle} would cost roughly 13% of the wages from spinning a single unica of purple silk. However, it would take over 10 years for a single laborer to earn enough to purchase a single pound of silk. This suggests that while labor was inexpensive, the control of the textile market, at least at the higher qualities of materials, was in the hands of the wealthy who could supply materials to the laborers.

I chose silk for this breakdown because it is the only material that has the cost for the material dyed and undyed, and has the variable wages for spinning. However, this is also a good example for why the items listed in Diocletian's Price Edict are misleading as well. As a list of maximum prices, it sets a cap for the highest prices for already luxury items. The silk itself would have been imported from China, raising the cost of materials because it was imported and also the scarcity of the material would lead to fewer @index['("quasillaria/spinner")]{spinners} with the knowledge of how to spin it properly.@cite-footnote[Wild1976 169] There is no way to get to the initial investment cost for standard, locally-sourced materials or the cost of labor for ordinary laborers from a list of @italic{maxima.} Furthermore, only one category of @index['("textrix")]{weaver} is gendered as a woman, and it reflects the lowest wage of all of the @index['("textor/weaver")]{weavers}. This suggests that even if the other maximum wages are infered as male by default, there was likewise a lower standard wage for women in similar positions.

@; NDC: This is a big 'at least' - how much of the textile market did silk make up? Not much, I imagine ... if you're interested in ordinary laborers, don't look at silk.

@;footnote{have Chris help me with calculations after I time spinning}

@;bold{Table 7.5:} "Cloak with clasp" by region from Diocletian's Price Edict
@;tabular[(list (list @bold{Region} @bold{Price in denarius} @bold{Chapter})
@;               (list "Rhaetia" "12,000" "XIX.1 65")
@;               (list "Treveri" "8,000" "XIX.1 66")
@;               (list "Petovionicum" "5,000" "XIX.1 67")
@;               (list "Africa" "2,000" "XIX.1 68"))]

@;Diocletian's Price Edict can also give us insight into the larger trade network of the textile market. The higher qualities of wool, as indicated by price, originate in Tarentum, Laodiceia, and Altinum (XXV.1 1-3), suggesting that these locations had premium pastures for raising sheep. Completed textile goods were also ranked using the locations of their production centers. If we look at a single category of clothing, a cloak with clasp (Table 7.5), we can see a variation of prices with cloaks from Rhaetia (XIX.1 65 Den 12,000), Treveri (XIX.1 66 Den 8,000) -- of which region Trier was the center -- Petovionicum (XIX.1 67 Den 5,000), and Africa (XIX.1 65 Den 2,000). The broader list of textile types in the price edict therefore gives a more complete list of textile production centers. This form of labeling garments based on their production center is likewise attested in the receipt for garments discussed above which lists 5 cloaks in the Antiochian style and an Antioch hood.@footnote{P. Mich. Inv. 1050}

@; NDC: Conclusion of all this? What have you learned about your fundamental topic?

@; While not directly framed in the context of the textile market, Diocletian's Price Edict lays out maximum prices for sea freight. These shipping networks would have been crucial to the large-scale textile trade market in the Roman Empire. Using our case study site of Ephesus, for example, the edict informs us that there were established freight routes into the port of to Ephesus from Alexandria (XXXVII 8 Den 8 per 1k. mo.), Syria (XXXVII 20 Den 10 per 1k. mo.), and Nicomedia (XXXVII 39 Den 6 per 1k. mo.). Established routes from Ephesus shipped to Rome (XXXVII 22 Den 16 per 1k. mo.), Africa (Carthago) (XXXVII 23 Den 8 per 1k. mo.), and Dalmatia (Salona) (XXXVII 24 Den 12 per 1k. mo.). @cite-footnote[DiocletianEdict]

@section{Iconographic evidence}

A will be discussed further in the performative associations with textile production chapter, textile-related imagery on women's graves tended to be more domestic and took the form of wool baskets, @index['("spindle/whorl")]{spindles}, or @index['("distaff")]{distaffs} presented as an attribute of femininity. Textile related imagery on men's graves was more professional.@cite-footnote[LarssonLoven2013-production 118] This distinction was also sometimes difficult to make or subjective to the interpretation of scholars. A pair of wool shears depicted on a man's grave is interpreted as a symbol of his professional career,@footnote{Find exact example in @cite[LarssonLoven2002]} whereas a @index['("spindle/whorl")]{spindle} depicted on a woman's grave is interpreted as an attribute of femininity.@footnote{Find exact example in @cite[LarssonLoven2002]} Both examples represent the tool on its own and not in use in the margins of the stele, with a portrait of the deceased at the center. Both of these tools could be used either within a domestic setting or a commercial setting. 
@; NDC: good -- how do you get beyond this presumption of masculine-professional/feminine-symbolic? esp. when there's a cultural bias in antiquity against women working outside the home? How do you get at reality?

The funerary plaque of Gaius Cafurnius Antiochus and his wife Veturia Deutera discussed above has a representation of a sheep with a pair of conjoined hands above it @figures["GaiusCafurniusAntiochus"].@cite-footnote[LarssonLoven2013-production "119-120"] The sheep is a representation of his job as a @italic{@as-index{lanarius}} while the joined hands, the @italic{dextrarum iunctio} is a common symbol used on the graves of freedmen and women to represent their marital status. In this case, the combination of the job title in the epitaph and the image of the sheep indicates a link directly with the larger textile market rather than an agricultural association. 
@; NDC: In general; try to come to conclusions from these examples. Too often we have a description and maybe some obvious observations but you aren't pushing the evidence, or showing how it helps answer your questions. What have you learned from this relief?

@; Kleijwegt: Seneca reports that in the earthquake of 62 which caused great damage to Pompeii 600 sheep perished; this is in book 6 of his Quaestiones Naturales **Where do I put this?

A relief from Rome represents two men displaying a length of cloth at the right of the panel to four men at the left @figures["vestiarius"].@cite-footnote[LarssonLoven1998 77] Unlike Gaius Cafurnius Antiochus, this relief is not preserved in association with an inscription, so we do not have epigraphic evidence to identify the profession of the deceased, though it likely represents a @italic{@as-index{vestiarius}} or tailor. The scene is rather similar to the @italic{Tuchladen}/Salesroom scene from the Igel column @figures["Igel5" "Zahn2"].

The Igel monument from Trier, of course, has the most complete representation of the commercial stages of the textile industry ranging from baling the cloth, to quality control, through shipping the product down the river and concluding with the salesroom @figures["Igel2" "Igel1" "Igel4" "Zahn1" "Igel5" "Zahn2" "Igel6" "Zahn3" "Igel7" "Zahn4" "Igel 8" "Zahn5" "Igel9" "Zahn7" "Igel10" "Zahn8" "Zahn6" "Igel11"]. Like the archaeological evidence, this monument is glaring in its omission of the production stages of the textile industry. The monument's focus is primarily on the managerial aspects of the industry and oversight than the production. In that same spirit, the omission of women on the monument is likewise telling. 

@; NDC:Telling of what, exactly? That women weren't involved in textile production, shipping, etc? That they were involved but the Romans didn't want to admit it? the evidence could be argued either way; how will you argue for one reading or the other?

@; A couple of frescoes from Pompeii

@; NDC: And even here, couldn't women have been weaving in Fullonicae at Pompeii, in the many un-identified spaces?

@;ToDo: IX.7.6 Pompeii. House of Verecundus, fresco of feltmakers (Inscription above the image: Vettium Firmum aed(ilem) quactiliari rog(ant)      [CIL IV 7838])

@;ToDo: Fuller's tombstone Sens, Yonne treading cloth and cropping nap (Wild1970 179)


@section{The solution to the mystery: women}

Sometimes the hole that delineates the absence of evidence is itself the shape of the evidence. Economic historians may decry the absence of evidence for spinning and weaving for textile production.@footnote{ @cite[Jones1960] @cite[Andreau2012 41]} However, this is not truly the case: archaeological evidence for these early stages of textile production exist in abundance in domestic contexts, just not in the centralized production centers that would fit within the narrative of commercial production. The most logical conclusion is that women still played an active role in textile production from within their homes. Jongman briefly proposes this possibility in relation to the lack of evidence for spinning and weaving in commercial contexts.@cite-footnote[Jongman2000 194] Flohr likewise admits the possibility of individual @index['("textor/weaver")]{weavers} working from home for a commercial market.@cite-footnote[Flohr2013] Wild directly states that spinning, at least, was a cottage industry that monetized chores of the housewife.@cite-footnote[Wild1976 169] Roth makes a solid argument for the economic potential of textile production by female slaves as part of a larger rural villa economy.@cite-footnote[Roth2007-Spinning 86]

With the notable exception of Roth, who makes the claim that women's labor in a Roman villa deserves the same economic consideration as men's labor,@cite-footnote[Roth2007-Spinning 85] these interpretations suggest that women's domestic labor had minimal impact on either the household income or the textile market. In part, this is related to the difficulty in assessing the output of a loom and a @index['("spindle/whorl")]{spindle} within a household. We cannot assume that all households which contained a loom participated in the textile market with any more assurance than we can assume that all households with a single loom only wove to produce clothing and linens for household consumption. Even within households that did produce textiles with the intent to sell them, the output could vary seasonally or based on other domestic duties such as childcare.@footnote{ The extent to which textile production could be managed alongside childcare has been a source of debate. In a first-person account, a textile worker and her husband from Kyoto recall that she worked in a factory until the birth of their first child, at which point she switched to weaving from home in a putting out system, see: @cite[Hareven2003 54]. In eighteenth century France, } The economic structure of a cottage industry (or 'putting out' system) would allow for women and domestic servants to continue to spin and weave within their households then sell their work by weight, length, piece, or other unit of measurement. This structure allows for a greater flexibility in production modes and leaves room for variable levels of output. As discussed above, these systems have taken on various types, forms, and scales throughout history and it is likely that several types occurred concurrently in the Roman Empire.

It is highly likely that wool was sheered, processed, spun, and woven, as part of the overall agricultural model of some rural Roman sites, such as the villa of San Rocco, which has archaeological evidence for both spinning and weaving.@cite-footnote[Roth2007-Spinning 77] However, many of the domestic sites that contained textile tools across the Roman Empire, particularly in urban environments such as Ephesus, Trier, or Pompeii, clearly did not have the space to raise their own sheep. These households potentially purchased unspun fiber which was imported from the hinterland.@footnote{ This practice was used in eighteenth century France, where the poorest laborers even purchased the wool on credit which was paid off once they sold their fabric. @cite[FauveChamoux2001 169]} Both of these options reflect a cottage industry model where the laborers, or at least the household to which they belonged, independently produced the thread or fabric then sold a finished product either directly at a local market or to a @italic{@as-index{lanarius}} who would introduce it into a larger trade network.  

In an alternate model, closer to a putting-out system, a @italic{@as-index{lanarius}} could control the production from distributing raw materials to @index['("quasillaria/spinner")]{spinners}, purchasing back the spun thread, and redistributing it to @index['("textor/weaver")]{weavers}, who then sell back a finished product.@footnote{ Similar to the @italic{Chinbata} system of late nineteenth and early twentieth-century Japan the manufacturer provided the materials and sometimes even leased looms to laborers so they could weave from their homes. @cite[Hareven2003 55] This putting-out system grew in popularity over the previous system where @index['("textor/weaver")]{weavers} lived and worked in the manufacturer's home because it saved the manufacturer the cost of maintenance of both the machinery and the laborers.} It is possible that the counting of coins in the @italic{tuchladen} and  @italic{kontor} scenes from the Igel Monument represent the @italic{@as-index{lanarius}} or his overseers paying for and inspecting textiles that were produced within the laborer's home @figures["Igel5" "Zahn2" "Zahn6"].@cite-footnote[Drinkwater1982 120] In this scenario, the @italic{@as-index{lanarius}} could be both manufacturer and merchant. If he has control of the process from raw materials to distribution it also allows for the potential of occasional centralized production centers where multiple stages of production occur.

Various forms of cottage industries or 'putting out' systems played a large role in European textile markets right up until the industrial revolution.@footnote{ See examples from England: @cite[Pinchbeck2004], France: @cite[Hafter1985], America: @cite[Ely1999]}  With such a convenient answer at hand, how has it been largely relegated to footnotes or stray conjecture in scholarship on the Roman textile market?

If we consider the existing evidence within the patriarchal structure of society in Rome, we can see the other side of this 'missing evidence.' Women's roles in public life were limited and the stages of textile production that are considered as evidence of 'commercial' activity, such as @index['("fullery/fullonica")]{fulling} and @index['("Dye Shop")]{dyeing}, were largely carried out in public production centers. Political graffiti referencing textile workers and dignitary positions in the @italic{Notitia Dignitatum} are obviously skewed toward male roles because men were eligible to serve in office whereas women were not. Literary references to the textile market focused on either luxury items or large-scale trade, which was carried out by men. Even funerary inscriptions are skewed toward men in cases where a husband and wife are commemorated together and only the husband's profession is mentioned.

@; NDC: was this gendered in reality or normatively?
Finally, we need to consider how extremely gendered the Roman conception of spinning was. As discussed at length in the chapters on domestic production of textiles and will be expanded in the chapter on performative aspects of textile production, spinning had a long history as the quintessential woman's work, the epicenter of the woman's contribution to domestic economy. This idea was carried to the extent that the mere notion of a man spinning was a matter of ridicule.@cite-footnote[LarssonLoven1997 92] In depictions of Hercules dressed as Omphale, @index['("spindle/whorl")]{spindles} and @index['("distaff")]{distaffs} were used as gender-markers for the hero dressed as a woman while Omphale bears his standard attributes of the lion's skin and club @figures["HerculesOmphale"]. In one late Republican @italic{terra sigillata} cup, for example, Antonius is lampooned as Hercules dressed as Omphale followed by a procession of maids carrying a @index['("distaff")]{distaff} and wool basket among other feminine objects.@footnote{I unfortunately do not have an image of this cup to share, but it is cited in @cite[PasztokaiSzeoke2011 128]} Cassius Dio informs us that the emperor Elagabalus enjoyed spinning wool as evidence of his failure to conform to standard Roman gender roles.@footnote{Cassius Dio, LXXX,14,4 and LXXX, 16, 7)@cite[PasztokaiSzeoke2011 128]} Therefore, any men that did serve this role would not be likely to commemorate that work. It is, however, impossible to identify the extent to which these extreme examples were grounded in reality or if they were just normative.@footnote{For example, it is entirely probable that male domestic servants and slaves would have participated in the traditionally feminine gendered tasks of textile production if their master's required it. The obvious lampoons listed here may be closer to modern mockery of male nurses.}

The roles that women played in the textile industry, spinning and weaving, were the foundation of the entire industry, but were also the least remarkable work. One rarely looks at a piece of cloth and comments on the quality of the individual threads or looks at an article of clothing and discusses the fabric without noting the cut or design. Furthermore, one of the key benefits of cottage industries throughout history was that it utilized the labor of people who could not be working outside of the home including women, children, and the elderly.@footnote{ See examples from 18th century France: @cite[Gullickson1981 183], post-colonial India: @cite[Kumarappa1944 109], and 20th century Japan: @cite[Hareven2003 51]} If women were working from inside the home the labor could easily be classed as domestic labor even if it was then sold into the larger economic structure. Therefore, we have the persistence of the dichotomy between women's work (domestic) and men's work (commercial) even if one supported the other.

@; NDC: Scale of production. Could the free women of the household produce enough cloth to keep up with market demand (army, etc)? Or did households, for instance, keep large #s of slaves to do this work for them -- equally invisible but nonetheless very important?

@; "one might suggest ... that weavers and spinners in Britain were expected to produce some 22,500 tunics, cloaks and blankets per year for the army alone ... 5-6 wks. weaving time might be a more appropriate figure to quote for the weaving of each of the 67,500 requisitioned items in Roman Britain"@cite-footnote[Wild2002 31]



@make-footnote[]

@(generate-chapter-bibliography)