This paper investigates what light P. Cornell 127 possibly sheds on the use of slaves as tarsikarioi, weavers of fabric in the Tarsian style, in late antique Egypt. Slave labor for activities not directly relating to agriculture or domestic service was relatively rare in Egypt between the 4th and 6th century. In his survey of slavery in this period, Roger Bagnall (1993: 229-33) finds only a handful of examples in published papyri: a factory worker (a freedman) in P. Mich. IX 574, temporary inn workers in P.XV Congr. 22, a tarsikariosin P. Lips. 97, and a baker and another tarsikariosin P. Lips 26. The paidaria, including a baker, a butcher, and a brick-maker, in SPP XX 106 might be added to this list, though their status as slaves is not a settled issue (Rathbone, 1991: 89-91). The unpublished P. Cornell 127 offers further evidence of slaves used as tarsikarioiin this period.
This document is a letter of the 4th or 5th century requesting the pursuit, capture, and rendition of five tarsikarioi. The damaged state of the text makes it difficult to establish securely why the individuals are being sought. Three likely scenarios for such detention and rendition come to mind: (1) the individuals are runaway slaves (as in, e.g., P.Oxy. XII 1423, XIV 1643); (2) they have defaulted on debts or otherwise defrauded business partners (as in, e.g., P.Lond. VI 1915, P.Abinn. 42); or (3) they have fled their village in order to avoid liturgies or taxes (as in, e.g., P.Col. VII 175). While any of the three explanations is possible here, certain aspects of the letter suggest that (1) is most likely. The first extant line of the letter preserves one word: οá¼° ο [ς. Although the isolated word cannot with certainty be taken with tarsikarioi, it is not unreasonable to suppose that oikogenesis part of the description of one the fugitives in lines 2-4, where the number of fugitives, their occupation and the physical description of at least one of them is recorded. Further, the recipient of the letter is instructed to investigate ατá½° πα τὸς τοῦ σÏŽματος. Ιn this context σῶμα must mean, as it often does elsewhere (Preisigke, Wörterbuch, s.v. σῶμα, 3), the body of slaves associated with a house or estate. Finally, the language describing one of the tarsikarioi(e.g., leptos, monophthalmos) is similar to that used to describe runaway slaves in, e.g., P.Oxy. LI 3617, 3616, and XLII 3054. Since the weight of the internal evidence argues for identifying the tarsikarioimentioned in it as slaves, the reason they are being sought in all likelihood is their status of run away, that is scenario (1).
If the tarsikarioiin P. Cornell 127 are correctly identified as slaves, the document augments the very small body of evidence for the use of slaves in skilled trades in late antique Egypt. Additionally, the instruction to interrogate the oiketesin the letter’s post-script indicates that the activities of the tarsikarioiin this text, like those of the tarsikarioiof P.Lips. 26 and 97, are carried out in connection with an estate. The presence of slave tarsikarioion agricultural estates intimated by these documents calls for an evaluation of the extent to which slaves were used on agricultural estates for non-agricultural production, especially for the manufacture of textiles. It further raises the question of whether such use of slaves was limited to non-urban areas since there is evidence from cities of tarsikarioientering into contracts and leases (P.Oxy. LXVI 4534, P. Heid. Inv. G 1669), and tarsikarioiguilds (P. Lips. 89, P. Iand VIII 153).