* I wish to express my appreciation to my colleagues, Professors H. R. Moehring, J. Neusner, and R. Mathiesen, for reading the manuscript and for their comments. The unfailingly prompt and efficient help of the Brown University Library staff has expedited greatly the completion of the documentation for the paper.

Abbreviations: Ap. Const. = Apostolic Constitutions; Baumstark, Geschichte = Anton Baumstark, Geschichte der syrischen Literatur (Bonn: Marcus & Weber, 1922); Brightman, Liturgies = Frank Edward Brightman, Liturgies Eastern and Western, vol. 1 (London: Clarendon, 1896); Graf, Geschichte= Georg Graf, Geschichte der christlichen arabischen Literatur, vol. 1 (Vatican: Bibliotheca Apostolica Vaticana, 1944).

¹ Editio princeps of Br. Mus. Or. 9271 by Horner, George, “New Papyrus Fragment of the Didaché in Coptic,” JTS 25 (1924) 225–31Google Scholar; the manuscript was collated again and published in a more satisfactory manner by Schmidt, Carl, “Das koptische Didache-Fragmentdes British Museum,” ZNW 24 (1925) 81–99Google Scholar, and, more recently, by Lefort, Louis-Théophile, Les Pères apostoliques en copte (CSCO 135; Louvain: 1952) 32–34 (text; translation in CSCO 136, 25–28).Google Scholar

² For a thorough survey of the problems connected with the text and the interpretation of the Didache, see Audet, Jean-Paul, La Didachè: Instructions des apôtres (EBib; Paris: Gabalda, 1958)Google Scholar; on the Coptic text, cf. pp. 28–34. The recent full-length study of the Didache by Giet, Stanislas(L'énigme de la Didachè [Paris: Ophrys, 1970] 213, n. 76)Google Scholar dismisses the problem of the fragment in a brief note opting, following Audet, for the inauthenticity of the passage in question.

³ For a recent detailed discussion of the “ointment” prayer, with critical evaluation of earlier literature, see Vööbus, Arthur, Liturgical Traditions in the Didache (Stockholm: Estonian Theological Society in Exile, 1968) 41–60. See also Audet, La Didachè, 67–70.Google Scholar

⁴ Schmidt, “Didache-Fragment,” 84–85, lines 15–20; ed. Lefort, Pères apostoliques (CSCO 135) 32, lines 16–20.1 have personally collated the papyrus in the British Library on May 24, 1976, and transcribed the passage which is quoted here. Only the clearly necessary editorial emendations are indicated. There is no need to emend mas to mmas as Lefort does, since the text is in fact characterized by the omission of the particle n as the sign of the accusative (see Schmidt, “Didache-Fragment,” 83, for examples).

The dialect in which the text is written is similar to Fayyumic, but has some interesting features which cannot simply be dismissed as due to careless copying. According toKahle, Paul Ernst, Jr. (BalaʾDizah: Coptic Texts from Deir El Balaʾizah in Upper Egypt, 1 [London: Oxford University, 1954] 224)Google Scholar, we have here a text in a sort of proto-Fayyumic, or more precisely “Middle Egyptian, with Fayyumic influence.” It should be noted that what Kahle isolates and describes as a new “Middle Egyptian” dialect of Coptic is the language of a relatively small number of texts (several of which have not been published yet) which were previously regarded as written in a Sahidicized Fayyumic. Kahle's discovery of the new dialect has met with the weighty approbation of Jozef Vergote (“Les dialectes dans le domaine ègyptien,” Chronique d'Egypte 26 [1961] 243). Kahle is of the opinion that “the manuscript of the Didache is probably the latest text in this dialect [i.e., “Middle Egyptian with Fayyumic influence”] being written about the beginning of the fifth century” (BalaʾDizah, 226). Kahle's study of the language of the Didache fragment shows that it has in certain respects more similarities with Bohairic than with pure Fayyumic (ibid.). In view of this sophisticated dialectological analysis, Lefort's earlier view (Pères apostoliques [CSCO 135] xiii-xiv) that the peculiarities of the text can be accounted for simply by the hypothesis that it is not directly translated from Greek but is the transposition of an older Sahidic translation into Fayyumic cannot be sustained.

⁵ Or “word” (seği, which corresponds more closely to Bohairic saği than to either Fayyumic šeği or Sahidic šağe). The reference is perhaps only to “the manner in which,” but could also indicate a more precise directive for the actual wording, peri tou logou, of the prayer which follows.

⁶ We shall presently comment at length on the meaning of this crucial term and therefore leave it for the moment untranslated.

⁷ ”New Papyrus Fragment,” 230: “The aroma (ointment).”

⁸ ”Didache-Fragment,” 85: “Salböl.”

⁹ Schmidt, “Didache-Fragment,” 85. On the Ap. Const, text see below, p. 71.

¹⁰ Pères apostoliques (CSCO 136) 26, n. 13. In the Sahidic NTsočn translates myron 12 times (Wilmet, Michel, Concordance du Nouveau Testament sahidique [CSCO 183; Louvain: 1958] 898)Google Scholar, whereas stinoufe renders the expression osmēeuōdias twice (Wilmet, Concordance, 827) and euōdia by itself once (2 Cor 2:15). Once, to be sure, it renders myron in the Sahidic NT (Luke 23:56), and in the Bohairic version of Ezek 27:17 (Cram, Coptic Dictionary, 363a). But Lefort is clearly correct in saying that a Coptic translator normally would not have chosen stinoufi to render myron.

¹¹ Pères apostoliques (CSCO 136) 26, line 6.

¹² Stinoufi can be analyzed as sti, “smell,” plus noufi, “good.” As Crum points out (Coptic Dictionary, 240a), this form of the adjective is usually found only in compounds.

¹³ Compare the Coptic text just cited with Did. 9.2–3.

¹⁴ It should be noted that some scholars, while accepting Schmidt's myron interpretation, do not regard the text as having a baptismal connotation, but rather take it as a reference to the oil of healing and as providing a proof text for what was later called the sacrament of extreme unction (so Poschmann, Bernhard, Penance and Anointing of the Sick [New York: Herder & Herder, 1964] 237, following E. Riebartsch).Google Scholar

¹⁵ Insistence on a “baptismal” interpretation fundamentally vitiates Vööbus' lengthy analysis. Though he accepts Lefort's comments on stinoufi, Vööbus concludes: “The papyrus fragment in Coptic does not go beyond the term ‘aroma.’ No explanation of its meaning is given and there is nothing to help here” (Liturgical Traditions, 45). Then, simply on the basis of the reference to the “immortal aeon” in the parallel text in the Ap. Const, (where in any case the “aroma” is unambiguously identified as “myron”: see below, p. 71), he concludes that we are dealing with “the imagery of the baptismal rite and experience” (p. 46). This of course assumes that the Ap. Const, provide the correct interpretation of the Didache aucta text; even more crucially, the argument ignores that stinoufi is not merely “imagery” but is as concrete, as material, as the cup and the bread!

¹⁶ Euōdia is the most obvious but not the only possibility (see n. 17). The word euōdia of course basically just means “good scent,” and can be connected with ointment as well as incense (cf. Stumpff, A., “εὐωδία”, TDNT 2 [1964] 809).Google Scholar

¹⁷ Stinoufi or its equivalents in other Coptic dialects can stand for thymiama (in the Bohairic version of Ezek 16:18) and thymiasma (in both Bohairic and Sahidic versions of Isa 43:24), terms which can only refer to burning incense proper, as well as for the more general arōma and euōdia (Crum, Coptic Dictionary, 363a). Stinoufi is not the only expression for incense; ŝouhēne is another, even more common one; it renders thymiama uniformly in the Sahidic NT, and is used in the Sahidic text of Ezek 16:18. In some liturgical texts stinoufe is used interchangeably with ŝouhēne. Our initial contention merely is that stinoufi = “incense” is lexically possible; but, as we will presently show, it also yields the most plausible interpretation of the text.

¹⁸ Greek text cited and discussed in Audet, La Didachè, 83–85.

¹⁹ E.g., “Instruction of Apa Pachomius,” ed. Thompson, Ernest Alfred Wallis Budge, Coptic Apocrypha in the Dialect of Upper Egypt (London: Longmans, 1913) 146–76.Google Scholar

²⁰ See above, n. 4.

²¹ The fundamental investigation is, despite some needed rectifications, still Funk, Franz Xaver, Die apostolischen Konstitutionen: Eine litterar-historische Untersuchung (Rottenburg: 1891; reprint Frankfurt/Main: Minerva, 1970) 28–179.Google Scholar

²² See Connolly, R. Hugh, Didascalia apostolorum (Oxford: Clarendon, 1929) xx–xxi.Google Scholar

²³ Cf. Harnack, Adolf, Die Lehre der zwölf Apostel (TU 2; [1884]) 170–92.Google Scholar

²⁴ Bousset, Wilhelm, “Eine jüdische Gebetssammlung im siebenten Buch der apostolischen Konstitutionen,” Nachrichten von der Königlichen Gesellschaft der Wissenschaften zu Göttingen, Philologisch-historische Klasse aus dem Jahre 1915 (Berlin: Weidmannsche Buchhandlung, 1916) 435–89.Google Scholar

²⁵ Dix, Gregory, The Treatise on the Apostolic Tradition of St. Hippolytus of Rome (London: Society for Promoting Christian Knowledge, 1937) lxxi–lxxvi.Google Scholar

²⁶ Brightman, Liturgies, xxix-xlvi; the text of the reconstructed liturgy is printed, with conventional rubrics added, on pp. 3–27. For a study of the materials to be used for a reconstitution of the fourth-century Antiochene liturgy, with a careful estimate of the evidence of the Ap. Const., see Shepherd, Massey H., “The Formation and Influence of the Antiochene Liturgy,” Dumbarton Oaks Papers 15 (1961) 23–44.CrossRefGoogle Scholar

²⁷ Funk, Konstitutionen, 180–206. For an attempt at a more accurate dating (the 380's) see Schwartz, Eduard, Über die pseudapostolischen Kirchenordnungen (Strassburg: Trübner, 1910) 12–27.Google Scholar

²⁸ Ed. Funk, Franz Xaver, Didascalia et constitutiones apostolorum (Paderborn: Schoening, 1905) 414, lines 10–14.Google Scholar

²⁹ Some comment is in order here on the potentially important argument which Alfred Adam adduces in favor of his hypothesis of a Syriac original for the Coptic text from the stinoufi passage (“Erwägungen zur Herkunft der Didache,” ZKG 68 [1957] 8–11).Google Scholar Adam reasons in the following way: The entry in Crum's Coptic Dictionary on stinoufi shows that the word is a usual equivalent of euōdia, and in particular for the OT expression ryḥ hnyḥwḥ, which in the Peshitta “in den alttestamentlichen Stellen mit rēḥā danjāḥā oder rēḥ njāḥā an den neutestamentlichen Stellen mit rēḥā ‘Duft’ wiedergegeben ist.” Thus nyāḥā is the equivalent of euoōdia. Therefore, Adam concludes, “Auf Grund des älteren Sprachgebrauchs, wie er in der Peschitta des Alten Testaments vorliegt, darf wohl die Schlussfolgerung gezogen werden, dass in der syrischen Vorlage des koptischen Fragments njāḥā gestanden hat” (“Erwägungen,” 9). This, in Adam's view, provides the clue to the stinoufi passage. He notes that nyāḥā has a variety of meanings, inter alia that of anapausis, “refreshment,” and that the feminine form nyāḥtā is used as a terminus technicus for the agapē feast. Therefore, the stinoufi prayer is simply a “Schlussgebet das den Dank für die ganze Feier in den einfachsten Worten ausspricht” (p. 10)—that is, stinoufi is the whole agape feast! Now, though the reasoning is ingenious, it cannot be sustained upon closer examination. First, Adam does not even note that stinoufi can stand for words other than euōdia. Moreover, the employment of the word nyāḥtā, “repose,” is not common in early Syriac as an expression for the agapē feast—it is so used first by Rabbula (5th century) for funerary feasts, and in the 6th-century Philoxenian translation in 2 Pet 2:13, Jude 13. Even more crucially, the whole argument again ignores the liturgical specificity of the prayer: the blessing is made over some actual substance, not simply for “the good cheer,” so to speak! But Adam rightly recognizes the secondary nature of the Ap. Const, “myron” text, and that it represents a misunderstanding of “das Hinweisen auf den wohlgefälligen ‘Duft’ der Agapefeier” (p. 10). Adam, with commendable reserve, does not commit himself on the question of the relationship of the hypothetical Syriac text to the Greek of the Constantinople MS. A detailed analysis of his other arguments for a Syriac Vorlage would take us too far afield, but they have little cogency either. Cf. Vööbus' brief critique of Adam's hypothesis (Liturgical Traditions, 44–45).

³⁰ Vööbus (Liturgical Traditions, 45–46) opts for a baptismal interpretation. But the wording is strange indeed, and has no affinity with the definitely baptismal peri tou mystikou myrou eucharistia in VII.44.

³¹ The Greek MSS used by Funk all have the phrase in question; but the Ethiopic Didascalia (a reworking of Books I-VII of the Ap. Const., not to be confused with the old Greek Didascalia which is the Vorlage of Books 1-VI!) has the following wording: “Thou shalt give thanks thus: We give thanks to Thee, Creator of all things, for the savour of this chrism, and for this oil of immortality, which Thou hast revealed to us by Jesus Christ Thy Son. For Thine is the glory, and the kingdom, and the power for ever and ever. Amen.” (transl. from Br. Mus. Or. 752, by Harden, John Mason, The Ethiopic Didascalia [London: S.P.C.K., 1920] 172).Google Scholar The Ethiopic is based on an Arabic version of the Ap. Const., but neither the Arabic nor the Ethiopic has been investigated with any thoroughness.

³² Since we have accepted Lefort's argument that myron would have been rendered by sočn rather than stinoufi, it is not necessary to review at length the various opinions concerning authenticity which have been based on the older myron interpretation. Incidentally, Erik Peterson's provocative hypothesis that the myron prayer is authentic, and that its suppression in the Bryennios text of the Didache is a sign of Novatian heresy which, inter alia, opposed postbaptismal anointing (“Über einige Probleme der Didache-Überlieferung,” in his Frühkirche, Christentum und Gnosis [Freiburg: Herder, 1959] 166Google Scholar; reprinted from Rivista di archeologia cristiana [1951] 37ff.) has been thoroughly refuted and needs no further comment. See Audet, La Didachè, 68, n. 3, and Vööbus, Liturgical Traditions, 31–33.

³³ Baumstark, Anton, “Aegyptischer Oder antiochenischer Liturgietypus in AK I-VII?” OrChr 1, 7 (1907) 388–407.Google Scholar

³⁴ Ibid., 395. The prayer in question is clearly a blessing over the water and oil of exorcism and healing (ed. Wobbermin, Georg, Altchristliche liturgische Stücke aus der Kirche Ägyptens (TU 17, 36 [1898] 7–8).Google Scholar There is a similar prayer at a later point which, however, does not seem to be connected with the eucharistic prayers (ibid., 13–14). See also Brightman, Frank Edward, “The Sacramentary of Serapion of Thmuis,” JTS 1 (1900) 108, 267–68.Google Scholar

³⁵ Thus the application of the expression antitypa (Ap. Const. VII.25) to the eucharistic elements is not peculiarly Egyptian (so Baumstark, “Liturgietypus,” 398): cf. my article, “The Eucharistic Doctrine of the Byzantine Iconoclasts and Its Sources,” Byzantinische Zeitschrift 68 (1975) 13, n. 46.Google Scholar Baumstark notes (“Liturgietypus,” 395) that the Testamentum Domini has a prayer over oil and water after the eucharistic prayers proper (1.24–25); (ed. Rahmani, Ignatius Ephraem, Testamentum domini nostri Jesu Christi [Mainz: Kirchheim, 1899] 48).Google Scholar But that this work (written in Greek, but [apart from some Latin fragments] extant only in oriental translations [Syriac, Arabic, Ethiopic]) is of Egyptian origin is by no means as certain as Baumstark assumes (“Liturgietypus,” 389–90). Thus, though transmission in Arabic (via Coptic) and in Ethiopic argues for Egypt, Baumstark himself (“Überlieferung und Bezeugung der διαθήκη το κυρίου μν Ἰησο Χριστο, RQ 14 [1900] 39)Google Scholar accepted earlier tne hypothesis of Syrian origin. Funk, Franz Xaver (Das Testament unseres Herrn und die verwandten Schriften [Mainz: Kirchheim, 1901] 87Google Scholar opts for Syria; A. J. MacLean, though with many misgivings, even proposes Asia Minor (in Cooper, James and MacLean, Arthur John, The Testament of Our Lord [Edinburgh: Clark, 1902] 42–45).Google Scholar But Theodor Schermann's refusal (Ägyptische Abendmahlsliturgien des ersten Jahrtausends [Paderborn: F. Schöningh, 1912] 97–98)Google Scholar to grant any cogency to Baumstark's arguments depends on an all too ready assumption of Syrian influence on well nigh all aspects of early Egyptian liturgical usage.

³⁶ The fundamental work on the whole subject is still Atchley, Edward Godfrey Cuthbert Frederic, A History of the Use of Incense in Divine Worship (London: Longmans, 1909).Google Scholar Another valuable collection of the material (even though one must dissent from some of the hypothetical constructions imposed upon it) is Schneider's, Carl“Studien zum Ursprung liturgischer Einzelheiten östlicher Liturgien, 2. ΘΘϒΜΙΑΜΑΤΑ,” Kyrios 3 (1938) 149–90, 293–311.Google Scholar See also E. Fehrenbach, “Encens,” DACL 5. 1, cols. 2–22, esp. cols. 6–11, and Hanssens, Jean Michel, Institutiones liturgicae de ritibus orientalibus, Tom. III (Rome: Gregoriana, 1932) 70–91.Google Scholar

³⁷ See Atchley, History, 81–96. The material is also collected in Warren, Frederick Edward, The Liturgy and Ritual of the Ante-Nicene Church (London: S.P.C.K., 1897) 129–31.Google Scholar See also Frere, W. H., “Notes on the Early History of the Use of Incense,” in Westall, Henry, ed., The Case for Incense (London: Longmans, 1899) 43–86.Google Scholar

³⁸ Cf. Atchley, History, 97–116.

³⁹ Only one 9th-century Latin version of the Acta of Peter, but none of the older Greek, Coptic and Arabic recensions mentions that, after the preparation for burial (odoriferis condientes aromatibus induerunt illum sericis indumentis), the saint was carried to his final resting place flammantibus cereis, concrepantibus hymnis, flagrantisque thymiamatibus (PG 18. 465C).

⁴⁰ “Burn incense [besmē] in the sanctuary; as for me, accompany me with prayer. … Go and burn [the incense] in the sanctuary, so that there may be good smell for those who go there” (ed. Beck, Edmund, Des heiligen Ephraem des Syrers Sermones IV [CSCO 334; Louvain: 1973] 51Google Scholar, lines 9–11). At an earlier point the text makes a very realistic reference to the corpse and “the stinking smell of Ephrem,” riḫeh saryā dAphrem (ibid., 48, lines 9–10). In this work the fumigatory use of incense seems to have as its pendant the (habitual?) use of incense in churches, though no connection is made with the eucharistic liturgy.

⁴¹ E.g., Horn, xli: de S. Pelagia (PG 50.585).

⁴² Late fourth or early fifth century; see Wilkinson, John, Egeria's Travels (London: S.P.C.K., 1971) 27–30Google Scholar for an attempt at an exact chronology.

⁴³ Dictis ergo his tribus psalmis et factis orationibus tribus ecce etiam thiamataria inferuntur intro spelunca Anastasis, ut tola basilica Anastasis repleatur odoribus (Itinerarium Egeriae XXIV. 10, ed. A. Franceschini and R. Weber, in Itinerariae et alia geographica, CChr Series latina 175.69, lines 81–84).

⁴⁴ This suggestion was first made by Mateos, Juan, “La vigile cathédrale chez Egérie,” Orientalia Christiana Periodica 27 (1961) 292Google Scholar: the resurrection theme explains, inter alia, “Fencensement du tombeau que rappellent les parfums portés par les femmes.” The explanation is accepted by Wilkinson, Egeria's Travels, 65, and Zerfass, Rolf, Die Schriftlesung im Kathedralofficium Jerusalems (Münster: Aschendorff, 1969) 27.Google Scholar

⁴⁵ “May your fasting be armor for our land, your prayer a shield for our city, may your censer [piromāk] purchase reconciliation [tarʿuṯā]. Blessed be he who sanctified your sacrifices” (ed. Beck, Edmund, Des heiligen Ephraem des Syrers Carmina Nisibena 1 [CSCO 218; Louvain: 1961] 46, lines 18–19).Google Scholar Hans Lietzmann (following the suggestion of the earlier editor, Bickell) sees in the passage a reference to the sequence of liturgical activities: fasting, prayer, censing of the eucharistic elements and the eucharistic sacrifice (Messe und Herrenmahl [Bonn: Marcus & Weber, 1926] 86–87).Google Scholar However, other evidence which merely associates the offering of incense with supplication (see note 46) makes this reasoning somewhat doubtful.

⁴⁶ Cf. Atchley, History, 118–19.

⁴⁷ Thus, in one recension of a Syriac collection of the miracles of the Virgin, some of her devotees, threatened by imminent shipwreck in a storm “remembered her, and placed incense on the fire [sāmu besmē^y al nurā] and immediately the sea became calm …” (ed. Budge, Ernest Alfred Wallis Thompson, The History of the Blessed Virgin Mary… [London: Luzac, 1899] 138, n. 1).Google Scholar

⁴⁸ The evidence is collected and subjected to an incisive, but not entirely convincing, analysis by Lietzmann, Messe und Herrenmahl, 86–93.

⁴⁹ Cf. Taft, R., “Evolution historique de la liturgie de saint Jean Chrysostome … L'encensement et le lavabo,” Proche Orient Chrétien 25 (1975) 275–86.Google Scholar

⁵⁰ PG 58.781, cited by Atchley, History, 200. Chrysostom contrasts the censing of the church to produce a pleasant odor with the reluctance to banish the evil smell of spiritual uncleanliness.

⁵¹ See Funk, Konstitutionen, 180–206, for detailed proof.

⁵² Funk, Didascalia et constitutiones apostolorum, 564, lines 8–10.

⁵³ Dionysius' Interpretatio prima has timiama (id est incensum) tempore quo sancta celebratur oblatio (ed. Turner, Cuthbert Hamilton, Ecclesiae occidentalis monumenta iuris antiquissima 1.1 [Oxford: Clarendon, 1909] 10, col. 1, lines 3–5). The Interpretatio altera differs only in giving the more correct transliteration “thymiama” (col. 2, line 2).Google Scholar

⁵⁴ The Sahidic version has ouŝouhēne mpnau nte prosphora etouaab (“incense at the time of the holy sacrifice”), ed. Lagarde, Paul Anton de, Aegyptiaca (Göttingen: Dieterich, 1883) 210.Google Scholar The Bohairic text (ibid.) uses ousthoi-noufi for “incense”; this text, according to the colophon translated from Sahidic (de Lagarde, Aegyptiaca 238; see also Tattam, Henry, The Apostolic Constitutions or Canons of the Apostles in Coptic [London: Oriental Translation Fund, 1848] 213)Google Scholar may reflect a Sahidic stinoufe as the word for incense. Rather interestingly, the marginal gloss in the Bohairic MS, nem ousouğen, “with myron,” shows an evidence of wanting to bring the work liturgically up to date.

⁵⁵ The version in the Epitome of Ebedjesus (cf. Baumstark, Geschichte, 324) has besmē bʿedānā dqurbānā ʾalahāyā, “incense at the time of the divine offering” (ed. Mai, Angelo, Scriptorum veterum nova collectio … tomus X [Rome: Apud Burliaeum, 1837] 175).Google Scholar The version incorporated in an older Syriac pseudo-Clementine canonical collection has only besmē bʿedānā dqurbānā (ed. Lagarde, P. de, Reliquiae iuris ecclesiastici antiquissimae [Leipzig: 1856Google Scholar; reprint Osnabruck: Zeller, 1967], p. mh, lines 5–6). Cf. Baumstark, Geschichte, 82, and Urbina, Ignazio Ortiz de, Patrologia syriaca (Rome: Pont. Institutum Orientalium Studiorum, 1958) 224–25.Google Scholar

⁵⁶ baṯūr fī waqt al-quddās al-ṭāhir, “incense at the time of the pure consecration,” from the version of the Apostolic Canons proper incorporated in the “127 Canons of the Apostles” (ed. Jean Perier and Augustin Perier, PO 8 [1912] 655). On this compilation, which with most likelihood is translated from Coptic, see Graf, Geschichte, 572–77.

⁵⁷ ʿeṭān bagizē q'erban neṣuḥ “incense at the time of the pure offering” (ed. Fell, Winand, Canones apostolorum aethiopice [Leipzig: Brockhaus, 1871] 13).Google Scholar

⁵⁸ PG 80, 284C.

⁵⁹ Trans. Connolly, R. Hugh, The Liturgical Homilies of Narsai (TextsS 8.1; Cambridge: Cambridge University, 1909) 12Google Scholar (=ed.Mingana, Alphonse, Narsai doctoris syri homiliae et carmina primo edita, vol. 1 [Mosul: Typis Fratrum praedicatorum, 1905] 281).Google Scholar There has been much doubt expressed about the authenticity of the homily, on stylistic and liturgiological grounds, and because of the attribution in some MSS to the thirteenth-century writer Ebedjesus of Elam (Connolly, Liturgical Homilies, p. xii). Cf. Baumstark, Geschichte, 112 and 348 (opting for inauthenticity!). See also Unnik, Willem Cornelis van (Nestorian Questions on the Administration of the Eucharist … [Amsterdam: 1937; reprint Amsterdam: Grüner, 1970] 422ff.)Google Scholar who concludes that the question is still sub judice (p. 57).

⁶⁰ For a tentative reconstruction of the pseudo-Dionysian liturgy see Brightman, Liturgies, 487–88.

⁶¹ De ecclesiastica hierarchia 3, PG 3. 425B.

⁶² Ed. Riedel, Wilhelm and Crum, Walter E., The Canons of Athanasius of Alexandria (London: Williams and Norgate, 1904).Google Scholar Though originally composed in Greek, the work is extant only in Sahidic and Arabic. The Sahidic version, though relatively old (one MS has been dated by Crum to ca. A.D. 600), is, however, fragmentary, and in particular these sections where the Arabic version mentions incense are missing. The Arabic text is not translated from the Sahidic, but rather, as Crum suggests, from a lost Bohairic version made independently from the Greek (ibid., 81). Though Riedel, with some hesitation, accepts the Athanasian authorship of the lost Grundschrift (ibid., pp. xiv-xxvi), it is quite clear that the translation into Arabic and the division into paragraphs is the work of Michael of Tinnïs (according to Graf [Geschichte, 605] perhaps to be identified with the 11th-century redactor of the History of the Patriarchs of Severus ibn al-Muqaffaʿ). At any rate, the final portion of the Arabic (which includes [par. 106] a detailed description of the censing of the gospel book) has, as Riedel himself notes “almost the appearance of a subsequent edition” (Riedel and Crum, Canons of Athanasius, p. xxvi). In view of the silence of Athanasius and other early Egyptian fathers concerning liturgical censing (cf. note 63) I hesitate to ascribe the two passages concerned with incense (par. 7fin., par. 106) to Athanasius, whatever else in the work may be of early provenience.

⁶³ For a convenient collection of liturgical allusions in this patristic literature, see Brightman, Liturgies, 504–09. The earliest explicit mention of liturgical censing, to my knowledge, in a datable narrative Egyptian text is from the 11th-century recension of the Arabic biography of the patriarch Shenouti I (A.D. 849–80), according to which in a nocturnal liturgy “while he went around the sanctuary (haikal) with incense, his eyes shed bitter tears” (trans. Yassā ʿAbdal-Masīḥ and Oswald Hugh Ewald Burmester, History of the Patriarchs of the Egyptian Church, 2.1 [Cairo: L'institut français d'archéologie orientale, 1943] 55; Arabic text not published). To be sure, a number of censers from Egypt have been dated, in particular by Strzygowsky, Josef (Catalogue général des antiquites égyptiennes du Musée du Caire … Koptische Kunst [Vienna: Holzhausen, 1904] 280ff.)Google Scholar as coming from the pre-Islamic period; he even dates two standing censers to the 2d to the 4th century (Nos. 9122–23). Though it is not within my competence to control the datings, they seem to be on the whole rather arbitrary, especially where one is dealing with objects acquired by purchase from private individuals or from antiquities dealers. In any case, there is no indication of the function these censers had, and for several there even seems to be no clear indication that they are of Christian origin. The archeological evidence is therefore of little moment, it seems to me, compared to the silence of the literary sources on censing in the eucharistic liturgy. Cf. Leclercq, Henri, “Encensoir,” DACL 5 (1922) cols. 21–33Google Scholar, and Schneider, , “Studien,” Kyrios 3 (1938) 171–74.Google Scholar

⁶⁴ For an example from “orthodox” Coptic literature, cf. the encomiastic Vita of bishop Pisentius (7th century), in which the saint is described as “one who was full of light and who spread forth stinoufe all the time” (ed. Ernest Alfred Wallis Thompson Budge, Coptic Apocrypha in the Dialect of Upper Egypt [London: Longmans, 1913] 74).Google Scholar

⁶⁵ See Ernst Lohmeyer's classic monograph, Vom göttlichen Wohlgeruch (Sitzungsberichte der Heidelberger Akademie der Wissenschaft, Phil-hist. Klasse 9; 1919), particularly pp. 15–22 on ancient Egyptian material; the latter is also discussed by the Bonnet, Egyptologist Hans, “Die Bedeutung der Räucherungen im ägyptischen Kult,” Zeitschrift für ägyptische Sprache und Altertumskunde 67 (1931) 20–28.Google Scholar See also Waldemar Deonna's more comprehensive study, “ΕϒΩΔΙΑ, Croyances antiques et modernes: l'odeur suave des dieux et des élus,” Genava 17 (1939) 167–263.Google Scholar

⁶⁶ In the Coptic Gnostic Gospel of Truth “the smell” (pstaei) is identified with the Spirit and is described as an attribute of the Father (ed. Malinine, Michel et al., Evangelium Veritatis [Supplementum; Zürich and Stuttgart: Rascher, 1961] 34).Google Scholar Much of the Gnostic material, and the classical background, is discussed à propos this passage in Jacques-É. Ménard, , L'Evangile de Vérité (Leiden: Brill, 1972) 158–63.Google Scholar In the Gospel of Philip, the “odor” (stoei) occurs, but only in connection with “ointment” (sočn) (ed. Till, Walter C., Das Evangelium nach Philippos [Patristische Texte und Studien 2; Berlin: de Gruyter, 1963] 55, lines 36ff.).Google Scholar Incense specifically does not seem to have played a demonstrable role in the sacramental system of the Gnostics, judging by both the patristic accounts and the Coptic texts.

⁶⁷ It should be noted here that ch. 38 of the “vulgate” recension of the Arabic Didascalia (cf. note 31) has a detailed description of the liturgy, with a preanaphoral triple circumambulatory censing of the altar “in honor of the Holy Trinity” by the bishop, with a subsequent censing of the congregation by a presbyter (translated, with conventional rubrics added, in Brightman, Liturgies, 510–11; for another translation, see Funk, Konstitutionen, 233–34). The “Abū Isḥāq” recension, represented by Borg. ar. 22 and the Ethiopic translation, though lacking this particular chapter along with others (cf. Graf, Geschichte, 565), does assume a liturgical use of incense. The Ethiopic text (tr. Harden, John Mason, The Ethiopic Didascalia [London: S.P.C.K., 1920] 92)Google Scholar says, “We command you, then, that no layman execute the office of the priesthood, neither offer incense, nor baptize, nor lay on hands, nor give the bread of blessing”—a reworking of Ap. Const. III. 10, which does not mention the offering of incense as part of the sacerdotal duties. This Egyptian Didascalia tradition does attest the use of incense, but still there is no cogent reason to assume that it preserves fourth- or fifth-century usage. MacLean's statement that the censing in the Arabic Didascalia text is consistent with a fourth-century date for the work, simply because “incense is mentioned in the Pilgrimage of Silvia, about 385 A. D.” (Cooper, James and MacLean, Arthur John, The Testament of Our Lord [Edinburgh: Clark, 1902] 34) is a misleading simplification, in view of the material we have presented earlier.Google Scholar

⁶⁸ One can only speculate about the time of and motives for the introduction of liturgical censing in Egypt; however, in view of the relatively early Syrian evidence and the strong influence of Syrian liturgical usages, especially after the monophysite schism (cf. Schermann, Ägyptische Abendmahlsliturgien, 97ff.), it seems that liturgical censing was such a Syrian import.

⁶⁹ By the twelfth century liturgical censing came to be extremely widespread in Egypt. The peculiar custom arose even of individuals confessing their sins before the censer carried around by the priest, at the time of the preliminary censing of the altar and nave (cf. n. 66), as a substitute for auricular confession. See Graf, Georg, “Über den Gebrauch des Weihrauchs bei den Kopten,” Dem Prinzen Johann Georg Herzog zu Sachsen zum 50. Geburtstag gewidmet: Ehrengabe deutscher Wissenschaft dargeboten von katholischen Gelehrten (hrsg. von Franz Fessler; Freiburg: Herder, 1920) 223–32Google Scholar, and idem, Ein Reformversuch innerhalb der koptischen Kirche im zwölften Jahrhundert (Paderborn: Schöning, 1923) esp. 55–59, 150–52.Google Scholar

⁷⁰ Baumstark reduces the contribution of the final redactor to little more than that of arranger and compiler (“Die Urgestalt der ‘arabischen Didaskalia der Apostel’ …” OrChr 1, 3 [1903] 208).Google Scholar

⁷¹ This can be easily seen by simultaneously comparing the Coptic translation and the relevant portion of the Ap. Const. with the Greek Didache text of chaps. 10–11.

⁷² It should be pointed out, however, that a recognition of the Syrian element in the liturgical portion of the Didache is not incompatible with the view that the extant Greek recension (represented by the Bryennios MS and the Georgian version) is of Egyptian provenance. This is the opinion, for instance, of Richardson, Cyril C. (Early Christian Fathers [Philadelphia: Westminster, 1953] 163)Google Scholar echoing in part Harnack's detailed arguments for an Egyptian Heimat for the Didache as a whole (Lehre der zwölf Apostel [TU 2; 1884] 15–18).Google Scholar If the final redactor of the Greek Didache was indeed Egyptian, then a fortiori the Greek Vorlage of the Coptic text would be an Egyptian MS, and the need to account for the influence of possible Syrian liturgical censing would disappear.

⁷³ See Jeremias, Joachim, Die Abendmahlsworte Jesu (3rd ed.; Göttingen: Vandenhoeck & Ruprecht, 1960) 111, n. 6.Google Scholar Cf. Clerici, Luigi, Einsammlung der Zerstreuten: Liturgiegeschichtliche Untersuchung … der Fürbitte für die Kirche in Didache 9,4 und 10,5 (Münster: Aschendorff, 1965) 1–2Google Scholar, and the most recent discussion by Rordorf, Willy, “Les prières eucharistiques de la Didachè,” Eucharisties d'Orient et d'Occident (ed. Bernard Botte, et al.; Paris: Cerf, 1970) 1. 65–82.Google Scholar

⁷⁴ All the possibly pertinent material, literary, ethnographical, and archeological, on the Jewish use of incense is marshalled by Goodenough, Erwin Ramsdell (Jewish Symbols in the Greco-Roman Period [New York: Pantheon, 1954] 4. 195–208).Google Scholar

⁷⁵ See Friedrich Pfister, “Rauchopfer,” PW2.1, col. 278; Atchley, History, 64–66.

⁷⁶ If they were sitting down to eat, each individually, he says a blessing for himself; if they have a banquet, one says the blessing for all. If wine is brought to them in the middle of the meal, each one says the blessing for himself; if after the meal, one says the blessing for all, and he says [the blessing] over the incense, also if they bring in the incense only after the meal” (ed. Albeck, Chanoch, [Jerusalem-Tel Aviv: Mosad Bialik, 1958] 1. 25).Google Scholar

⁷⁷ The Mishnaic Hebrew word mûgmār is cognate to Jewish Aramaic gûmrā, “burning coal,” and it refers to burning incense, unlike the word bōsem (pl. běsāmim), which has the broader connotation of aromatic materials, spices, perfumes, etc. On the subject, cf. Krauss, Samuel, Talmudische Archaeologie (Leipzig: Fock, 1910) 1. 237–38.Google Scholar It is not at all certain that the mûgmār of m. Ber. 6.6 is to be identified with the aromas, běsāmim, in the disputed sequence in m. Ber. 8.5 (“light, food, běsāmim, habdālāh” [School of Shammai]; “light, běsāmim, food, habdālāh” [School of Hillel]). For a commentary on this passage, see Neusner, Jacob, Invitation to the Talmud (New York: Harper, 1973) 49–50.Google Scholar That the burning of mûgmār was a routine part of the meal is made clear also by m. ʿEd. 3.11, where R. Gamaliel [I?] allows that on a festival day the floor between the couches may be swept, and that “they may place the mûgmār [on the fire]” (ed. Albeck, Mishna, 4.297). The same ruling is found in m. Beṣa 2.7 (ed. Albeck, Mishna, 2.293).

⁷⁸ Tosepta Ber. 6.6 does not mention mûgmār, but does note a difference of the schools of Shammai and Hillel concerning the order of the benedictions of the light and the běsāmim (a dispute similar to m. Ber. 8.5 but different in detail). Is this a sign that at the time of the compilation of Tosepta Ber. the custom of burning incense has already fallen into desuetude?

⁷⁹ “Rabbi Zērā said [that] Rabba bar Jeremiah said, ‘When do they say the blessing over the fragrance? When its column [of smoke] rises’” (b. Ber. 43a [bottom]). Since an objection from Zērā follows, and then a final solution from the master based on the distinguo of “intention,” it seems that the text is in slight disarray; the initial question (When do they say the blessing?) must be in fact Zērā's. The parallel text in the Palestinian Gemara is briefer but more coherent: “Rabbi Zeʿirā [said] in the name of Rab Jeremiah, ‘[Over] the mûgmār, as soon as the smoke rises, it is necessary to say the blessing’” (y. Ber. 6.6, p. 10b).

⁸⁰ Over all mûgmārôt the correct form of the benediction is “[Blessed be You], who do create fragrant woods” (b. Ber. 43a).

⁸¹ This connection was surmised by Robert A. Kraft in his commentary Barnabas and the Didache (The Apostolic Fathers 3, ed. Grant, Robert M.; New York: Nelson, 1965)Google Scholar, but not explored in detail. I will quote his comment in full: “An important clue… might be discovered if the significance of the ‘ointment’ or ‘perfume’ (fragrant oil) of 10:8 were known. It is not impossible that this too is a vestige from Christian Love Feasts, since the Jewish fellowship meal ritual included a blessing on the aromatic spices (‘ointment’?) which usually were burned” (p. 167). As his source for Jewish usage Kraft refers only to Dix, Gregory, The Shape of the Liturgy (Westminster: Dacre, 1945) 425–6.Google Scholar Dix alludes to the burning of spices at ḥaburah meals, and notes the section from m. Ber. which we have analyzed previously. However, then Kraft seems to retract his suggestion: “But ointment/oil was used in many connections in early Christianity” and lists various uses such as the oil of baptism, episcopal ordination, etc. In commenting on the text, Kraft does not seem to know Lefort's edition of the Coptic fragment and the important rectification of the translation of stinoufi; otherwise he would not be faced by the dilemma of “ointment” being burned, which forces him back to the myron interpretation.

⁸² Cf. Delling, Gerhard, “σμή,” TDNT 5 (1967) 493–96.Google Scholar