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

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

@title{Karanis}

@section{History and Excavation}

Karanis is a @index["rural"]{rural village} in the Fayum region of Egypt that offers a wide variety of evidence for every stage of textile production and consumption. The site was first settled in the 3rd century BCE. It was multi-ethnic from the beginning, populated with Egyptians and Greek mercenaries following the conquests of Alexander the Great.@cite-footnote[Gazda1983 8] By the first century CE, the waterworks which supported agriculture in the Fayum had begun to malfunction but repair efforts from the Roman army after Egypt was annexed into the Roman Empire revitalized the economy of the village just as it expanded to accommodate Roman settlers, further diversifying the population.@cite-footnote[Thomas2001 5] The maintenance or neglect of these dikes and canals would ebb and flow with the prosperity of the town – and indeed the empire at large.@cite-footnote[Thomas2001 4] Whether the decline of the town led to the neglect of the waterworks or their state of disrepair contributed to the demise of the town, by the sixth or seventh century CE the town was abandoned.@footnote{Following the initial findings of Boak and Peterson, the abandonment has been traditionally cited as the 4th century CE based on a lack of coinage and other artifacts finds from after that period. @cite[BoakPeterson1931]. Pollard's in-depth analysis of the pottery from Karanis suggests habitation at least until the 6th century, possibly into the 7th.@cite[Pollard1998]} Because of the disruption of the site prior to excavation, the specific dating of artifacts is difficult even within stratigraphic layers.@cite-footnote[Thomas2001 21] However, production peaked in the mid third century and most artifacts cited in this study were given the broad range of first through third centuries CE.@cite-footnote[Thomas2001 4]

Desertification of the area caused the city to be largely submerged in the sands and between that and the hot and arid conditions of Egypt, many artifacts survive in Karanis that do not survive in other areas of the Roman Empire such as wood, raw wool, textiles, and papyri. However, in the late 19th and early 20th centuries the site fell prey to an agricultural practice that granted local farmers permits to mine archaeological sites for soil. When ancient cities were buried, the decaying materials from the settlement produced a nitrogen-rich soil, @italic{sebbakh}, that could be used for fertilizer.@cite-footnote[Gazda1983 8] While this severely disrupted major portions of the site, the minor finds that the farmers sold on the antiquities market piqued the interest first of Burnard Pyne Grenfell and Arthur Surridge Hunt then Francis W. Kelsey. As director of excavations at Karanis carried out by the University of Michigan, Kelsey’s ambitious vision for Karanis included “the reconstruction of the environment of life in the Greco-Roman period ... [and the] increase of exact knowledge rather than the amassing of collections.”@footnote{This quote is from an unpublished manuscript: Kelsey, Francis W 1926, University of Michigan Near East Research Committee: Memorandum 14} The analysis of artifacts at present in this study will focus primarily on the Michigan excavations because Kelsey's meticulous documentation of small finds primarily from domestic contexts provide a wealth of textile tools.@footnote{The records published by the Kelsey Museum of Archaeology in the @italic{Record of Objects, Karanis,} include both the nearly 44,000 artifacts in the Kelsey collections as well as those in the Cairo Museum and have been transcribed into a CSV database, @cite[Karanis1929]} Access to this database, graciously provided to me by the Kelsey Museum of Archaeology, has been crucial in my analysis of the textile tools and artifacts from Karanis.@footnote{Michael Koletsos and Drew Cabaniss, PhD students at the University of Michigan are currently working with the Kelsey Museum to write a catalog of the textile implements from Karanis. In addition to the catalog, they are conducting research on the physical construction of the surviving textiles and using experimental archaeology to confirm that the implements found on site are compatible with the textiles found on site. These publications are forthcoming.}

@section{Archaeological Evidence of Textile Production}

@; NDC: Are they producing textiles for purely local consumption or for a wider market?
@; NDC: And not just tools but faunal remains; has anybody figured out how many sheep were living near Karanis? Is this a big place for sheep? good wool?

The tools uncovered in Karanis suggest that all stages of the textile process occurred within this city. A pair of iron wool @as-index{shears} with bronze handles from a domestic context would have been used to annually shear the wool from sheep @figures["KM3638"]. The presence of this artifact in addition to faunal remains suggest that at least some sheep were kept in-town or nearby.@cite-footnote[Gazda1983 14] The single set of @as-index{shears} on its own does not give us much information about the scale of sheering in Karanis; however, as these objects were made of bronze and were portable, it is likely that other @as-index{shears} were simply packed for future use by the owners when the town was abandoned. 

The evidence for the processing of wool within the site is more abundant.  The wool itself survived in various stages of processing. A mass of unwashed @as-index{fleece} was found in an undesignated area of the site @figures["km13095"]. This sample retains the original locks as well as particulates of vegetal matter and feces. Another mass of wool excavated from a domestic context shows a more advanced stage of processing as it was washed, combed, and ready to be spun @figures["km29768"]. These two finds together suggest that at least some portion of the wool supply for the town was cleaned, washed, and processed on site.

A total of 477 @index['("spindle/whorl")]{spindles} and spindle whorls in various states of preservation were recorded in the excavations at Karanis. The variation in size, weight, and shape of the @index['("spindle/whorl")]{spindle whorls} suggest that various weights and types of thread were produced.@footnote{The records published by the Kelsey Museum of Archaeology in Record of Objects, Karanis, include both the nearly 44,000 artifacts in the Kelsey collections as well as those in the Cairo Museum and have been transcribed into a CSV database. Access to this database, graciously provided to me by the Kelsey Museum of Archaeology, has been crucial in my analysis of the textile tools and artifacts from Karanis.} In most ancient contexts outside of Egypt, the only evidence for spindles that survive are spindle whorls of ceramic, stone, glass, or ivory because the perishable materials do not survive. This can give us a skewed representation of materials and distribution of the spindles used. The climate of Karanis preserved a much larger sample of textile tools of varying materials and types. Twenty-four @index['("spindle/whorl")]{spindles} survive with a whorl or fragment of a whorl attached to a shaft or fragment of a shaft @figures["KM3801ab" "KM88638"]. Spindle shafts were uniformly made of wood with the occasional addition of an iron hook @figures["KM3674"]. Spindle whorls came in a wider variety of materials : wood, stone, clay, glass, ivory, and bone (Table 1, Figs. @figures["KM3791" "KM5975" "KM21889" "KM21890"  "KM7641ab" "KM23970ab"].@footnote{The excavation database does not consistently list the materials for all of the spindle whorls, so I cannot report a full breakdown of the proportions of each material. However, I will include here a breakdown of the materials of the 74 spindle whorls listed on the Kelsey Museum’s website. Wood: 46; Glass: 12; Bone; 6; Ivory: 4; Stone: 4; Clay: 2} In spite of this impressive array of materials, 62% of the spindle whorls in the Karanis database which note the materials are made of wood, whereas glass, it’s closest contender, represents only 16%. @index['("spindle/whorl")]{Spindle whorls} of wood, ivory, bone, and glass were decorated with simple geometric patterns, most frequently crossed lines intersecting the hole of the whorl and arrangements of concentric circles.@footnote{Variations on this motif can be found on spindle whorls from the first through eight centuries CE in Egypt**} The product of these tools, hanks of spun thread and yarn, survive as well @figures["km22605"].

@; NDC: Can you suggest therefore that outside of Egypt where we have only whorls made of imperishable materials we're missing about half?

@tabular[(list (list @bold{Material} @bold{Quantity})
               (list "Wood" "46")
               (list "Glass" "12")
               (list "Bone" "6")
               (list "Ivory" "4")
               (list "Stone" "4")
               (list "Clay" "2"))]
@bold{Table 1:} @index['("spindle/whorl")]{Spindle Whorls} from Karanis by material}

@; NDC: 239 loomweights is not all that many, though. Suggests that most looms weren't warp-weighted? Your footnote just refers the reader to other discussions of looms; can you go further about Karanis based on the # of LW and other artifacts? If a house didn't have LW could you suggest that they used a different type of loom? How have your studies helped answer your basic questions?

Archaeological evidence for weaving at Karanis is simultaneously extensive yet inconclusive. There were likely multiple types of looms in use at Karanis including warp-weighted (the standard form of loom used in Ancient Greece), two-beam upright looms (the standard form of loom used in late Republican and Imperial Rome), and horizontal looms (the traditional form used in Ancient Egypt).@footnote{For the types of looms used in Roman Egypt see @cite[Wipszycka1965], 49-50. For more in-depth discussions of the types of looms used in the ancient world see @cite[Hoffman1964], @cite[Wild1976]} @index['("loom weight")]{Loom weights} were found in abundance;@footnote{In the Kelsey Museum’s Karanis database, at least 239 objects were identified as loom weights (three entries merely indicated loom weights with no specific number)@cite[Karanis1929]} however, many were individual weights or small groups @figures["KM3338" "KM7699" "KM24011" "KM25788" "KM25789"]. Only seven sets of ten or more loom weights were found and only two groups were large enough to service a loom (29 and 56 weight respectively). As the majority of loom weights were made of unfired clay,@footnote{The Karanis database uses the term 'unfired mud' but for the purposes of this study I will use the more common 'unfired clay' @cite[Karanis1929]} it is probable that more of these were once serviceable sets but some had deteriorated beyond recognition.@footnote{Of the 239 objects listed as loom weights in the Karanis database, 219 total and all sets of ten or more loom weights found together were made of clay. Other objects possibly identified as loom weights were made of varied materials, however none of these come close to representing a usable set (Table 2). It is important to note the difficulty in distinguishing between loom wights and other objects such as weights for fishing nets. The clay weights were unlikely to be used for fishing as they would disintegrate rapidly in water, however the five weights from Karanis that have rope or cord still attached suggest some function other than weaving clothing.@cite[Karanis1929]} Three loom weights were preserved with string still attached @figures["KM3338"]. Only two @index['("loom weight")]{loom weights} from Karanis had any noted decoration and in both cases it was minimal. ­
@; figure out context/house#s for setsofloom weights.
@; Figure out what decoration on loom weights
@; NDC: mud or clay? geologists don't use mud as a material; where does this word come from? Morgan: It comes from the Karanis database based on the original excavation reports. Look into this further
@; NDC: Are none of the other publications about Karanis any use with regard to weaving utensils? Morgan: At this point I was primarily listing finds from the database, but should incorporate more sources

@tabular[(list (list @bold{Material} @bold{Quantity})
               (list "Clay" "219")
               (list "Limestone" "2")
               (list "Pebble" "1")
               (list "Stone" "9")
               (list "Plaster" "3")
               (list "Pottery" "1")
               (list "Marble" "1"))]
@bold{Table 2:} @index['("loom weight")]{Loom Weights} from Karanis by material}

At most Roman sites, the most we can expect for archaeological evidence of weaving are @index['("loom weight")]{loom weights} for the warp-weighted loom since they are made of clay or stone and the frames of the looms were wooden and therefore perishable. However, several wooden textile tools which are often absent from the archaeological record at other sites survive at Karanis. One loom shuttle, used to pass the weft thread through the warp was discovered in a domestic context @figures["km3905"]. This long, flat ovoid piece of wood has one complete and one partially drilled hole to hold the thread in place, and shows evidence of wear on both sides. A high quantity of weaver’s combs and fragments thereof, 298 total, were uncovered at Karanis @figures["KM3352" "KM3787" "KM3788" "KM3789"].@footnote{This quantity is seemingly disproportionate compared to the numbers of @index['("spindle/whorl")]{spindle whorls} and @index['("loom weight")]{loom weights} found at the site, perhaps they were left behind when the site was abandoned} This wooden implement is typically around 24cm across constructed of three or four pieces of wood held together with dowels with the center piece extended into a handle. The comb has short, evenly spaced tines cut out of the front that were used to evenly pack the weft strands when weaving heavy fabrics without disturbing the spacing of the warp strands.@cite-footnote[Thomas2001 17] Wooden @index['("heddle")]{heddles} – used to separate the warp strands and create a shed for the weft to pass through – and @index['("heddle jack")]{heddle jacks} – blocks used to support the heddle – also survive in Karanis @figures["KM26501" "KM3779"]. The variation in numbers between these types of tools is difficult to interpret. If we assume that in use each loom would require a shuttle, a @index['("weaver's comb")]{weaver's comb}, a heddle, and two heddle jacks, it is unclear why only a single shuttle, six @index['("heddle")]{heddle} fragments, and seven @index['("heddle jack")]{heddle jacks} were recovered in comparison to the 298 @index['("weaver's comb")]{weaver's combs}.

In spite of both archaeological and written records that indicate that weaving was done in the city there is a notable absence of looms in the archaeological record of Karanis. This is surprising given the fact that other large-scale wooden objects survive, such as a complete wooden door with a lock.@footnote{ KM8151, not pictured here} Some larger wooden fragments were cataloged in the @italic{Record of Objects} as 'loom fragments;' however, upon closer inspection they appear to be independent pieces such as @index['("heddle jack")]{heddle jacks} and support beams as opposed to pieces of the loom frame itself @figures["KM24863" "KM24866" "KM26544"].

It is possible that while certain tasks such as preparation of fibers and spinning were performed in the house weaving could have been done in a centralized location. If such a production center had existed, it likely would have been located in the city center, presumed to be the location of most of the public activities of the village. Unfortunately that location was too damaged by farmers digging for @italic{sebbakh} to be methodically excavated.@cite-footnote[Husselman1979] The presence of other weaving tools, such as wool combs, @index['("loom weight")]{loom weights}, and shuttles in domestic contexts, however weakens that explanation. It is unlikely that these artifacts would be stored in the home unless weaving was likewise done there. A more likely explanation is that when the village was abandoned, looms were disassembled and moved with their owners – a similar fate to much of the furniture of Karanis.@cite-footnote[Gazda1983 19] As an integral part of the domestic economy of a household, a loom likely would have been a considerable investment and held continued value to the family.@footnote{ In later cultures which relied on cottage industry, looms were passed on to the next generation. For this practice in eighteenth century France, see @cite[FauveChamoux2001 167]}

@; NDC: What would a loom look like? Would it be recognizable if disassembled?
@; NDC: Is there evidence for weaving in centralized locations?
@; NDC: For other types of furniture do we have evidence that it existed?

@; NDC: Maybe mention in text that it is not yet possible to say which households wove, where they were found, etc. Can you say anything more about this here? Can you give a minimum # of houses, or say whether those groups of LW were found with spindle whorls or not? Go further here. Or were you not able to glean this information from your visit to the Kelsey or other sources?

The majority of these textile implements found in Karanis came from houses and were often grouped in assemblages.@footnote{For example, many individual or small groupings of @index['("loom weight")]{loom weights} and some @index['("weaver's comb")]{weaver's combs} were found in unknown contexts or in the street, but of the seven sets of ten or more loom weights, six were discovered in domestic contexts. @cite[Karanis1929]} This indicates that textile production on some level continued to be practiced in the home in Karanis.@footnote{The Karanis Housing Project has made a GIS map of the site of Karanis and is currently populating the map with artifacts from the University of Michigan excavations. When their data entry is complete, this project will provide better visualizations for artifact dispersal and object groupings in Karanis. @cite[KaranisHousingProject]} As is often the case with small portable artifacts, isolated finds occur in streets and unknown contexts. In addition to these finds, three @index['("loom weight")]{loom weights} and a wooden @index['("heddle")]{heddle} were found in the temple.@cite-footnote[Karanis1929] These numbers are far too low to comprise a working set and were therefore likely votive offerings rather than evidence of religious production of textiles. 

@; The majority of houses **% contained textile tools, and **% of these houses contained five or more tools.

Textiles of all types and uses were found across the whole site of Karanis @figures["km0482" "km3409" "KM10479" "KM11192"]. Most of the nearly 3,500 pieces of textiles from Karanis were from domestic contexts. Overwhelmingly, the textiles from Karanis show extensive signs of use-wear and are often mended or fragmentary from extended use.@cite-footnote[Gazda1983] Textiles were a commodity and since most of the inhabitants of Karanis were unpretentious @index["rural"]{rural} families, textiles would be mended when possible and re-purposed as they were woren beyond use or torn into smaller sections. Several cloth dolls survive from Karanis, offering us a glimpse of the non-utilitarian uses of textiles, though they were likely made of scraps of fabric that had exceeded their use elsewhere @figures["KM3648"].@cite-footnote[Wilson1933] Most textiles that were more serviceable at the time the village was abandoned would likely have been taken with.

@; NDC: Is it possible to distinguish home-made from professionally-woven textiles here? Is there evidence for professional weaving at Karanis?

@;Graves often produced as tunics and shawls @figures[""] 24-25). Although funerary contexts preserve more complete examples of textiles – they were created and used as grave goods and therefore don’t have signs of use and wouldn’t have been taken when the village was abandoned.@cite-footnote[Thomas2110 3]

@section{Textual Evidence of Textile Production}

@; NDC: You don't want to simply ignore evidence about male weavers, though: they will have had an important influence on female weavers, and you could compare their experiences etc. (bring in P. Mich. 81

Perhaps more fascinating than the tools themselves are the texts about textile production from Karanis. Two apprentice contracts survive, the first contracting the writer's son as an apprentice to a male @index['("textor/weaver")]{weaver} (P. Mich. 81), the other apprenticing a slave girl to a @index['("textrix")]{female weaver} (P. Mich. Inv. 5191).

@nested[#:style 'inset]{P. Mich. 81:}

@nested[#:style 'inset]{To Theon, collector of the weavers’ tax, from Pausiris, son of Ammonios, resident of the city of Oxyrhynchus, of the Cavalry Camp Quarter. I wish, from the present ninth year of Nero Claudius Caesar Augustus Germanicus Imperator, to apprentice my son, Pausiris, a minor, to the master @index['("textor/weaver")]{weaver,} Epinikos, son of Theon, of the Temple of Hermes Quarter, that he may learn the art of weaving and pay the tax paid by persons of his class. Therefore I ask that my son be registered in the class of apprentices, as is fitting. Farewell. The 9th year of Nero Claudius Caesar Augustus Germanicus Imperator, Phaophi 22.}

@nested[#:style 'inset]{(2nd hand) I, Theon, have affixed my signature. The ninth year of Nero Claudius Caesar Augustus Germanicus Imperator, Phaophi 23.}

@; **Give summary of this document. Emphasis on tax, reference to the 'class' of apprentices. This is a tax registration that references an apprentice contract rather than the contract itself. This is absolutely commercial

@nested[#:style 'inset]{P. Mich. Inv. 5191:}

@nested[#:style 'inset]{Aurelius Ision, son of Nilammon, a resident of Karanis, has given over to Aurelia Libouke, a resident of the quarter of the Bithynians and other areas, a @index['("textrix")]{weaver}, acting without guardian by right of her children, the slave child of the same Ision, to learn with Aurelia Libouke the indicated craft in the period of one year from the first of the ensuing month Mecheir, the child being fed and clothed by her ... may receive from the weaver .... And for as many days as she is idle because of sickness or any other cause, she is to remain available an equal number of days as compensation after the end of the period. When the slave child has completed the agreed time without fault, the teacher shall return her after she has learned the craft with skill equal to those of her own age.}

@nested[#:style 'inset]{Neither party shall have authority to alter either one or another stipulation nor to transgress any part of the written agreement, but let whosoever does transgress give to the one abiding by it, as penalty, two hundred silver drachmai. The apprentice contract is valid, and when questioned, they reciprocally agreed. Aurelia Libouke, about 58 years of age, with a scar on her left shin: the slave child is receiving at the end of the time,T to the account of Ision, sixty drachmai.}

@nested[#:style 'inset]{Year one of Lucius Domitius Aurelianus and Septimius Vaballathus Athenodorus, Tybi 26.@footnote{As translated in @cite[Thomas2001]}}

This contract gives us an unusually detailed account of the relationship between two women @index['("textrix")]{weavers} in Karanis, the craftswoman Aurelia Libouke and her pupil, an unnamed slave girl from the household of Aurelius Ision. The document reveals quite a bit about Aurelia Libouke. In addition to her age, the location of her residence, and the location of her scar, the phrase "by right of her children" implies that she had acquired the @italic{ius trium liberorum}, which allowed women who had three sons the privilege of acting in her own interests without guardianship of an adult male relative.@footnote{ This was one of the 'rewards' of the @italic{lex Iulia de maritandis ordinibus} to promote women to have more children, see @cite[Milnor2005 152]} This contract is of further interest to this study since it stipulates that the slave girl will acquire skills in weaving which she will then put to use to produce textiles in her owner's household. Between these two women, it therefore provides evidence for women participating in weaving that blurs the commercial and domestic distinctions.

@; NDC: What does 'craftswoman' mean? Is there a technical or specific term, or is she just called a woman operating on her own? What's the significance of 'by right of her children'?

Another contract barters a woman's domestic labor in exchange for money (P. Mich. Inv. 2819). 

@nested[#:style 'inset]{Aurelia Taesis, daughter of Asklepiades and Sarapous, from the city of Memphis, has acknowledged to have received from Aurelia Thaisarion, daughter of Komon, from the village of Karanis, the capital sum of one myriad eight thousand [18,000] silver drachmai, i.e. three talents, which, as Aurelia Taesis has further acknowledged, have been paid for a debt of her aforementioned father Asklepiades; and that she, the first party, will of necessity stay by Aurelia Thaisarion, performing the weaving and household tasks that she knows in place of the interests of the capital sum. If she wants to give up, [she acknowledges that] she will also of necessity repay the aforementioned three talents of silver without delay, and that the right of execution on demand shall rest with Aurelia Thaisarion against the first party, Aurelia Thaesis, or against all her property, as if in accordance with a legal decision. This document written as a single copy shall be valid anywhere it may be produced. And in response to the formal question, she has so agreed.}

@nested[#:style 'inset]{(2nd hand) I, Aurelia Taesis, have received the aforementioned three talents of silver and shall stay [by her] for service of my craft and other household tasks. If I give up, I shall repay the aforementioned money, as aforementioned. And in response to the formal question, I have so agreed. I, Aurelius Horion son of Soterichos, from the quarter of Phremei, have written for her because she is illiterate.}

@nested[#:style 'inset]{[In the ... year] of our lord Probus Augustus, Tybi 15.@footnote{As translated in @cite[Thomas2001]}}

Again we have two women at the center of this contract incidentally both also named Aurelia. The lender, Aurelia Thaisarion of Karanis has paid a sum of three silver talents in order to discharge the debts of Asklepiades of Memphis. In lieu of interest on the loan, his daughter, Aurelia Taesis, will work off the debt through weaving and other domestic labor. The contract does not set a length of time to completion although it is rather specific as to the immediate repayment of the loan should Aurelia Taesis leave. This implies that the contract is indefinite until the full sum may be paid. The effect of this contract is that A. Taesis is effectively a domestic slave in A. Thaisarion's house until the loan is paid off.

@; NDC: How much is 18,000 silver drachmai/three talents at the time? If this were Classical Greece this would be an impossibly huge amount of money, almost 50 years' wages. Morgan: I need to look into this, but the slave girl in the above contract received 60 drachmai for her one year apprenticeship

The generally accepted paradigm for women’s roles in textile production in the Roman empire posits that women produced textiles within their own homes primarily to provide for the needs of their own @italic{domus} then sold the surplus to aid in the domestic economy.@cite-footnote[Wipszycka1965 84] A parallel narrative to this places predominantly male professional laborers working in larger-scale commercial textile production workshops.@cite-footnote[Thomas2001 18] While the high percentage of homes containing textile tools discussed above supports the narrative of domestic production, these contracts verify that women served as @index['("textrix")]{weavers} outside of their production for their own homes in Karanis. These women participated in the commercial production of textiles as skilled artisans, not merely wives providing home-spun cloth for their own families. They also provide evidence for women's active roles in the economic world of Roman Egypt with a central focus on textile production. The only textile-related professional association we find in the Karanis tax rolls is a guild of sheep shearers associated with the temple.@cite-footnote[Gazda1983 15] @; try to get hands on a copy of this, there might be more useful info

An even more abundant type of written sources for textiles at Karanis are receipts (P. Mich. Inv. 1050) or letters with descriptions of or requests for clothing items (P. Mich. Inv. 121, P. Mich. Inv. 5638, P. Mich. Inv. 5390, P. Mich. Inv. 5391, P. Mich. Inv. 5389, 5401). While these do not directly speak to the production of textiles in Karanis, they illustrate the market base and demand for textiles both purchased locally or imported from Rome or elsewhere in the empire. They demonstrate that while domestic production was certainly present, and perhaps even prevalent, in Karanis, the rural site was still ingrained into the large-scale trading networks of the empire. 

@;Outside of these two paradigmatic groups of textiles, those produced for domestic use and those produced or imported for the market, a third group remains. As members of the Roman Empire, Karanis as a village was required to provide clothing for the Roman military (P.Mich.Inv.5065a).@cite-footnote[Gazda1983 15]

@;**beef up with sources outside of Gazda and Karanis, useful for other sites as well**



@section{Visual Evidence of Textile Production}

@; NDC: Is this a standard pose -- so many of these statues have 'types'

Although there are no direct visual representations of textile production in Karanis, a sculpture of Isis-Aphrodite offers an interesting possibility @figures["KM10728"]. The copper-alloy statue was found in a domestic context with a hoard of 2944 coins. The female figure is smaller than life size, nude, and wearing the distinctive star crown of Isis-Aphrodite. Her arms are both raised and her left hand posed with three fingers curled in as if wrapped around a cylindrical object while her index finger and thumb are somewhat extended. Her right hand is posed with her thumb and index finger pinched together and her other fingers splayed out. This pose is consistent with her holding a @index['("distaff")]{distaff} in her left hand and drafting thread with her right. Even with one missing digit on her right hand, the poses of the hands are strikingly similar to images of women spinning on Greek vases. The positioning of her fingers is nearly identical to an Attic red-figure white-ground oinochoe from the British Museum depicting a spinning woman @figures["Beazley204379"]. There is, however, a notable difference in the position of her arms. The vase from the British Museum, as well as many other such depictions, holds the @index['("distaff")]{distaff} in her left hand significantly higher than her right drafting hand, whereas the Isis-Aphrodite from Karanis holds her right drafting hand only slightly higher than her left @index['("distaff")]{distaff} hand. While a pose where the @index['("distaff")]{distaff} hand positioned higher is more common in Greek depictions of women spinning, an Attic white-ground Lekythos of a woman spinning at the Yale University Art Gallery reflects a similar pose to the Isis-Aphrodite from Karanis @figures["Beazley204379"]. @footnote{ For other vases representing women spinning with their hands at nearly even height, see the following vases from the Beazley Archive Pottery Database: Beazley 209084, Beazley 208940, Beazley 209043, Beazley 208920, Beazley 202937. @cite[BeazleyArchive].}

While this pose is fairly ubiquitous in Greek depictions of women spinning on vases, evidence for such sculptural depictions is less common. In her book @italic{Women’s Work,} Elizabeth Wayland Barber notes a similarity in musculature and pose between this same iconography of spinning women and the Venus de Milo @figures["VenusDeMilo" "BarberVenus"]. Barber argues that the unnatural position of her missing arms would be consistent with the pose of spinning wool.@cite-footnote[Barber1994 ", 236-238"] Several statues of Aphrodite of the same type – including the Venus of Capua and the Venus of Arles – display similar poses @figures["VenusArles" "VenusCapua"]. The Venus of Capua is a rare example that retains her arms. The left is raised above and to the left while her right hand is lower and extended in front of her. Both hands feature the thumb and index finger pinched together. The three fingers on her left hand are wrapped as if gripping a cylindrical object such as a @index['("distaff")]{distaff}. The three fingers on her right hand are loosely curled as if she were drafting. Like the Venus de Milo, the arms of the Venus of Capua do not survive, only the right shoulder and the left arm to the drapery covering the elbow to indicate arm position. The statue was reconstructed in the reign of Louis XIV, approximating the position of the arms and adding an apple and mirror in her hands as attributes of Venus. Barber argues that in this sculpture type, the @index['("distaff")]{distaffs} and @index['("spindle/whorl")]{spindles} they potentially held would have been made of more perishable materials and were therefore lost to time. The symbolic association between female deities and textile production is a common theme in Greco-Roman antiquity. Barber notes that spinning is a fitting association with Aphrodite’s role as the goddess of love and procreation since it is often associated with the beginning of life and formation of new beginnings. If this is a continuing tradition, it may also be something of a bawdy joke, such as the so-called "spinning @italic{hetaira}" on Greek vases, which are representations of nude or provocatively dressed women that are often interpreted as prostitutes.


Although the idea of the Isis-Aphrodite from Karanis spinning is particularly tempting for a location with such strong evidence for female textile production, it is also tenuous. The statue does not have a @index['("spindle/whorl")]{spindle} and thread in her hands and no spindles or whorls found in the same context that could have belonged to the statue. Since the statue is made of costly metal, and was found in the same context as a hoard of coins, ii is possible that this space served as something of a domestic treasury for the owners of the house and the associated @index['("spindle/whorl")]{spindle} was missing before it was stored.
@; Should I move the spinning Isis-Aphrodite discussion to the Ceremonial chapter?

@; NDC: A bigger problem is why are her hands on close to the same level? Doesn't one hand hold the ball of wool high up, and you need the vertical space between the upper and lower hands for the thread being created? Can this be done more or less horizontally?

@section{Conclusion}

In comparison to the urban centers of Trier and Ephesus discussed below, the village of Karanis played a relatively insignificant role in political and economic networks of the Roman Empire. However, the remarkable level of preservation at Karanis gives us a greater breadth of archaeological and papyrological evidence for textile production than any other Roman site. The wealth of textile tools provides us with objects that typically do not survive in the archaeological record such as @index['("weaver's comb")]{weaver's combs}, @index['("heddle jack")]{heddle jacks}, and @index['("spindle/whorl")]{spindle} shafts. Even for tools that do survive elsewhere, such as spindle whorls and @index['("loom weight")]{loom weights}, the range and ratio of materials that survive at Karanis likely give us a better-rounded idea of the missing evidence from other sites. The fact that over half of the @index['("spindle/whorl")]{spindle} whorls from Karanis were made of wood and the majority of the @index['("loom weight")]{loom weights} were made of an unfired clay, neither of which would have survived outside of the arid conditions of Egypt, suggests that similarly simple and cost-efficient tools were likely used elsewhere as well. The corresponding raw materials, textiles, and contracts for weaving apprenticeships, and receipts give us a reasonably well-rounded understanding of the local textile market.

Even with this wealth of material culture, Karanis does leave us with missing evidence, most notably the absence of any looms. Unlike the sites of Pompeii and Herculaneum, Karanis does not leave us with a snapshot of a living city. The village went through a gradual decline in prosperity and ultimately was abandoned. Therefore what survived archaeologically were the objects left behind. It is possible that the higher ratio of perishable tools may have reflected the worn, mundane objects that were left behind while higher-quality stone, glass, and ceramic equivalents were packed away when the inhabitants removed from the site. Likewise, the damage to the city center and the paucity of funerary evidence likewise limit the types of evidence we have if there was a separate purpose-made market for hihgher quality tools as grave goods. 

@make-footnote[]

@(generate-chapter-bibliography)