The Mandaeans and the Question of Their Origin

ARAM, 16 (2004) 47-60 C. MÜLLER-KESSLER 47

THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN

CHRISTA MÜLLER-KESSLER1 (Friedrich-Schiller-Universität Jena)

In memory of my teacher in Semitics Rudolf Macuch

Browsing the Internet for the Mandaeans one comes across the website “Mandaean World”, where one can read under the logo of the Uroboros snake, the gnostic serpent encircling the world: “The origins of both the peo- ple, and of the religion are one of the continuing mysteries of Mandaean research.” This quotation describes exactly the difficulty by which a hundred and fifty years of scholarly discussion was led mostly within Europe concerning the “Mandäerfrage”. A scholarly dispute more at home in theological cir- cles or in research groups whose main interest is the study of the world religions than among linguists.

I have a feeling that time has come to re-open the issue of this unsolved mystery. Where would it be more appropriate than within this elected circle, whose interest is the culture, the religion, the literature, and the language of a very fascinating religious-ethnic group, the Mandaeans, a small community who has survived nearly two thousand years, fighting persecution and other calamities. Like the Jews they have stuck to their ancient rites and religious books, helping them through several waves of near extinction. Their literary works have been handed down in a surprising close tradition and might be the reason for their opportunity to overcome all kinds of external difficulties. Only our much-praised achievements of the modern world made such an ancient religious sect partially leave their former homeland. Its exile might lead to an end of a two thousand years history of the last remaining gnostics. The recent so-called liberation of Iraq came too late for the Mandaeans, since most of them were already driven out from one of their homesteads with other marsh inhabitants after the first Gulf war in 1990.2

1 This paper was originally presented in a modified version in the Habilitation-Colloquium at the Philolosophische Fakultät, Friedrich-Schiller-Universität, Jena, and was read again at the Sec- ond International Aram conference on the Mandaeans, Oxford 10th, July 2002. 2 A figure of 30 000 Mandaeans living around Nasariye (Ur) given in the English Newspaper The Independent on Easter Sunday 21st of April 2003 is doubtful. 48 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN

SOURCES OF THE MANDAEANS

The issue from where the Mandaeans originate and how the elements of their religious doctrine can be geographically localised, can be followed up in plenty of scholarly disputes. Some of them have been forgotten in the mean- time. The basis for most discussions were the translations of the Mandaeans’ literary corpus by the semitist and eminent epigrapher M. Lidzbarski, edited between 1915 to 1926. It was then continued from the thirties till the sixties by Lady E. S. Drower who contributed her exact observations of modern Man- daean cultural practise in Iraq and Iran which she laid down in her famous book The Mandaeans of Iraq and Iran, their cults, customs, magic, legends and folklore,3 and published other Mandaic text genres, among them The Ca- nonical Prayerbook of the Mandaean,4 astronomical ominas,5 and incantation series6. This stands in contrast to the 19th cent. scholars like W. Brandt and K. Kessler who had only the hand written copy of the Mandaean Ginza by H. Petermann at hand7, but at least a precise and so far valid grammatical treat- ment of the Mandaic language. When the only existing scholarly Mandaic dictionary was published in 1963 by Lady E. S. Drower and R. Macuch,8 the major text editions had been edited.

THE STATE OF RESEARCH CONCERNING THE “MANDÄERFRAGE”

The divergent scholarly works on the question of the Mandaeans cannot be summed up in a short paper. The reason for this are the diverse sources of their literature. Some parts of their literary major works were composed and redacted only after the Islamic conquest, e.g. the major part of the Ginza, some of them already during the final period of the Parthian rulership.9 One of their religious books, the Book of John which carries already an Arabic title drasa ∂-yahya, foreshadows an intensive dispute with the Christian religion which cannot be assumed for the older Mandaic literary stage. With the compilation of the Ginza or also called Sidra rabba, the most important religious work, presumably finalised shortly after the defeat of the Sasanians by the Arabs 3 Op. cit., (Oxford, 1937). 4 Op. cit., (Leiden, 1959). 5 Drower, E. S., The Book of the Zodiac, (London, 1949). 6 The most relevant articles are: “+afta ∂ Pi,ra ∂ Ainia”, Journal of the Royal Asiatic Soci- ety, 1937, pp. 589-611; 1938, pp. 1-20 and “A phylactery for rue”, Orientalia Nova Series, 15 (1946), pp. 324-346, since they give more hints to the Mandaean legacy than the so-called “high literature”. 7 Petermann, H., Thesaurus sive Liber Magnus vulgo “Liber Adami” appellatus, opus Man- daeorum summi ponderis, (Leipzig, 1867). 8 A Mandaic dictionary, (Oxford, 1963). 9 That certain religious concepts were already fixed can be followed up in the earliest text material, the Mandaic incantations on lead sheets and in the later text corpus, the magic bowls. The supposed first ever published Mandaic lead roll gives some good examples and was chosen on purpose from the Pir Nukraya archive by M. Lidzbarski, see n. 10. C. MÜLLER-KESSLER 49 near al-Qadisiya in 637 – when the Mandaeans were granted the status of a book religion – and helped them to obtain religious freedom under the Islamic dominance. However, one has to discriminate in the chronology of Man- daeans’ major literary works between different layers. Already M. Lidzbarski, the brilliant translator of the Ginza, had to admit that there is lack of distinc- tive indications to distinguish between older and younger textual layers. Lidzbarski considered a Mandaic incantation text on a lead strip which he published in 1909 as the oldest textual evidence, and dated it to the 4th cent. AD.10 It is more than conceivable that the oldest sections in the Mandaean literary works were composed several centuries before their actual final redaction. This fact induced wild assumptions and speculations concerning the first ap- pearance of the Mandaeans, and opinions ranged from some centuries BC up to the 6th cent. AD. All this came to a standstill with the energetic and con- tinuous publication activity of the well known “Religionshistoriker” K. Ru- dolph. According to his publications, the question of the Mandaeans seemed to be solved.11 From that time on the “Mandäerfrage” attracted hardly any attention. In principle the results and opinions of K. Rudolph go back to the reflec- tions of M. Lidzbarski who as an unparalleled translator and with his unques- tionable exemplary competence in Semitic languages, took in general an idi- osyncratic attitude towards any form of criticism. Lidzbarski favoured a West- ern origin, meaning Palestine and denied a Babylonian background. Following Lidzbarski, K. Rudolph is of the opinion that such Mandaic central terms as manda and naÒorayye (by which the Mandaeans describe themselves), and the two religious terms kus†a12 and gupna13 support enough evidence for a Chris-

10 Lidzbarski, M., “Ein mandäisches Amulett”, in Florilegium ou recueil de travaux d’éru- dition dédies à Monsieur le Marquis Melchior de Vogüé, (Paris, 1909), pp. 349-73, esp. 349. 11 Rudolph, K., Die Mandäer I. Das Mandäerproblem, (Göttingen, 1960); –, Die Mandäer II. Der Kult, (Göttingen, 1961); –, Theogonie, Kosmogonie, and Anthropogonie in den mandäischen Schriften, (Göttingen, 1965); –, “Die Religion der Mandäer”, in Gese, H., Höfner, M. and Rudolph, K. (eds.), Die Religionen Altsyriens, Altarabiens und der Mandäer, (Stuttgart, 1970), pp. 403-469; –, “Die Mandäer”, in Widengren, G. (ed.), Der Mandäismus, (Darmstadt, 1982), pp. 468-471; and many other contributions in reference works. 12 The dissimilation of two emphatics in one root in Aramaic occurs only under the influence of Akkadian, which is the case in the earliest Aramaic inscriptions (Zincirli, Sfire, Tell Fekheriye, Neirab), and in the imported administrative language from Mesopotamia to Egypt, known as Of- ficial Aramaic or Imperial Aramaic. Therefore kws†ˆ could only develop in an Akkadian phonetic environment. There are other cases where /q/ is not dissimilated to /g/, but to /k/ in Mandaic, e.g. in the Akkadian loanword k}rÒ}, kyrÒˆ < qrÒ ‘slander’, see Drower and Macuch, p. 216, and Baby- ˆ†kamÒa/ < /qamÒa/ ‘locust’. The Greek transliteration of kws/ כמצא lonian Jewish Aramaic kousta, koustijl given in Lidzbarski, M., Das Johannesbuch, (Giessen, 1916), p. XVIII, are based on the expected transliteration of Semitic languages into Greek q = k and † = t, and are no arguments for a Western origin. 13 The word gwpnˆ is attested with the same spelling in the Babylonian Talmud as well, see Sokoloff, M., A dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic periods, (Ramat-Gan, 2002), p. 271, and does not need to be imported from Palestine. A magic bowl with Mandaean religious concepts is BM 91709, published as 082M in Segal, J. B., Catalogue of the Aramaic and Mandaic incantation bowls in the British Museum with a contribution by E. C. D. Hunter, (London, 2000), pp. 108-10. 50 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN tian Jewish sect in the Jordan valley.14 A splinter group broke away and immi- grated to the East, meaning Mesopotamia. Western elements of the Mandaean doctrine were created in a very vaguely described geographical environment in Palestine or Syria. For example, the Mandaean ritual baptism, the so-called maÒbuta15 was imported from there. Also the clear Iranian and Babylonian components of the Mandaean texts belong already to the first layer, the western phase of Mandaism. In pre-Christian times they already reached Palestine and Egypt. The migration thesis is corroborated by a legendary source from the Islamic period, the Haran Gawayta legend which speaks of the sojourn of the Mandaeans in the north east of Mesopotamia, meaning Harran, in the time of the Parthian king Artaban III (12–38 AD), and also their trip through the Median hills.16 The terminus ad quem for a stay in the West would be the 2nd cent. AD, while in the 3rd cent. the Mandaeans were already present in Babylonia. In this connection the colophons of the Mandaean literary texts are always referred to. These mention as the oldest Mandaic scribe one Zazay ∂-Gawazta bar Hawa from the town ™ib, who according to the numerical counting is dated into the year 272 AD.17 Further, interrelations are pointed out between the Mandaic hymns and the Manichaean Thomas Psalms, where one can assume the integration of the Mandaean elements in the 3rd cent. AD. Another debat- able issue has been on the one hand the connection between the Nabataean script and the Mandaean one, and on the other hand the Elymaean inscriptions from Tang-e Sarvak, i.e. the western region west of the Persis next to the Per- sian Gulf.18 The wandering of the Mandaeans from Palestine or Transjordan to Harran in Upper Mesopotamia, from there even into the eastern mountainous region, and only then to Central and South Babylonia, was considered from now on as definite fact, and was not questioned any more. R. Macuch was able

14 Rudolph, K., “Probleme einer Entwicklungsgeschichte der mandäischen Religion”, in Le Origini dello Gnosticismo, Colloquio di Messina, 13–18 Aprile 1966, (Studies in the History of Religions, Suppl. of No. 12; Leiden, 1967), pp. 583-96. 15 The root from which the noun maÒbuta derives is Òb¨, Mandaic Òbˆ meaning ‘to dip, dye’ and gets a special meaning in the cleansing ceremony ‘to baptize’. The root is Common Aramaic and does not need necessarily to be imported from Palestine. It is also not an argument that Syriac employs instead ¨md. Syriac has lexically to be kept separate form the Central and South- eastern group Mandaic and Babylonian Talmudic Aramaic. 16 Drower, E. S., The Haran Gawaita, (Studi e Testi, 176; Rome, 1953); Macuch, R., “Alter und Heimat des Mandaismus nach neuerschlossenen Quellen”, Theologische Literaturzeitung, 82 (1957), pp. 401-408. 17 The colophons as historical witnesses were suggested by R. Macuch in his Handbook of Classical and Modern Mandaic, but considered as inadequate for the pre-Islamic period and the “Religionsgeschichte” by K. Rudolph, “Probleme einer Entwicklungsgeschichte”, p. 596, n. 37. 18 Concerning the Mandaic script “The origin of the Mandaic script”, in Vanstiphout, H. L. J. et al. (eds.), Scripta Signa Vocis. Studies about scripts, scriptures, scribes and languages in the Near East presented to J. H. Hospers, (Groningen, 1986), pp. 111–20, by A. Klugkist, is a sig- nificant contribution. C. MÜLLER-KESSLER 51 to make the point from that time on that no one any longer supported the thesis that the Mandaean community originated from Babylonia.19

REVISION AND NEW EVIDENCE

So far so good. Is it possible for a comparative semitist of our time to con- tribute something to the question of the Mandaeans? These days a semitist is not trained in the field of religious and historical hermeneutics. He will not have come across the term gnosis during his research or will have any desire to go deeper into this controversial and confusing discussion. But for sure, he will be able to judge the linguistic values of the repeated arguments. He will find it striking that the language elements concerning the question of the Mandaeans are only employed sporadically as secondary arguments to support the “religionswissenschaftliche” thesis and are taken as finalised. From the methodological point of view he will find fault with the proce- dure that the origin of the Mandaeans has hardly ever been distinguished from the actual background of the doctrine. The Aramaic scholar K. Beyer wrote in his wide spread opus magnum with its title Die Texte vom Toten Meer or in the English translated version The Aramaic Language in a more concrete fashion about the language situation of the Mandaean Aramaic: “Mandaic was adopted by the Nasoraeans, a gnostic/ baptist community, after they had left Palestine (Jordan area) in the 1st cent. A.D. as a result of the hostility of contemporary Judaism and had migrated at the latest in the middle of the 2nd cent. AD via northern Mesopotamia (Harran/Charrhae) to southern Babylonia (Mesene, Khuzistan). There, to judge from the Mandaic script (the script of the 2nd cent. AD Elymais inscriptions is closely related to it: and orthography, their oldest poems were recorded during the Arsacid period and probably in Arsacid Aramaic” – Beyer means by Arsacid Imperial Aramaic –, “from which after 224 AD they were gradually rendered into central south eastern Aramaic, which had meanwhile been adopted by the Mandaeans.”20 K. Beyer like my mentor in Semitics R. Ma- cuch were first of all trained theologians who found their way to Semitic studies in a roundabout way. It was natural for them to adjust the language devel- opment to the theological opinion of the “Religionswissenschaft”, and even to subordinate it. This means in other words that when the first Mandaeans developed their early rites and doctrines at the beginning of the 1st cent. AD, they composed and canonised them in a Western Aramaic idiom. Their following stay in Harran in Upper Mesopotamia where a central Aramaic dialect, Syriac,

19 Macuch, R., “Anfänge der Mandäer”, in Altheim, F. and Stiehl, R. (eds.), Die Araber in der Alten Welt II, (Berlin, 1965), pp. 76-190. 20 Beyer, K., Die aramäischen Texte vom Toten Meer, (Göttingen, 1984, 1994); –, The Ara- maic language, (Göttingen, 1986), pp. 45-46. 52 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN was in use, did not leave any linguistic traces. Then, when they reached Babylonia they recorded their literature in Arsacid Imperial Aramaic and after fifty years they translated their text corpus into a Central and Southeastern Babylonian Aramaic dialect. I think that it is not necessary to point out that such a chameleon-like frequent change into different Aramaic forms of dialects in such a short period of time is more than unlikely. To the best of my knowledge I am not aware of any similar case within the Semitic language field. In this instance of comprehensive translations from one Aramaic dialect into another, one finds always traces of grammatical and lexical features of the “Vorlagentexte” or of the linguist background of the scribe. Such indications are missing from the Mandaic text sources. Having a closer look at the Mandaic texts the linguistic shortcomings within the discourse of the issue of the Mandaeans are very striking. There have never been any attempts to ana- lyse the oldest text material, the Mandaic incantation corpus in relation to the so-called classical Mandaic “Hochliteratur” or to start reading and editing the vast sources in museums and collections. It has been restricted to quoted re- marks by Lidzbarski that the Mandaic lead rolls contained old features or that these texts are related to the popular language. Such arguments have never been systematically and closer investigated. To stay with the diction of K. Ru- dolph, the Mandaic incantation corpus belongs to a sunken higher religion. The epigraphic shortcomings in the publications of Mandaic incantations, their late copies from the end of the Sasanian period 4th-7th cent. made them appear unintelligible. It has not helped to draw to them, the attention they deserve in my opinion, since one can trace plenty of elements of the basic doctrines. The low opinion which can be found in frequent statements concerning the Mandaean incantation material, reflects according to my point of view the part of the Mandaean written corpus that has not been approached in the way it deserves.21 The Mesopotamian and Iranian elements of the Mandaean texts, and their placement are of major importance when it comes to the question of the Mandaeans’ homeland. The early magic corpus shows variants of several morphems which appear in the later Mandaeans text in different forms.22 The

21 Buckley, J. J., Review of Gündüz, S., The Knowledge of Life, (Journal of Semitic Studies Monograph, 3; Oxford, 1993), in Journal of the American Oriental Society, 116 (1996), pp. 301- 302. 22 Apart from the spelling of Mand¨a ∂-Hiyya with an etymological ¨ which is referred to often as a sign of an older Mandaic text, see Lidzbarski, “Ein mandäisches Amulett”, p. 349, the new magic corpus provides us with more details. Suddenly forms turn up which are later aug- mented by ˆ- like in Babylonian Talmudic Aramaic which is also of later date, one comes across twtyˆ 2Ac40 (BM 135793 II), 13Aa37 (BM 135791 I), 1Ba58 (BM 132947+); BM 91715:II2, BM 91724:7, BM 91748:10, BM 91708:7, BM 91777:14, 15, 17, BM 132168:6/7 instead of later ˆtwtyˆ or the preposition mn†wl ‘on account of’ 1Ba209 (BM 132947+), 6Ba15, 58, 67, 70 (BM 132948), the conjunction mn†wl ∂- ‘because, since’ 2Ab20 (BM 135793 II), 1Ba207 (BM 132947+), 6Ba15, 58, 67 (BM 132948), 5Ba32 (BM 132955+), for the later variant in bowls ˆmyn†wl ∂- 5L12 (AO 2629), YBC 2364:8, 23; the until now unattested demonstrative pronouns instead of the later contracted hˆk, hˆzˆk ‘that (m.)’, hzˆk 1Aa68 (BM 135793 I), hˆzˆk 6Ba14 (BM 132948); hˆzyk ‘that (f.)’ in the correlative hˆzyk ∂- 13Aa42 (BM 135791 I). C. MÜLLER-KESSLER 53

Mandaean, or better the Aramaic incantation type does not express the popular substrate or any kind of popular language. I can state here again that we do not have any kind of idea of the colloquial language type of this Late Antique period. We are dealing here with pure literary idioms. Mandaic incantations are works of literature. With the Akkadian, here the Babylonian incantation type, they belong to a specific literary level with its own vocabulary and literary style. Despite their external Akkadian influences, an independent Aramaic incantation type existed for several centuries, running parallel to the Akkadian one. Such low opinion of magic can be found as well in a recent publication of the British Museum Collection of magic bowls by J. B. Segal: “The incantation texts are written in a popular language. We should not, then, expect them to conform to the standard norms of grammar.”23 or “Neither in their date nor in their quantity have they much significance when set within the panoramic range of Mesopotamian Aramaic literature”.24 In a just published article I tried to point out that the unique Hellenistic Aramaic incantation from Uruk, South Mesopotamia, written in cuneiform, can be taken as a typical prototype of the Mandaic incantation type or better Aramaic incantation type as to be found in the published and also unpublished Mandaic lead rolls, and bowls, and some- times in Aramaic square script incantations of pre-Islamic time. In this first attested Aramaic incantation dated ca. to 150 BC, a demon story in the first person singular is related and shows already a few Mandaic linguistic idioms or better Central and Southeastern Babylonian Aramaic ones.25 At the time of the 2nd cent. AD when the Mandaeans dwelt for sure in Babylonia like other gnostic sects, e.g. the Elchasaites and when Christianity began to spread over the Near East, the traditional Babylonian temples were still functioning in Central Babylonia, especially in towns like Babylon, Borsippa and Kutha. Among the respected sciences of the time Mesopotamian incantation literature and astronomy belonged next to medicine. According to the information of the Babylonian Talmud Jewish Rabbis discussed with Ba- bylonian scholars in a period of intellectual exchanges and the flourishing of the economy in Babylonia.26 Even the Mandaean scribes of this period belong to the setting. Mandaic incantations, astronomy tractates, and omens as in the 23 Segal, Catalogue of the Aramaic and Mandaic incantation bowls, p. 30. A similar point of view is taken in Beyer, Aramaic language, p. 46 “The magic texts on rolls of lead and bowls (4th–7th cent. A.D.) are closer to the colloquial”. Neither statements can be proven, since the colloquial idioms of that period are not known to us. 24 Segal, J. B., “Review of Yamauchi, E. M., Mandaic incantation texts, (American Oriental Series, 49; New Haven, 1967)”, in Bulletin of the American Schools of Oriental Research, 33 (1970), p. 609. 25 Müller-Kessler, C., “Die aramäische Beschwörung und ihre Rezeption in den mandäisch- magischen Texten am Beispiel ausgewählter aramäischer Beschwörungsformulare”, in Gyselen, R. (ed.), Magie et magiciens, charmes et sortilèges, (Res Orientales XIV; Louvain, 2002), pp. 193-208. 26 Greenfield, J. C. and Sokoloff, M., “Astrological and related omen texts in Jewish Palestin- ian Aramaic”, JNES, 48 (1989), pp. 201-14 [213]; Geller, M. J., “The last wedge”, Zeitschrift für Assyriologie, 87 (1997), pp. 56-57. 54 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN later compiled book asfar malwasia “The Book of the Zodacial Signs” have not been considered in this issue as Mandaic high literature, only the Mandaic religious texts. However, in that time they belonged to the every day life of the Mandaean community and other groups, and represented the intellectual knowledge of scribes and priests.27 The individual groups of the Mandaeans’ magic corpus developed in different periods. Chronologically, they can be judged according to their different layers of orthography, grammar, and contents. It is for sure that the first material goes back to the second or the first half of the 3rd cent. AD, meaning that it is to be dated into the late Parthian period.28 This can be assumed by the mentioning of the Istar of Hatra in one lead roll belonging to the Pir Nukraya archive, housed today in the British Museum, which is on the brink of being published.29 The Parthian city Hatra was destroyed by the Sasanian ruler Sabuhr I. towards 249 AD, and was never populated again.

LEGACY OF BABYLONIAN DEITIES IN MANDAEAN SOURCES

At the beginning of the 20th century H. Zimmern a scholar, in Assyriology and Semitics, looked for Mesopotamian elements in the Mandaic doctrine. He wanted to see a connection between the baptismal rituals of the Mandaeans and the Akkadian lustration rituals performed for the deities Ea and Enki.30 Such less convincing attempts were already an ideal base for M. Lidzbarski, and subsequently K. Rudolph to minimise the Mesopotamian elements in the major literary works of the Mandaeans, and to look for alternatives outside the geographical boundaries of Babylonia. For example the role of the seven Ba- bylonian planet deities Bel, Nabu, Nerig, the Babylonian Nergal, Sin, Samis, Istar-Delibat, and Kewan, the Babylonian Kajjamanu is noteworthy. For Rudolph and others these gods have been only a sign of an older “Baby- lonischer Hellenismus”, whose elements were transferred to the West some time ago. It has always been pointed out that several of the deities mentioned can also be traced outside of Babylonia, since Bel and Nabu were worshipped

27 Drower, The Book of the Zodiac. And details on the placing of such texts, see Müller- Kessler, C., “Aramäische Beschwörungen und astronomische Omina in nachbabylonischer Zeit. Das Fortleben mesopotamischer Kultur im Vorderen Orient”, in Renger, J. (ed.), Babylon: Focus mesopotamischer Geschichte, Wiege früher Gelehrsamkeit, Mythos in der Moderne. 2. Inter- nationales Colloquium der Deutschen Orient-Gesellschaft 24.–26. März 1998 in Berlin, (Saar- brücken, 1999), pp. 427-43. 28 A student of K. Rudolph has recently pointed this out for the polemic concerning the practise of the sacrifices, see Koch, C., “Der religionsgeschichtliche Hintergrund der Opferpolemik im mandäischen Ginza”, in Flasche, R. (ed.), Religionswissenschaft in Konsequenz, (Marburger Religionswissenschaftliche Beiträge, 1; Münster, 2000), pp. 81-95. 29 See 2Ba78 [= BM 132956+] in Müller-Kessler, C., Incantations for the House of Pir Nukraya. Mandaic lead rolls in the British Museum. Part I, (Leiden; in preparation). 30 Zimmern, H., “Das vermutliche babylonische Vorbild des Pehta und Mambuha der Mandäer”, in Bezold, C. (ed.), Orientalische Studien. Theodor Nöldeke zum siebzigsten Geburts- tag gewidmet von Freunden und Schülern II, (Giessen, 1906), pp. 959-67. C. MÜLLER-KESSLER 55 in Northern Syria and Upper Mesopotamia, meaning Palmyra, Edessa and one of them till Late Anquity31 even till the early Arabic period in the case of the moon god Sin in Harran.32 Contemporaneous situations in Babylonia have been constantly neglected. However, one has to come to a radical reconsidera- tion of this very issue. Analysing the Mandaean texts, one can find that the demonizied Mesopotamian deities, which are also mentioned in the later Mandaic text corpus prove without doubt that they were recruited from a Late Parthian cultural setting of Central Babylonia. In the case of two prominent female goddesses, Mulit and Mammitu, the first is the Mylitta of Herodot I, 131, a later demonic figure appearing in certain sequences of former Babylo- nian gods in Mandaic incantation formulas, lists which are not identical to the always cited seven planetary deities sequence. An entry mulita is already in the Mandaic dictionary without recognition.33 The identification of Herodot’s Mulit is due to two Assyriologists S. Dalley and S. Parpola who noticed about the same time that the correct reading of the Assyrian goddess Ninlil is Mulissu. With the help of a poetic word play in a cuneiform text published in Revue d’Assyriologie Karlheinz Kessler and I could predict a Babylonian Mulit as well.34 It goes without saying that Herodot referred to a Babylonian Mulit cult and not an Assyrian one. The second goddess I would like to draw attention to is Amamit, the Ara- maic variant spelling of the Babylonian Mammitu – the latter a Mandaic figure of the Netherworld in the Ginza – comparable to Babylonia where she is one of the spouses of the Netherworld god Nergal since Old Babylonian times (18th cent. BC). Recently, K. Kessler was able to emend a cuneiform passage in the astronomical diaries from the Hellenistic period on the basis of the background of the early Mandaic texts – where imagination started to run wild among Ancient Historians specializing in Hellenistic cuneiform to interpret this line.35 It becomes clear that both deities were only at home in Babylonia and never crossed the boundaries of Babylonia in that time.36 Perhaps one may go one step further and suggest another identification of a Babylonian goddess called Tarbusnita who occurs with the previous men-

31 Rudolph, K., Die Mandäer I, pp. 203-204, where he claims “Aus aramäischen Urkunden in Assur erfahren wir, daß noch im 3. Jh. n. Chr. babylonisch-assyrische Gottheiten verehrt wurden. Aber alle diese Zeugnisse zählen nicht mehr zur eigentlichen babylonischen Religion, sondern gehören in den „babylonischen Hellenismus“, (Schaeder), wie aus den Ausgrabungen in Palmyra, Dura-Europos hervorgeht wo Nabu, Bêl, Bêlti, Nanai, Tammuz und die Planetengott- heiten neben den syrischen, arabischen, griechischen u.a. Göttern begegnen.” 32 Gündüz, S., The Knowledge of Life: The origins and early history of the Mandaeans and their relationship to the Sabians of the Qurˆan and to the Harranians, (Journal of Semitic Studies Monograph, 3; Oxford 1993), pp. 192-207. 33 Drower and Macuch, Mandaic dictionary, p. 261. 34 Müller-Kessler, C. and Kessler, K., “Spätbabylonische Gottheiten in spätantiken mandäi- schen Texten”, Zeitschrift für Assyriologie, 89 (1999), pp. 65-87, esp. 70-72. 35 Scurlock, J., “167 BCE: Hellenism or reform?”, Journal for the Study of Judaism, 31 (2000), pp. 125-61. 36 Müller-Kessler and Kessler, “Spätbabylonische Gottheiten”, pp. 80-84. 56 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN tioned Babylonian deities in the same list. It is an account which is already known from the unpublished Drower Collection no. 43 Aa42/43 and was presented by A. Caquot within the publication of an earlier variant of the same formula on a lead roll in 1972. It reads ll. 29-34: ¨syr nbw w¨syr nyryg w¨syrˆ nnˆy wsdyˆ w¨syrˆ ¨strˆ w¨syrˆ ˆmˆmyt w¨syrˆ trbwsnytˆ w¨syrˆ mwlyt “bound is Nabu and bound is Nerig and bound is Nanay and Sadya and bound is Istar37 and bound is Amamit and bound is Tarbusnita and bound is Mulit”.38 Another passage in the early Mandaic demon list on an unpublished lead roll in the British Museum gives the two goddesses a geographical setting: ¨syrˆ ¨strˆ trbwsn}ytˆ wmwlyt ∂-nhˆr kmˆr “bound is the goddess Tarbusnita and Mulit of the river Kamar” 1Ba155–156 (= BM 132947+).39 From the spelling point of view Tarbusnita is not reminiscent of any former Babylonian deity and cannot be taken here as an Iranian one, since she is mentioned in the group of already well established Babylonian deities, except for Sadya.40 Therefore, the only possible Babylonian goddess surviving into the late Hel- lenistic period who comes close to Tarbusnita might be Z/∑arpanitu, the spouse of the city god Bel-Marduk in Babylon who is also known under her other name Belti.41

LEXICOGRAPHIC ARGUMENTS

Another shortcoming concerning the question of the Mandaeans is the Mandaic lexicography which can be only tackled from the Semitic point of 37 It should be pointed out that the name of the deity Istar is rarely mentioned in her function as a goddess in Mandaic texts, since she appears more frequently under her hypostase Delibat or other terms, e.g. Qaddista. The noun ¨strtˆ or ˆ(y)strtˆ is only generic. The occurrence of Istar is restricted to incantation formulars which took the demon names from the great demon account and to the demon list itself referring to various Istar cults, ¨strˆ rbtyˆ ∂-… “the great Istar of …” CBS 2941:10; ¨strˆ ∂-ˆkˆt “Istar of Akkad” BM 91724:6; <¨st>rˆ rbtyˆ ∂-krkˆ b¨rtˆ rbtyˆ ∂-hwÒyˆ “the great Istar of Karka, the great capital of Khuzistan” 2Ba:1 = BM 132956+, ¨strˆ ∂-b¨rtˆ rbtyˆ ∂-yˆtbˆ {b} bdbr} ¨syrˆ ¨strˆ rbtyˆ ∂-byt ˆbwgdˆnˆ “Istar of the great capital who dwells in the desert; bound is the great Istar of Bit Abugdana” 2Ba:62–64; ¨stˆr h†ryˆ wgwkˆyˆ The Ha†rian and Gaukian Istar” 1Bc:76–77 = BM 132947+ and a few others in Müller-Kessler, Incantations for the House of Pir Nukraya. See also Fauth, W., “Lilits und Astarten in aramäischen, mandäischen und syrischen Zaubertexten”, Welt des Orients, 17 (1986), pp. 66-93, esp. 70-71. 38 Caquot, A., “Un phylactère mandéen en plomb”, Semitica, 22 (1972), pp. 67-87, esp. 77. 39 Müller-Kessler, Incantations for the House of Pir Nukraya. 40 Preliminarily, see Müller-Kessler, C., “A Mandaic gold amulet in the British Museum”, Bulletin of the American Schools of Oriental Research, 311 (1998), pp. 83-88. 41 See Müller-Kessler and Kessler, “Spätbabylonische Gottheiten”, pp. 69-70. Depending on the cuneiform sign value zar or Òar of the first syllable one will have to predict that the Aramaic reading of Z/∑arpanitu might have been ™arpanitu, since /Ò/ and /†/ are phonetically closely related as in the well established sound shift from Akkadian and Hebrew /Ò/ to Aramaic /†/. That Tarbusnita is written with voiceless /t/ might be based on the fact that one did not distinguish any more at that time between voiceless or emphatic sounds like e.g. in ¨sqwptˆ or ¨skwptˆ ‘threshold’. That leaves still the unexplained additional s which makes the suggested solution nearly impossi- ble. For the time being it sounds quite fantastic and I ventured to suggest it only in a non Assyriological circle. One should leave it as a possible explanation of this unique former deity name Tarbusnita in the Mandaic sources. C. MÜLLER-KESSLER 57 view in relation to the neighbouring dialects and languages of that geographical area. The influence of the Akkadian-Aramaic culture to which I drew attention before can be traced abundantly in the lexical environment of the Cen- tral and Southeastern Babylonian Aramaic dialects. Next to very specific Akkadian loanwords sˆptˆ ‘incantation, here: incantation series’, pysrˆ ‘solving of an exorcism’ for the magic issue42 or nˆndbyˆ ‘sacrifices’ for the question of the cult, one comes across idiomatic phrases which can only be understood through the knowledge of Akkadian or such an idiom like rkybnˆ ¨l ywmˆ “I am riding on the storm” and would be translated in the Aramaic understanding “I am riding on the day”.43 Such lexical items dominate in the early Mandaic incantation corpus and show the influential heritage of Akkadian and of early Iranian stages on the Mandaic language. The lack of Greek loanwords noticed before is also striking, but when it came to evaluation it was judged differ- ently. A vast lexical gap separates Syriac around Edessa and Harran, which is rich in Greek terms from the Central and Southeastern Babylonian Aramaic. A big difference is also noticeable in the linguistic features. All this makes it dif- ficult to explain why a group which comes first from Palestine or Transjordan, resides then in Harran, in the neighbourhood of Edessa, does not show more Greek words or lexical items in their early compiled text corpus from their former places of residence. That the essential elements of the Mandaean doctrine developed parallel to the Qumran Aramaic text corpus does not help to solve the problem either. The study of the earliest Mandaic vocabulary leads very often to the central issues of the Mandaean doctrine. To pick out just one example, one can take the Mandaean baptism ritual, the maÒbuta which is always considered to be the basis of Mandaean doctrine, but cannot be traced in the early text corpus at all. The verbal root Òby ‘to baptize’ does not occur like the river Jordan. The latter appears only in texts that are considered late according to criteria of contents, grammar and lexicography. Even the part of the Mandaean hymns from the 3rd cent. AD which appear in the Manichaean Thomas Psalms do not mention the maÒbuta and the river Jordan, as one editor admitted.44 Therefore, one has to question in how far is the maÒbuta of major importance in comparison to other cultural lustration rites and can be considered to belong to the original doctrine of the Mandaeans at all? It is conceivable that the cultural importance came up as competition for the rising Christianity in the East. According to the early Mandaic sources, there is no doubt that the animosity towards Jews and

42 Müller-Kessler, “Aramäische Beschwörungen”, p. 431. Also rightly pointed out in Ford, J. N., “Notes on the Mandaic incantation bowls in the British Museum”, Jerusalem Studies in Arabic and Islam, 26 (2007), p. 250. 43 Müller-Kessler, “Aramäische Beschwörungen”, pp. 433-35. 44 Adam, A., Die Psalmen des Thomas und das Perlenlied als Zeugnisse vorchristlicher Gno- sis, (Beihefte zur Zeitschrift für die neutestamentliche Wissenschaft, 24; Berlin, 1949), pp. 39- 41, 76-80. 58 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN

Christians is of later date, that is added by the animosity of Islam and cannot be taken as belonging to the early doctrine.

LINGUISTIC ARGUMENTS

Similar shortcomings next to the lexicographic and Babylonian elements exist for the Iranian influences on the Mandaic sources. That the long rule of the Sasanians must have left Iranian traces in contents and language has already been pointed out by G. Widengren. What particular part the Parthian loanwords take in the vocabulary of Mandaic, is not yet clear. Indications to particular Parthian spellings and forms show that one needs to go deeper into it.45 I would like to draw attention to the fact that S. Shaked could explain unsolved safel forms and other questionable words appearing in the Mandaic dictionary with convincing Iranian etymologies.46 The as yet unpublished early Mandaic incantations contain plenty of unknown theophorous Iranian proper names from the contemporary Sasanian period. And finally one should come to a clear definition of the older pre-Parthian part in the Mandaic text corpus which is predicted by the Religionswissenschaft for a geographical localisation in the West, but from the linguistic point of view has never been proved con- vincingly. In my opinion it does not even exist. Concerning the Aramaic dialect itself, in Mandaic the historical spellings of early Aramaic q and z have survived. q is now attested in the Aramaic Uruk cuneiform incantation from the Seleucid period, a prototype of the Aramaic incantation type.47 The phonetic state of Mandaic is totally overlaid by the breakdown of all the gutturals in the Akkadian script. The direct object is indi- cated always by the proclitic preposition l_ since Imperial Aramaic. These dia- lectal phenomena except the historical spellings Mandaic shares with Babylo-

45 Concerning the Iranian element, i.e. loanwords, this became much clearer recently with the updated publication of the Babylonian Talmudic vocabulary in Sokoloff, Dictionary of Jewish Babylonian Aramaic. Here the contemporary Middle Persian layer dominates whereas the Mandaic sources tend to have an earlier substrate of Parthian forms. The Babylonian language influence prevails in the Mandaic texts. These recent insights require further investiga- tions. 46 Shaked, S., Appendix in Greenfield, J.C. and Naveh, J., “A Mandaic lead amulet with four incantations [Hebrew]”, Eretz Israel, 18 (1985), pp. 106-107, e.g. shrz, srhz = MIran. passive participle sahrid ‘to scare, tremble’; p¨†yˆrˆ = MIran. petyarag ‘enemy, adversary’; ¨s†wnˆ = MIran stunag ‘Säule’; prhz or phrz = MIran. pahrez- ‘to keep off, prevent, abstain’, further Sokoloff, Dictionary, 928-29; prˆmˆnˆ = MIran. framan ‘decree’ = Babylonian Talmudic Aram- which was already attested as pwrmˆnˆ and pˆrmˆnˆ in Drower and Macuch, Mandaic הרמאנא aic dictionary, pp. 368–69, 378, and was overlooked by Shaked. Unfortunately, this rather useful article is hardly mentioned, but I had plenty of opportunity to go into it in my review article on the Catalogue of the Aramaic and Mandaic magic bowls from the British Museum, see Müller- Kessler, C., “Die Zauberschalensammlung des British Museum”, Archiv für Orientforschung, 48-49 (2001-2002), pp. 115-45. 47 Müller-Kessler, “Die aramäische Beschwörung und ihre Rezeption in den mandäisch- magischen Texten”, pp. 193-201. C. MÜLLER-KESSLER 59 nian Talmudic Aramaic, the latter is of a later language stage. 48 Since there are no western traces in the Mandaic idiom, a Palestinian origin is more than doubtful.

CONCLUSIONS

To summarize and conclude one can say that what a learned Mandaean scribe and priest knew in the Parthian times remains to be revealed by the still unpublished lead roll archives in the British Museum which provide us with plenty of new information. The Mandaeans took up and transmitted knowledge which was still around when the last temples in the Central Babylonian triangle Babylon, Borsippa and Kutha were still functioning. The Pir Nukraya archive of which I selected a few passages for the Mesopotamian magic vol- ume contains two similar demon lists with an account of mostly former demonized deities of Babylonian or Iranian provenances. Some of those cults are unknown and have never been mentioned in later Syriac, Arabic and Mandaic sources. Apart from these new gods, all kinds of geographical locali- ties in Khuzistan and Babylonia to Upper Mesopotamia are mentioned, but Harran is not among them.49 Another archive of lead rolls written for the family of Mah-Adur-Gusnasp of which I have published a few specimens (mostly only single magic stories) includes a long ritual after a unique Mandaean demon historiola.50 When I presented this ritual to cuneiform scholars in the University Museum of Pennsyl- vania, I was asked during the discussion, whether I had been translating an Aramaic or an Akkadian text, since the arrangement and contents is so close to what we are acquainted with from some Akkadian incantation series, e.g. sa.zi.ga. There occur termini technici which can only be understood through a profound knowledge of the Akkadian magical literature. Other texts in preparation like the Kelsey Museum lead roll contains an intriguing charm against

48 On the dialect distribution in Central and South Babylonia, see now Müller-Kessler, C., “The earliest evidence for Targum Onqelos from Babylonia and the question of its dialect and origin”, Journal for the Aramaic Bible, 3 (2001), pp. 181-98; –, “Die Stellung des Koine- Babylonisch-Aramäischen auf Zauberschalen innerhalb des Ostaramäischen”, in Nebes, N. (ed.), Neue Beiträge zur Semitistik (Jenaer Beiträge zum Vorderen Orient, 5; Wiesbaden 2002), pp. 91- 103. 49 Müller-Kessler, C., “Interrelations between Mandaic lead rolls and incantation bowls”, in T. Abusch, T. and Toorn, K. van der (eds.), Mesopotamian magic, (Ancient Magic and Divina- tion, 1; Groningen, 1999), pp. 197–209. The dependent magic bowl texts from the British Mu- seum were transliterated in a corrected version in Müller-Kessler, “Die Zauberschalensammlung des British Museum”, pp. 134, 136–38. 50 Müller-Kessler, C., “Aramäische Beschwörungen und astronomische Omina”, pp. 432-35, 439-43; –, “Die aramäische Beschwörung und ihre Rezeption in den mandäisch-magischen Texten”, pp. 204-205; –, “A Charm against demons of time”, in Wunsch, C. (ed.), Mining the Archives. Festschrift for Christopher Walker on the occasion of his 60th birthday, (Babylonische Archive, 1; Dresden, 2002), pp. 183-89. 60 THE MANDAEANS AND THE QUESTION OF THEIR ORIGIN death, and a lead roll from the Ligabue Collection, Venice throws light on the meaning of the often discussed demon name shrˆ. Although not of Akkadian origin, it shows an alternative usage for the name of the moon god Sin and the word for moon shrˆ as in Late Akkadian texts.51 Even a small corpus of magic texts in Aramaic square script on bowls drew their texts from Mandaic Vorlagen 52 which means that the Mandaic texts must have been finalized at the end of the Parthian Period. It seems that the Mandaean magic text corpus brings us nearer to the solving of the origin of the Mandaeans. Apart from the Mandaeans’ awareness of Ba- bylonian magic and astronomy, their Aramaic idiom can only have developed in Central Babylonia. Finally, one can now state that the Mandaeans recruited from an Aramaic population in Babylonia and therefore, could transmit information for which we still have gaps in the Late Babylonian cuneiform sources. With the help of the editions of new text material from both language areas we shall be able to close these gaps in the near future and prove far more satisfactorily the question of the Mandaeans’ Heimat.

Addenda et Corringenda to Müller-Kessler, C., “Phraseology in Mandaic incantations and its rendering in various eastern Aramaic dialects. A collection of magic ter- minology”, Aram, 11–12 (1999–2000), pp. 293–310: n. 13, read: no apocope of /l/, /n/, /r/; metathesis in certain words with glottal /h/; n. 18, read: See n. 8; p. 304, 1. 11, read: “and they drink blood of bones; n. 46, read: Lidzbarski; p. 306, l. 18, read: with those three mysteries; p. 310, l. 7, read: independ- ently.

51 Müller-Kessler, C., “Mehr zu den Mondämonen Sidrus-Sira und Sin-Dew”, Orientalia Nova Series [in preparation]. 52 Kessler, C., Die Stellung des Koine-Babylonisch-Aramäischen innerhalb des Aramäischen. Kontext, Texte, Grammatik. Habilitationsschrift der Philosophischen Fakultät der Friedrich- Schiller-Universität Jena, (Jena 2002). Will be printed in a shortened version as Müller-Kess- ler, C., A Handbook of magic bowls in Koiné Babylonian Aramaic, (Semitic Handbook Series I; Leiden).