06 - GREEK TRANSCRIPTIONS OF THE TETRAGRAMMATON.

VI.

GREEK TRANSCRIPTIONS OF THE TETRAGRAMMATON.

καὶφοβηθήσονταιτὰ ἔθνητὸὄνομάσουκύριε. IN a notice of Professor W. Dindorf’s edition of Clement, Professor P. de Lagarde¹⁰⁵⁸ reproaches the editor, in reference to the passage Strom. v. 6 34 (Dindorf, p. 27 25), with having “no idea whatever of the deep significance of his author’s words, or of the great attention which he must pay to them in this very passage”. Dindorf reads there the form Ἰαού as τὸτετράγραμμονὄνοματὸμυστικόν. But in various manuscripts and in the Turin Catena to the Pentateuch¹⁰⁵⁹ we find the variants Ἰὰοὐαί or Ἰὰοὐέ.¹⁰⁶⁰ Lagarde holds that the latter reading “might have been unhesitatingly set in the text; in theological books nowadays nothing is a matter of course”. The reading Ἰαουέ certainly appears to be the original; the ε was subsequently left out because, naturally enough, the name designated as the Tetragrammaton must have no more than four letters.¹⁰⁶¹ The form Ἰαουέ is one of the most important Greek transcriptions of the Tetragrammaton usually referred to in seeking to ascertain the original pronunciation. F. Dietrich in a letter of February, 1866,¹⁰⁶² to Franz Delitzsch, makes the following collection of these transcriptions:—

It is an important fact that nearly all the transcriptions which have thus come down from the Christian Fathers are likewise substantiated by “heathen” sources. In the recently-discovered Egyptian Magic Papyri there is a whole series of passages which—even if in part they are not to be conceived of as transcriptions of the Tetragrammaton—merit our attention in this connection. As early as 1876 W. W. Graf Baudissin,¹⁰⁶⁵ in his investigation of the form Ἰάω, had referred to passages relating to it in the Magic Papyri in Leiden¹⁰⁶⁶ and Berlin.¹⁰⁶⁷ Since that time the edition of the Leiden Papyri by C. Leemans,¹⁰⁶⁸ and that of the Paris and London Papyri by C. Wessely,¹⁰⁶⁹ the new edition of the Leiden Papyri by A. Dieterich,¹⁰⁷⁰ the latest publications of the British Museum,¹⁰⁷¹ and other works, have rendered still more possible the knowledge of this strange literature, and an investigation of these would be worth the trouble, both for the historian of Christianity¹⁰⁷² and for the Semitic philologist.¹⁰⁷³ The Papyri in their extant form were written about the end of the third and beginning of the fourth century A.D.; their composition may be dated some hundred years before —in the time of Tertullian.¹⁰⁷⁴ But there would be no risk of error in supposing that many elements in this literature belong to a still earlier period. It is even probable, in view of the obstinate persistence of the forms of popular belief and superstition, that, e.g., the books of the Jewish exorcists at Ephesus, which, according to Acts 19:19, were committed to the flames in consequence of the appearance of the Apostle Paul, had essentially the same contents as the Magic Papyri from Egypt which we now possess.¹⁰⁷⁵ In the formulae of incantation and adjuration found in this literature an important part is played by the Divine names. Every possible and impossible designation of deities, Greek, Egyptian and Semitic, is found in profuse variety, just as, in general, this whole class of literature is characterised by a peculiar syncretism of Greek, Egyptian and Semitic ideas. But what interests us at present are the forms which can in any way be considered to be transcriptions of the Tetragrammaton. For the forms which are handed down by the Fathers, in part still questioned, are all verified by the Papyri, with the sole possible exception of Clement’s Iaove.

Ιαω. To the examples given by Baudissin there is to be added such a arge number from the Papyri since deciphered, that a detailed enumeration is unnecessary.¹⁰⁷⁶ The palindromic form ιαωαι¹⁰⁷⁷ is also frequently found, and, still more frequently, forms that seem to the author to be combinations of it, such as αρβαθιαω.¹⁰⁷⁸ The divine name Ιαω became so familiar that it even underwent declension: εἰμὶθεὸςθεῶνἁπαντωνιαων σαβαωθαδωναια[βραξ]ας (Pap. Lugd. J 384, iii. 1).¹⁰⁷⁹

Ια.

Likewise not infrequent. Without claiming exhaustiveness we cite the following:—

ὁἐπὶτῆςἀνάγκηςτεταγμένοςιακουβιαιαωσαβαωθ αδωναι [α]βρασαξ (Pap. Lond. cxxi. 648, 640),¹⁰⁸⁰ with which compare the gem-inscription ιαιαιαωαδωναισαβαωθ,¹⁰⁸¹ the combinations ιαηλ (Pap. Lond. xlvi. 56 6 Pap. Paris. Bibl. nat. 961 and 3033¹⁰⁸²), and ιαωλ (Pap. Paris. Louvre 2391 151),¹⁰⁸³ as also a whole mass of other combinations.

Ιαωια.¹⁰⁸⁴

(read) ἐπὶτοῦμετώπουϊαωϊα (Pap. Paris. Bibl. nat. 3257).¹⁰⁸⁵ Ιαη occurs more frequently; in particular, in the significant passage:—

ὁρκίζωσεκατὰτοῦθεοῦτῶνἙβραιωνἸησου˙ιαβα˙ιαη˙αβραωθ˙αϊα˙θωθ˙ελε˙ελω˙αηω˙εου˙ιιιβαεχ˙αβαρμας˙ϊαβαραου˙αβελβελ˙λωνα˙αβρα˙μαροια˙βρακιων (Pap. Paris. Bibl. nat. 3019;¹⁰⁸⁶ again, in the same Papyrus, 1222 ff.¹⁰⁸⁷κύριε ιαωαιηιωηωιηωιηιηαιωαιαιουωαηωηαιιεωηυωαηιαωαωα αεηιυωαευιαηει˙. One might surmise that the form ιαη in the latter passage should be assigned to the other meaningless permutations of the vowels.¹⁰⁸⁸ But against this is to be set the fact that the form is authenticated as a Divine name by Origen, that in this passage it stands at the end of the series (the ει of the Papyrus should likely be accented εἶ), and thus seems to correspond to the well-known form ιαω at the beginning. Nevertheless, too great stress should not be laid upon the occurrence, in similar vowel-series, of purely vocalic transcriptions of the Tetragrammaton.

Further, in the same Papyrus, 1564¹⁰⁸⁹ and 1986¹⁰⁹⁰; also in Pap. Lond. xlvi. 23.¹⁰⁹¹ This form is also found in W. Frohner’s¹⁰⁹² issue of the bronze tablet in the Museum at Avignon: the last two lines should not be read καὶσύσυνέργειἈβρασάξιληἸαώ, as Frohner reads them, but καὶσὺσυνέργειαβρασαξιαη¹⁰⁹³ιαω. The reverse combination ιαωιαη is found in a leaden tablet from Carthage, CIL. viii. Suppl. i., No. 12509. We may, finally, at least refer to the passage ὅτιδισύλλαβοςεἶαη (Pap. Paris. Bibl. nat. 944).¹⁰⁹⁴ According to Dieterich,¹⁰⁹⁵αη is “simply a mystical Divine name,” and “it is possible that it should be read αω”. We consider this alteration quite unnecessary. Either αη is an indistinct reminiscence of our ιαη, or else we must definitely conclude that the ι of ιαη coining after et has fallen out by hemigraphy.¹⁰⁹⁶ Αϊα.

Theodoret’s form Αϊα, for which the Augsburg Codex and the ed. print. of Picus read Ια,¹⁰⁹⁷ is found not only in the above-cited passage, Pap. Par. Bibl. nat. 3019 ff., but also in Pap. Lugd. J 395, xvii. 31,¹⁰⁹⁸ as—a fact of special interest—the correction of the αιρα which originally stood in the MS.

Jaoth. The Latin codices of Iremeus yield the form Jaoth.¹⁰⁹⁹ Irenus distinguishes one pronunciation with a long, and another with a short, o (ii. 35 3, Massuet: Jawth, extensa cum aspiratione novissima syllaba, mensuram praefinitam manifestat; cum autem per o graecam corripitur ut puta Jaoth, eum qui dat fugam malorum significat). F. Dietrich has erroneously questioned this form.¹¹⁰⁰ The following should be added to the citations given by Baudissin:— Pap. Lond. xlvi. 142 (ιαωτ),¹¹⁰¹ " " xlvi. 479 (ιαωθ),¹¹⁰² Pap. Par. Bibl. nat. 3263 (ιαωθ),¹¹⁰³ Pap. Lugd. J 395, xxi. 14 (αβρατιαωθ),¹¹⁰⁴ Pap. Lond. xlvi. 56 (αρβαθιαωθ),¹¹⁰⁵ Pap. Berol. 2 125 (αμβριθιαωθ).¹¹⁰⁶ With reference to the agglutination of a T-sound to ιαω, cf. the literature cited by Baudissin.¹¹⁰⁷ The Papyri yield a large number of examples of similar forms in -ωθ. Similar forms with Greek terminations (e.g., Φαραώθης), in Josephus and others.¹¹⁰⁸ Ιαουε.

Regarding Clement’s form Ιαουε, the author calls attention to the following passages:—

θεὸςθεῶν,ὁκύριοςτῶνπνευμάτων¹¹⁰⁹ὁἀπλάνητοςαἰὼν ιαωουηι,εἰσάκουσόνμουτῆςφωνῆς˙ἐπικαλοῦμαίσετὸνδυνάστηντῶνθεῶν, ὑψιβρεμέταΖεῦ,Ζεῦτύραννε,αδαιναι^sicκύριειαωουηε˙ἐγώεἰμὁἐπικαλούμενόςσεσυριστὶθεὸν μέγανζααλαηριφφουκαὶσὺμὴπαρακούσῃςτῆςφωνῆςἑβραϊστὶ αβλαναθαναλβααβρασιλωα˙ἐγὼγάρεἰμισιλθαχωουχλαιλαμ βαασαλωθιαωιεωνεβουθσαβιοθαρβωθαρβαθιαωιαωθσαβαωθπατουρηζαγουρηβαρουχαδωναιελωαιιαβρααμβαρβαραυωναυσιφ ὑψηλόφρονε . . . (Pap. Lond. xlvi. 466-482).¹¹¹⁰

ἀκουσάτωμοι^sicπᾶσαγλῶσσακαὶπᾶσαφωνή,ὅτιἐγώ εἰμιπερταω [μηχχαχ] μνηχσακμηφιαωουεηωηωωηωιεουωηι ηιαηα [corrupt] ιηωυοει¹¹¹¹. . . (Pap. Lugd. J 384, vi. 12-14).¹¹¹²

σὺεἶὁἀγαθοδαίμωνὁγεννῶνἀγαθὰκαὶτροφῶντὴν οἰκουμένην,σοῦδὲτὸἀένναουκομαστήριον,ἐνᾧκαθίδρυταί σουτὸἑπταγράμματονὄνομαπρὸςτὴνἁρμονίαντῶνζ’φθόγαραφαιαβρααρμαραφααβρααχπερταωμηχακμηχιαωουεη ιαωουεειουαηωεηουιαω. . . (Pap. Lugd. J 395, xvii. 25-32).¹¹¹³

ὅτιπροσείλημμαιτὴνδύναμιντοῦἈβραὰμἸσὰκκαὶτοῦ Ἰακὼβκαὶτοῦμεγάλουθεοῦδαίμονοςιαωαβλαναθαναλβα σιαβραθιλαωλαμψτηριηιωω.θεέ,ποίησον,κύριε,περταωμηχ χαχμηχιαωουηειαωουηειεουαηωεηουιαω (Pap. Lugd., J 395, xviii., 21-26).¹¹¹⁴

It might appear at first sight very natural to assume that these forms are related to Clement’s Ιαουε. In consideration of the great freedom with which the Hebrew vowels were transcribed in Greek, it need not seem strange that the E-sound at the end of words is rendered by ηι, ηε and εη in the Papyri; in point of fact the strengthening or lengthening of the ε by the addition of η would give a more distinct rendering of the ה,- than the bare ε of Clement. The coming of ω before ου is the only strange feature. Still, even this peculiarity might be explained by the preference for law, the most popular transcription, which it was desired should have a place also here. For these reasons Kenyon maintains that the form Ιαωουηε is actually the Divine name, and, indeed, that it is an expansion of the form Ιαω.¹¹¹⁵ Notwithstanding, we must not trust entirely to plausibility. We must first of all investigate whether the said forms do not belong to the manifold permutations of the seven vowels,¹¹¹⁶ which are all but universally considered to be capricious and meaningless, mocking every possible attempt at explanation, and which can therefore, now less than ever, yield a basis for etymological conjectures. An instructive collection of these permutations and combinations of the seven vowels for magical purposes is found in Wessely’s treatise, Ephesia Grammata.¹¹¹⁷ That writer elsewhere¹¹¹⁸ passes judgment upon them as follows: “other [names] again appear to have no special meaning, for, just as magical formulae are formed from the seven vowels αεηιουω and their permutations and combinations . . . , so in all probability there were magic formulae formed from the consonants also, now Hebraising, now Egyptianising, now Graecising, and without any definite meaning”. We are unable to decide whether this assertion concerning the consonantal formulae is correct. But certainly when the chaos of the vocalic formations is surveyed, the possibility of accounting for the great majority of the cases may be doubted.¹¹¹⁹ If, then, it were established that the forms cited above should also be assigned to this class, they could, of course, no longer be mentioned in the present discussion. We should otherwise repeat the mistake of old J. M. Gesner,¹¹²⁰ who believed that he had discovered the Divine name Jehovah in the vowel series ΙΕΗΩΟΥΑ. But in the present instance the matter is somewhat different, and the conjecture of Kenyon cannot be summarily rejected. To begin with, the form twooune or mammy, in the first passage quoted, does not stand among other vowel-series; on the contrary, it is enclosed on both sides by a number of indubitable Divine names. Further, the same form with insignificant modifications is found in various passages of various Papyri; from this we may conclude that it is at least no merely hap-hazard, accidental form. Finally, its similarity with Clement’s Ιαουε is to be noted. At the same time, wider conclusions should not be drawn from these forms—none, in particular, as to the true pronunciation of the Tetragrammaton: for the fact that in three of the quoted passages the form in question is followed by vocalic combinations in part meaningless, constitutes an objection that is at all events possible. The value of the vocalic transcriptions of the Tetragrammaton for the determination of its true pronunciation appears to us, by reason of the diffuse and capricious usage of the vowels which we find throughout the Magic Literature, to be at most very small. The very great uncertainty of the traditional texts must also be urged as an objection to its being so employed. Nowhere could copyists’ errors¹¹²¹ be more easily made, nowhere are errors in reading by editors more possible, than in these texts. Let any one but attempt to copy half a page of such magic formulae for himself: the eye will be continually losing its way because there is no fixed point amidst the confusion of meaningless vowels by which it can right itself.

Ιαβε.

It is thus all the more valuable a fact that the important consonantal transcription of the Tetragram, Ιαβε, given by Epiphanius and Theodoret, is attested likewise by the Magic Literature, both directly and indirectly. The author has found it four times in the collocation ιαβεζεβυθ:—

ἐξορκίζω ὑμᾶςτὸἅγιονὄνομ[α

ερηκισθαρηαραραραραχαραραηφθισ . . . .

ιαωιαβεζεβυθλαναβισαφλαν . . .

εκτιπατμμουποφδηντιναξο

ὁτῶνὅλωνβασιλεὺςἐξεγέρθητι

(leaden tablet of cent. 2 or 3 from a Cumean tomb, CIG. iii., No. 5858 b). J. Franz¹¹²² has correctly explained this form: habes in ea formulaΙΑΩJudaicum satis no turn illud ex monumentis Abraxeis, deinde IABE, quo nomine Samaritanos summum numen invocasse refert Theodoretus Quaest. in Exo. xv. On ζεβυθ see below [↓]. Wessely¹¹²³ conjectures that law ΣΑΒΑωΘ appears in the third line. But ζεβυθ is vouched for by the two following passages which give the same magic precept as a precept, which is actually put in practice in the Cumman tablet:— On a tablet of tin shall be written before sunrise among other words the λόγοςει. . .σιφθήιαβεζεβυθ (Pap. Lond. cxxi. 410),¹¹²⁴ On a chalice one shall write besides other words ερηκισιθφηλόγονιαβεζεβυθ (Pap. Par. Bibl. nat. 2000),¹¹²⁵ Similarly ἐπικαλοῦμαίσου . . . τῷ μεγάλῳ σου ὀνόματι . . . ερηκισιθφηαραραχαραραηφθισκηρειαβεζεβυθ ιωβυθιε (Pap. Par. Bibl. nat. 1784 ff.).¹¹²⁶

How are we to explain the form ζεβυθ¹¹²⁷ which thus occurs four times in union with ιαβε? F. Lenormant¹¹²⁸ maintains that it is the names Beelzebuth and Jao which are found on the tablet. He reads ἰαὼἰᾶβεζεβὺθθλαναβὶ σαφλαν . . .¹¹²⁹ Leaving aside the fact that the form Beelzebuth can be nowhere authenticated,¹¹³⁰ it is very precarious to see it in the βεζεβυθ of the Inscription. The mere absence of the λ, indeed, would not be decisive¹¹³¹ against Lenormant’s idea, but certainly the φ, which cannot be read as u,¹¹³² is decisive, and above all the great improbability of the assumption that the names of God and the Devil stand thus closely together. We consider it to be much less objectionable to explain¹¹³³ζεβυθ as a corruption of צְבָאוֹת, and to see in ιαβεζεβυθ the familiar צְבָאוֹתיְהוָֹה. With reference to this identification, the author’s colleague, Herr P. Behnke, Pastor and Repetent at Marburg, has kindly given him the following additional information:—¹¹³⁴ “υ = Heb. o is frequently found. The examples, however, in which this vowel-correspondence appears before ρ should not be taken into account (מֹר = μύῤῥα, צֹר = Τύρος, תּבוֹר = ἸταβύριονἈταβύριον, כּוֹרֶשׁ = Κῦρος,כִּנּוֹר = κινύρα. In מֹר, צֹר, כּוֹרֶשׁ, תּבוֹר [?] the ō is a lengthened ŭ and the ordinary transcription of Sem. u, is υ. But a difference appears in כִּנּוֹר, which goes back to an original kannār; here therefore the υ corresponds to an ō which has been derived from ā as would be the case with -υθ = ־וֹת). But it seems to me to be of greater consequence that the Phoenician pronunciation of Heb. ō (and ô ) is y. Thus we have in the Poenulus of Plautus (ed. Ritschl) [chyl = כֹּל = kull], מוֹצָאִי (= ma͡uṣāi ) given as mysehi ; אוֹת (sign, original form ath) as yth, זֹאת as syth. Moreover, overs (Phöniz., ii., 1, p. 110) has identified Berytos with בְּאֵרוֹת, and Lagarde (Mitteil., i., p. 226) has acknowledged the identification. It is thus quite possible that צבאות could have become ζεβυθ in the mouth of a Phoenician juggler. Still, the omission of the ā before oth in the pronunciation remains a difficulty.” Perhaps Ιαβε is also contained in the word σεριαβεβωθ (Pap. Lond, xlvi. 8)¹¹³⁵; but the text is uncertain and the composition of the word doubtful. Reference must finally be made to a number of forms, in respect of which the author is again unable to allow himself a certain conclusion, but which appear to him to be corruptions of the formιαβε, and therefore in any case to merit our attention:—

ιαβοε, Pap. Lond. xlvi. 63;¹¹³⁶ιαβα¹¹³⁷ is frequently found: ὁρκίζωσεκατὰτοῦθεοῦτῶν ἙβραίωνἸησοῦ˙ιαβα˙ιαη˙. . . .αβαρμας˙ϊαβαραου.αβελβελ . . . (Pap. Par. Bibl. nat. 3019 ff.),¹¹³⁸ἐπικαλοῦμαίσετὸν μέγανἐνοὐρανῷ . . . . βαθαβαθι˙ιατμων˙αλει ˙ιαβα θαβαωθ¹¹³⁹σαβαωθ ˙αδωναιὁθεὸςὁμέγαςορσενοφρη (Pap. Par. Bibl. nat. 1621 ff.),¹¹⁴⁰ὑμᾶςἐξορκίζωκατὰτοῦϊαωκαὶτοῦσαβαωθ καὶαδωναι. . .βαλιαβα (Pap. Par. Bibl. nat. 1484 ff.),¹¹⁴¹ιαβαεδδιαω (a gem-inscription)¹¹⁴²; ιαβαωθ¹¹⁴³: ιαωθιαβαωθ (Pap. Par. Bibl. nat. 3263),¹¹⁴⁴διὰ τὸμέγα ἔνδοξονὄνομααβρααμἐμεινααεουβαωθβαιθωβεσια ιαβαωθ (Pap. Lond. cxxi. 314 f.);¹¹⁴⁵

ιαβας : σὺεἶιαβαςσὺεἶιαπως (Pap. Lond. xlvi. 104).¹¹⁴⁶ A. Dieterich¹¹⁴⁷ thinks it superfluous “to seek a Ἰάβης or similar name” in this; it is but “mystical play-work set down at random”. But the supposition that ιαβας and ιαπως are not mere capricious forms, but rather corrupt Graecisings of Ιαβε, is supported by the context of the whole passage, which belongs to those that are most strongly permeated by Jewish conceptions.

There may also be mentioned another series of forms, chiefly verbal combinations, in which this transcription appears, in part at least, to be contained. We mention only the examples: ιαβω (Geoponica, ed. Niclas, ii., 42 5);¹¹⁴⁹ιαβουνη (Pap. Lond. xlvi. 340);¹¹⁴⁹ the names of angels βαθιαβηλ and αβραθιαβρι (Pap. Lond. cxxi. 906 f.);¹¹⁵⁰ further, ιαβουχ and ιαβωχ (Pap. Par. Bibl. nat. 2204).¹¹⁵¹ Even putting aside the last-quoted series of forms, we consider it to have nevertheless been made plain that Ιαβε must have enjoyed an extraordinary popularity in the Magic Literature. Now this may appear strange if we remember the observation given by the Fathers that it was the Samaritan pronunciation of the Tetragram: how did it get to Egypt and the land of the Cumaean Sybil? The question, however, does not appear to the writer to be unanswerable. We must not of course so conceive of the dissemination of the form as if it had been consciously employed, in such various localities, as the true name of the Mighty God of the Jews; the writer of the Cumaean tablet simply copied it along with other enigmatic and, of course, unintelligible magic formula from one of the numerous books of Magic, all of which, very probably—to judge from those still extant—point to Egypt as their native region. But Egypt was just the country which, because of the ethnological conditions, was most ready to transfer Jewish conceptions into its Magic. One may therefore not unjustifiably suppose that here especially the Tetragrammaton was used by the magicians as a particularly efficacious Name in its correct pronunciation, which was, of course, still known to the Jews, though they shrank from using it, up to and into the Christian era. Thus we have been using the Ιαβε not necessarily for the purpose of indicating the specifically Samaritan pronunciation as such, but rather as an evidence for the correct pronunciation. But we consider it quite possible to account for the occurrence of Ιαβε in Egyptian Papyri by “Samaritan” influence. Besides the Jews proper¹¹⁵² there were also Samaritans in Egypt. “Ptolemy I. Lagi in his conquest of Palestine had taken with him many prisoners-of-war not only from Judaea and Jerusalem but also ‘from Samaria and those who dwelt in Mount Gerizim,’ and settled them in Egypt [Joseph. Antt. xii. 1]. In the time of Ptolemy VI. Philometor, the Jews and Samaritans are reported to have taken their dispute concerning the true centre of worship (Jerusalem or Gerizim) to the judgment-seat of the king [Joseph. Antt. xiii. 34].”¹¹⁵³ Some Papyri of the Ptolemaic period confirm the relatively early residence of Samaritans in Egypt. As early as the time of the second Ptolemy we find (Pap. Flind Petr. ii. iv. 11)¹¹⁵⁴ mention of a place Samaria in the Fayyûm, and two inhabitants of this Samaria, θεόφιλος and Πυῤῥίας,¹¹⁵⁵ are named in Pap. Flind. Petr. xxviii.¹¹⁵⁶ Even more important, in this connection, than such general information, is a passage in the supposed letter of Hadrian to Servianus, in which it is said that the Samaritans in Egypt, together with the Jews and Christians dwelling in that country, are all Astrologers, Aruspices and Quacksalvers.¹¹⁵⁷ This is of course an exaggeration; but still the remark, even if the letter is spurious, is direct evidence of the fact that magic and its allied arts were common among the Egyptian Samaritans. We may also refer here to Acts viii.: Simon the magian was altogether successful among the Samaritans: “to him they all gave heed, from the least to the greatest, saying, This man is that power of God which is called Great”.¹¹⁵⁸ As the Divine name played a great part in the adjurations, we may conclude that the Samaritan magicians used it too—naturally in the form familiar to them. From them it was transferred, along with other Palestinian matter, to the Magic Literature, and thus it is explained why we should find it in a remote region, scratched by some one unknown, full of superstitious dread, upon the lead of the minatory magical tablet.