‘Like a Virgin’: A Sogdian Recipe for Restoring Virginity and the Sanskrit Background of Sogdian Medicine

Abstract

Among the fragments of Sogdian medical texts is what seems to be a collection of gynaecological prescriptions, including a recipe for the restoration of virginity. In addition to providing an edition and translation of the text, we attempt to set it within a comparative context including recipes for virginal simulation occuring in Arabic, Chagatai, Greek, and Latin medieval works. Finally, we identify the text as, like much of Sogdian medicine, possible Sanskrit origin, and give an overview of the Sogdian medical fragments so far identified.

Full Text

§ 1. A Sogdian recipe for restoring virginity Every fragment of a Sogdian manuscript alludes to a lost literary and historical world that may be but glimpsed through the surviving, partiallypreserved words and phrases. When two or more fragments may be joined together to yield a longer passage, scholars may gain a clearer image of the text and the context in which it was produced. One of the longer medical texts surviving in Sogdian is obtained in just this way, by joining together the fragments So 10100k, So 20249, So 20250, © Reck, Christiane, Persian Manuscripts, Goettingen Academy of Sciences and Humanities, Marburg (reck@bbaw.de) © Adam Benkato, Department of Near Eastern Studies, University of California, Berkeley (benkato@gmail.com) ∗ The authors wish to thank Dieter Maue for suggestions on the Sanskrit origin of several Sogdian botanical terms and Yutaka Yoshida for his helpful comments on a draft of this article. and So 20251 of the Berlin Turfan collection, along with So 20235, which belongs to the same manuscript but does not join the previous fragments 1 directly. The medical text, in fact, is written on the verso side of the paper; on the recto side is what seems to be a Manichaean narrative, still unpublished. The order of the sides was previously undetermined: in the catalogue of RECK (2006, 32-34 #23), it had not yet been fixed but in a review of the catalogue, YOSHIDA (2008, 57) suggested that the Manichaean text was written first and would thus be the recto side, one indication being that this side shows faint ruling lines and margins while the other does not. It is worth mentioning that the opposite situation is present in another group of fragments, So 10100b+So 10102(1), So 10102(2), So 15501 (RECK 2006, #16, #27, #228; RECK 2016, #463, #478, #762). There, the blank verso side of a Sogdian Buddhist scroll was re-used for a Sogdian Manichaean text. In the present group of fragments, the text of the verso is written in the ‘formal’ variety of the Sogdian script, in black ink, in a somewhat careless but legible hand. The name ‘formal’ was introduced by SIMS-WILLIAMS (1976, 44-5) to describe a type of Sogdian script characterized by the consistent distinction of the individual letters in contrast to the so-called ‘cursive’ script, where multiple letters have the same shape and can distinguished only by interpretation of the content. In fact the formal script is also cursive in the proper meaning of the word, since the letters are connected in most cases, but the main point of difference between the ‘formal’ and ‘cursive’ varieties is the distinction of the letters, most remarkably that of the aleph which is marked by a significant stroke. Only a few punctuation marks are preserved in this text and they take the form of two black strokes with a red stroke above and below. Its dating is a difficult question as that of Sogdian texts in general is rather unclear. If we assume that the bulk of Manichean texts were written between the 8th and the 11th cc. CE, we can assign a plausible range for the recto with the verso of course having been written at some unclear point afterwards. In the Turfan area evidence for the use of Sogdian begins to fade at the beginning of the 11th c. CE, and there is thus far no evidence later than that. The manuscript as a whole, or at least the section of it preserved in these fragments, seems to have contained medical treatments of a gynaecological nature. The preserved text begins with a prescription for restoring virginity 1 A montage of the four fragments So 10100k, So 20249, So 20250, and So 20251 joined together is at the end of this article. (xyδ ms δβtyw pwrʾych βwtkʾm ‘then (the woman) will become a virgin 2 again’). Although such recipes occur in several medical texts, this is the first occurence of such, to our knowledge, in medicinal texts from Central Asia before Islam. The recipe instructs one to prepare a mixture of smʾnk, krkrwγn (‘ghee’), ptʾnk-δʾrʾwk (a kind of wood), swrxyc (‘red clay’?), rwtr (‘saffron’?), srcrs (‘sal-resin’), cmprs and ʾnprs (both resins), and kʾwʾrtyʾsprγmʾy (a type of flower), to boil and pound it, to mix it with water, and finally to clean the vagina (the Sogdian text employs the euphemism cʾδr γrʾyw ‘lower-body’; in other texts similar euphemisms for body parts are encountered, e.g. cʾδʾr pδyk ‘lower-part’ and pšyy δβryʾ ‘back door’ for ‘anus’, cf. SUNDERMANN 2007, 410 n. 42, 45) and insert the mixture. This apparently would lead, as the Sogdian text states matter-of-factly, to the woman becoming a virgin again. We will comment on this recipe in more detail in the following section. Given that we know little about pharmacological and medicinal terms in Sogdian, many of the ingredients contained in this recipe are unknown. However, as many of the ingredient names seem to have a Sanskrit origin, an adaption from an Indic (i.e. Sanskrit or Prakrit) medical tradition is quite plausible-we discuss this possibility in more 3 detail following the text edition. The following part of the text seems to contain remedies for foul smells of the ‘lower body’ (again probably a euphemism for vagina). In it, the three kinds of myrobalans (Skt. triphala) well-known in Indian medicine are used. In Sogdian these are ʾʾrʾyrʾy, βrʾyrʾy, ʾʾmδʾy (ārirē, virīrē, and āmaδē), corresponding to Sanskrit harītakī, vibhītaka, āmalaka respectively. Two of these three, ʾʾrʾyry and ʾʾmδʾy were already attested in Sogdian fragments in Brāhmī script, while the third, βrʾyrʾy, is attested here for the first time. 4 These three myrobalans are mentioned at least two times more in this text. 2 The recipe was first recognized by YOSHIDA (2008, 57) who cited this phrase. 3 At least one example of a Sogdian translation of an Indian medical text has been identified so far, that of Vāgbhaṭa’s Aṣṭāṅgahṛdaya-saṃhitā, see RECK and WILKENS (2015). There also seem to be, as Gudrun Melzer informed us, some unpublished Sanskrit dhāraṇīs for gynaecological problems. 4 For discussion of ʾʾmδʾy see YOSHIDA (1984, 146), for ʾʾrʾyrʾy see MAUE and SIMSWILLIAMS (1991, 493). The form of βrʾyrʾy in Sogdian had already been predicted in MAUE (2009, 300 with n. 64-67) though unattested until now. Text 1 (pl. 1): So 10100k + So 18249 + So 18250 + So 18251 (verso)[20] /1/ [ ½ l. ](.) [....](k/p)[.](m) (.)[.](.) (k/p-k/p..δ) /2/ [. . . . . . . . . . . . . . . . . ](.) xyδ ms δβtyw pwrʾych βwt kʾm ǁ /3/ [. . . . . . . .](.) smʾnk ZY krkrwγn ptʾnk δʾrʾwk ZY /4/ [. . . . . . . .]ZY swrxyc ZY rwtr ZY srcrs ZY cmprs /5/ [. . . . . . . .] ʾsprγmʾy ZY ʾnprs ZY kʾwʾrty (ʾ)sprγmʾy /6/ βyʾxš ZY ʾyny wʾβʾyδ rwrʾ ẓʾm nxwʾy ʾyδ ʾʾpyh prw /7/ ptryδ ZY wβyw cʾδr γrʾywy snʾyʾt ZY cʾδr γrʾywyh /8/ cyntr sʾr δʾrʾt rty δβtyw pwrʾych βwt ǁ rtykδʾ /9/ cʾδr γrʾywh γncnʾk βʾt ZY rʾkh šwšmy γwnʾy pδwβsʾt /10/ [. . . . . .](.) šʾw ẓmʾyx ZY wrẓʾ ZY rwtr ZY ktypr ZY /11/ [. . . . . .]ʾʾrʾyrʾy ZY βrʾyrʾy ZY ʾʾmδʾy ZY tkkr ZY [. . .] /12/ [. . . . . .]x ZY ʾyny sʾt ʾyw zʾyh βyʾx š ZY xyδ ʾ(.)[. .] /13/ [. . . . . . . .] snʾy ZY (l)ym kwnʾ γrm γrm cyntr[ . . .] /14/ wrʾy(δ) ZY [. . . . .]kwʾy γwt mʾt wʾt Lʾ pyẓʾt [. . . . .] /15/ kyʾ cʾδr γrʾ(y)[w]h γntʾk [ ½ l. ] /16/ Lʾ ẓγtʾ kwnʾt δβtyw β(.)[ ½ l. ] /17/ nyẓʾt ZY ẓnʾkh xwʾʾt βʾt [ ½ l. ] /18/ [s]nʾyʾt ZY βy cyntr sʾr r[ ½ l. ] /19/ [. . . . . . . . r](w)tr prmʾʾ ZY (β.)[ ½ l. ] /20/ [. . . . . . . . ]kwšt ZY r(y)nk m(y)[ ½ l. ] /21/ [. . . . . . . . .](.) ʾyny ẓʾm nxwʾy ZY [½ l. ½ l.] /22/ [. . . . . . . . . . . .](k/pry) prʾyw kcy (.)[½ l. ] /23/ [. . . . . . . .] nyẓʾt wʾt Lʾ p(y)[ẓʾt ½ l.] /24/ [. . . . . . . . . .](β) rʾyrʾy (ZY)[ ½ l. ] Pl. 1. So 10100k + So 18249 + So 18250 + So 18251 (verso) /1/ ... /2/ ... and then she will become a virgin again. /3/ ... samānak and ghee and patānak-wood and /4/ ... red clay(?) and rodhra and sal-resin and chamba-resin /5/ ... flower and amba-resin and kāwārti-flower /6/ boil and this (with) so much herb pound finely. Mix this /7/ with water so that she would wash the vagina (lit. lower-body) (with it) and also let her hold (it) /8/ inside the vagina (lit. lower-body). She will become a virgin again. If /9/ the lower-body becomes odorous and (if) a šwšmy-colored vein sticks to [...] /10/ ... black clay and waržā and rodhra and katīpar and /11/ ... harītakī and vibhītaka and āmalaka and takkola and ... /12/ ... and all this boil altogether and this ... /13/ ... wash lim, warm up slowly inside ... /14/ mix and ... is necessary so that wind should not strike (it) ... /15/ whose vagina (lit. lower-body) [would be] bad [-smelling(?)] ... /16/ would not be able to hold [...], again ... /17/ should go out and (if) the body becomes weak ... /18/ she would wash, and you inside ... /19/ ... rodhra, a measure, and ... /20/ ... costus and rynk ... /21/ ... pound this finely and ... /22/ ... together with [...] kcy ... /23/ ... should go out, wind should not s[trike] ... /24/ ... vibhītaka and ... Commentary /3/ smʾnk [samānak?] is unknown, and a connection with smʾn ‘sky, heaven’ naisreP weN ot detaler spahreP .luftbuod smees سمنک, a sweet paste made from wheat (one of the ‘haft sin’ prepared for Nowruz celebration). - krkrwγn [karkrōγan] ‘ghee’. It is in a Sogdian version of the Mahāyāna Mahāparinirvāṇasūtra, fragment So 15289/r/15/, the Sogdian translation of Chinese 醍醐 tíhú (T.T. 374, vol. 12, 394c23) rendering Sanskrit sarpirmaṇḍa (SUNDERMANN 2010, 79). The word was borrowed into Old Turkic as kakruγn (in the fragments *U 9216 und *U 9224, see RACHMATI 1932, 408, text 1, l. 61 and 432, text 5, l. 3). - ptʾnk- δʾrʾwk [patānak?-δārūk] ‘patānak-wood’ is unknown but could possibly be connected with Sanskrit paṭṭikā ‘species of lodhra’ or paṭṭeraka ‘Cyperus Hexastachyus Communis’ (M-W, 579-580). /4/ swrxyc [surxēč] is a hitherto unattested word; it may be connected with New Persian surx ‘red’, MP suxr, since ‘red’ is usually krmʾyr in Sogdian. With the suffix -yc that forms nouns and adjectives, perhaps swrxyc means ‘the red (thing)’ and could have been a name for red clay or another red-colored ingredient. - rwtr [rūtar] may represent one of two Sanskrit words: Skt. rodhra ‘Symplocos Racemosa (a kind of tree)’ (M-W, 889) or Skt. rudhira ‘saffron’ (M-W, 884, also appearing in Tocharian B as rutir). This latter occurs as Old Turkic lutır in fragment *U 9126 in Brāhmī script (MAUE 2015, 378-379; RACHMATI 1932, 406-407, text 1, l. 41) and has been partially reconstructed in Sogdian as [l]uttar in a Sogdian fragment in Brāhmī script (MAUE and SIMSWILLIAMS 1991, 491-492). The present text makes no distinction between <r> and <l> (i.e. an <r> with a diacritic mark beneath), so it is difficult to know which possibility is more likely. - srcrs [sarčaras] is likely to represent Sanskrit sarja-rasa-, the name for resin (rasa-) from the Sal tree (Shorea robusta). J. Wilkens draws our attention to an Old Turkic passage from the Altun Yaruk Sudur (ed. CEVAL KAYA 1994, 266, § 476: 13-14) satčarasi yig hobık “Sarjarasa, (das heißt) frisches Harz” (MAUE and SERTKAYA 1986, 98, no. 31). - cmprs [čambaras] is another kind of resin, but the first part cmp- is unknown. /5/ At the beginning only the second part of a compound containing ʾsprγmy ‘flowers’ is preserved. Alternatively, ʾsprγmy ‘flower’ alone could be interpreted as ‘fragrant herb’ (as in Middle Persian sprahm). - ʾnprs [ambaras] seems to be yet another kind of resin, if -rs again represents Sanskrit rasa-. The word seems comparable with MP ʾbwrs ‘juniper’, but the latter’s etymology is unknown; furthermore the -n- of ʾnprs is clearly readable here. - At the end of the line kʾwʾrty (ʾ)sprγmy is another type of flower or herb, but the first part kʾwʾrty is unknown. /8/ The punctuation in this line (two vertical black strokes with a red stroke above and below) appears in both Sogdian Buddhist and Christian texts, though in the former it is not the most common form. /9/ šwšmy γwnʾy may be a compound with second element γwnʾy ‘colored’. /10/ wrẓʾ [waržā] seems to be an unknown word; wrz ‘miracle, magic’ (SIMS-WILLIAMS and DURKIN-MEISTERERNST 2012, 205a) would make little sense in this context. - ktypr [katīpar?] is unknown. J. Wilkens draws our attention to Old Turkic k(a)t(a)p(a)l = Skt. kaṭphala (RACHMATI 1932, 443b), which is a kind of nut, Lat. Strychnos potatorum, used in Ayurvedic medicine. /11/ tkkr [takkara?] may represent Skt. tagara- which refers to both ‘Tabernaemontana coronaria (i.e. East India rosebay or pinwheelflower)’ and the powder produced from it (M-W, 432). A word tkr also occurs in a list in the fragment Ch/So 14842a+, line 5, on which see RECK 2018 #1045, and #3 in the list of medicinal texts appended. See also Old Turkic tagar (CEVAL KAYA 1994, 250 § 440c3) and the discussion in MAUE and SERTKAYA 1986, 91, Nr. 15. /12/ If zʾyh is taken as ‘place’ and ʾyw zʾyh as ‘(in) one place’, the context suggests the colocation is an adverb meaning something like ‘altogether, all at once’. /13/ (l)ym kwnʾ is difficult to understand in the context, and the reading of the first letter is uncertain. However, the diacritic below the first letter suggests the reading [l]; this diacritic, a subscript <r>, is used sporadically in Sogdian. Yutaka Yoshida (personal communication) compares this lym to the Chinese word lin 痳 (Middle Chinese *lim) ‘venereal disease, gonorrhea’, no doubt a relevant word in this text. However, the order of the words is still unusual, and the following kwnʾ γrm γrm may mean something like ‘make warm slowly’. /18/ ZY βy looks like the enclitic form of the 2sg. gen. pronoun ‘you’ but its function is unclear if so. /19/ prmʾʾ is unknown, but could represent Skt. pramā- ‘basis, measure’ (M-W, 685). /20/ ]kwšt is only partially preserved; it may be part of a word such as ʾnkwšt ‘finger’ or ʾkwšty ‘costus (plant)’ (RECK and WILKENS 2015, 436/r/5/), the latter without an initial ʾ and final -y, but the context is unclear. - r(y)nk is unknown; a reading rβnk or even r(δ)nk may be possible. /22/ kcy is unknown. Comparable words in Sanskrit are kaca- ‘hair’, kacu- ‘Arum Colocasia’, kaccha- ‘riverbank’, ka-ja- ‘lotus flower’ (M-W, 242-43), but these are only guesses. Text 2 (pl. 2): So 20235 (verso) /1/ p]twʾyš ʾy(w)[. . . .](. . .)[.](y) (γrʾyw . .)[ /2/ . . .] sʾr γrʾyw pʾty γwt rt(y)[ /3/ ]y ʾʾrʾyrʾy ZY βrʾyrʾy[ /4/ . . nx]wʾy ZY ʾʾpyh p(rw)[ ptrʾyδ /5/ . . . cʾ]δr γrʾywy tyn m[ /6/ . . . . βw]δʾntk xwrt xwrʾy[ /7/ . . . . .]ʾyt kʾn ZY k(.)[ ] char, one [...] body [ ] to [...] the body must be protected. And [ ] harītakī and vibhītaka [ pr]ess and [mix] with water [ ] insert (in the) vagina (lit. lower-body) [ ] should eat [fra]grant food [ ] will be and [ Pl. 2. So 20235 (verso) /1/ [p]twʾyš may be only the second occurence of a verb with the meaning ‘to char, burn’, in the 2sg.impv/, otherwise attested only in a Manichaean cure for migraines (M568+M746c, line R1 qpyyʾtyy pṭwyš ‘char fishflesh’), possibly to be derived from an Old Iranian cognate of Sanskrit oṣ ‘to burn’. In Christian Sogdian ptwyš, attested once, translates a Syriac verb meaning ‘to kill off’ (SIMS-WILLIAMS 2016, 155). /5/ tyn in this context might be a 2sg.impv. tyn of the verb tny-/tyn ‘to bring in, insert’. /6/ [βw]δʾntk [vōδantē] ‘fragrant’ is a hypothetical restoration. § 2. Medieval recipes for restoring or simulating virginity Though so far the only known example of a recipe for the restoration of virginity from the pre-Islamic Central Asia, this Sogdian text is far from being the only example thereof in medieval literature more generally. Recipes for simulation of virginity preserved in works from other medical traditions may be compared, namely: those in the Latin compendia of women’s medicine known collectively as the Trotula first compiled in Salerno in the 12th c.; those in the Arabic medical encyclopedia Firdaws alḥikma compiled by Abū Ḥasan ʿAlī b. Sahl Rabban aṭ-Ṭabarī in the 9th c. In addition, a Greek source attributed to the Greek physician Galen, is brought into discussion. Finally, a somewhat later Chagatai medical treatise, the Ṭabīblik kitābï (or Khulāṣat al-ḥukamāʾ in Uzbek publications) by Sayyid Subḥān Qulï Muḥammad Bahādur Khan in the 17th c., also contains several similar recipes. It is worth examining these recipes for virginal restoration or simulation in some detail. The name Trotula refers to a group of three Latin treatises on gynaecology stemming from 12th c. Salerno, the most popular works on women’s medicine in Europe until the 15th c., and ones which drew heavily from not only the Greco-Roman medical tradition but also from the Arabic medical texts which had just begun to be widely translated.[21] The texts contain a number of methods for “restoring” virginity: this is usually done by employing constrictives, while one recipe simply employs leeches in order to draw blood, all in order to simulate rather than truly restore virginity. It is worth citing them in full here for the sake of comparison. [190] Constrictorium ad uuluam ut quasi puelle inueniantur. Accipe albumina ouorum et distempera cum aqua in qua coctum sit pulegium et huiusmodi herbe calide et panno nouo lini intincto bis uel ter in die uulue impone. Et si nocte minxerit, iterum inpone. Et nota quod prius abluenda est bene cum eademi aqua calida cum qua fuerint ista distemperata. [191] Accipe corticem ylicis renatium, et tritum distempera cum aqua pluuiali, et cum panno lineo uel bombace predicto modo uulue inpone, et hec omnia ante horam accessionis coitus remoue. [192] Item. Accipe puluerem nitri uel mori et inpone; mirabiliter constringit. [193] Item quedam sunt immunde et corrupte meretrices que plus quam uirgines cupiunt inueniri, et faciunt constrictorium ad idem, sed inconsulte, quoniam se ipsas reddunt sanguinolentas et uirgam uiri ulcerant. Accipiunt nitrum puluerizatum et uulue inponunt. [194] Aliter. Accipe gallas, rosas, sumac, plantaginem, consolidam maiorem, bolum armenicum, alumen, chimoleam, ana unciam .i. In aqua pluuiali decoquantur hec, et cum aqua illa fomententur pudibunda. [195] Quod ut melius fiat una nocte antequam nubat, ponat sanguissugas in uulua, sed tamen caute ne subintrent, ita ut sanguis exeat et in crustulam conuertatur, et ita uir decipitur propter sanguinis effusionem. [190] A constrictive for the vagina so that they may appear as if they were virgins. Take the whites of eggs and mix them with water in which pennyroyal and hot herbs of this kind have been cooked, and with a new linen cloth dipped in it, place it in the vagina two or three times a day. And if she urinates at night, put it in again. And note that prior to this the vagina ought to be washed well with the same warm water with which these things were mixed. [191] Take the newly grown bark of a holm oak. Having ground it, dissolve it with rainwater, and with a linen or cotton cloth place it in the vagina in the above-mentioned manner. And remove all these things before the hour of the commencement of intercourse. [192] Likewise take powder of natron or blackberry and put it in; it constricts [the vagina] marvelously. [193] Likewise, there are some dirty and corrupt prostitutes who desire to seem to be more than virgins and they make a constrictive for this purpose, but they are ill counseled, for they render themselves bloody and they wound the penis of the man. They take powdered natron and place it in the vagina. [194] In another fashion, take oak apples, roses, sumac, great plantain, comfrey, Armenian bole, alum, and fuller’s earth, of each one ounce. Let them be cooked in rainwater and with this water let the genitals be fomented. [195] What is better is if the following is done one night before she is married: let her place leeches in the vagina (but take care that they do not go in too far) so that blood comes out and is converted into a little 7 clot. And thus the man will be deceived by the effusion of blood. The Firdaws al-ḥikma fī aṭ-ṭibb “The Paradise of Wisdom concerning Medicine” was a voluminous encyclopedia of medicine compiled by Abū Ḥasan ʿAlī b. Sahl Rabban aṭ-Ṭabarī (c. 780-c. 860) in the year 850 AD at 8 Sāmarrā’. Unique is the sketch of Indian medicine at the end of the text (Part 7, Discourse 4), drawing from the works of Suśruta, Caraka, the Nidāna, and the Aṣṭāṅgahṛdayasaṃhitā of Vāgbhaṭa.9 This encyclopedia preserves four recipes for simulating virginity. The first occurs in a section on treatments of the uterus (part 4, discourse 9, chapter 9). Another three occur in the above-mentioned sketch of Indian medicine, in a chapter entitled min kutub imra’ah hindiyyah fī tanqīyat al-wajh wa ‘ilāj famm ar- 7 Edition of the Latin text and English translation from GREEN 2001, 144-147. 8 For information on Abū Ḥasan al-Ṭabarī, see D. Thomas, “al-Ṭabarī” in Encyclopedia of Islam (2nd edition), and MEYERHOF 1931, 7-12. The Firdaws al-ḥikma was edited by Siddiqi (1928), and is surveyed in MEYERHOF (1931, 12-14) as well as in the introduction to Siddiqi’s edition (in Arabic, Siddiqi 1928). 9 On this part of the work see MEYERHOF (1931, 42-46) for more details. Both the major works of Suśruta and Charaka contain gynaecological treatments, though not, as far as we can tell, ones for the restoration of virginity, cf. Suśruta’s Saṃhitā, ch. 38 “Yonivyāpat-pratiṣedha” and Caraka’s Saṃhita, ch. 30 “Yonivyāpat-cikitsitam”. raḥm ‘from the books of an Indian woman on the cleaning of the face and the treatment of the opening of the uterus’ - “probably from a book on midwifery” according to MEYERHOF (1931, 29). In a 1942 article, Siggel translated many of the sections concerning gynecological matters on the basis of Siddiqi’s edition, including the recipes for virginal simulation just mentioned. These we give here in Arabic and English: 1) From Part 4, Discourse 9, Chapter 19 fī ‘ilāj ar-raḥm wa tashīl al-wilādah‘On treatment of the uterus and facilitation of childbirth’. علاج يصير المسنة شبه بكرٍ تأخذ رامك و عفص و ھليلج اصفر و قشور رمان حامض و صمغ السوس و دم الاخوين من كل واحد بالسوية تدق و تنخل و تعجن بماء الخرنوب او بماء الآس و تصير منھا شيافاً طوالاً فتمسكھا المرأة معھا و تصير الى فراش الزوج ليلاً ان احتملته نھاراً او نھاراً ان احتملته ليلاً فانه جيد بالغ إن شاء9 “A treatment which makes an older woman like a virgin. One takes 10 musk, oak apples, yellow myrobalan, peels of sour pomegranate, gum of licorice, dragon’s blood, equal quantities of each, pounds it, sieves it, and prepares it into a dough with carob syrup or myrtle syrup, and makes from that a collyrium. The woman then inserts with it and repairs to the marital chamber at night, if she has inserted the collyrium by day, or in the day, if she has inserted it by night. It is then entirely 11 good, God-willing”. 2) From Part 7, Discourse 4 min jawāmi‘ kutub al-hind ‘From the compendia of Indian books’, chapter 35 min kutub imra’ah hindiyyah fī tanqīyat al-wajh wa ‘ilāj famm ar-raḥm ‘From the books of an Indian woman on the cleaning of the face and the treatment of the opening of the uterus’. علاج المكتھلات حتى تكن مثل الابكار، يوخذ من شحم الجرذان و دھن سمسم غير مقشر و من البادنجان اجزاءً سواءً يدق و يسحق و يرفع في قارورة يطلى منه على فم الرحم و يرفع منه في فرزجة صغيرة في كل وقت فانه يعيد العجوز بكراً بأذن 9، او يوخذ من العفص و عظام محرق و من البادنجان اجزاء سواء يدق مثل الكحل و تذر على فم الرحم قبل الجماع ،او تأخذ من فلفل و دار فلفل و زنجبيل و ھليلج و زعفران و عفص و ورق الآس و الجلنار 10 The word rāmik was left untranslated by SIGGEL (1942). 11 Ed. SIDDIQI 1928, 284-285, German tr. in SIGGEL 1942, 259, our English rendering. و ورق الاترنج و ورق الزيتون و المسك اجزاء سواء اربعة مثاقيل، و من ماء عروق الرمان و ماء عروق الصفصاف المطبوخ مائة و عشرون استاراً يصب ذلك في قدر نحاس و يجعل فيه من دھن سمسم اربعين استاراً و من البان البقر الحليب ثمانين استاراً، يطبخ فيه الدوية و يساط يوقد تحته بنار لينة حتى يبقى الدھن به و يذھب الماء و يصفى بخرقة كتان و يرفع في جرة خضراء و يدھن به النافية داخلھا و خارجھا ليلاً و نھاراً فانه ينفع من جميع ما فيھا و يجففھا إن شاء9 “Treatment for older women, so that they become like virgins: one takes rat-fat, oil of unpeeled sesame seeds, and aubergine, equal quantities of each, pounds and pulverizes it, and preserves it in a bottle; one applies this to the vagina. Or, one applies this to a pessary which is inserted at any time. The old woman becomes a virgin again, Godwilling. Or, one takes equal quantities of oak apples, charred bones, and aubergine, pounds it like eyebrow-powder, and spreads it on the vagina before intercourse. Or one takes pepper and pepper-husks, ginger, myrobalan, saffron, oak apples, myrtle leaves, pomegranate-blossom leaves, lemon leaves, olive leaves, and musk, four miṯqāls each, 120 istārs each of the juice of pomegranate root and the juice of cooked willowroot, pours that into a copper vessel and adds 40 istārs of sesame-oil and 80 istārs of cow’s milk, cooks the medicament, stirs it, and lights a small flame, to the point that the oil remains and the water cooks away. One strains it through a linen cloth and preserves it in a green jar. One uses it to lubricate the useful inner and outer parts, at night and during the day. This helps against everything that is in her 12 and makes her dry, God-willing”. It is notable that three of the recipes cited by aṭ-Ṭabarī are found in the section of his book drawing on “Indian” medicine. We assume that the Sogdian recipe has a similar origin on the basis of the many Sanskrit words written in Sogdian transcription. In fact, this is probably true of many Sogdian pharmacological texts thus far recognized, and the discovery of a Sogdian translation of part of Vāgbhaṭa’s Aṣṭāṅgahṛdayasaṃhitā proves at least that some Sogdian medicinal works came directly from Sanskrit. A pseudo-Galenic recipe preserved in Greek has also come to our attention. The recipe occurs in a section entitled Πρὸς τὸ μὴ 12 Ed. SIDDIQI 1928, 591-592, German tr. in SIGGEL 1942, 263, our English rendering. καθυγραίνεσθαι τὸ αἰδοῖον ἐν ταῖς συνουσίαις τῶν γυναικῶν, recipes ‘against the drying of the vagina during copulation with women’. Although it likewise aims not at restoring, but at simulating, virginity, it seems to be prescribed for women who have been violated. It too calls for oak apples as an ingredient. [ιβ΄. Ὡς γυνὴ ἡ βεβιασμένη παραφαίνηται παρθένος.] Λαβὼν ὀμφακιτίδων, κύπρου, ξυλοκασσίας, ῥόδων ξηρῶν, πεπέρεως λευκοῦ καὶ κόκκου γνιδίου ἀνὰ ἐξάγιον α΄. προλελουμένῃ ὑστέρᾳ προσθέτω ταῦτα·ὠφελεῖ δὲ καὶ σπέρμα ὀξυλαπάθου τετριμμένον καὶ 13 πρόσθετον. “How a woman who has been violated appears like a virgin again: Take one [measure] of unripe oak apples, Cyprian (tree), wood-cinnamon(?), dried roses, white pepper, and cnidium seed (seed of the daphne laureola). Insert this mixture after having washed the uterus. The pulverized and inserted seed of the common sorrel (rumex acetosa) is 14 also useful”. Last but not least, the much later Chagatai medical treatise Ṭabīblik kitābï (“Book of Medicine”, also known as Khulāṣat al-ḥukamā in Uzbek publications) by Sayyid Subḥān Qulï Muḥammad Bahādur Khan (1624/5- 1702 CE) contains a chapter ḫātūn-lar farǰïnï tar etmäk-ni bayānïda ‘on narrowing the vagina of women’ (chapter 25) which details five different 15 methods for either restoring or simulating virginity. The aim of the procedures varies; sometimes the text states that the woman will “become a virgin”, sometimes “like a virgin”. yigirmäbešinči bāb ḫātūn-lar farǰïnï tar etmäk-ni bayānïda: ḥukamā aytïb tururlar kim ḥayż-dïn arïġandïn soŋ üč kün har kündä yigirmi diram angubīn-ni otuz diram boyanïŋ süti birgä qošub ašasa wa andïn sïgïr öti kim qurutġan bolsa šafa qïlïb kötärsä asr wa tar olġay farǰ-ï<n> yaḫšï aq olġay agar balïġ öti <birlä> ni yāsmin yaġïġa šāfa qïlïb kötärsä 13 [Galen], De remediis parabilibus XIV, ed. KÜHN 1827, 478. 14 For the tentative translation of this passage from the Greek we are grateful to Roland Wittwer of the Berlin-Brandenburgische Akademie der Wissenschaften. 15 See WILKENS 2016, 186, 188. qïz olġay … agar ḫātūn boya süti ašab boya sütidin kötärsä qïz däk bolġay agar qïz oġlan-nïŋ bakāratï zāyil bolġan bolsa bātingān-nï alïb suw bilä bišürüb eškü{niŋ ič} yaġï bilä qošub šāfa qïlïb kötärsä qïz bolġay agar köz ašmaġan it balasini qaynatsa andaq kim yaġï šïqsa ol yaġïnï alïb ḫātūn kötärsä bikr qïz däk bolġay “Chapter twenty-five on narrowing the vagina of women: Physicians say that if someone mixes twenty dirhams of honey with thirty dirhams of madder juice, eats it when becoming clean after menstruation for three days every day, and after that applies the dry gall-bladder of a cow with a tampon, her vagina will become firm, narrow and white. If someone [mixes] the gall-bladder of a fish with jasmine oil and applies it with a tampon, she will become a virgin. … If a woman drinks madder juice and applies madder juice [with a tampon], she will become like a girl. When ther virginity of a girl has been violated: take aubergine, boil in water and mix with the tallow of a goat. If she applies it with a tampon, she will become a virgin. If a woman boils a puppy which has not yet opened its eyes until its fat comes out, takes that fat 16 and applies [with a tampon], she will become like a virgin girl”. The recipes cited here all make use of different ingredients in these recipes with a common goal: the restoration or simulation of virginity. Some ingredients occur in multiple recipes: clay or special types of earth in the Sogdian and the Latin texts; saffron in the Sogdian and Arabic; aubergine in the Arabic and Chagatai; fat of animals in the Arabic and the Chagatai texts; some kinds of trees and flowers in the Sogdian and the Greek; and oak apples in the Greek, Arabic, and Latin. Though several ingredients in the Sogdian text are still unknown (samānak, patānak-wood, amba-resin, chamba-resin, kāwārti-flowers), the diversity of recipes means that a comparison cannot necessarily help identify those words - though perhaps the Sogdian word for oak apples (Sanskrit māyāphala or vanamūrdhajā) or aubergine (Sanskrit raktapākī) is to be found there. The desired outcome of these recipes seems to be achieved mostly through the use of ingredients, such as oak apples, which have an astringent effect. Since the Sogdian text is fragmentary, we cannot tell whether a type of commentary or explanation may have accompanied the recipe. Did the 16 Ed. KÁROLY 2015, 94 (Chagatai text), 177 (English translation). author of this text believe that restoring virginity was actually possible, or did the text acknowledge that the point was rather to have the woman appear to be a virgin? The matter-of-fact statement xyδ ms δβtyw pwrʾych βwtkʾm ‘indeed she will become a virgin again’ seems to imply the former. Did the text approve of or condemn such methods? At the very least, the straightforward presentation of the ingredients and instructions seems to indicate that this text, or this part of the text, was just a listing of recipes without further commentary. The Trotula, for example, simply states that certain of the recipes are for “a constrictive for the vagina, so that women may be found to be as though they were virgins” (§ 190 in GREEN 2001). One recipe more clearly notes that the point was to deceive the man (§ 195). Likewise, the Arabic parallels which we have cited here give recipes for “women, so that they become like virgins” (ḥattā takun miṯl alabkār). The editor of the Trotula points out that the very transmission of this information, as well as its placement next to a disapproved method of ‘virginity restoration’ apparently used by women who had been violated and prostitutes, implicitly approves the other methods of “virginal simulation” (GREEN 2001, 42). Regarding the Trotula, Green points out that “the desire of women, ‘honest’ or ‘dishonest,’ to ‘restore’ their virginity suggests acknowledgment by at least some medical practitioners that women’s honor ... to a degree that would never have been true for men, was bound up intimately with their sexual purity. If successful, these recipes may well have made the difference for some women between marriage and financial security, on the one hand, and social ostracization and poverty, on the other” (ibid.). Unfortunately, little can be said about the context in which this Sogdian fragment was produced and it is therefore difficult to situate it within a specific social or religious community. The fact that this text is situated on the verso of a Manichaean scroll may indicate that the Manichaean text on the recto was no longer important and may point to a context postdating the Manichaean community in Turfan. But nothing more can be concluded about the society which produced the text and which may have used the medicinal direction within. Nevertheless, the mere existence of a recipe for “virginal simulation” in Sogdian strongly implies that what Green says about the Trotula is equally applicable: it attests to the social importance of women’s purity in the social context in which it was produced. § 3. Medical Texts in Sogdian The Turfan collections contain a number of medical or pharmacological text fragments, and it will be useful to present all such fragments so far 17 identified in a numbered list in order to facilitate future research. These are nearly all written in Sogdian script (both regular and ‘formal’ variants) and on scrolls or pustaka-leaves. Indeed, only a few medicinal texts on codex pages have been found, one in Manichaean script (no. 12 below) and two in Syriac script (nos. 19-20 below). This may indicate that, even if the Manichaean or Christian communities were producing medicinal texts, these were not compiled in their books but rather on more usable writing supports. The vast majority of these fragments are preserved in the Berlin Turfan collection, where they have recently been identified and catalogued. Otherwise, the Turfan collections of London, St. Petersburg, Paris, and Kyoto possess one or two fragments each. Manuscript Signature Find Signature Catalogue No. 1. So 10100k+So 20249+So 20250+So 20251 (verso) T I D/TM 394 1026 A text containing, among other things, a recipe for restoring virginity. Published here. 1a. So 20235 K 35 1067 Another copy of part of the text of the above. Published here. 2. Ch/So 14842a + So 14645 T II Y 17/T II S 21 1045 So 14480 + So 14841 T II D 201/T II Y 17 1043 So 14481 T II D 201 1035 Ch/So 14840(2) verso T II Y 17T 1042 U 5735 T II Y 17 1109 18 Otani 9133 - - 17 We also give the catalogue numbers corresponding to each fragment in order to simplify looking them up in Reck’s catalogues of the fragments in Sogdian script; see RECK 2018, concordance 3.4. 18 This fragment was published in the catalogue of the Otani fragments, see KUDARA et al. 1997, 159-160. Fragments from a scroll. Ch/So 14842a+So 14645 were found pasted together for secondary usage and were separated by the restoration department of the Turfanforschung. The other fragments would have originally belonged to this scroll as well (RECK 2014, 547 n. 18). Its script is rather unique, written-out broadly with almost clearly distinguishable letters and colons between some single words which are not used in that form in other Iranian texts. The content of the text seems to have a background in Indian medicine, as it mentions eight mammals whose milk or urine is used in the Indian medical tradition. 3. So 20167-So 20171 T II Toyoq 1059-62 Fragments of a bilingual group of pustaka leaves written with a brush. The fragments So 20167-So 20171 (note that So 20170 and So 20171 can be joined), contain several medical precepts and mention magic formulas (ptsrwm) as well, but do not preserve enough continuous text to clarify the content. The script is the same as that found in some Sanskrit fragments of apotropaic magic (vidyā) such as SHT 2058, where the page numbers and some comments are in Sogdian script and language. In So 20167 there are also parts of Brāhmī akṣaras visible on the blank verso side, meaning that the fragments should be read horizontally. This fact as well as their size (6.7×13 cm), which is shared with some Sanskrit pustaka leaves, suggests a Buddhist background. 4. So 15900 T III K 268 1047 So 10789(3) T I D 1029 A unique text, these are fragments from a Sogdian translation of Vāgbhaṭa’s Aṣṭāṅgahṛdayasaṃhitā, a comprehensive and well-known book of Indian medicine which was translated into several languages as Tibetan, Old Turkic, and Arabic. The fragments may be part of a codex page or a short-lined pustaka leaf. The script is similar to the Sogdian script used in Manichaean texts. Publ: RECK and WILKENS (2015). 5. So 20211 - 1064 A fragment of a pustaka leaf or codex, 11 lines. Contains prohibitions against different types of pain. The personal name Serguis (srkys) which occurs in the text may connect it to a Christian context. See RECK (forthcoming). 6. So 14822 T II T 35 724 Completely preserved pustaka leaf, which contains treatments of diseases of the lower abdomen and of ten kinds of water illness (ʾʾph rʾβ δsʾ znkʾn). 7. So 10006 MIK III 106 T II Toyoq 445 Completely preserved pustaka leaf, with two different texts. The first, on the recto side, is written in the formal script and contains prohibitions of various foods and meals. The second text, on the verso side, is written with a brush and contains dhāraṇīs partly in Sanskrit. 8. Ch/U 7187 T III 1078 Chinese Buddhist scroll with Sogdian text on the verso. The ten kinds of water illness (ʾʾph rʾβ) are mentioned here as well. 9. Ch/U 7211 T II T 1079 Scroll fragment of 9 lines containing an unidentified medical text. 10. Ch/So 20207 T II T 1063 Scroll fragment containing a medical text. 11. So 10339 T I α 1027 Fragment of a pustaka leaf, containing fragmentary remedies. Noteworthy is the abbreviation δr for the measure ‘drachma’, suggested by N. SimsWilliams. Publ: RECK 2014, BENKATO (forthcoming). 12. M 568 + M 746c - - Page from a codex containing a remedy for migraines (nymy sʾrxwyc). The only known example of a pharmacological text in Manichaean script. Publ: BENKATO (forthcoming). 13. So 14460+So 14427+So 14428+So 14462 / M 142 T II D II 179 155 / -155 / - A medical calender text about the Way of the Spirit of Life in the human body, known from Chinese folk calenders and preserved in Old Uighur as well. The fragment M 142 in Manichaean script is parallel to those in Sogdian script. Publ: MORANO and RECK (forthcoming). 14. P19 - - Paris collection, part of a scroll written in the formal script listing recipes for an emetic, purgative, and aphrodisiac. Publ: BENVENISTE 1940, 150 (without translation), HENNING 1946, 713 n. 5 (translation of one part). 15. L47+L48 - - St. Petersburg collection, two joining fragments probably from a scroll, containing 9 lines of a remedy for piles (ʾrsʾx). Publ: RAGOZA 1980, joining and further comments SIMS-WILLIAMS (1981:235). 16. BL Fragment 34 (Or. 8212/1811) - - London collection, three lines containing a fragment of an unidentified remedy. Publ: SIMS-WILLIAMS 1976, 73-4. 17. Otani 1159 - - Kyoto collection, a medical fragment containing numerous Sanskrit terms. Publ: KUDARA ET AL. 1997: 55-56. 18. E38 (n303) [T II] B 62 + [T II] B 13 - Pharmacological fragment in Syriac script, thus from a Christian context, 23 total lines. Publ: thus far unedited, see SIMS-WILLIAMS 2012, 187 for details. 19. E39 (SyrHT 343) 1876 - Fragment in Syriac script containing only one damaged line, see SIMSWILLIAMS 2012, 187 for details. 20. Mainz 639 - - A bilingual Sanskrit-Sogdian pustaka-fragment in Brāhmī script contai-ning parts of four remedies for diseases of the eye. Publ: MAUE and SIMS- 19 WILLIAMS 1991. 19 There are other medicinal texts in Sogdian in Brāhmī script, but are very fragmentary and remain unpublished, see SIMS-WILLIAMS 1996. § 4. Glossary to the edited Sogdian fragments Abbreviations: adj. adjective impv. imperative adv. adverb n. noun conj. conjunction obl. oblique dem. Demonstrative postp. postposition fthc. forthcoming pp. past participle gen. genitive subj. subjunctive ʾʾmδʾy n. ‘āmalaka, Emblic myrobalans’ 1.11 ʾʾpyh n. ‘water’ obl. 1.6, 1a.4 ʾʾrʾyrʾy n. ‘harītakī, Chebulic myrobalans’ 1.11, 1a.3 ʾ]kwšt ? 1.20* ʾnprs n. ‘amba(?)-resin’ 1.5 ʾsprγmʾy n. ‘flowers’ 1.5 ʾyδ dem. ‘this’ 1.6 ʾyny dem. ‘this’ 1.6, 1.12, 1.20 ʾyw number ‘one’ 1a.1 ʾyw zʾyh adv. ‘altogether’ 1.12 βrʾyrʾy n. ‘vibhоtaka, Belliric myrobalans’ 1.11, 1.24, 1a.3 βwδʾntk adj. ‘fragrant’ 1a.6* βw- ‘to be, become’. вʾt 3sg.subj. 1.9, βwt 3sg.pres. 1.8, βwt kʾm 3sg.fut. 1.2 βy ? 1.18 βyʾxš ‘boil’ 2sg.impv. 1.6, 1.12 cʾδr γrʾywy n. ‘lower-body, (euphemism for vagina?)’ 1.7, cʾδr γrʾywyh 1.7, 1.9, 1.15*, 1a.5 cmprs n. ‘chamba(?)-resin’ 1.4 cyntr postp. ‘inside’ 1.13* cyntr sʾr postp. ‘in, inside’ 1.9, 1.18 δʾr ‘to have, hold’, δʾrʾt 3sg.subj. 1.8, Lʾ ẓγtʾ kwnʾt 3sg.subj.tr.pot. 1.16 δβtyw adv. ‘again, a second time’ 1.2, 1.8, 1.16 γncnʾk adj. ‘bad-smelling’ 1.9 γntʾk adj. ‘evil, bad’ 1.15 γrm adj. ‘warm’ 1.13 (2) γrʾyw n. ‘body’ 1a.1, 1a.2 γw- ‘to be necessary’ γwt 3sg.pres. 1a.2, γwt mʾt 1.14 kʾn future suffix of lost verb 1a.7 kʾwʾrty ʾsprγmʾy n. ‘kāwārti-flowers’ 1.5 kcy ? 1.22 krkrwγn n. ‘ghee’ 1.3 ktypr ? 1.10 kwnʾ ‘do’ 2sg.impv. 1.13 kyʾ relative pronoun obl. ‘whose’ 1.15 lym n. ‘venereal disease’ 1.13 mʾt conj. ‘that, so that’ ms adv. ‘also’ 1.2 nxwʾy ‘pound’ 2sg.impv. 1.6, 1.20, 1a.4* nyẓʾt ‘go out’ 3sg.subj. 1.17, 1.23 pʾty pp. ‘protected’ 1a.2 pδwβsʾt ‘to stick to, be fastened to’ 3sg.subj. 1.9 prmʾʾ n. ‘measure(?)’ 1.19 prw postp. ‘with’ 1.6, 1.22 (prʾyw), 1a.4 ptʾnk δʾrʾwk n. ‘patānak-wood’ 1.3 ptrʾyδ ‘mix’ 2sg.impv. 1.7 ptwʾyš ‘char, burn’ 2sg.impv. 1a.1 pwrʾych n. ‘virgin’ 1.2, 1.8 Lʾ pyẓʾt ‘strike’ neg.3sg.subj. 1.14, 1.23* rʾkh n. ‘vein’ 1.9 rty conj. ‘and, so’ 1.8, 1a.2 rtykδʾ conj. ‘and if’ 1.9 rwrʾ n. ‘herb’ 1.6 rwtr n. ‘rodhra?’ 1.4, 1.10, 1.19* rynk ? 1.20 sʾr postp. ‘to’ 1a.2 sʾt adj. ‘all’ 1.12 smʾnk n. ‘samānak (unknown)’ 1.3 snʾy ‘wash’, snʾy 2sg.impv. 1.13, snʾyʾt 3sg.subj. 1.7 srcrs n. ‘sal-resin’ 1.4 swrxyc n. ‘red (clay?)’ 1.4 šʾw ẓmʾyx n. ‘black clay?’ 1.10 šwšmy-γwnʾy adj. ‘šwšmy-colored’ 1.9 tkkr n. ‘tagara?’ 1.11 tyn ‘insert’ 2sg.impv. 1a.5 wʾβʾyδ adv. ‘so much’ 1.6 wʾt n. ‘wind’ 1.14, 1.23 wβyw adv. ‘also’ 1.7 wrʾyδ ‘mix’ 2sg.impv. 1.14 wrẓʾ ? 1.10 xyδ adv. ‘then’ 1.2, 1.12 xwʾʾt adj. ‘weak’ 1.17 xwrt n. ‘food’ 1a.6 xwrʾy ‘eat’ 3sg.opt. 1a.6 ẓʾm adv. ‘finely’ 1.6, 1.20 ẓnʾkh n. ‘body’ 1.17 ZY conj. ‘and’ 1.3 (2), 1.4 (3), 1.5 (2), 1.6, 1.7 (2), 1.9, 1.10 (4), 1.11 (4), 1.12 (2), 1.13, 1.14, 1.17, 1.18, 1.20, 1.21, 1.24, 1a.3, 1a.4, 1a.7 Abbreviations Pl. = Plate