INTRODUCTION

Fragment M 4 consists of two double sheets containing both Middle Persian and Parthian texts. It was published by Müller¹ and republished by Salemann,² both times with the sheets in the wrong order. The mistake was corrected by Reitzenstein, who made a detailed study of the Parthian text.³ This is of an unusual character, being made up of the first line or lines of a series of hymns, grouped according to subject matter. The number of lines in a group varies. Each group has its own heading, but some of the headings are themselves obscure.⁴ Two can, however, be readily interpreted: 'zg'myg b'š'h'n' ('death hymns') and fršygyrdyg b'š'h ('end-of-the-world hymn(s)'). From these, Reitzenstein inferred that the whole text was a death-mass for the World-Soul, whose deliverance from matter 'only becomes complete with the end of the world' (original: "erst mit dem Ende der Welt vollständig wird").⁵ He explained its unusual character by supposing it to be an abstract of a larger work, namely, the hymn-cycle Angad Rōšnān. The title of the sixth (and last) section of M 4 is nys'r'd 'ngd rwšn'ny b'š[']: 'Begun the hymn(s) connected with Angad Rōšnān.' Beneath this only one line is preserved, which is the opening line of the hymn-cycle. Nothing else survives to show the nature of the angad rōšnānī hymns or their relation to the cycle itself. It was on this evidence that Reitzenstein sought to establish a close connection between the whole of M 4 and the handām text.

By chance, two of the longest fragments of the hymn-cycle available to Reitzenstein bear the title 'Sixth Limb—Angad Rōšnān'.⁶ He assumed, therefore, that the title Angad Rōšnān was peculiar to the sixth section both of the hymn-cycle and of M 4;⁷ and thus that each section of M 4 corresponded with a canto of the hymn-cycle. The cantos he believed to be composed of sets of short hymns;⁸ and he therefore further supposed that each hymn was

¹ H.-R. ii, pp. 49 ff.
² Man. Stud., pp. 4 ff.
³ In these pages, M 4 is used with reference to the Parthian section of the fragment only.
⁴ Thus the title of tngy'nyg b'š'h'n ('body-soul hymns') is recorded in only one other place (see Henning, BBB., p. 47, text d); and those of gy'nyg b'š'h'n ('soul hymns') and mwqr'nyg b'š'h occur only here. The literal meaning even of the latter is unknown. The lines beneath it appear to have a Semitic original (see M. Lidzbarski, 'Ein manichäisches Gedicht' original: "A Manichaean Poem", N.G.G.W., 1918, pp. 501 ff.).
⁵ Erl. Myst., p. 18.
⁶ These are M 96 and M 175.
⁷ This assumption was particularly rash in the case of M 4, which is a fragment without beginning or end. It is unlikely, therefore, that the numbering of the sections as they survive is significant. This was pointed out by Lentz; see W.-L. i, p. 67.
⁸ Reitzenstein was misled by the fact that on M 91 and M 175 there is a