we already reviewed:
we already reviewed:
(The hour of) morning prayer is till midday; Rabbi Yehuda says, “Till the fourth hour. afternoon prayer is till evening;” Rabbi Yehuda says, “…
evening prayer has no set time;” Rabbi Yehuda says… (…) 47. see alon, Jews in Their Land, 119–31; shmuel safrai, “Bikkureihem shel hakhamei Yavne be-Roma,” in In Times of Temple and Mishnah, ed. shmuel safrai (Jerusalem: Magnes, 1996), 365–81. There are serious doubts about this interpretation, however.
48. for this argument, see shaye J. D. cohen, “The significance of Yavneh: Pharisees, Rabbis, and the end of Jewish sectarianism,” in The Significance of Yavneh and Other Essays in Jewish Hellenism, TsaJ 136 (Tübingen, Mohr siebeck, 2010), 44–70, 55.
49. see the forthcoming article mentioned above n. 23. The point was made in relation to the Birkat haminim by William horbury, see below n. 60.
50. for this analytical method, see Goldberg, “Tosefta,” 211–44. for discussion of the material, see G. stemberger, Einleitung in Talmud und Midrasch, 8th ed. (Munich: Beck, 1992), 113–52; see 47–54 on oral transmission.
TOMsON: The LORD’s PRaYeR
181
Rabbi Nehonya ben hakana used to say a short prayer when entering and leaving the house of study. They asked him: “What is the place (sense) of this prayer?” he said to them (…)
Rabban Gamliel says: “every day, a person is to say eighteen benedictions.”
Rabbi Yoshua says: “(Only) a summary of the eighteen.”
Rabbi akiva says: “if the prayer is fluent in his mouth, he prays eighteen, if not, a summary of the eighteen.”
Rabbi eliezer says: “he who makes his prayer fixed, it is no supplication.”
Rabbi Yoshua says: “One who is underway at a dangerous place (at the hour of prayer), says a short prayer.” (m. Ber. 4:1–4) The introductory sentence gives diverging specifications of the hours of prayer; the repeated mention of R. Yehuda’s name ascribes this discussion to his generation, the one of usha. The principle of three prayers a day itself is not disputed. hence this represents the consensus of the preceding Yavneh generation of Gamaliel and colleagues. We saw that the principle is also shared by the Didache, presumably around the same time, ca. 100. in the continuation Gamaliel’s ruling with ensuing discussion is sandwiched between two passages on “short prayers.” a narrative introduction tells of the short prayer said by Nehonya, an older contemporary. Gamaliel’s clause seems to be formulated in reaction, lapidarily. it provokes three objections, and finally Yoshua’s short prayer follows.
The intention of the final editor with this seemingly undecided discussion may be read from the disposition of opinions.51 eliezer rejects the ruling on principle. he is against the whole idea of a form of prayer fixed for all: prayer needs to be personal (“his prayer”), and it must be
“supplication” and include variation. incidentally, this emphasis brings him close to what is told of Jesus (see Matt 6:6; Mark 11:22–24; Luke 18:1).52 his opinion, however, is preceded by the ones of Yoshua and akiva. Yoshua objects to the rule prescribing eighteen benedictions three times daily for all: he wants to go no further than a “summary,” a short prayer inspired by the eighteen. akiva, the younger contemporary and former disciple of Yoshua and eliezer, intermediates and adds a conces-51. fleischer, “Beginnings of Obligatory Jewish Prayer,” 426–28 simply supposes Yehuda the Prince reckoned with persisting resistance.
52. see n. 42 above. The same opinion is ascribed to another earlier contemporary, shimon ben Nataneel, m. avot 2:13.
The DiDache: a MissiNG Piece Of The PuzzLe
sion: when someone’s prayer is fluent, the eighteen are to be said, but if not, only a summary. The discussion ends with Yoshua’s opinion, already quoted, that in times of distress only a “short prayer” is to be said. This is open to the assumption that, in normal circumstances, the eighteen Benedictions are the rule.
Thus analyzed, the course of the discussion subtly suggests a gradual softening of the initial protest against Gamaliel’s initiative. Yoshua’s fundamental objection is accommodated by akiva in his habitual role of mediator. The radical rejection by eliezer stands isolated, and the resulting implicit understanding is that, barring exceptional circumstances, the eighteen Benedictions are the rule. This is confirmed by later discussions in the Talmud that take the eighteen Benedictions for granted. in spite of initial resistance,53 Gamaliel’s initiative has won out. The Talmud summarizes the process succinctly: “shimon ha-Pekoli arranged the eighteen Benedictions in their proper place before Rabban Gamliel at Yavneh.”54
if we now combine these results with our findings about the early christian sources, a conclusion of great importance follows. Gamaliel’s institution of the eighteen Benedictions seems to have met with considerable resistance both in his own rabbinic circle and in the community of the Didache and Matthew. This conclusion may be spelled out in seven steps.
(1) Matthew’s polemics, on prayer among other things, is aimed at the
“hypocrites” or Pharisees who recently had begun calling themselves “rabbi.”
(2) The Didache, closely related to Matthew, turns against the “hypocrites” who have the second and fifth weekday for fast days instead of the fourth and sixth ones and who pray another prayer than the Lord’s Prayer.
(3) fasting on the second and fifth days, Monday and Thursday, is the rule in rabbinic tradition, while the preference for Wednesday and friday is associated with an alternative Jewish calendar the Didache-Matthew community seems to have supported.
(4) The similarities of Matthew’s polemics against the Pharisaic rabbis and the Didache’s resistance against Pharisaic-rabbinic fast days suggest that both texts also militate against the prayer text propagated by the Pharisees-rabbis.
(5) The objection in the Didache and Matthew to this rabbinic daily prayer parallels the initial protest registered in the Mishnah against the 53. see the memory of the violent discussion between Yoshua and Gamaliel about the status of the daily evening prayer, b. Ber. 27b–28a.
54. see b. Ber. 28b; b. Meg. 17b. for the sequel, see below.
TOMsON: The LORD’s PRaYeR
183
institution of the eighteen Benedictions. as regards Gamaliel’s rule of eighteen, the three rabbis cited reject it altogether or propose a “short prayer” as a “summary of the eighteen” instead, while the editors of the Didache and Matthew simply stick to Jesus’s short prayer in its Matthean form.
(6) The rule to pray thrice daily presupposed by the Didache shows that a similar custom did exist in pre-70 Judaism. in addition, Mark 11:25
makes the pre-70 existence of some form of the Matthean “ecclesial” version of the Lord’s Prayer likely and by implication also the pre-70 existence in whatever form of Pharisaic daily synagogue prayers.
(7) all of this underscores the historical plausibility of the rabbinic reports of Gamaliel’s institution of the eighteen Benedictions as a daily prayer for all.
6. The confrontation of Judaism and christianity The above results on the matter of prayer have important implications for the communities involved. On a closer look the three named rabbis and the authors of the Didache and Matthew not only share their protest against the eighteen Benedictions in common. There is also a significant difference between them. eliezer, Yoshua, and akiva remained members of the rabbinic community.55 But the anti-Pharisaic, antirabbinic polemics of Matthew and the Didache makes it abundantly clear that their churches were not a part of this community. Nevertheless, Matthew shows many similarities with rabbinic vocabulary and terminology. We have noted some examples in passing. The implication is that the close ties that had existed between the Didache-Matthew churches and the Pharisaic-rabbinic community were now broken.
More precisely, the community we may see reflected in Matthew and the Didache now feels cornered by the apparent imposition of the eighteen Benedictions as a daily prayer and by the rule that fasting must be done on Mondays and Thursdays.56 The novel rabbinic movement led by Gamaliel the Younger, which Matthew and later church fathers keep indicating as
“Pharisees,” seems to have had had the means to impose its rulings on the nascent christian community and to encroach on their independence. Put 55. The banning of R. eliëzer is not connected to this; see b. B. Meṣ. 59b.
56. concerning this item, we have no tradition of a novel rabbinic rule. fast days were the subject of early rabbinic discussion, however, see Vered Noam, Megillat Ta’anit: Versions, Interpretation, History (Jerusalem: Yad Ben-zvi, 2003) (hebrew).
The DiDache: a MissiNG Piece Of The PuzzLe
differently, the Didache and Matthew register the impact of the rabbinic regime headed by Gamaliel.
as we have stated, recent discussion has included severe doubts about the real power of this regime. Particular attention was given to the Birkat haminim, the “Benediction about the heretics.” This concerns the second criticized element of elbogen’s theory that we must now discuss. We quote once again the relative Talmudic tradition about the institution of the eighteen Benedictions, this time with the sequel: shimon ha-Pekoli arranged the eighteen benedictions in their proper place before Rabban Gamliel at Yavne. said Rabban Gamliel to the sages:
“is there anyone who could put the benediction about the heretics in order?” shmuel the smaller rose and put it in order.57 (b. Ber. 29b; see also b. Meg. 17b; y. Ber. 4:3 [8a])
elbogen took this tradition simply to mean that indeed shmuel the smaller added the benediction at Gamaliel’s behest. While minim can indicate any kind of heretics, the implication would be “that this prayer actually referred to christians, and that it was one of the means to achieve full separation of both religions.” in support, elbogen referred to patristic reports about deprecations of christians in synagogues and to rabbinic traditions showing how Jewish christians would be damning themselves while pronouncing the Birkat haminim.58 in recent decades, this part of elbogen’s description has been rejected on the grounds that explicit mentions of “christians” occur only in textual witnesses of rabbinic prayers beginning in the fourth century.59 On the other hand, other 57. according to the Yerushalmi, shmuel the smaller’s Birkat haminim complemented the number of benedictions to become eighteen; according to the Bavli, he made them nineteen. cohen, “Nature of shim’on,” 551–53 suggestively proposes eighteen was seen as a round figure, not an exact number.
58. elbogen, Jüdische Gottesdienst, 36–39. see the more explicit presentation in alon, Jews in Their Land, 288–307.
59. schäfer, “sogenannte synode”; G. stemberger, “Die sogenannte ‘synode von Jabne’ und das frühe christentum,” Kairos 19 (1977): 14–21; Reuven Kimelman,
“Birkat ha-minim and the Lack of evidence for an anti-christian Prayer in Late antiquity,” in Aspects of Judaism in the Graeco-Roman World, vol 2. of Jewish and Christian Self-Definition, ed. e. P. sanders (London: scM, 1981), 22–44; Pieter W. van der horst, “The Birkat ha-minim in Recent Research,” in Hellenism, Judaism, Christianity: Essays on Their Interaction (Leuven: Peeters, 1998), 113–24.
TOMsON: The LORD’s PRaYeR
185
studies have upheld elbogen’s thesis specifically on the basis of the early christian evidence.60
for our purposes, we merely need to establish that the Birkat haminim plays no visible role in the conflict with the “Pharisees” or rabbis reflected by Matthew and the Didache. Nevertheless, those documents testify to a painful conflict. apparently the Pharisees-rabbis already wielded sufficient power without such an adaptation of the eighteen Benedictions.
Rabbinic and christian sources indicate that around the turn of the first century the incipient rabbinic movement took other measures, ostra-cizing “heretics” including Christians. Rabbinic rules effectively excommunicating minim appear to have been adapted or put in place, and by this time their applicability to christians must have become explicit.61
Of prime importance for dating this development is again the christian source material, especially the repeated communication in John that “the Jews” or “the Pharisees” had “already decided” (συνετέθειντο) to make christians ἀποσυνάγωγοι (John 9:22; 12:42; see also 16:2), which must be translated “put out of the community.”62 This source must be dated to around 110 ce.63
60. see esp. William horbury, “The Benediction of the Minim and early Jewish-christian controversy,” in Jews and Christians in Contact and Controversy (edinburgh: T&T clark, 1998), 67–110, and the introduction, 3–14.
61. The central rabbinic passage is t. ḥul. 2:19–24. see adiel schremer, Brothers Estranged: Heresy, Christianity, and Jewish Identity in Late Antiquity (Oxford: Oxford university Press, 2010); Joshua schwartz and Peter J. Tomson, “When Rabbi eliezer was arrested for heresy,” JSIJ 10 (2012): 1–37, http://www.biu.ac.il/Js/JsiJ/10-2012/
schwartzandTomson.pdf.
62. for this meaning of ἀποσυνάγωγος, see shaye J. D. cohen, “Were Pharisees and Rabbis the Leaders of communal Prayer and Torah study in antiquity? The evidence of the New Testament, Josephus, and the early church fathers,” in cohen, Significance of Yavneh, 266–81 (esp. 275–76); Peter J. Tomson, “The Wars against Rome, the Rise of Rabbinic Judaism and of apostolic Gentile christianity, and the Judaeo-christians: elements for a synthesis,” in The Image of the Judaeo-Christians in Early Jewish and Christian Christian Literature, ed. Peter J. Tomson and Doris Lambers-Petry, WuNT 158 (Tübingen: Mohr siebeck, 2003), 14–18, although there incorrectly restricted to the Birkat haminim.
63. On internal literary-historical grounds, this dating is made by Raymond e.
Brown, An Introduction to the New Testament, aBRL (New York: Doubleday, 1997), 376.
The DiDache: a MissiNG Piece Of The PuzzLe
7. historical and Theological conclusions
Let us sum up in some general conclusions, first on the level of history. The classic form of the “short prayer” taught by Jesus is extant in Matthew and the Didache. These documents were finalized when the “rabbinic” regime of Gamaliel the Younger took shape, and they signal resistance to the rules it issued concerning prayer and fasting. initial resistance involving prayer was also registered in the Mishnah. The influence of this regime that is here reflected proves its power, the doubts of modern scholars notwithstanding. The question of the sources of this power is beyond the purview of the present study.64
Matthew and the Didache do not contain echoes of a “Benediction about the heretics” being applied to christians, nor do they even show signs of the excommunication of christians evidenced in early rabbinic traditions and the Gospel of John. These developments must have occurred not long after the final redaction of Matthew and the Didache.
Without any polemics, Luke relates how Jesus taught his disciples a brief prayer, just as John the Baptist and other Jewish teachers did. We saw the potential authenticity of this report confirmed in rabbinic traditions about short prayers. also, the short version of the Lord’s Prayer that Luke copied from an unknown source sounds more authentic and more typical of Jesus. The Matthean text sounds more “rabbinic” and seems to have been taking shape within the community, a process that may have begun before the redaction of Mark. it must have been the subsequent suprem-acy of Matthew that made the Lukan version fall into disuse, although we could also speculate that the “ecclesial” features of the Matthean version correspond with the christian positioning in the emerging conflict with rabbinic Judaism.
We should not miss the opportunity also to draw some theological conclusions. first, theologically, there is no fundamental difference between the two versions of the Lord’s Prayer. By content and wording, both texts are closely related to the various Jewish prayers from the period.
Decisive differences are found only in the context. in both preserved versions, Jesus’s short prayer is a Jewish prayer.
second, in Matthew and the Didache, this Jewish prayer is framed in a setting reflecting conflict with the Jewish leaders called Pharisees or 64. see above in nn. 47–48.
TOMsON: The LORD’s PRaYeR
187
“rabbis.” here a theological dissonance is felt between the Jewish prayer text and the anti-rabbinic and ultimately anti-Jewish context in which it is mounted.65 By contrast, Luke’s framing of the short prayer of Jesus makes his own voice heard among those of other Jewish teachers. There is no dissonance between the Jewish text of his prayer and its context in Luke.
a third theological conclusion concerns present-day readers even more directly. in the framework of the New Testament canon, christianity’s break with Judaism is not an inherent necessity but is contingent on particular circumstances. Theologically, the continuity of Judaism and christianity depicted by Luke carries as much weight as the rupture reflected by Matthew. Judging by the way the prayer of Jesus is transmitted in the New Testament, the relationship of Jews and christians remains an undecided issue. it is the extracanonical text of the Didache that helps us realize this.
65. On Matthew’s eventual anti-Judaism, see Peter J. Tomson, “Das Matthäusevangelium im Wandel der horizonte: Vom ‘hause israels’ (10,6) zu ‘allen Völkern’
(28,19),” in Judaistik und neutestamentliche Wissenschaft: Feschrift, ed. Lutz Doering, hans-Günther Waubke, and florian Wilk, fRLaNT 226 (Göttingen, Vandenhoeck & Ruprecht, 2008, 313–33; and, “If This Be from Heaven,” ch. 6.
formalized speech in Didache 9–10
Jonathan Schwiebert
There is widespread agreement that certain phrases in the Didache’s meal prayers are fixed or formulaic. a short list by any measure would include the phrase “through Jesus your servant” and the doxologies that conclude each prayer unit. But a cursory glance at the following chart will reveal that, at least in three of the six prayer units, more than these two expressions are formulaic.1
Did. 9.2
Did. 9.3
Did. 10.2
We thank you, our
We thank you, our
We thank you, holy
father, for the holy vine father, for the life and father, for your holy
of David your servant,
knowledge,
name, which you made
to reside in our hearts,
and for the knowledge
and faith and immortal-
ity
which you made known which you made known which you made known to us through Jesus
to us through Jesus your to us through Jesus your your servant. To you be servant. To you be glory servant. To you be glory glory forever. [amen.]
forever. [amen.]
forever. [amen.]1
i have pointed out elsewhere that these fixed expressions tend to cluster toward the beginning and ending of these prayer units, leaving a central variable phrase (or two).2 a similar phenomenon can be seen in two other 1. Translations of the Didache are my own.
2. Jonathan schwiebert, Knowledge and the Coming Kingdom: The Didache’s Meal
-189 -
The DiDache: a MissiNG Piece Of The PuzzLe
prayer units, which (in the extant text of the Didache) conclude the pre-and postdinner prayers:
Did. 9.4
Did. 10.5
Just as this bread fragment was scat-
Remember, Lord, your assembly, to
tered upon the hills and, gathered,
rescue it from all evil, and to com-
became one, so let your assembly
plete it in your love, and to gather it
be gathered from the corners of the
from the four winds, [after it is made
earth into your kingdom.
holy]a into your kingdom, which you
prepared for it.
Because yours is the power and the
Because yours is the power and the
glory forever.b
glory forever.
[amen.]
[amen]
Notes
a. The words are lacking in the coptic and may therefore be a secondary addition. see Kurt Niederwimmer, The Didache: A Commentary, translated by Linda M. Maloney, hermeneia (Minneapolis: fortress, 1998), 161.
b. i have emended the text of the doxology. see schwiebert, Knowledge and the Coming Kingdom, 64–66.
although these two prayer units are verbally distinct, they do share key vocabulary and also conclude on a strikingly similar note—and with the same doxology (distinct, however, from the shorter doxology above). as i have pointed out elsewhere, if this longer doxology is a congregational response, then the final words of these two prayers (“into your kingdom”) is not open to change: it is formulaic, or fixed. The doxology, at any rate, is fixed.3
finally, even in the outlier (the only prayer unit not yet cited), we encounter the formulaic phrase “through Jesus your servant” (without, however, the phrase “which you made known to us”) and the (shorter) doxology that always follows it in Did. 9–10:
Ritual and Its Place in Early Christianity, LNTs 373 (New York: T&T clark, 2008), 88–91.
3. see schwiebert, Knowledge and the Coming Kingdom, 68–69. ironically, the doxology is not “fixed” in the textual tradition, but then different doxologies are in usage in the New Testament and other liturgical texts, one of which likely affected the manuscript tradition here.
schWieBeRT: PRaY “iN This WaY”
191
You, almighty Master, created all things for the sake of your name, and have given nourishment and drink to all humanity, for their enjoyment, so that they would thank you. But us you have favored with spiritual nourishment and drink, and eternal life though Jesus your servant. for all things we thank you, because you are powerful.4 To you be glory forever. [amen.] (Did. 10.3–4)
i take it as uncontroversial that formulaic speech exists in these prayers, then: certain phrases that are fixed in advance of the liturgy’s enactment through word and gesture and action. The degree of formulaic speech is disputed, but its presence, almost never.5
The more difficult question concerns how to interpret such fixed expressions. and here critics divide at best into observable tendencies. On the one side, and relatively few in number, stand critics who read the text of the prayers almost as if they were a theological treatise to be exegeted by standard hermeneutical methods. Representative here are Willy Rordorf and andré Tuilier. Observe that two fixed expressions in the first set of prayers above employ (1) παῖς (“child/servant”) and (2) γνωρίζω
(“make known”). Rordorf and Tuilier argue that the juxtaposition of these two terms is intended to mean that Jesus “revealed the profound significance” (2) of the Jewish notion of the messiah (1). Moreover, since both Jesus and David are called παῖς in the text of the prayers, Jesus is clearly a Davidic messiah, and the community’s prayers are themselves “Davidic.”6
in their hands the phrase “vine of David” (9.2), which adolf von harnack famously found impervious to interpretation, further shows that the church is “essentially founded on the messianic vocation of israel.”7 for 4. The text is somewhat suspect here; see schwiebert, Knowledge and the Coming Kingdom, 70–71.
5. The parallelism is so extensive that it has given rise to the suggestion that we have two versions of one liturgy, one in chapter 9 and the other in chapter 10. see Maurice Goguel, L’Eucharistie: Des origins à Justin Martyr (Paris: fischbacher, 1910), 237, 241–42; alan J. P. Garrow, The Gospel of Matthew’s Dependence on the Didache, JsNTsup 254 (London: T&T clark, 2004), 27. More plausible to me is a prayer pattern that the community follows in its meal prayers that bracket the meal.
6. Willy Rordorf and andré Tuilier, La doctrine des douze Apôtres (Didachè), sc 29 (Paris: cerf, 1998), 43–44. Note that to make this claim, Rordorf and Tuilier must directly link the wording of a formula used in several prayer units to the wording of a phrase that appears only once.
7. ibid., 45.
The DiDache: a MissiNG Piece Of The PuzzLe
them the prayers “proclaim” the parousia of the resurrected christ “not only proper to the end of time, but that it is already realized in the eucharistic synaxis.”8
Observe the procedure: authorial word choice, syntactical arrangement, and repetition of key terms are used by the interpreter to disclose the intention of the text, its meaning, or interpretation. This is essentially the manner in which exegetes traditionally tackle a Pauline epistle, for example. Put another way: what the reader should understand from the text is gleaned from grammatical-linguistic observations. Needless to say, most interpreters of the Didache’s prayers have been highly trained in such analysis, and the desire to employ their considerable powers of observation is entirely understandable.
it is on the face of it somewhat puzzling then that this exegetical approach is relatively rare: on another side stands the more common tendency merely to assert the formulaic nature of these phrases and assume that as such they are impervious to interpretation. so, for instance, Martin Dibelius in his classic source-critical study does not venture beyond a claim that the phrase “which you made known” means that Jesus reveals salvation.9 Regarding the doxologies of Did. 9, he states that each “has so little connection with the [preceding] request, that it is best understood as a stereotyped response.”10
Yet Dibelius does not entirely neglect exegetical tactics. consider his treatment of “the holy vine of David.” for him, if not for us, this is an opaque symbol whose meaning had become well known to those who used these prayers. he traces a tradition-historical lineage through Jer 2:21; Ps 79(80):9, 15; 4 ezra 5:23, and a (postulated) hellenistic synagogue prayer tradition. in light of this history, as a symbol the phrase can in effect be translated into another word: “salvation.”11 The meaning or intention of the text’s author then remains important, or at least relevant, to a proper interpretation of certain phrases within the prayers.
Kurt Niederwimmer’s approach is in general along the same lines. he calls the choice of the verb εὐχαριστέω only “formal,” “without theologi-8. ibid., 101.
9. Martin Dibelius, “Die Mahl-Gebete der Didache,” in Zum Urchristentum und zur hellenistischen Religionsgeschichte, vol. 2 of Botschaft und Geschichte, ed. heinz Kraft and Günther Bornkamm; Tübingen: Mohr siebeck, 1956), 117–27.
10. ibid., 122.
11. ibid., 119–20.
schWieBeRT: PRaY “iN This WaY”
193
cal significance.”12 even briefer is his treatment of the concluding doxologies: “The community gathered for the meal responds to this prayer by saying….”13 Yet he also occasionally shades into treating the language of the prayers as theological speech:
in place of the [traditional Jewish] address “Lord our God, Ruler of the universe,” we read, “our father.”… This predication certainly expresses the self-concept of the group, who understand themselves as the table company of children who receive the goods of time and eternity from the hand of the heavenly father.14
for Niederwimmer, a self-concept is expressed in the choice of the word
“father” over the choice of some other phrase (“Lord our God”). here at least the verbiage rises above “formulaic.” his reading of the phrase “holy vine of David” is even more in the vein of an exegetical treatment: i understand this to mean that in the presence of the cup and the wine, which are earthly gifts, the thoughts of the community are directed farther, to the heavenly good, which God has given to God’s own and which can be understood through the metaphor of the “holy vine.” The concepts are thus transparent: the subject of revelation is God, and its object is the “holy vine,” that is, salvation, and more specifically eschatological salvation, which was once promised but has now been given.15
interestingly, Niederwimmer once again derives (or at least defends) this interpretation from the juxtaposition of this phrase with the formulaic expression “which you revealed through Jesus your servant.” arrangement of phrases reveals meaning; word choice plays a secondary role in Niederwimmer’s reading.
a few more recent interpreters, paying closer attention to the ritual and oral-performative features of the text, have strained against these two options. aaron Milavec, reading the prayers through an oral-performative lens, sees symbolic reservoirs in the puzzling expressions of the prayers, which are to be tapped and verbally expounded by a prayer leader. even so, he ends by rendering these phrases into other, fuller ones meant to 12. Niederwimmer, Didache, 145.
13. ibid., 152.
14. ibid., 145.
15. ibid., 146.
The DiDache: a MissiNG Piece Of The PuzzLe
capture and bring out their deeper “meaning.”16 in that sense, though not in others, his procedure still resembles Niederwimmer’s.
Jonathan Draper, attuned to the ritual setting of these prayers, offers a somewhat analogous avenue of interpretation via Victor Turner’s symbolic approach. for instance, Draper takes the “vine of David” as a symbol (indeed a primary symbol) of the liturgy and finds that as symbol it alludes to “a rich symbolic tradition in the culture of israel.” This tradition is shared for Draper in the fifteenth benediction of the shemoneh esreh, isa 11:1–12, and various other texts within the Jewish scriptures and the Dead sea scrolls. The tradition reveals that the strange phrase “vine of David,”
taken as equivalent to “sprout of David,” “has both dynastic-messianic significance and also communal covenant dimensions.”17 in developing this interpretation, however, Draper too resorts to an exposition of the fixed expression, “which you revealed through Jesus your servant,” taking it to mean (in this connection) that Jesus “reveals the vine to those initiated into the community as king of the renewed kingdom.”18
under the heading of symbolism, then, we seem to move back into the domain of exegesis. Ritual formulas are parsed as units of language, however dense, and become again carriers of meaning not far different from the way in which normal language operates.
if it is possible to extrapolate from these few examples, scholars appear to differ on the question of whether the fixed expressions in these prayers can bear meaning as words. The reigning approach is apparently to acknowledge the essentially uncommunicative nature of formulaic speech. But wherever an opening seems to present itself, interpreters are in the habit of looking for (and finding) thoughts and ideas expressed in the language of the prayers. The terminological decisions, the grammatical constructions, and the juxtapositions of phrases remain exegetical keys by which we might grasp after the denotative meaning of the prayers’ language. as language, the assumption goes that we can translate 16. e.g., his treatment of the “holy vine”; see aaron Milavec, The Didache: Faith, Hope, and Life of the Earliest Christian Communities, 50–70 CE (New York: Newman, 2003), 361–64.
17. Jonathan a. Draper, “Ritual Process and Ritual symbol in Didache 7–10,” VC
54 (2000): 121–58 (150).
18. ibid., 148. Like Rordorf and Tuilier, then, Draper’s interpretation requires a direct link between phrases that occur repeatedly and a phrase found only once.
schWieBeRT: PRaY “iN This WaY”
195
the prayers into other equivalent verbal forms, thereby accessing a fuller sense of their meaning.
i would suggest that these two tendencies exist in unexamined tension within the literature. if, on the one hand, formulaic speech is impervious to interpretation, then its juxtaposition with nonformulaic speech (or simply, puzzling expressions) cannot disclose some deeper meaning in those expressions. Put another way, a verbal object that people use but do not think of as “words” cannot become “words” in the hands of the interpreter who needs them to be “words.” and this procedure is all the more problematic when the same interpreters treats the expressions as objects (and not “words”) elsewhere in their treatment.
Let me illustrate the two tendencies with a trivial example: in my culture, the expression “goodbye” is used mostly (i gather) as a kind of verbal object. suppose someone said something nasty to me on the telephone and then concluded the conversation with “goodbye.” Would it be legitimate to take the “good” from “goodbye” and use it to soften the nasty comment?
could i drag out the ancient origins of the phrase in “God be with ye” and argue that the nasty comment was a joke and the person’s true intentions were positive? failing these, could i at least claim that by using the polite term “goodbye” the speaker meant to maintain a relationship, despite the nasty remark? in fact, none of these interpretations of “goodbye” would be valid or at all plausible. What then is a more useful approach to fixed formulas like “goodbye”? Do they mean nothing at all? This conclusion would also be unwarranted, i think. Yet these are the two tendencies that a closer attention to the uses and properties of formulaic speech can help us move beyond.
in what follows i have selected one scholar of ritual and, in particular, ritual speech whose account seems to me both persuasive and capable of clarifying central issues at stake. Others could be chosen. This is my choice and is not put forward as the last (but rather the almost first) word on the subject.19
19. One thing in the above account i find both plausible and useful: the idea that certain turns of phrase have a prior history in the scriptures and traditions of Judaism and early christianity. But what use we must make of these tradition-historical observations is not yet clear until we know how such formulas function.
The DiDache: a MissiNG Piece Of The PuzzLe
1. The Limitations of formalized Language
Maurice Bloch has argued with respect to religious rituals that symbols, songs, and dance embody an extreme form of tradition. choreography and script reveal a tradition so fixed that the ritual participant must accept that tradition in advance or abandon the performance prior to beginning it. But in agreeing to a preformulated script, he argued that language undergoes severe limitation.20
The Didache’s meal prayers contain not only formulaic speech, but arguably a kind of script. The fixed phrases identified above, which conclude each of the six prayer units, are in every case followed by predictable doxologies. critics disagree whether the community would respond to the prayer units with the doxology itself or with an “amen” (found only in the coptic). either solution advances the main claim: the wording of the prayers is signaling a verbal exchange between the participants. The language is fixed. it is even conceivable that these prayers are entirely fixed, that they were meant to be performed verbatim. such a conclusion seems unwarranted, however, precisely because they vary in predictable places.21
When we encounter a script, we are immediately confronted with the notion of “choice,” so important for exegetical treatments of texts. Bloch points out several categories of choice that a speaker has in “everyday speech acts.” These include loudness, intonation, choice of vocabulary, flexibility in the sequence of utterances, and a full range of stylistic variation. in formalized speech acts, on the other hand, each of these categories can be curtailed and often is. for instance, only certain tones of voice may be appropriate, only a partial vocabulary may be available, and importantly, the sequence of speech acts may be fixed in advance.22
Bloch further points out that this surrendering of choices has “a geo-metrical effect on the restriction of forms of speech”: if we think of the sentence “the cat sat on the mat” and assume that there are three alternative nouns for each noun locus we find that there are nine possible sentences. if we say that the verb can be in one of three 20. Maurice Bloch, “symbols, songs, Dance and feature of articulation: is Religion an extreme form of Traditional authority?” EuroJS 5 (1974): 55–81.
21. see my discussion in Knowledge and the Coming Kingdom, 66–71, 78–83, 88–89, and sources cited there.
22. Bloch, “symbols, songs, Dance,” 60.
schWieBeRT: PRaY “iN This WaY”
197
tenses there are then twenty-seven possible sentences and then that these sentences can be said in one of three intonations, there are then eighty-one possible sentences. Now if we look at this backwards we see that by just specifying one restriction, say intonation, saying that only one intonation is suitable, the number of possible sentences drops from eighty-one to twenty-seven. If, however, we bring in two restrictions occurring at the same time, say that with a certain intonation only certain nouns are acceptable, the effect on the choice of sentences is dramatic. in fact it falls from eighty-one possible sentences to three.23
Given our exegetical predilections, it is important to linger here a moment.
Remember that we humans use all of these choices (intonation, word choice, verb tense) to convey our meaning to one another. Not surprisingly, therefore, when we come to interpret what someone has said or written, we are prone to weigh the writer’s or speaker’s choices to interpret their meaning insofar as we can. in the case of texts, we have no access to intonation (a sad loss, but largely unavoidable), and therefore we tend to put additional emphasis (in our interpretations) on the word choices and syntactical arrangement and verb tenses we find.24
Meanwhile, there is another kind of restriction to observe: sequential communication. Bloch writes: “The first and the most obvious of the implications of abandoning linguistic choice is that an utterance instead of being potentially followed by an infinity of others can be followed by only a few or possibly only one.”25 an almost infinite number of responses can follow an everyday speech act, but in situations where only a few responses are appropriate: “with increased formalisation A predicts to an ever greater extent B.” he sums up:
in terms of the experience of an individual it means that as he uses formalized language he very largely implies his last words by his first, since once he is speaking in the right way there is only one predetermined line along which he can proceed.26
in fact, one speaker’s words can (within the accepted code) predict the next speaker’s words, including the appropriate tone, the choice of vocabulary, 23. ibid., 61 (emphasis mine).
24. a glance at a standard biblical commentary will illustrate my point admirably.
25. Bloch, “symbols, songs, Dance,” 62–63.
26. ibid., 63 (emphasis mine).
The DiDache: a MissiNG Piece Of The PuzzLe
and so on. in these situations, the language’s communicative power is radically reduced. The speaker’s ability to “intend” and “express” (precisely what we normally exegete) is lost.
Put another way, what may appear to be a conversation between two parties becomes in a formalized context completely unlike an actual conversation. in everyday speech acts, speakers are communicating information and other things, like emotional states and relational connections (“how are you?”—“Not all that well, honestly.”—“Really, what’s wrong?”).
in formalized speech acts, this kind of exchange is ruled out by the “rules,”
by the obligation of the second speaker to say a certain thing and not something else. That loss of true conversation, with its lost freight of new information, ought to impact how we interpret a formalized exchange.
Take, for instance, a conversation in which “how are you?” can only (politely) be answered with “fine. how are you?” There the conversation says nothing about the actual state of either speaker and only communicates acceptance of a polite code, an acknowledgement of the other’s existence within that code, or some such thing. The “meaning” of the words has nothing to do with the “content” of the words.
as this illustration demonstrates, among the things “lost” in a formalized exchange is the expectation of a neat fit between language and particular situations or circumstances. Two routine functions of language, Bloch contends, “disappear” in formalized speech: “the ability of language to communicate messages concerning particular events and its ability to convey specific messages leading to particular action.” But formalized speech excels in another arena, namely, asserting unity. as words and utterances lose their flexibility, their capacity to adjust to the particular needs of the moment, they gain the ability to make “all events … appear as though they were all alike.” Put another way, words are uttered in a given context that are appropriate to that context. They do not express by any necessity the actual state of affairs but rather the polite or accepted utterance befitting the current occasion. They assert, for whatever reason, that this is how it ought to be and consequently say nothing about how it actually is.27
27. ibid., 62. examples from political rhetoric abound, including for instance expressions of praise heaped upon one politician by another, which are unreliable pre-dictors of actual esteem or respect. Bloch uses his observations also to think about political speech acts and traditional utterances and their occasions before turning to the matter of ritual utterances, which is our present concern.
schWieBeRT: PRaY “iN This WaY”
199
for these reasons and others, Bloch affirms that “to look for ‘meaning’ in a fixed utterance without qualification is clearly misleading.” consider the differences between a script and an argument. in an argument, a statement B that follows a statement a is (1) open-ended and (2) open to contradiction. Neither of these properties holds in formalized contexts: we have already seen that, there, statement B is not open-ended, but predicted in advance by statement a. The second point follows from the first: you cannot contradict statement a if your answer is a predetermined statement B and one of agreement (like an “amen”). “formalised language is therefore non logical and any attempt to represent it as such, whether by a paraphrase into ordinary language which implies ‘explanation’ or by the use of tabular representation containing a logical form, is misleading.”28
Given Bloch’s argument (and if it is sound, as i think it is, at least in its leading points), we must acknowledge that a large number of efforts to interpret the vocal utterances of ritual speech, including at points the texts of the prayers in Did. 9–10, are incompatible with the observation that these are “fixed utterances” wherein the speaker (whether the prayer leader or the one who responds to the prayer leader) has little or no choice about such “everyday speech” elements as selection of vocabulary, tense of verbs, syntactical structure, juxtaposition of phrases, and (probably) intonation. in other words, if Bloch is correct, the patterns that almost all students of these prayers acknowledge, discussed briefly above, rule out recourse to a grammatical-linguistic interpretation. The person who prays
“in this way” accepts a code, and the person who responds in kind either with the doxology or the “amen” also accepts that code.
Does this mean that we cannot interpret the prayers at all, that the phrases are meaningless? in fact, it happens that Bloch’s observations also challenge the other tendency in the literature, to which we now turn.
2. interpreting the Prayers through Their context Part of what interests Bloch is the question of how formalized language
“communicates without explanation,” especially since language in ritual has often been taken to “explain” some aspect of reality. his argument challenges us to consider that the expressions in our prayers are not trying to 28. ibid., 66.
The DiDache: a MissiNG Piece Of The PuzzLe
explain anything. as he puts it, in formalized speech “communication has stopped being a dialectic and has become a matter of repeating correctly.”29
This little quote is a nice clue to the drift of his argument. “Repeating correctly” is about compliance, not communication. for Bloch, the
“force” of formalized language is “traditional authority, but disguised in that it has been accepted unconsciously before the event by the acceptance of the proper, of the polite, or the appropriate way of behaving.”30 One accepts not the “truth” of the thing being said, but the authority behind its formalization. Drawing from J. L. austin, Bloch refers to this authority as
“illocutionary force,” “not to report facts but to influence people.”31 how, then, do we interpret it? Or can we interpret it at all?
a useful place to begin is with Bloch’s remark that “the experience of the ritual is an experience fused with its context and therefore only an attempt to explain what this event as a whole is for is an explanation of the content.”32 here we encounter several insightful leads, each worth exploring. The first is that ritual is something one experiences only in a particular context. Our first task when we come to interpret any particular ritual object or verbal formula is to clearly define (insofar as possible) where and when and by whom it is used. Bloch’s remark offers also a second lead, a complement to the first: a ritual’s purpose, its “context,” is already an explanation for the ritual’s content. Bloch does not clarify whether this is an all-encompassing explanation, ruling out other modes of interpretation, but he does lead us to suppose that context deserves pride of place. Moreover, he adds, we must consider not a particular moment in isolation but rather the whole ritual within which this particular moment occurs. here alone will we find the “context” that provides a privileged insight into the ritual speech we aim to understand.
Turning to the Didache’s prayers with this advice in hand, we discover that a fair amount of information is available to us regarding this ritual’s larger context.33 We can infer, for instance, that this formulaic speech 29. ibid., 72.
30. ibid., 66–67.
31. ibid., 67 (citing austin, How to Do Things with Words, 234).
32. Bloch, “symbols, songs, Dance,” 77. he goes on: “because the units of meaning are changed so that they can serve purposes only for religious events, and since they are inseparable from the context, there can be no explanation without an explanation of what the event or the context is.”
33. for the purposes of this present analysis, i forego any discussion of redactional
schWieBeRT: PRaY “iN This WaY”
201
comes immediately before and after a meal, and one in which people are expected to eat until their appetite is sated (10.1). Peeking outward into the rest of the Didache, we know (or can infer) that this meal alternates with a cycle of fasting: regular twice-weekly fasting and occasional additional fasting with novitiates (7.4–8.1). We know that (at least ideally) everyone eating the meal has received a minimum of instruction in the way of life, followed by immersion in water or pouring from a cup (7.1–3). We know too that diners can expect to encounter spirit-inspired speech at their meal (11.7–12). and last of all, we know that the meal is called a “thanksgiving”
and is considered “holy” (9.1, 5).
When we come to interpret the content of the meal prayers, then, we can (and should) bring to bear this whole network of associations of context. at the level of abstraction that Bloch seems to recommend, the ritualized, authoritative speech we encounter in these prayers is about the community and its gathering. By embarking on the thanksgiving with its prayers and responses, the community celebrates and reaffirms its gathering. it, so to speak, asserts the reality of its existence as a meaningful group in the midst of an environment that must surely entail rival allegiances: to family, to city, to empire, and so on. fasting from food “out there” contrasts to partaking fully “in here,” making the meal around which these acceptable words are said into a “thanksgiving” for a kind of community that is not obvious or given.
The extent of tension with alternate allegiances, which can here only be suggested, may shed interesting light on how freighted this particular fasting-eating alternative is for these ritual participants. for instance, fasting from meals implies (does it not?) abstinence from family or other group gatherings. But this is only partial. The fasting occurs only on two days of the week. it is possibly tied to the preparatory rite for inclusion in the community. at least the instructions lay side-by-side. so then, the meal partly displaces family membership or identity, but not completely.
Meanwhile, fasting on days that do not coincide with “the hypocrites”
(that is, other Jews) seems to entail a more radical kind of dissociation.
The cycle of fasting completely disagrees with these others’ cycle of fasting layers in the text. What i say here applies best (and perhaps only) to the final form of the text. earlier layers (for which we must subtract this or that detail) might require adjustments to the “context” and hence the interpretation i provide. My own effort to relegate each relevant detail to a particular redactional phase can be seen in Knowledge and the Coming Kingdom, 148–82.
The DiDache: a MissiNG Piece Of The PuzzLe
(both occur only on two days, but these two days are entirely distinct).
There is no overlap except in the idea that it must be done.34
We can add another layer of context by picking up the observation from above that the formulas used in the prayers are often (though not always) drawn from Jewish tradition. it might be possible to say that Jewish tradition is the silent authority that has been accepted by those who gather at this meal were it not for the intrusion of phrases that cannot be ascribed to that tradition. The most obvious of these is “through Jesus your servant,” which obviously does not arise from mainstream Judaism. We are thus left with a community with Jewish traditions that is not coterminous with the Jewish community, that practices a Jewish sort of fasting discipline but on different days, and that views these other Jews as “hypocrites.” The traditional language of the prayers seems to reinforce this community’s Jewish identity, even while it disrupts it by asserting a difference, a peculiarity, reinforced by the language of “holiness” and the other rites associated with this one.
We can go further. Philo of alexandria claims that “(it would be irreverent) … and equally unlawful to enjoy and partake of any form of food for which thanks had not been offered [εὐχαριστήσαντας] in the proper and rightful manner” (Philo, Spec. 2.175).35 This remark suggests that the idea of offering verbal thanksgiving for a meal is a peculiarly Jewish instance of
“the appropriate” and “polite” thing to do. The prayers in this context mean something like “we too are pious.”
But again Bloch presses us to answer the question of what this entire ritual is for. Only then will we understand its content. When we say “entire ritual,” in this case i think we must include the cycle of fasting, instruction, and immersion or pouring: the preparatory rites, if you will. We must also bring in the ecstatic speech, the sense of holiness, the hedging about of the meal: its immediate rites and usages. Next, we must not forget that the ritualized speech in this case is prayer, speech addressed to God (not 34. Would agreement that it should be done have been seen by members of either community as a point of commonality? it is difficult to say. Jews, Muslims, and christians agree that God should be worshiped in community one day a week, though again on different days. Do they see this point of agreement as meaningful, binding them into one community?
35. cited in huub van de sandt and David flusser, The Didache: Its Jewish Sources and Its Place in Early Judaism and Christianity, cRiNT 3.5 (assen: Van Gorcum; Minneapolis: fortress, 2002), 321. This was evidently a customary term for meal prayers (so Jean LaPorte, Eucharistia in Philo [New York: Mellen, 1983], 53–55; Paul in Rom 14:6; and the eighteenth Benediction of the amidah).
schWieBeRT: PRaY “iN This WaY”
203
the people) and not in the persona of an elder or some ancestral figure but rather in the voice of the whole assembly: a “we.” someone (we do not know who) speaks for the whole assembly, and the assembly permits this, because this is at least a part of what the gathering means: they are there, among other reasons, to speak as one to God. finally, there is the food. The prayers justify the sharing of the meal and make it something more than the obvious: a holy, quasisacrificial feast (9.5; 14).
i have already addressed the fasting. The content of the instruction is also relevant: it consists almost entirely of ethical rulings, guidelines for a way of life that stands in sharp contrast to a way of death. This instructional content evidently culminates in ritual washing, after which the meal is open: the washing acts as a sentinel on the ritual of the meal. Being immersed (or poured over) “in the name of the Lord” suggests within a Jewish cultural context an association with that Lord characterized by belonging, perhaps an allegiance or maybe discipleship.36 interestingly, the washing is apparently also the gateway to ecstatic speech, which can (and cannot) be critiqued by the community (11.7–12). apparently the one-time recipient of instruction can become by possession of the spirit a purveyor of knowledge and insight. But the Didache says very little about
“spirit” or its possessors, the “prophets.” The danger associated with spirit-talk is above all that it might lead people to do things that contradict the surer teaching received from the tradition. But even the ritual formulas used at the meal setting can be supplemented by prophets (10.7).
The larger context then has begun to come into view. The ritual speech under investigation takes place at a regular communal meal. as such it asserts the (Jewish or traditional) piety of the group, its adherence to what is “proper.” at the same time the meal is uncommon (“holy”), reserved for those who have undergone a period of instruction that clarifies for them in what ways they are (however fictitiously) distinct from outsiders.
This instruction has a clear terminus, marked ritually by a water rite that involves fasting. fasting reappears also in the weekly cycle of behavior that binds the community in a common calendar (an experience of time) and contrasts in a concrete way with the eating that takes place at the meal.
The community that fasts together also eats together. They are “together”
in time and then again literally together at the meal. The ritual speech 36. Paul (himself a Jewish man, of course) uses the phrase “baptized into Moses”
in a similar way (1 cor 10:2).
The DiDache: a MissiNG Piece Of The PuzzLe
we are looking at is not so sacrosanct that it should rule out spirit-talk.
This is a community that holds a role for spirit-people (“prophets”), but all spirit-talk must be held to the standard of the tradition, however toler-ant of spirit-people one may be. finally, the whole ritual serves (among other things) to reassert the existence of the community that gathers for the meal and the legitimacy of the meal itself as something more than a common, everyday meal. all that the community is, is bound up with it: yoked to the Lord Jesus (and his teachings), partially distinct from the surrounding world and other allegiances, more radically distinct from certain Jews, on the path to life, open to spirit.
all this we learn, not from an examination of the “meaning” or
“syntax” of the prayers themselves, nor by translating them into an exposition. all this rather comes from the “context” that the ritual inhabits within the Didache as it now stands and within the world from which it was written. its assertion of the essential existence, the meaningful unity and identity of the group, is simply that: an assertion. We have no way of knowing whether “on the ground” community members “believed” or felt invested in this identity above all others. But in some important ways (though bets are hedged here, interestingly) the community’s existence and unity asserted by virtue of the prayers taking place are made primary, a reference point for all of life: ethical, ritual, and eschatological.
Perhaps some readers may wonder if this is enough, or if we have traded away too much of the “meaning” of the words in this kind of contextual interpretation. for on this reading, why say these words and not some other words? Why the insistence on particular formulas, if their denotative meaning is lost? here then i think we must come back to a question left unanswered above: whether the context of the ritual is coterminous with its interpretation or whether there remains a role for words to communicate in something more like the manner of normal speech. in other words, we must critique Bloch’s theory.
3. The Role of Tradition or Possible Limits to Bloch’s Theory for Bloch, “the interesting question is the disconnection between the religious statement and the real world, a disconnection which is produced by the mode of communication of ritual.” he goes further: ritual speech, for him, “serves to hide the actual situation and preserve authority.”37 here 37. Bloch, “symbols, songs, Dance,” 77 (emphasis mine).
schWieBeRT: PRaY “iN This WaY”
205
that would mean the assertion (fiction?) that this is more than a meal and, in fact, somehow embodies an extraordinarily meaningful community. in coming to terms with his theory, we must meet this final question, then, and decide whether ritual speech is ultimately an extreme form of control or “authority.” Bloch ascribes just such a potent force to “tradition.”
Let me sketch out two possible directions we may take going forward.
On the one hand, with Bloch we might conclude that what is happening in the Didache’s ritual choreography is an assertion from above, from tradition, of the existence and purity of a community that does not exist in any essential sense. The ritual formulas serve to hide the multiple identities and mixed allegiances of the participants as well as the material aspects of the food. They do so at the loss of flexibility in language, in other words, at the loss of actual give-and-take and negotiation that may result in more conditioned but realistic agreement among the participants. for Bloch the Didache’s ritual can succeed in asserting this kind of unified identity and allegiance, so long as the participants perform their roles, saying the lines scripted for them in advance. and in many cases they will do this, having no choice.
But (continuing with the first alternative) insofar as this kind of ritualized, formalized speech is practiced, it cannot help being a top-down assertion or, if you prefer, control from without. it entails a loss of freedom, not only of expression, but of self-definition, self-assertion, and defiance of tradition. Bloch’s interpretation then is a fairly dismal view of ritualized speech and its capacity for human fulfillment. it seems in some sense to make ritual into control or power and on some level to exist outside of the individual who participates in the ritual. it might perhaps even make the ritual participant into a colluder, one who participates in their own restriction, self-limitation, and even individual annihilation.
in seeking refuge from such a dismal view, we might look to the example of marriage. here too scripted lines are found that once begun must inevitably lead to the end (for example, “i now pronounce you”). in a traditional wedding each line would be carefully prescribed. The things spoken, the objects used, all may be dictated from without, from tradition.
This all might well be described as a fiction, a cultural imposition, and a form of self-limitation. But the individual participant is likely to feel quite differently about the ritual. for instance, the ritual speech may well come after a lengthy period of prior negotiation between two equal parties. it might be accompanied by intense emotional investment, and as a result it might reflect the “true feelings,” the deepest aspirations and hopes of
The DiDache: a MissiNG Piece Of The PuzzLe
the ritual participants. Thus, even though countless generations before me have uttered the words “to have and to hold from this day forward, for better for worse, for richer for poorer,” i might well feel a profound investment in precisely these words that i utter to my beloved. i might believe them to be true, hope that they will prove true. i might invest myself fully in the ritual utterances as though they were words that i chose from an infinite number of possible choices to express my particular love for and commitment to this particular person.
in other words, ritual participants might well so invest in or agree with the “context” of a particular ritual occasion that they (from their own resources) fill the ritual speech used within that context with a meaning that is more akin to “everyday speech acts.” But there is one difference in that case: the speech acts take from tradition an added weight of authority. One becomes in such an instance infused with the context, the setting, the “polite” or “appropriate” in such a way that words here can mean more (not less) than words can mean in any other place. so it may be for the true believer, at least sometimes, if not always.
The ritual does not prescribe such an emotional state, but where the stakes are high (as often in marriage), it does invite such an investment.
and where it occurs, i am not convinced that the critic, standing outside the liturgy and looking on from an analytical perspective, can rightfully dismiss such an investment as collusion in oppressive authoritative structures. for perhaps ritual participants really do experience the best form of themselves in the saying of these words, which carry more meaning here than words can possibly carry in the rough-and-tumble mundane world.
in any case, insofar as members of the community mean what they say in these prayers, they will be asked to invest in the fragile community that gathers here, as nowhere else, and that speaks these like no other words. To some readers, this may perhaps be a subtler form of control. But to those who believe, it may instead unleash the freedom to hope for a better world to come.
in other words, and by way of conclusion, the choice to embark upon the ritual is not trivial, and the choice of which ritual to embark upon is likewise not trivial. The choice of this ritual script over some other possible ritual script, whether from Jewish tradition or the emerging christian movement, can indeed tell us something significant about the community that prays “in this way.” The choice of words, the juxtaposition of particular phrases, the verb tenses, cannot perhaps be translated into other words like those of a homily or an argument, but they have meaning in
schWieBeRT: PRaY “iN This WaY”
207
that they embody traditional, if fixed and scripted, hopes and aspirations of the community. The formulas may not act exactly as symbols pregnant with meaning, at least not in a clear or obvious sense, but the participants in the ritual would have been free to invest something of their identity in the words themselves as if they were choosing just what to say at just this moment. No doubt that choice could at other times have amounted to nothing more than a verbal object, the polite, proper, accepted thing to say before and after partaking of food, a way of pretending the meal meant more than the ingestion of food. But whether out of politeness or with genuine investment, those who chose to embark on this ritual thereby asserted the meal to be a gathering of unusual significance, a place where
“we” are set apart and “our” coming together is about something more than the obvious food on the plate.
The Ritual Meal in Didache 9–10:
Progress in understanding
John J. Clabeaux
The meal rituals presented in Did. 9–10 and referred to in Did. 14 have long eluded consensus as to how the meal was celebrated and what exactly it involved. it has given rise to speculative reconstructions that are presented as the true understanding of the ritual. in the last two decades a greater possibility of consensus is emerging, which is informed by liturgical studies such as those of Paul Bradshaw and the applications of ritual studies by Jonathan Draper and Jonathan schwiebert. schwiebert’s Knowledge and the Coming Kingdom1 represents a significant advance in comprehending not only what the prayers in Did. 9–10 meant, but how they fit within the development of early christian liturgical practice. in the “afterword” of his classic commentary on the Didache, Kurt Niederwimmer mused: it is not easy to locate the Didache within the whole sweep of the history of early christianity. still, one can form a more or less complete picture of this document, and from that alone is it possible to give a rather accurate estimate of the place occupied by the Didache.2
i believe schwiebert to have succeeded in this, and i focus on his work for three additional reasons: it is one of the most sustained and detailed treatments of Did. 9–10. second, schwiebert has been an active participant in the “Didache in context” group of the society of Biblical Literature. he 1. Jonathan schwiebert, Knowledge and the Coming Kingdom: The Didache’s Meal Ritual and Its Place in Early Christianity, LNTs 373 (London: T&T clark, 2008).
2. Kurt Niederwimmer, The Didache: A Commentary, trans. Linda M. Mahoney, hermeneia (Minneapolis: fortress, 1998), 228.
-209 -
The DiDache: a MissiNG Piece Of The PuzzLe
built well on foundations laid by Draper3 and with direct support from helmut Koester (harvard university), a New Testament scholar and early church historian who spent the greater part of his career in the study of the sayings of Jesus traditions. The help schwiebert received from Jewish scholars Paula frederickson (Boston university) and Jonathan Klawans (Boston university) enabled him to avoid the mistakes often made in Didache research involving Jewish prayers and customs. finally, in this book schwiebert has significantly clarified that the central focus of the thanksgiving prayers and ritual are the knowledge communicated in the tradition of the sayings of Jesus and the immanent onset of the eschaton.
he accomplishes this by careful textual analysis, with attention to oral features of ritual development and ritual dynamics, and then by a study of the trajectory of the Didache meal prayers in later christian literature.
he brings this off well, showing restraint and nuance in demonstrating that: (1) the Didache meal ritual does not invoke the Last supper or deal directly with the death and resurrection of Jesus, as is often assumed until relatively recently in discussions; (2) the sayings material in the Didache relates most closely, but not perfectly, to the sayings traditions in Q more than to the Gospel of Thomas or to the Gospel of John; (3) the meanings of the Didache meal prayers were sometimes absorbed by compromise but in the end were misunderstood and overridden both in orthodox works such as the apostolic constitutions and canons and equally badly by ascetic and gnostic egyptian texts, resulting in what schwiebert calls “a dead end that has two faces”4 for the Didache meal prayer tradition.
The story of the use of the Didache meal prayers in different, even conflicting, traditions is the subject of the last three chapters of the book.5
it is one of the most instructive sections of the book and demonstrates that the later use of a text is not a reliable indicator of the original meaning of the text. The meaning of the Didache eluded both the orthodox appropriation, with their explicit Last supper and passion connections, and the Valentinians, among others, who made of it what they would. in the end the meaning of the Didache’s prayers must be interpreted from what is said in Did. 9 and 10. We simply do not yet have sufficient insight into what the Didache community was actually like to be confident of the accuracy 3.3 Jonathan Draper, “Ritual Process and Ritual symbol in Didache 7–10,” VC 54
(2000): 121–58.
4. schwiebert, Knowledge and the Coming Kingdom, 239.
5. ibid., 183–250.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
211
of our readings of these prayers. But thanks to schwiebert and those on whom he depends, we are getting closer.
in this essay, i will address these three demonstrations by schwiebert in greater detail and highlight matters of special significance for the advancement of Didache studies. i conclude by commenting on theological implications of the results under discussion for those who wish to understand christian meal prayers today.
1. The Didache Prayers Themselves
in his first four chapters schwiebert provides a detailed “state of the question” on the challenges Did. 9–10 presents to understanding how it relates to what was to develop as christian eucharistic prayers. The first chapter (“Ritual history of an idiosyncratic Tradition”) makes a cogent plea to consider it likely that, in the earliest days of the Jesus movement, there could be markedly different understandings of what the community meal involved and communicated. for years research was taken up with the discussion of whether this was a eucharist or an agape, presuming a perfect agreement on what exactly those two terms described, where or whether there were “words of institution” (there are none), at what point the bread and wine were changed into the body and blood (neither body nor blood are mentioned in Did. 9–10), and whether the Didache eucharist involved an actual “filling” meal or only the “eucharistic elements.” it is made clear in the chapters of schwiebert’s book that follow (chs. 2–4) that the actual meal came after the prayers of Did. 9, and the meal was then followed by the blessings of Did. 10. This is not like present mainstream christian practices, but it is what a plain reading of Did. 10.1 requires when it says,
“Then after you are full, give thanks like this.”6
One thing has been and remains for many people not a question, namely, that the prayers of Did. 9–10 were based on Jewish meal prayers.
This assumption almost always specifies the Birkat hamazon with which the Didache prayers have only a few words in common. i have yet to speak to a Jew familiar by experience with the Birkat hamazon who sees any significant similar content. But year after year, this is passed on as what 6. Much as people would like to see the liturgical dialogue in Did. 10.6 (“if anyone is holy, let him come”) as some sort of invitation to come forward to receive, the plain meaning of 10.1 rules out the possibility. Translations of the Didache are from schwiebert, Knowledge and the Coming Kingdom, 63 (10.1) and 72 (10.6).
The DiDache: a MissiNG Piece Of The PuzzLe
schwiebert describes “tongue in cheek” as “a ‘golden link’ between Jewish meal prayers and the christian eucharistic anaphoras.”7 There are more terms and concepts in parallel with the prayers of Did. 9–10 in the Lord’s Prayer as it appears in Did. 8 than there are to any Jewish meal prayers we can actually establish as in use in this period. To this we will shortly return.
in chapter 2 (“The Last supper Tradition”), schwiebert presents the requisite parallel column displays of the Markan, Matthean, Lukan, and Pauline (1 cor 11) eucharistic prayers and then their later interpretation in ignatius of antioch, John 6, and Justin Martyr. This is familiar ground, but the way that schwiebert presents it is very effective in that it sets a baseline by which one can see that almost everything in Did. 9–10 is different from the canonical Last supper tradition. in the four eucharistic texts from the Gospels and epistles, and in all three of the later church fathers, one sees “body and blood” and connection to the passion narrative, neither of which is mentioned in Did. 9–10. The assumption that it must be implied in Did. 9–10 is undone by schwiebert’s thorough ritual analysis in chapter 3 (“The Logic of the Didache’s Meal Ritual”). This is an important chapter. i am not convinced about certain details of his argument, but on the whole i find it quite convincing. i noted after reading it how much he owed to Draper’s “Ritual Process and Ritual symbol in Didache 7–10.”8 in that essay, Draper showed how Victor Turner’s three stages of initiation rites (“separation, liminality, and aggregation”) are modeled in Did. 1–10. it is a highly satisfying explanation not only of the logic of Did. 7–10, but actually of Did. 1–10 in that the teaching of the Two Ways will separate the gentiles seeking to enter into the community.
They are in a liminal state just prior to baptism in Did. 7, for which they and those carrying out the rite with them must fast. Then in Did. 8 they are given the prayer that, in their praying it three times a day from that day on, will remind them of that rite of liminality, instantiate their separation from other gentiles and the community’s prerabbinic competitors that the author calls “the hypocrites,” and aggregate them with their fellows who also pray it three times a day. The meal ritual is the consummate means of enacting and remembering their aggregation. i will return to this. schwiebert’s interpretation is affected by Draper’s, but rather than repeat Draper’s arguments, schwiebert provides further evidence from 7. schwiebert, Knowledge and the Coming Kingdom, 10. he is no more convinced of this “golden link” than i.
8. Draper, “Ritual Process and Ritual symbol.”
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
213
his analysis of the “fixed elements” that appear in the prayers in Did. 9–10, namely, (1) the “we thank you father,” which occurs exactly three times (9.2, 3; and 10.2), (2) the “which you made known” clauses, which also occur three times in the same verses (9.2, 3; and 10.2), (3) the “gather into your kingdom” petition, which appears at the end of each set of prayers (9.4 and 10.5), and (4) the doxologies throughout the prayers, four short (9.2, 3; 10.2; and 10.4) and two long (9.4 and 10.5).9 schwiebert indicates the ritual effects of each of these. The “we thank you” prayers provide “a formal realization that gratitude is being required by some specific situation—for example eating.” But the last one “constructs a relationship of dependence between the speaker(s) and the deity … namely as children to a father. This relationship of dependence is not communicated or explained by these words, but assumed or inculcated by them.”10
in the three “which you made known” clauses, “the ritual participants are being scripted in the role of a group who has received knowledge from God.”11 The “gather into your kingdom” petitions “take for granted the existence (if only in the future) of a ‘kingdom’ belonging uniquely to God and defining the fate of the community.”12 and then in the doxologies, “the community signals its acceptance of each prayer spoken by the leader.”13 all of this comes to a climax in the dramatic liturgical dialogue of 10.6.14 This reading of the ritual cues substantiates schwiebert’s argument that the prayers are primarily focused on “knowledge and the coming kingdom” as opposed to the body and blood of christ or the saving death and resurrection.
My only serious reservations about schwiebert’s arguments in this chapter involve his emendation of the doxology in Did. 9.4. he agrees with Klaus Wengst that “through Jesus christ” is a later addition, as that title appears nowhere else in the Didache. But schwiebert adds that there was also a corruption of the “power and glory” from 8.2 to the reversed order of “glory and power” in 9.4. he bases this argument in part on the fact that, without the “through Jesus christ” clause in 9.4, both 9.4 and 10.5 have 9. schwiebert, Knowledge and the Coming Kingdom, 78.
10. ibid., 79 (emphasis original).
11. ibid., 80.
12. ibid., 81.
13. ibid., 81.
14. ibid., 82.
The DiDache: a MissiNG Piece Of The PuzzLe
exactly twice as many syllables as the shorter doxologies.15 My objection is that text criticism must be carried on with great caution in the Didache, given the limited number of manuscripts. The coptic fragment does not include Did. 9. schwiebert’s argument from syllable count and scansion is trumped by a pattern of reversals that Draper noted in these prayers.
Draper’s discussion of the structure of Did. 9–10 is more elegant and does not need to resort to textual emendation. Draper sees a clean structure of three units before the eating of the meal (two blessings and one petition) and three units after the meal (again two blessings and a petition), all of which is concluded with the liturgical dialogue in 10.6.16 The short and long doxologies mark off the parallelisms in this two-fold structure and some of these parallelisms are chiastic; that is, they involve reversals of order. in Did. 9 we encounter cup (9.2) and bread (as fragment, 9.3), but in Did. 10.3 we see the order “food and drink” and “spiritual food and drink,” reversing the order of Did. 9. Draper further points out that in 9.3
we see “life and knowledge” and then in 10.2 “knowledge and faith and immortality,” which, if we take immortality as equivalent to life, reverses the order of “life and knowledge” that appeared in 9.3. This is important, because it can be seen again in the pattern of the long doxologies in 8.2, 9.4, and 10.5. “Power and glory” appears in 8.2, “glory and power” in 9.4, and “power and glory” in 10.5. Given the other reversals mentioned by Draper, the reversal of order in 9.4 does not require textual emendation, and this eliminates schwiebert’s supporting argument for dropping the
“through Jesus christ” in 9.4.17
The pattern of the doxologies in Did. 8–10 brings us back to Draper’s arguments about structure in Did. 7–10.18 in “Ritual Process and Ritual symbol in Didache 7–10,” Draper provided a helpful three-parallel-column presentation of the text of Did. 7–8, then 9, and then 10. Draper followed up on Niederwimmer’s suggestion that the eucharistic prayers of Did. 9–10 may have been the Sitz im Leben for the development of the doxology in the Lord’s Prayer in Did. 8.2, commenting that “most of the 15. ibid., 65.
16. Draper, “Ritual Process and Ritual symbol,” 139–40.
17. Niederwimmer, Didache, 138, presents an important chart of the short and long doxologies in Did. 8, 9, and 10. i recall showing it to Reuven Kimelman in a discussion on the Didache prayers. he immediately noted the change of word order in 9.4
and remarked “that helps establish an inclusio from 8.2 to 10.5.”
18. Draper, “Ritual Process and Ritual symbol,” 139–40.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
215
ritual symbols of the eucharist Prayers are found also in the Lord’s Prayer”
and that the Lord’s Prayer is integral to the structure of the Didache, and in fact that the Lord’s Prayer in Did. 8 may have been formulated the way it was precisely for the purpose of the initiation process.19 i wish to follow up on this by revising his parallel column presentation of Did. 7–10 by restricting it to the prayers in Did. 8–10 to see what it reveals about the relationship of the Lord’s Prayer to the eucharistic prayers.
Did. 8
Did. 9
Did. 10
first about the cup:
We give thanks to you,
We give thanks to you,
Our father,
Our father,
holy father,
who are in heaven
hallowed be your name.
for the holy vine of
for your holy name,
Your kingdom come,
David your servant,
which you caused to
Your will be done on
dwell in our hearts, and
earth as in heaven.
for the knowledge and
faith and immortality,
which you have made
which you have made
known to us through
known to us through
Jesus your servant.
Jesus your servant.
To you be glory into the To you be glory into the ages.
ages.
You, almighty Master,
have created all things
on account of your
name,
Give us today our daily
Then, about the frag-
food and drink you have
bread and forgive us our ment: We give thanks
given to the sons of men
debt, as we forgive our
to you, Our father for
for refreshment. But to
debtors.
the life and knowledge,
us you have graciously
which you have made
given spiritual food
known to us through
and drink and eternal
Jesus your servant.
life through Jesus your
servant.
19. ibid., 137–38. The earlier article was “christian self-Definition against the
‘hypocrites’ in Didache 8,” in The Didache in Modern Research, ed. Jonathan a.
Draper, aGJu 37 (Leiden: Brill, 1996), 223–43.
The DiDache: a MissiNG Piece Of The PuzzLe
fora (or before) all
things we give thanks
to you because you are
mighty.
To you be glory into the To you be glory into the ages.
ages.
and lead us not into
Just as this fragment
Remember, Lord, your
testing but deliver us
was scattered over the
church, by delivering
from the evil one,
mountains and, when
it from every evil and
gathered, became one,
making it complete in
so let your church be
your love, and gather
gathered from the ends
it from the four winds
of the earth into your
into your kingdom,
kingdom.
which your prepared
for her.
for yours is the power
for yours is the glory
for yours is the power
and the glory into the
and the power through
and the glory into the
ages.
Jesus christ into the
ages.
ages.
Note
a. Ms h reads πρό for a translation of “before.” it is an easy misreading of the more sensible ὑπέρ (Lectio difficilior preferendum est, sed non sine sensu).
When the prayers are aligned in this way, many parallels and reversals “pop out.” The underlines on the chart indicate the strongest direct parallels in vocabulary or concepts. Note particularly the very last lines of each column: the long doxology appears in all three, but chapter 9 has
“power and glory” reversed and may have had the special name indicated by διὰ Ἰησοῦ Χριστοῦ. This could have been added later, but there is no manuscript evidence for the text without it, and its presence is attested (admittedly weakly) by apos. con. 7.25.4 by διὰ αὐτοῦ. Then notice the very first lines: only chapters 9 and 10 have the “we thank you” formula, but all three have “father” followed by something being “hallowed.” Only chapter 10 has the unusual “holy father,” which appears elsewhere only in John 17:11. Both 8 and 10 have the “holy name,” but 9 has the unusual
“holy vine of David.” The greater extent of word for word repetition in 9
and 10 is due, i think, to the predominance of formulaic language in the meal prayers, as in the “we thank you” prayers, the “which you have made known” clauses, and the short and long doxologies. schwiebert refers to these as the “fixed elements” of the Didache prayers and makes a powerful
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
217
argument for their importance in stressing the significance of the coming kingdom in these prayers.20
The bold-face type indicates a long-range, reverse parallel. The “kingdom” occurs in the first element of chapter 8 but in the last of both 9 and 10. This may seem like reaching on my part, but it may suggest a deliberate progression in the movement from the Lord’s Prayer to the end of the meal prayers. That progression may also be attested by the way in which the simple “daily bread” (or “bread of tomorrow” or “bread of the age to come”) in the central unit of 8 is paralleled in the other center units by
“fragment” in 9.3 and then doubly paralleled by “food and drink” and
“spiritual food and drink” in 10.3.21
The petition of the third element of the Lord’s Prayer in Did. 8 (“and lead us not … but deliver us from evil”) is strongly paralleled in 10, which even picks up the word “evil.” The word “evil” does not appear in Did. 9, but perhaps this is to indicate that the evil to be feared is the scattering of the “assembly” or whatever the author means by ἐκκλησία.
20. schwiebert, Knowledge and the Coming Kingdom, 78–82.
21. “Bread of tomorrow” and “bread of the age to come” are alternative translations for ἐπιούσιος, traditionally rendered as “daily.” They are mentioned here because they correspond with the intensely eschatological tone of Did. 9–10. hans Dieter Betz, Sermon on the Mount: A Commentary on the Sermon on the Mount, including the Sermon on the Plain (Matthew 5:3–7:27 and Luke 6:20–49), hermeneia (Minneapolis: fortress, 1995), 397–400. Betz discusses the possibilities in detail, noting at the outset
“The problem of the meaning of the term ἐπιούσιος has been debated since antiquity.”
he is least impressed by the arguments from semitic roots and the arguments from the versions that render it by the more mystical “supersubstantialem (Vulgate) or per-petuum (curetonian syriac). he gives serious consideration to meanings derived from ἐπιέναι (such as, “for the following day”) and those derived from ἐπι ουσία, such as
“what is necessary for existence (398). frederick Danker (BDaG, 376–77) pushes in the direction of a translation that emphasizes eschatology, but it takes time to discover this. he lists in order four major meanings: (1) “necessary for existence”; (2) “for the current day”; (3) “for the following day”; and (4) “for the future,” which is subdivided into five “sub-meanings.” he cites thirteen articles later than 1500 for the first meaning, six for the second, and eighteen for the third. in support of the fourth meaning, he cites thirty-one articles later than 1500, with twenty-five of these articles supporting the last of the five “sub-meanings”: “referring to the coming kingdom and its feast.”
Also sprach Danker! he seems to be pushing the argument toward a translation that is very favorable to schwiebert’s argument that “the coming kingdom” is a primary matter of importance to the Didache prayers.
The DiDache: a MissiNG Piece Of The PuzzLe
it is difficult to determine whether the meal prayers were the source of the Lord’s Prayer in Did. 8 or the other way around. if the Lord’s Prayer were constructed from the meal prayers, it is difficult to understand why the “we thank you” formula and the creational references are missing from the Lord’s Prayer. But if the Lord’s Prayer were the source of the meal prayers, it is hard to understand why forgiveness of sin is absent from the meal prayers. still, the single closing doxology of the Lord’s Prayer could have served as the model for both the short and long doxologies in the meal prayers, but the reverse is also possible.
The web of connections between the prayer in Did. 8, which is placed directly after the baptismal instruction and just before the meal prayers, supports Draper’s conclusions about the interconnectedness of 7–10 as leading to an experience of aggregation. The rather high percentage of formulaic words supports schwiebert’s conclusions about the significance of the fixed parts of the prayers signaling the primary importance of the experience of the community members being placed in a situation of dependence on the father for supplying the knowledge and life and not only awaiting but nearly “living into” the coming kingdom.
2. The Relationship of the Didache Meal Ritual
to Other early christian Texts
schwiebert’s fifth chapter (“The Original Milieu of Didache 9–10”) begins with a well nuanced expression of the Jewish background of Didache, indicating that it was a community of a Greek-speaking background with a strong desire to bring in gentiles. he does not take particular rabbinic prayers and argue that they affected the Didache prayers. his discussion of the possible effect of Q on the Didache is quite compelling. One might hesitate to accept a relationship between the Didache and Q, since the Didache shares so many features and expressions with Matthew that the use of Q would seem unnecessary. More importantly, positing a relationship between the Didache and Q is bedeviled by the many unresolved questions about both. We know precious little about the actual community that produced the Didache, when it existed, where it existed, and over how much time its text was used or ignored. There are also many questions difficult to resolve that surround the exact makeup of Q. What was the full extent of Q? are the roughly 6,500 words in an english translation that we now possess about 90 percent of Q, or merely 20–50 percent? The higher percentage is reasonable, since that is about the percentage of Mark that
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
219
Matthew and Luke manifest. But i have no way of knowing for certain that Q was not two to four times as long as our recovered 6,500 words or even if the extent of Q could vary in every performance of it. There could well have been many things in Q that Matthew and Luke did not feel were necessary or helpful.
also problematic is the likelihood that Q, like the Didache, is subject to questions about stages of composition (early Q, Q1, Q2, etc.). This seriously complicates developing a convincing theory, since every proportional possibility must be multiplied by all other proportional possibilities in the argument. Yet what i found compelling about schwiebert’s chapter is that he demonstrates a number of solid parallel passages in the Didache and early Q, but he does not from this conclude that Did. 9–10 is the Q
meal ritual. What he concludes is that from among the set (Q, Gospel of Thomas, and Gospel of John) the best parallels are with Q. That is no small finding in that it presents us with two first century witnesses (Q and the Didache) for Jesus groups for whom the word “gospel” means the sayings of Jesus rather than the kerygma, as it did in the Pauline epistles and other early christian literature.22 The importance of this for our interpretation of the Didache meal ritual is that we must not read it presuming direct reference to the Last supper and the paschal mystery as in canonical New Testament texts. But i am equally impressed by the fact that schwiebert does not spell out more exactly the relationship between the Didache and Q. a number of important features of Q are entirely missing from the Didache, for instance the title “son of Man,” the figure of John the Baptist, and the theme of “being opposed to this generation.” But it is true that the Two Ways teaching in the Didache does reflect Q’s Deuteronomic theme, and the Didache’s emphasis on active prophecy also supports this connection. But in my examination of Q, i found many passages closer to Matthew than to reconstructed Q. The baptismal formula in Did. 7.1, which is repeated in 7.2, is like Matt 28:19 and so according to some ought to be eliminated from the Didache as a later addition that was meant to conform it to canonical formulae.23 The Didache Lord’s Prayer is closest to 22. Niederwimmer, Didache, 50. he cites helmut Koester, Synoptische Überlieferung bei den apostolischen Vätern, TuGaL 65/5.10 (Berlin: akademie, 1957), 10.
23. i am reluctant to dispatch these Trinitarian formulae in Did. 7.1 and 7.2 to a much later revision of Didache as schwiebert does (Knowledge and the Coming Kingdom, 152 n. 8). schwiebert sees these formulae in Did. 7.1–2 as contradicting Did. 9.5
“baptism in the name of the Lord.” he cites arthur Vööbus, huub van de sandt and
The DiDache: a MissiNG Piece Of The PuzzLe
Matthew’s; Did. 8 has parallels to Matt 6:1–18 (fasting and praying not as
“the hypocrites” do); Did. 16 has been well argued as possibly dependent on an apocalypse very like that of Mark 13 (possibly a source of Mark 13) and Matt 24.24
Particular phrases from Matthew appear:
• Matt 5:41: the command to go a second mile with one who forces you, though not in Luke, appears in Did. 1.4.
• Matt 5:39: the specification of the “right cheek” is in Did. 1.4
but not in Luke 6:29.
• Matt 5:48: “be perfect” seems to be reflected in Did. 1.4 immediately after the command not to strike back.
• Matt 5:26: “the last penny” appears in Did. 1.5d, but “penny”
is missing in Q.
• Matt 19:18 and 5:33: prohibitions appear in Did. 2.2–4 but not in Q.
• Matt 24:31: “gather from the four winds” appears in Did. 10.5b but not in Q.
sadly, the exact wording of Q is even more elusive than the exact wording of the Didache. also, just as the Didache is known to have likely endured
“updating” similar to what is obvious in manuscripts of 1Qs, so it seems that Q, reconstructed in the first place from two different gospels, developed in stages. Dieter Lührmann’s observation that “the community from which Matthew obtained Q had already reworked Q” would explain some David flusser, and alan Garrow as being in agreement. But is there any evidence from the period that formulae of baptism were so restricted or that Paul, who generally uses the language of “baptized into christ” (Gal 3:27), would object to the Trinitarian formula? it may be that both 7.1 and 7.2 were glossed, but is there any evidence beyond a requirement of consistency in sacramental formulae? a similar tendency is in evidence in the removal of “through Jesus christ” in Did. 9.4. should we also remove
“holy father” from 10.2, since it appears nowhere else in the Didache, and may have been added by a scribe familiar with the Gospel of John?
24. alan J. P. Garrow, The Gospel of Matthew’s Dependence on the Didache, JsNTsup 254 (London: T&T clark, 2004), 190–215. i do not find his emended ending of the Didache (29–66) convincing. schwiebert points out in a note (Knowledge and the Coming Kingdom, 121 n. 35) that Koester long ago (Synoptische Überlieferung, 190) argued against the availability of any of the synoptic Gospels to the composer of the Didache.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
221
of these special connections between the Didache and Q by providing a Q
particular to the Matthean community. But it is a hypothesis. The point of these remarks is to express agreement with the caution schwiebert shows by saying Didache is closer to Q than any other source. We are not entirely clear on what texts or oral traditions were available to the Didache community. Yet Q, or a sayings source related to it, was certainly among them.
3. schwiebert’s analysis of the Didache
Meal Ritual Trajectories: Theological implications a warning from Niederwimmer: The Didache is “written without theological claims and entirely focused on praxis and the order of community life.… a reconstruction of the “theology of the Didache” would therefore be a foolish enterprise.”25 Without a doubt, the Didache is not systematic theology; it is better described as halakah. But there can be theology implicit in halakah. Thomas O’Laughlin notes that from the time of its discovery the Didache triggered outrage both from ultra-montane catholics and from “extreme Protestants,” because it calls into question fixed theological positions about how the eucharist ought to be conducted, but also demanded fixed times of prayer and fasting, thus annoying people on both ends of the spectrum.26 Yet a significant dedication to the study of the Didache perdures. in fact, some joke that the Didache does not merely attract readers, it takes prisoners. it is important to examine this appeal of the Didache. consider that the Didache is deceptively familiar to those who know the early christian canon, but is often even more outspoken than the canonical works that it resembles. it has robust even strident apocalyptic expectation, as in the Pauline epistles and the book of Revelation, but also radical moral commands like in the synoptic Gospels. it uses the title ὁ παῖς σύ (with Davidic associations) as in the early chapters of acts, but also contains the Johannine emphasis on the mystical indwelling of the holy name, and in Did. 9.1 the elusive “holy vine of David” may be related to John 15:1–11. Then there is the liturgical dialogue in Did.
10.6 that compares well with 1 cor 16:22 and Rev 22:17–20. finally, there is the elusive relationship to the Gospel of Matthew. it lacks direct reference to the Last supper and passion/resurrection tradition that we see in 25. Niederwimmer, Didache, 228.
26. Thomas O’Laughlin, The Didache: A Window on the Earliest Christians (Grand Rapids: Baker; London: sPcK, 2010), 8–10.
The DiDache: a MissiNG Piece Of The PuzzLe
Matthew, but it has the teaching of the Two Ways that appears in a different form in Matt 7:13–27, the demands for forgiveness of Matt 5:23–26
and 6:14–15, and rules for community order as in Matt 18:15–35 (but again in a very different form). Because of references to εὐαγγέλλιον in the Didache, it was long presumed that the Didache postdated that gospel.
alan Garrow argues persuasively that the Didache was a source for Matthew.27 But it must be admitted that, in spite of all of these similarities to canonical traditions, “knowledge and the coming kingdom” are forcefully proclaimed in the Didache without direct reference to the Last supper or the passion and resurrection of Jesus, which are presumed in all the canonical references cited above. To be fair, there is no indication that the Didache community did not believe in the passion and resurrection, but there is no assurance that the community did. People bothered by the fact that the same holds true for Q can deal with the problem by saying that, since we do not have the entire text of Q, we cannot make firm asser-tions about what precisely was or was not held by that community. But the Didache is nearly complete, a written text, with echoes and resonances in other traditions and texts. These echoes and resonances are examined by schwiebert in his last three chapters. he describes possible relationships between the language and ritual logic of the Didache and the meal rituals of other groups of christians from the second to fourth centuries.
schwiebert has engaged the Didache with sensitivity and openness to its uniqueness. his monograph is a tribute to what is amazing about this text, but it will not let those of us who would understand this text rest, satisfied that we comprehend it.
in chapter 7 (“Didache’s Meal Ritual and early Prayer Traditions in Rome and asia Minor”),28 he begins with a marvelous description of the development of liturgy in the first four centuries of the christian movements. he shows that the “great liturgical compromises” that we see in evidence in the fourth century were already happening in the second and 27. Garrow, Gospel of Matthew’s Dependence. Not every argument and reconstruction by Garrow needs to be accepted for him to have succeeded in shifting the basic presumption of Matthean priority to the Didache. as seen in n. 21 above, Koester demonstrated the likelihood of that long ago in (Synoptische Überlieferung, 190).
The Didache is a community rule in many ways more primitive and less refined than the Gospel of Matthew. Matthew may have fused some of its traditions with the passion/resurrection material of the other synoptic Gospels.
28. schwiebert, Knowledge and the Coming Kingdom, 186–90.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
223
third. Traditions initially different from and even competing with positions that came to be orthodox were incorporated, even if it meant their subordination to the emerging orthodox position. schwiebert shows us Luke connecting Jesus as servant, the teacher/revealer of Q and the Didache, to Jesus who suffered, died, and rose from the dead, as in Mark.29 Then schwiebert shows that the παῖς title in combination with the doxology so frequent in the Didache appears in 1 clement, Martyrdom of Polycarp, and the apostolic Tradition associated with hippolytus. The apostolic Tradition is especially notable for its use of similar prayer language to that of Did. 10.30 Justin Martyr seems to have fewer clear and distinct echoes of the Didache, but even in Justin we see
praise and thanksgiving for creation, perhaps in great detail; thanksgiving and petition for preservation from evil, including expectancy for the coming kingdom; and thanksgiving for being made worthy of the eucharistic gifts given by christ (his flesh and blood) which call to mind his sufferings.31
here we see quite clearly the combination of the Didache features with the eventually prevailing association with the passion.
schwiebert describes a similar situation in chapter 8 (“The influence of the Didache’s Ritual Logic in egypt”), but the texts he cites are far less familiar. in the sacramentary of serapion of Thumis (fourth century) and the papyrus from Dêr Balizeh (third–fourth centuries), we see the image of the church as grain once scattered and now gathered. But instead of “into the kingdom” as in the Didache, it is “into the one catholic church.” But each of these texts presupposes an institution narrative and Last supper connections. in the papyrus from Dêr Balizeh, not only is grain used as a symbol of ingathering, but also water and wine. it is gathered now rather than in the future as in the Didache. cyprian further to the west reflects this image of gathering in the wine.32
schwiebert cites the ascetic meal prayer in the Pseudo-athanasian, Virg. 13, which resembles Did. 9.4:
29. ibid., 186–90.
30. ibid., 190–201.
31. ibid., 206.
32. ibid., 209–11.
The DiDache: a MissiNG Piece Of The PuzzLe
When you recline at table and begin to break bread, sealing (it) three times, in this way give thanks and say,
“We thank you, our father, for your holy resurrection, for through Jesus your servant you have made it known to us. and just as this bread was scattered, which is upon this table, and being gathered became one, so let your assembly be gathered from the corners of the earth into your kingdom. Because yours is the power and the glory forever and ever. amen.”33
There can be no question regarding the influence, if not of the Didache text itself, of an orally transmitted tradition of its meal ritual. unlike the echoes from Rome and asia Minor, we see here “gathering from the four corners into the kingdom” rather than “into the church.” in the very next section (Virg. 14), we see elements of Did. 10 with giving thanks, doxology, reference to creation, the “name above every name,” “fleshly nourishments”
contrasted with “heavenly food,” and “through your beloved servant our Lord Jesus christ, with whom to you be glory, honor, dominion, forever and ever, amen.”34 Once again the echoes of the Didache are clear, but so are the touches of the Last supper/passion tradition with the reference to resurrection and the fact that these prayers are to be prayed according to a schedule of the third hour, sixth hour, and ninth hour that seems to have been set by the passion narrative.
Next schwiebert discusses traces of the Didache in Barnabas, which are more thoroughly connected to the passion and then echoes the epistle to Diognetus, which refers to the “servant” figure, but also shows possible influence from Paul’s letter to the Romans and the sermon on the Mount.35
a Valentinian prayer from the Tripartite Tractate (late 200’s) exhibits clear gnosticizing tendencies, yet a source of the prayer seems to be Did.
9–10, as elements of its ritual logic are present.36 The Tripartite Tractate parallels are rather faint, but some Valentinian expositions (Xi 43, 20–38
and 44, 1–37) show striking parallels with the Didache prayers, but also introduce gnostic features.
in clement of alexandria’s Stromata, schwiebert shows echoes of the Didache’s ritual logic. in Strom. 7.6.31.7–8, Jesus is “him through whom 33. ibid., 213.
34. ibid., 214.
35. ibid., 217–22.
36. ibid., 222–28.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
225
we receive our knowledge,” and it moves immediately to prayers of glorification and petition. he shows several other passages that express the combination of “glorification and knowledge.” But schwiebert further points out that the actual eucharistic prayers in these texts do not look like Did. 9–10 except for the use of doxologies. But unlike the Didache, these prayers make direct reference to the saving death of Jesus.37
his discussion of Strasbourg Papyrus 254 (fourth–fifth centuries) is the most striking. The prayer begins with reference to creation “through your wisdom,” which turns out to be “the light of your true son.” We see prayers of “giving thanks,” described as “sacrifice” (but the language is that of Rom 12), multiple references to “the holy name,” and then “remember your holy and only catholic church.” There follows an open ended peti-tionary prayer, and it ends with “through our Lord and savior, through whom be glory to you through the ages.” schwiebert wisely concludes that here we see both “kinship and distance.” Yet he is able to conclude the chapter with the well-grounded assertion that “the phrases and formulae, as well as the ritual logic of the Didache are better attested in egypt than elsewhere.”38
Less needs to be said of schwiebert’s last chapter (“The end of a Trajectory”), since the texts and arguments of ignatius of antioch, irenaeus, and the apostolic constitutions and canons are more familiar and widely known. clearly in this period, battle lines were drawn and firm efforts made to assert the Last supper-related meal prayers (with unambiguous and repeated reference to the body and blood) and with almost every possible reference to the distinctive features of Did. 9–10 prayers being altered or removed. This is most clearly displayed in apos. con. 7.25–26 in which:
• the cup prayer is removed
• thanks is offered for “life” but not “knowledge”
• “knowledge, faith and immortality … made known” is replaced with “knowledge and faith and love and immortality that you gave”
• some doxologies are removed
• reference to “the fragment” is removed
37. ibid., 229–34.
38. ibid., 235–37.
The DiDache: a MissiNG Piece Of The PuzzLe
• thanks for “the precious blood of Christ and the precious body is added”
• the liturgical dialogue of Did. 10.6 calling for the coming of the Lord and this world passing away is replaced “with an invitation by the priest to the congregation to ‘come forward’
and partake of the elements.”39
schwiebert’s arguments are presented in detail because the story of the trajectory and final near extinguishing of the voice of the Didache that it evidences is an important story. The Didache, which survives in full in only one manuscript, goes back to a very early time and affected North africa, the eastern Mediterranean, and Rome, but was eventually nearly eliminated. it is worthwhile to deal with the potential theological or at least ecclesiological implications of this process.
The Didache meal that schwiebert describes was a ritual in which (1) the initiates along with the community expressed gratitude for knowledge that came from a loving father, “mediated by his servant, Jesus”; (2) the gathered community “was scripted in the role of an eschatological potentiality” (i.e., the coming kingdom, which was incredibly important to them); and (3) “this religious and social posture was achieved above all by the periodic repetition of the fixed formulae of the community’s meal ritual; and it also must have been fostered by (and helped to foster) the type of instruction transmitted in these circles.”40 if schwiebert is correct in this description of the distinctive features of this ritual, i am not so surprised that, even though other christian groups had meal rituals with direct links to the Last supper and passion, the features peculiar to the Didache remained valuable, honored, and celebrated in as many different times and places as they were.
What can be disturbing to a person steeped in a eucharistic tradition is to imagine how a meal prayer that did not relate directly to the Last supper and passion could have perdured. More importantly, it raises another question: Was anything lost in the eventual elimination of such a ritual as we see in Did. 9–10?
it is questionable how serious a role apocalyptic eschatology plays in the consciousness of the average participant in many forms of present day 39. ibid., 246.
40. ibid., 248.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
227
christian liturgical worship. in the popular imagination, “heaven” is far more often the language of that for which one hopes, as opposed to the age to come, the transformation of a universe (and a people) being freed from its slavery to corruption. The more prevalent view of the afterlife as
“heaven” seems to diminish any serious importance to the coming kingdom as described in the Didache, the Pauline epistles, and other apocalyptic sections of the New Testament. “heaven” is readily sentimentalized and individualized. The community that prayed the Didache seemed to cry out in its own time as though the age to come was nearly in reach and that, just as glorification took place in the pronouncement of the doxologies, so was the kingdom nearer and the people more united in sober expectation as it cried out “come!”
further, as an educator i am concerned with the dearth of serious religious education for adults in mainline congregations and parishes. in its simplicity, Did. 9–10 (although really 1–10) expresses the absolute character of and the need for the nurture and development of the “gift of knowledge.” it calls for continuous rather than occasional prayer (even fasting) and a far more integrated and pervasive engagement in instruction. Was the general success of the liturgical tradition that emphasized the elements (bread and wine) and their association with “body and blood” given for the forgiveness of sins accomplished at the expense of a robust understanding of the presence of God in the Word?
The strong case that schwiebert made for relationship with Q and the basic agreement with that tradition in its lack of direct connection to the Last supper, death, and resurrection tradition invites one final comment on theological implications. What is clear is that, while the Didache makes no direct reference to the death and resurrection, neither does it expressly deny them. The rather consistent pattern of authors later than the Didache merging its ritual logic with clear passion and resurrection traditions suggests that they did not view the Didache’s theology as directly opposed to their own. This may not entirely be a matter of misunderstanding. “Resurrection” is never mentioned in the Didache, but some sort of concept of the resurrection fits with the Didache’s references to the gift of “life” in 9.4 and “immortality” in 10.6. This is not evidence that the framers of the Didache proclaimed the resurrection, but it should be noted that the “life”
that they did proclaim has many possible referents. in the Gospel of John,
“life” is connected to the role of Jesus as restoring “life” by his passion and resurrection. in addition to this interest in the gift of “life,” John shares with the Didache a mysticism of the indwelling of the “name.” Besides the
The DiDache: a MissiNG Piece Of The PuzzLe
reference to “baptism in the name” in Did. 9.5, we see that expression in 10.2: “for your holy name, which you caused to dwell in our hearts, and for knowledge, and faith, and immortality, which you made known through Jesus your servant.” The mystical workings of “word” (and “the words”) and “the name” in John’s theology appear in John 17:6–8: i have made your name known to those whom you gave me from the world. They were yours, and you gave them to me, and they have kept your word. Now they know that everything given me is from you, for the words that you gave to me i have given to them, and they have received them and know in truth that i came from you.
and more dramatically in 17:11b:
holy father, protect them in your name that you have given me, so that they may be one as we are one.
in 17:12 this name, which protects them, keeps them from being “lost.”
here we have the convergence of the giving of knowledge, the giving and retaining of the word, and the protection by the name. This is related to John 15:1–11 on the “true vine.” in John 15:1–11 we also see the word as active in “cleansing” and contributing to the “remaining on the vine” (15:4–5) rather than being “cast into the fire” (15:6). These expressions may bear no direct relation to the Didache prayers, but what we do not know about the location and origin of the Didache community should make us hesitate to deny any relationship to the Johannine tradition.
The connection between the Didache and the Gospel of John as demonstrated by Johannes Betz41 may shed light on how the Didache’s ritual logic is likely to have influenced later canonical christian writings. Betz’s arguments are especially relevant to the point being made in this part of my essay, because they deal precisely with a “Lord’s supper theology” of the Didache, which he sees as related to that of the Gospel of John.42 Betz sees the following themes in both works: (1) Jesus as a revealer, (2) the 41. Johannes Betz, “The eucharist in the Didache,” in Draper, Didache in Modern Research, 244–75. Betz’s argument (274) that Did. 9–10 is not an agape and a eucharist but a eucharist based on an agape text does not need to be accepted in order to appreciate the strength of his analysis of the verbal and material relationship between the Didache and the Gospel of John.
42. Betz, “eucharist in the Didache,” 255.
cLaBeauX: The RiTuaL MeaL iN DiDache 9–10
229
image of the vine associated directly with Jesus, (3) the association of the bread with life and immortality, and (4) prayer for sanctification and gathering.43 he presents a chart with nineteen different terms in Did. 9–10, which appear in seven chapters of the Gospel of John. With support from Rudolf schnackenburg, Betz concludes that “the fourth Gospel knew and used the eucharistic prayers and formulations of the Didache tradition, even if it was from the oral stream of tradition.”44 Betz argues persuasively that the Didache was not dependent on John and also that it is doubtful that the author of the fourth Gospel “had Didache as a literary entity before him.” finally, he acknowledges that “what the Didache says prelimi-narily and briefly, is amplified and clarified, and more strongly christologized in the fourth Gospel.”45
Betz also notes the connection between the Didache and John on the issue of “the vine.” starting with the use of the image of vineyard in isa 5 as israel, Betz sees the vine in Did. 9.2 as christ and all christians, as in 1 cor 10:17. he states that in that passage “‘holy vine of David’ refers primarily to the person of Jesus as the Davidic Messiah, but secondarily to the meal fellowship constituted and actualized by him: so we come in fact close to the Johannine understanding of the vine.”46
These points of contact between John and the Didache tradition are quite important in terms of understanding the relationship of the Didache to canonical New Testament works. in its relation to John, the Didache seems to be part of a stream of tradition that was taken seriously by the author of the fourth Gospel, but it represents important steps in an earlier stage in the development of eucharistic theology on which the Gospel of John was to build. The fourth Gospel has increased the theological complexity, but it seems that important contributions were made by the early syrian meal tradition, which, while it did not proclaim the death and resurrection, still spoke of the significance of the meal in ways that were formative for the later christological development of the author of the fourth Gospel, who could be very suspicious of unsympathetic views. What we do see in the Gospel of John, which is much later than the Didache, is a concentration of certain language and images that for John do not exist “over 43. ibid., 255.
44. ibid., 256. Rudolf schnackenburg, The Gospel According to St. John, 3 vols.
(London: Burns & Oates, 1968), 1:152.
45. Betz, “eucharist in the Didache,” 256.
46. ibid., 267.
The DiDache: a MissiNG Piece Of The PuzzLe
against” a death and resurrection emphasis. as was seen in the second to fourth centuries, both sets of ideas can be brought together and not always with the violence that was done to the Didache in apos. con. 7.25–26.
as a final expression of gratitude for schwiebert’s Knowledge and the Coming Kingdom, it should be said that the readiness of so many traditions to bring the ritual logic of the Didache into their streams of theology and ritual traditions suggests that these traditions saw something valuable in the Didache. in spite of the Didache’s final suppression, can we not yet see a value in it for christianity today? if we can, then by studying the early centuries of christian liturgy, and this very early work in particular, we might yet be like “the householder who brings out of his or her storehouse new things, but also old things” (Matt 13:52c).
Leadership and Liturgy in the Didache
Joseph G. Mueller, S.J.
The four chapters to which i have the privilege of responding present us with rich perspectives on ritual and prayer in the central portion of the Didache, chapters 7–10. Of these four essays, Jonathan schwiebert’s asks the most basic question about the interpretation of this passage. he challenges us to think deeply about how the ritual character of these chapters should determine the way readers interpret them. it is for this reason that in what follows i will respond first to schwiebert’s essay. his contribution has helped direct my thinking about huub van de sandt’s contribution on prerequisites for participation in the eucharist of the Didache and about Peter Tomson’s reflections on the place of the Lord’s Prayer in this document. i give over the last part of my response to John clabeaux’s review of schwiebert’s marvelous book on the meal ritual of the Didache.
chapters 3–6 of that book, Knowledge and the Coming Kingdom: The Didache’s Meal Ritual and Its Place in Early Christianity, drew inspiration from Maurice Bloch’s theory of ritual in order to build an interpretation of this ancient eucharist. schwiebert’s contribution in this present volume begins by helpfully summarizing the main findings of those chapters, but he goes on to examine more explicitly than he did in his book the way Bloch’s theory can help and hinder interpretation of the meal ritual presented in Did. 9–10. in the beginning of his essay, schwiebert rightly takes as settled the notion that these prayer texts transmit to us formulaic speech, phrases fixed in advance of the liturgy’s enactment. he judiciously points out that figuring out how to interpret this ritual, with its fixed and variable elements, provides more of a challenge than determining which of its elements are fixed or variable.
-231 -
The DiDache: a MissiNG Piece Of The PuzzLe
he calls on Bloch’s theory of ritual to invite the students of the Didache to change the ways they usually have interpreted the fixed formulae of chapters 9 and 10. schwiebert tells us that most seem at least to imply that these expressions do not mean much of anything, but when they can, scholars find ideas and thoughts expressed in these formulae by comparing them to other ancient texts, sometimes nonritual ones. supposed authorial decisions on the terms, grammatical constructions, and word order in these fixed formulae guide the search for this sort of meaning in them, a search that aims at displaying their meaning by translating the fixed lines of the ritual into equivalent expressions.
Bloch’s theory tells us that this way of finding meaning in a ritual text like Did. 9–10 cannot reveal the relationship people have to the words of the rite while they are practicing it, for no one makes decisions about the terms, grammar, and phrasing of fixed formulae of a ritual while they are executing that ritual with other people in a communal assembly. indeed, while they are engaged in this eucharist itself, the leader and the rest of the assembly make no decision even about the order in which they take their respective turns to pronounce the various fixed formulae. Thus, no participant’s decisions on grammar, vocabulary, and textual structure can provide a foundation for an interpretation of what the leader and the assembly mean while they are performing this ritual. it is precisely by seeing things from the point of view of an individual actually using formalized language that Bloch can tell us that such a person’s activity of meaning or explaining something by his or her words tends to shrink to the vanishing point. The participants in a ritual do use the fixed formulae of the rite to assert both the unity of the assembly’s members and the notion that, at least, during the performance of the ritual, things are going as they should for the assembly.
schwiebert deftly uses this last insight into the performative force of ritual formulae to found a number of perceptive observations on the purpose and meaning of the Didache meal ritual. in the process, he wisely follows Bloch’s counsel to interpret a rite by attending to where, when, and why the ritual words are used, as well as by whom and in the context of which other rites and activities. attention to this ritual context allows him to see in the meal ritual an expression of the assembly’s relationships with family, other Jews, non-christian gentiles, Jesus, and God, as well as an assertion of the assembly’s divinely given health and unity founded on the meal ritual itself. space limitations kept schwiebert from mentioning two elements of the ritual context of the Didache eucharist that have their importance. first, chapter 14 shows us that reconciliation defines, in
MueLLeR: ResPONse
233
part, the type of holiness that schwiebert rightly notes characterizes the meal ritual. i will say more about this part of the eucharistic context of the Didache when i respond below to huub van de sandt’s essay. second, the Didache eucharist takes place on the Lord’s Day (14.1), whatever that is.1
schwiebert asserts that his interpretation of the Didache eucharist comes entirely from examining the ritual context of that meal and “not from an examination of the ‘meaning’ or ‘syntax’ of the prayers themselves, nor by translating them into an exposition.”2 But does he not somewhat overdraw this contrast? at one point in his interpretation of the meal ritual, schwiebert does well to note,
We must not forget that the ritualized speech in this case is prayer, speech addressed to God (not the people) and not in the persona of an elder or some ancestral figure but rather in the voice of the whole assembly: a
“we.” someone (we do not know who) speaks for the whole assembly, and the assembly permits this, because this is at least a part of what the gathering means: they are there, among other reasons, to speak as one to God.… The prayers justify the sharing of the meal.3
The word “God” does not occur in Did. 9–10. how, then, do we know these prayers are addressed to God? The meaning of the words “father”
and “Lord” tells us that these prayers are directed to God.4 Thus, his interpretation does seem to come, at least in part, from the meaning of the 1. see Jonathan schwiebert, Knowledge and the Coming Kingdom: The Didache’s Meal Ritual and Its Place in Early Christianity, LNTs 373 (London: T&T clark, 2008), 165 n. 49 and 176. see also ignatius, Magn. 9.1; Justin, Dial. 41; Apol. 1.67. ignatius links life and Jesus’s role as giver of knowledge, themes of the Didache eucharist, to a reference to the κυριακὴν, seen precisely as the occasion of Jesus’s resurrection from death. for Justin, Jesus’s death and resurrection connect the sunday eucharist to Jesus’s function of conveying saving knowledge through teaching. is the Lord’s Day Didache meal ritual, then, as distant from a tradition of eucharistic commemoration of the Jesus’s death and resurrection as schwiebert and others believe? Or does Did. 14.1
refer to the Day of atonement? see Jonathan a. Draper, “Pure sacrifice in Didache 14
as Jewish christian exegesis,” Neot 42 (2008): 228–29.
2. see in the present volume, Jonathan schwiebert, “Pray ‘in This Way’: formalized speech in Didache 9–10,” above 189–207 (204).
3. ibid., 203.
4. in Did. 4.11, we learn that slaves are to see in their “lords” types of God, but never is the word “God” explicitly linked to the word “father” in the Didache. Knowledge of ancient Jewish prayer forms and of their use—among others, the Lord’s Prayer
The DiDache: a MissiNG Piece Of The PuzzLe
prayers, that is, from the meaning of their words. One indication of this fact is that he quite naturally translates some of the terms of the prayer,
“father” and “Lord,” into the equivalent expression “God.” furthermore, the block quote above shows schwiebert deriving his interpretation from the syntax of the prayers. The syntactical fact that the prayers use throughout the first person plural pronoun leads him to conclude that these are prayers of an entire assembly. finally, schwiebert’s interpretation notes that the prayers of Did. 9–10 justify something, namely, the sharing of a meal. since justifying is a form of explanation, his interpretation attributes an explanatory function to the prayers, even though Bloch’s theory criticizes such an interpretation. indeed, schwiebert tells us that Bloch “challenges us to consider that the expressions in our prayers are not trying to explain anything.”5 My point here is not to criticize any of schwiebert’s interpretations of the Didache eucharist that i have quoted here; indeed, they seem astute to me. i merely wish to indicate that his interpretation differs less than he seems to think from those accounts of Did. 9–10 that he tells us Bloch’s theory would rule out. it makes total sense, then, that schwiebert ends his essay with a critique of Bloch’s theory.
This critique notes that ritual participants can so commit themselves to a particular ritual occasion that they fill the rite’s verbal formulae with a meaning much like that of everyday speech, characterized as it is by individual expressive choices about terms, syntax, phrasing, and conversational distribution of speech roles. in fact, he notes, the investment in the tradition conveyed by the ritual can make what the ritual has us say mean even more to us than do our everyday communicative speech acts. The ritual invites, even it if does not prescribe, such investment in its words and in the act by which we mean what they say. schwiebert points out that this investment in the act of meaning what the ritual says opens us to invest in the community performing the ritual. This investment in the community is, to use schwiebert’s apt expression, “the choice to embark upon the ritual,” and it “is not trivial.”6 it is in considering this investment and this choice that schwiebert’s own interpretation is itself justified in not following all of Bloch’s strictures, for he understands more expansively and accurately than Bloch does the meaning act people can accomplish by in Matt 6 and Did. 8—help us to see that the Didache meal ritual is calling God
“father” and “Lord.”
5. schwiebert, “Pray ‘in This Way,’ ” 199.
6. ibid., 206.
MueLLeR: ResPONse
235
enacting the words, syntax, phrasing, and alternation of speaking roles in a ritual.
schwiebert goes on to note that “the choice of which ritual to embark upon is likewise not trivial. The choice of this ritual script over some other possible ritual script … can indeed tell us something significant about the community that prays ‘in this way.’”7 i take this comment as the most helpful element in his critique of Bloch’s theory. While schwiebert uses this theory to study fruitfully the ritual words of Did. 9–10 from the point of view of those who use these formulae in their ritual context, one can also legitimately interpret those words in the context in which we find them in the Didache, that is to say, as objects of rubrical instruction. indeed, the immediate context of these formulae in the Didache is that of the objects of the specification of what a group should say or do on a certain ritual occasion:
Now concerning the eucharist, give thanks in this way.… Let no one eat or drink from your eucharist, except those baptized.… Now after you have been filled, give thanks in this way.… Now let the prophets give thanks as much as they wish. (9.1, 5; 10.1, 7)8
Thus, it is totally legitimate to interpret these prayers as the object of a decision recorded in the Didache, a decision about who will say what and when. This is not the decision made by a person as he or she is repeating the fixed formulae during the performance of the rite, nor is it the act of meaning that results from some person’s or group’s particularly strong investment in a particular performance of the rite. The decision in question is, rather, made by those who determine (or the one who determines) what the rite will look like, or, at least, what form of the rite should be written down. One can fruitfully and reasonably interpret the rite from the point of view of these persons or this person and therefore take into account all of the linguistic choices that ritual performance can lead people not to make, for the fixing or recording of a rite is not the same action as participating in it.
understanding the fixing or recording of a rite will look different, then, from understanding that rite’s performance. Therefore, the type of interpretation that schwiebert criticizes earlier in his essay—one that sees 7. ibid., 206.
8. all translations are my own.
The DiDache: a MissiNG Piece Of The PuzzLe
the rite as the object of choices about words, syntax, phrasing, and script-ing—can be a totally legitimate understanding of the rite, if it sees the rite as something chosen by the tradition or the individual recording the rite.
a scholar legitimately considers the theology expressed by the choice of one ritual script over another, even if, when performing either script, one would not intend the theological meaning of the script in the way one does when choosing which script is the better. schwiebert’s approach has the merit of helping us to distinguish clearly these two interpretive points of view, while allowing us to conclude to the legitimacy of both in our interpretation of the Didache meal prayers or of any other ritual text.
schwiebert’s attention to ritual context and to points of view from which ritual texts can be interpreted can help us to clarify the way in which van de sandt’s chapter in this volume treats the textual units at the base of his interpretation of the Didache community’s limitations on who could participate in its eucharist:
Now let no one eat or drink of your eucharist but those baptized into the name of the Lord, for the Lord has said concerning this, “Do not give what is holy to the dogs.” (9.5)
if someone is holy, let that person come. if someone is not, let that person convert. (10.6)
Van de sandt sees both of these texts as “directives” and bits of “rubrical instruction.”9 These qualifications fit well with 9.5, but less so with 10.6.
The former verse comes right after a doxology ending the thanksgiving prayers that are performed before people are filled with a meal (10.1).
The second word of 9.5 (δέ) often functions in the Didache to introduce a new unit of the text and so fits well with the transition from prayer text to rubric. Yes, in its only nonrubrical appearance in Did. 9–10 (in 10.3), this Greek word clearly has the adversative function of indicating a contrast between food and drink from the almighty Master for people’s enjoyment, on one hand, and spiritual food and drink and eternal life through Jesus for the community’s “us,” on the other hand. But apart from 10.6, which is in question here, no ritual rubric in the Didache goes without δέ at the 9. see in the present volume, huub van de sandt, “Baptism and holiness: Two Requirements authorizing Participation in the Didache’s eucharist,” above 139–64
(140).
MueLLeR: ResPONse
237
front. for rubrics on permitted and forbidden foods, see 6.3; on baptism see 7.1, 2, 3, 4 (prebaptismal fast); on fast days see 8.1; on daily prayer see 8.2; on when and how to begin the eucharist, see 14.1 and 2.
Didache 10.6 begins and ends with formulaic acclamations of prayer:
“Let grace come, and let this world depart. hosanna to the God [or house, or son] of David.… Maranatha. amen.” in between them, van de sandt would locate a rubric about the eucharistic participation of the holy and the conversion of the unholy, a rubric that has nothing much to do with the prayer formulae immediately before and after it. every other rubric adjoining a text of prayer in the Didache tells people to say the prayer in question or to let certain people pray it and others not. from all of this, i conclude that 10.6 contains no rubrics and that the verse is composed of two sets of prayer acclamations,10 each one beginning with two sentences with a third-person singular imperative verb and ending with an acclamation using hebrew or aramaic. The whole prayer unit ends with the amen.
(a) Let grace come, and let this world depart. hosanna to the God [or house, or son] of David.
(b) if someone is holy, let that person come. if someone is not, let that person convert. Maranatha.
(c) amen.
if these conclusions are correct, another difficulty in van de sandt’s interpretation of Did. 10.6 is removed. his essay tells us that, with the words
“if anyone is holy, let him come,” the “celebrant bids the participants to leave their seats in order to take part in the communal meal.”11 however, 10.1 presupposes that those present at the celebration have already eaten their fill: “Now after you have been filled, give thanks in this way.” is van de sandt distinguishing here between a satiating meal eaten before the prayer in 10.1–5 and a communal meal taken after this prayer? if not, why invite the assembly to a meal after it has taken place? The oddness of inviting people to a meal after they have just had one disappears when 10.6 is read as a series of acclamations and when all the imperatives are seen, not 10. i thus agree on this point with Jonathan schwiebert’s and John clabeaux’s essays in this volume, as well as with Jonathan Draper’s interpretation of Did. 10.6.
see his “Ritual Process and Ritual symbol in Didache 7–10,” VC 54 (2000): 121–58
(130–31, 141–42); see also schwiebert, Knowledge and the Coming Kingdom, 72–79.
11. Van de sandt, “Baptism and holiness,” 142.
The DiDache: a MissiNG Piece Of The PuzzLe
as directions for a holy person or this world to do something or other, but as expressions of desire for, or agreement with, the way things are or soon will be.
finally, in the spirit of schwiebert’s essay, i point out that the fact that 10.6 contains no rubrics, but only acclamations, shows that the “argument”
van de sandt wants to see in the acclamation on eucharistic participation for the holy and conversion for the unholy is no argument at all. acclamation and argument constitute two distinct activities. Whether from the point of view of those who say this acclamation during a performance of the Didache eucharist, from the angle of those choosing to put this acclamation at this place in the rite or from the outlook of those who choose to engage in this rite rather than to do something else, we should not see this acclamation as an argument. it does express convictions, however, and van de sandt’s essay illuminates those convictions admirably by pointing out their connection to the arguments in Did. 9.5 and in 14.1–3, both of which passages contain rubrics the justifying arguments for which are signaled by the use of γάρ to introduce a quotation of the Lord’s words.
Van de sandt presents a very plausible interpretation of the limitation of the Didache eucharist to the baptized in 9.5. his argument moves from that limitation to the explanatory dominical logion that follows: what is holy is for not dogs. he establishes that what is holy here is food for sacrifice like that in the temple, in accord with LXX and other Jewish sources.
On that basis he interprets baptism in the Didache as a presacrificial purification rite. The incorporation of the Two Ways treatise into baptismal purification shows that that purification is ethical as well as ritual. This fact explains the less than punctilious attitude in Did. 7 about the type of water used for baptism. The prescriptions of this chapter are interpreted in terms of the Mishnah treatise on baths (m. Miqw. 1:1–8). Van de sandt thus artfully reads the restriction of eucharist to the baptized in its context within the Didache, as well as in a wider Jewish context.
his argument for these broad lines of his interpretation convinces me, but some of the details of his discussion leave me with questions. first, he emphasizes rightly that the rubric in 9.5 has the effect of excluding gentiles from the eucharist, but would it not also exclude Jews who were not baptized into the name of father, son, and holy spirit (7.1)? second, van de sandt’s comparison between Did. 7 and m. Miqw. 1:1–8 leads him to conclude that the Didache here represents a lessening concern for the details of which sorts of water are apt for ritual cleansing. But is this really what is going on here?
MueLLeR: ResPONse
239
This mishnaic passage is unlike Did. 7 in various ways. No word for
“pure,” “impure,” “clean,” or “unclean” occurs in Did. 7, while these words occur frequently in this part of the Mishnah. in m. Miqw. 1:1–4a, what differentiates the different types of water is contact with something unclean or impure, but this issue never comes up in Did. 7. in the rest of m. Miqw.
1:4, what differentiates the different types of water is whether they are near places frequented by people, but this question never surfaces in Did. 7.
There is a preference for flowing water over contained water in m. Miqw.
1:5–6, as we see in Did. 7, but while the mishnaic text reports different views on how such water is purified, Did. 7 never mentions purification of water as an issue. Perhaps the most interesting parallels with Did. 7 are in m. Miqw. 1:7–8, verses that allow living water, spring water, and a pool of standing water all to be used in purification and in that order of preference. The Mishnah text is later than the Didache. simeon bar Yochai, mentioned in m. Miqw. 1:3 and 1:5, lived in the mid-second century; the houses of shammai and hillel, both of which were functioning in the first century, are mentioned in m. Miqw. 1:5, but this passage mentions them in connection with the issue of how water is purified, a concern never mentioned in Did. 7. for all of these reasons, it is unclear to me that m.
Miqw. 1:1–8 allows us to conclude that Did. 7 represents a slacker ritual observance than what we see elsewhere. i therefore cannot be so sure as van de sandt is that Did. 7 shows a lessening of concern for ritual purity, in favor of an increased concern for ethical purity as a prerequisite for eucharistic participation. While the text itself in Did. 7.1c–3a might “reflect a development that abandoned an originally strict ritual practice,”12 it might not. it might just represent a compendium of different contemporaneous practices, all of which the Didache community had always thought valid.
all of that said, i find myself admiring van de sandt’s clear exposition of Did. 10.6 and 14.1–3. he shows well that confession and reconciliation form part of the holiness required for the purity of the Didache eucharist.
The parallels he shows between these parts of the Didache and the Qumran documents indicate persuasively a certain priority of the ethical over the ritual in the purity concerns of the Didache community. however, i am not sure that i would agree when he seems to affirm that baptism in the Didache has more to do with introducing someone to eucharistic participation than it does with initiating that person into the community itself.
in view of the meaning of that meal as an affirmation of the community’s 12. ibid., 150.
The DiDache: a MissiNG Piece Of The PuzzLe
unity, identity, and defining belief and expectation—so ably portrayed by schwiebert’s essay—van de sandt’s contrast between initiation into the community and initiation into the eucharist is hard to understand. Van de sandt hits the right note when he lets go of this contrast near the end of his essay: “exclusion from the meal simply means exclusion from the community, and full membership is achieved when the candidate is allowed to share in the food.”13
in the Didache, baptism is mentioned in connection with the eucharist only once (9.5). The noun or verb for baptism occurs eight times in Did. 7, and none of these occurrences is part of a link between baptism and eucharist. a number of these occurrences are related to fasting. The contrast between the fasting of the hypocrites and that of the Didache community provides the opening for the mention made of the Lord’s Prayer in this document. Tomson’s essay in this volume situates this prayer with respect to its context in the Didache, to the versions of Matthew and Luke in their respective literary contexts, and to the larger historical context for all of these documents.
he notes that the patristic and New Testament manuscript record indicates that the Matthew-Didache version of the prayer came early on to predominate over the Lukan version. i agree with him that even Mark 11:25 does seem to show traces of the former’s use. While Luke places the prayer easily next to a mention of the prayer instruction of another Jewish teacher, John the Baptist, Matthew, and the Didache set it off in opposition to prayer practices of the hypocrites. showing that the opposition in Did.
8.1 to the hypocrites on the matter of fast days is a Jewish position against a Pharisaic or early rabbinic practice attested in the Mishnah, Tomson sensibly reads the option of this document for the Lord’s Prayer as a similar, Jewish move against the same opponents. he adduces passages from the Mishnah and Tosefta that do converge enough with the Lukan narrative about Jesus teaching the Lord’s Prayer (12:1–5) to warrant the conclusion that that prayer, like those taught by other Jewish teachers of the time, was a short model prayer of variable length and wording to be used when brevity was called for. Tomson shows clearly from m. Ber. 4:1–4, that the Didache shares the thrice-daily rhythm of prayer with the customs inculcated by a number of first-century rabbis.
13. ibid., 164.
MueLLeR: ResPONse
241
Tomson rightly asserts that the Didache preference for fasting on Wednesdays and fridays agrees with the ancient Jewish calendar attested in Jub. 6:23–28 and other Qumran documents. But there is nothing in this passage of Jubilees about which days of the week are proper for which activities. Jubilees 50:13 does forbid fasting on the sabbath, but this issue never arises in Did. 8. Thus, while the fast days of the Didache do fit with the calendar of Jubilees, so do the fast days the Didache identifies as those of the hypocrites. Tomson could, therefore, have explained further how the Jubilees calendar clarifies the relationship between the Didache community and the hypocrites. More precisely, it is unclear how the relationship between Did. 8.1 and Jub. 6 helps to show, as Tomson thinks it does, that the Didache community opposes Gamaliel’s institution of the eighteen Benedictions. Yet i would grant, in agreement with Tomson, that it makes sense that the rabbinic option for the eighteen Benedictions represents at least part of what the Didache opposes by proposing the text of the Lord’s Prayer for daily use.
however, i am not convinced, as Tomson is, that objections in Did. 8
and Matt 6 to the prayer of the hypocrites parallels the protest registered in the m. Ber. 4:1–4 against the institution of the eighteen Benedictions.
in fact, the only thing we know that the Didache does not like in hypocrite prayer is that it does not use the Lord’s Prayer. Yes, the Lord’s Prayer is a short prayer, like those proposed in m. Ber. 4 as alternatives to the eighteen Benedictions, but if the Didache would think of all the rabbis mentioned there as hypocrites, its rule to use the Lord’s Prayer could be critiquing Gamaliel together with his rabbinic opponents. The only opposition to the hypocrites over prayer in Matt 6 is the preference for private over public prayer, while this matter is an issue in neither the Didache nor in m. Ber. 4. Meanwhile, Matt 6 does not contrast the text of the Lord’s Prayer with hypocrite prayer, as the Didache does, but with prolix gentile usage. i conclude that Tomson might not have found as much evidence in Did. 8 and Matthew as he would like for the historicity of the accounts in rabbinic literature of Gamaliel’s establishment of the eighteen Benedictions as a universal daily Jewish prayer. still, Tomson makes a firm case for the Jewish contextualization of the thrice-daily use of the Lord’s Prayer taught in Did. 8.2–3.
Because clabeaux has contributed to this volume a reflective review of Jonathan schwiebert’s Knowledge and the Coming Kingdom, i will spare the reader a review of that review. Those who want to evaluate clabeaux’s solid essay would best begin by reading schwiebert’s impressive book. here i
The DiDache: a MissiNG Piece Of The PuzzLe
will respond to clabeaux’s closing reflections on the potential theological or ecclesiological implications of the influence the Didache’s eucharist had on the history of christian prayer, as schwiebert presented that influence in the last chapters of Knowledge and the Coming Kingdom.
having asked what might have been lost as the Didache’s eucharistic ritual fell into desuetude, clabeaux notes first the absence from today’s christian liturgy of an obviously strong emphasis on eschatological expectation of God’s coming kingdom. Of course, the sidelining of the Didache eucharist is only one factor among very many that account for our time’s muted christian hope for the perfection of God’s assembly at the end of the world. Throughout christian history, the tensed muscles of an imminent eschatological expectation have very often accompanied deep disaffection with the status quo viewed with the wide-angle lens that takes in whole empires and world systems. Thus, where christians evidence a slackened interest in the final perfection of the world-historical process through God’s transcendent initiative, their implicit or explicit judgment or enthusiasm about the most basic cultural frameworks of daily life might be in need of deep revision, if not reversal. The study of an ancient text like the Didache is available perhaps mostly to those too comfortable with the way of the world. Those of us who need it should try to benefit from the tonic of studying its eucharistic ritual.
clabeaux points out that when the Didache meal ritual ceased to be practiced, christians lost one expression of gratitude for the knowledge brought to us from God by Jesus, for example, in the form of catechetical moral instruction and the sayings of christ. he then asks whether the dominant emphasis on associating the bread and wine with Jesus’s body and blood given over for the forgiveness of sin cost christians “a robust understanding of the presence of God in the Word.”14 My reading of christian history leads me to answer in the negative. Yes, over the centuries at various times and places and to varying degrees, christians have needed the reform that brings them back to God’s word. But the New Testament gives ample witness to the need for such reform even in the days when some christians still practiced the Didache meal ritual. The Didache itself shows us that its community knew that christians kept a tenuous hold on the instruction on which their life depended (6.1; 11.2, 10; 16.2–5).
On the other hand, i agree with the great french Jesuit teacher Édouard 14. see in the present volume, John J. clabeaux, “The Ritual Meal in Didache 9–10: Progress in understanding,” above 209–30 (227).
MueLLeR: ResPONse
243
Pousset (1929–1999), who liked to repeat that, throughout christian history, the gospel returns in a fresh way where it is needed. sometimes we might think that we have to wait too long for that return. Perhaps a lively eschatological expectation like that nourished by the Didache meal ritual can help to make us hunger so much for the gospel that we become instru-ments of its return in our own time and place.
at the end of his essay, clabeaux presents arguments for the idea that the Gospel of John concentrates, likely much after the Didache is written, language and images related to a number of themes in the Didache eucharist, without seeing these images and language as opposed to an emphasis on Jesus’s death and resurrection, a thematic complex absent from that eucharist. he adds this point to the evidence that schwiebert’s book collects of aspects of the Didache meal ritual explicitly linked in other ancient christian sources to Jesus’s death and resurrection. Leaving to others more qualified than i the task of evaluating his case for the Didache-John connection, i underline that clabeaux’s point here is more historical than theological. he tries to show that a number of ancient sources bring into unproblematic coexistence two thematic complexes: on one hand, knowledge through Jesus and the coming kingdom; on the other hand, Jesus’s death and resurrection. But clabeaux does not ask what, if anything, about the theologies respectively implied by these thematic complexes allowed for this coexistence in a single ritual or a single gospel. That is a precisely theological question, one that takes us to a point of view different from the history or sociology of religious texts and movements.
i can only adumbrate a response to this question at the close of this response to four stimulating studies. Our father gives us knowledge through Jesus his servant, he gives us spiritual food and drink and eternal life through the same Jesus, and it is again through Jesus that glory and power will belong to our father for his gathering of the church in perfect love at the end of the world. God gives us knowledge through Jesus’s teachings, as well as through that servant’s mighty deeds, the greatest of which is his death and resurrection, after which he instructed his followers with a new efficacy. Jesus’s death and resurrection confirm that his teachings and servanthood had God as their origin and goal. his death and resurrection confirm this fact in part because it is through them that God gives spiritual food and drink and eternal life to christians. after having preached eternal life, after having performed marvelous signs of it through healings and exorcisms, God’s servant showed himself to be the very heaven-sent nourishment of this life by dying and rising from the dead. Because this
The DiDache: a MissiNG Piece Of The PuzzLe
life is perfected in love for God and neighbor, the nourishment of this life must be shared with thanksgiving at the center of a community living the way of life. Knowing that God’s servant, their teacher, has come to perfection in immortal risen life, the community awaits the consummation of its loving unity in the eternal life that comes through its spiritual food and drink. This waiting is active and so requires following the way of life and continual reconciliation as the preludes to sharing in the meal that concretizes the community’s defining expectation. These few statements do not outline the theology of the Didache. They sketch a theology that can be fed by the Didache, the record of a first-century christian Jewish community eager to set itself off from nascent rabbinical currents in its adhesion to the teaching of a servant whose name and prayer proclaim that israel’s God saves.
The Didache and Matthew
E. Bruce Brooks
in this paper i will argue that the Didache is a growth text, that it makes contact with Matthew only at the end of its formation period, and that its core is best understood if it is related to the period before Matthew and Luke: the time of Mark and of other texts that seem to reflect a preresurrection form of belief and practice that i have elsewhere called alpha christianity. since at several points my exposition departs from current scholarship on the Didache, it may assist readers (and avoid many separate and complicated footnotes later on) if i begin by briefly introducing some of the considerations that lie behind those departures.
1. Prolegomena
This is a set of mutually consistent solutions to problems otherwise solved by many, bearing upon the existence and nature of a christianity arising before Paul and persisting during his lifetime and afterward.
1.1. Growth Texts
Not all texts are written at one sitting, and authority texts in particular are likely to extend themselves over time in order to remain valid under changing conditions. One physically obvious example is the Gortyn law code from early crete, literally cut in the stones comprising the walls of the room in which law cases were heard, but with two extensions at the end, the second of which is in a different “hand” than the preceding text.1 The Gospels, which evidently aspire to tell the truth about Jesus in 1. for a complete description, see Ronald f. Willetts, The Law Code of Gortyn, Ksup 1 (Berlin: de Gruyter, 1967); for the addendum, especially 4, 27, and 78.
-247 -
The DiDache: a MissiNG Piece Of The PuzzLe
narrative form, are in their nature authority texts, and for John in particular a three-stage composition process has recently been proposed by urban c. von Wahlde.2 These and like extensions are not corruptions in the usual scribal sense, and they are not forgeries. They are an integral if later part of the text and reflect continued management of that text by its original author (or if we envision a school or municipal text, by its original proprietor or his successors). i propose in this paper to consider the evidence for the Didache as a growth text.
1.2. Mark
That Mark is a growth text3 is readily shown by considering its interpolations, many of them signaled by the fact that they interrupt the context, that when they are removed the context closes up to make a perfectly con-secutive text, and that the interpolations themselves can be combined on the basis of similar content to make plausible accretion layers. Mark 14:28
and 16:7 are of this type.4 in each case the verse following ignores the verse in question and addresses a comment or responds to a situation in the verse preceding. The two passages are related (the second refers back to the first) and thus form a layer of their own. What they add to the previous story is not the forthcoming appearance of Jesus to the disciples in Galilee, which was already implicit in the narrative, but Jesus’s foreknowledge of that appearance. i take this as an instance of the progressive divinization of Jesus, a process easily seen in the four Gospels taken together,5 but 2. The Gospel and Letters of John (Grand Rapids: eerdmans, 2010).
3. a complete if tentative reconstruction was presented in e. Bruce Brooks,
“structural evolution in Mark” (paper presented at the New england regional meeting of the society of Biblical Literature, 21 april 2006); see now “The Resurrection of Jesus in Mark,” WSP 3 (forthcoming).
4. Noticed by heinrich J. holtzmann, Die Synoptiker, hcNT 1/1 (Tübingen: Mohr, 1901), 174; ernst Lohmeyer, Das Evangelium des Markus, KeK 1.2 (Göttingen: Vandenhoeck & Ruprecht, 1937), 312; frederick c. Grant, “The Gospel according to st. Mark,” IB 7:627–917 (879); more guardedly by Vincent Taylor, The Gospel According to St. Mark (London: Macmillan, 1959), 549. The insecurity of these interruptive passages in context passes without comment in many recent commentaries. This i attribute to the Zeitgeist (the same is true of all canonical New Testament texts) and not to any defect in the evidence.
5. for a brief account, see e. Bruce Brooks, “Gospel Trajectories,” WSP 1 (2010): 171–72.
BROOKs: BefORe aND afTeR MaTTheW
249
which, as these interpolations show, is also in progress within Mark itself.
The characteristic of earliest christianity is that it lies at the beginning of this trajectory, and divinized Jesus not at all.
Other passages in Mark that have been convincingly analyzed as layered include the Markan apocalypse (by Vincent Taylor)6 and the Markan passion narrative (by adela Yarbro collins).7 The latter reconstruction ends at Mark 15:38 and thus does not include the resurrection.8 Notice also that the atonement doctrine (the salvific value of Jesus’s death) is later in Mark than the resurrection doctrine. Whereas the resurrection has been thoroughly worked into the text, the atonement doctrine is exiguous: it appears in only two passages: Mark 10:45 and 14:24b, both of which may be removed without detriment (in the latter case, arguably with benefit) to the context. stratified Mark thus contains a sort of archaeological history of the evolution of early christian thought.9
6. see Taylor, Gospel According to St. Mark, 636–41; e. Bruce Brooks, “Mark’s apocalyptic Discourse,” WSP 3 (forthcoming).
7. i should add that Yarbro collins prefers to call the earlier form the pre-Markan passion narrative, whereas i consider it to be an early state in a continuous process of additions to Mark and thus within the Markan formation process.
8. adela Yarbro collins, Mark: A Commentary, hermeneia (Minneapolis: fortress, 2007), 819 and preceding commentary. for an independent argument that the resurrection is not original in Mark, see Peter Kirby, “The case against the empty Tomb,” JHC 9 (2002): 175–202. The resurrection predictions in Mark evidently stand or fall together, and the second of them (Mark 9:30–32) is in conflict with Mark 9:33
as to whether Jesus talked with his disciples on the way, or did not do so and had to ask afterward what they had been saying. Of the two, it is the latter that makes better sense in the immediate context. it follows that 9:30–32, and with it the other resurrection predictions, are late in Mark.
9. a common response to evidence for interpolation in Mark is the edwards
“sandwich” thesis. The prize example of this supposed authorial structure is the Woman with the flow of Blood (Mark 5:24b–34) interrupting the Jairus’s daughter narrative (Mark 5:22–24a and 5:35–43). Of this and all the “sandwiches,” edwards observes, “almost always the insertion is the standard by which the flanking material is measured, the key to the interpretation of the whole”; see James R. edwards,
“Markan sandwiches,” in The Composition of Mark’s Gospel, ed. David e. Orton, BRBs 3 (Leiden: Brill, 1999), 215. a similar and stronger statement could be made of an interpolation, whose purpose is to update the “flanking material.” The point of the Jairus story was the power of Jesus to heal. The point of the inserted Woman story is quite different: it is the power of faith in Jesus to heal, even without the personal intervention of the historical Jesus (Mark 5:34: “and he said to her, daughter, your faith has made you well” [unless otherwise noted, biblical citations follow the asV]). By
The DiDache: a MissiNG Piece Of The PuzzLe
Mark is usually dated to after 70 by interpreting Mark 13:14 as a reference to the destruction of Jerusalem by the army of Titus. But desecration is not destruction, and this passage (with its clear reference to Dan 11:31 and thus to the desecration of the temple by antiochus epiphanes)10
is more plausibly taken as referring to the threatened desecration of the temple by caligula in the summer of 40.11 When a gospel writer wishes to portray the army of Titus besieging Jerusalem, the result is likely to read more like Luke 21:20, which in that gospel replaces the parallel Mark 13:14. The latest firm terminus post quem in Mark is the reference to the execution of James zebedee by herod agrippa i, probably in 44, which is predicted in Mark 10:39. There is no reason in the text to date any of Mark after around 45, and the majority of Mark would then have been earlier.
as a whole then, Mark belongs to the early, not the middle, period of early christianity. Given the evident similarity between certain Didache passages and Matthew, scholarship on the Didache has naturally focused on Matthew, sometimes to the extent of excluding Mark and other seemingly early texts from consideration. i propose in this paper to bring Mark (and the also early epistle of James) back into the picture.
1.3. Luke
This text is also stratified, as is most easily seen in the passages that occupy different positions in the narrative from their counterparts in Mark and create inconcinnities in their new positions. These can only have been relocated at a later point during the formation of the text.12 One of the most inserting this story, Mark updates what had been merely a story of Jesus and directly addresses the concerns of the churches of his own time.
10. The reference to Daniel is made explicit in the Matthean version of this passage; see Matt 24:15.
11. This has often been pointed out. it may be the basis of the Paschal chronicle’s assigning to Mark the date 40 and the early tradition, expressed at the end of several uncial manuscripts, that Mark was written in the tenth year after the ascension, which on some views about the chronology of Jesus would again give 40. among more recent treatments, see for example James G. crossley, The Date of Mark’s Gospel: Insight from the Law in Earliest Christianity, LNTs (London: clark, 2004), 29–30. The threat was removed by the death of caligula in January 41, so that Mark 13:14 can only have been written in 40. This is one of two passages in Mark for which an absolute terminus post quem can be determined. for the other, Mark 10:39, see below.
12. see e. Bruce Brooks, “Prolegomena to Proto-Luke” (paper presented at the
BROOKs: BefORe aND afTeR MaTTheW
251
obvious is the Nazareth episode (Luke 4:16–30) in which Jesus’s hearers demand that he do wonders such as he had done at capernaum. But at this point in the present Lukan narrative, Jesus has not yet been to capernaum.
in his Gospel and also in the acts, Luke takes a highly revisionist view of Paul. Paul’s most characteristic doctrine is probably the atonement, asserted with typical vehemence in Paul’s last letter, Rom 3:20–24 and (with the example of abraham) in 4:1–3. Luke in his Gospel suppresses the two Markan passages in which this doctrine occurs, and in acts he never shows Paul as preaching that doctrine.13 it was apparently Luke’s intention to transmit both Jesus tradition and later christian history in a form that would be more suitable for the future than the actual facts. among the more upsetting of those facts was the bitter exchange between Paul in Romans and the last addition to James,14 which in 2:20 holds Paul’s doctrine of faith (in the atonement) up to scorn (“Do you want to be shown, you foolish fellow, that faith apart from works is barren?”) and in 2:21–24
goes on to refute Paul’s example of abraham, arguing that abraham was an instance of deeds, not merely of faith (“Was not abraham our father justified by works, when he offered his son isaac upon the altar?”). These disputes among christians were probably divisive, and Luke may have felt that some sort of peace needed to be achieved within christianity so as to save energy for the conflicts that it was experiencing from outside. i propose in this paper to bring this conflict between Paul and earlier christianity into sharper perspective.
1.4. Matthew
Matthew 24:15 repeats without modification Mark 13:14, with its caligula desecration prediction.15 The text thus does not appear to be aware of annual Meeting of the society of Biblical Literature, san Diego, ca, 17 November 2007) and “acts-Luke,” Alpha (forthcoming). The latter argues for a three-stage composition process for Luke, the second and third of which correspond to a two-stage formation process for acts.
13. it is mentioned only in acts 21:28 as part of Paul’s farewell to the ephesians, but as a personal idea of Paul’s, not as something that Luke represents him as preaching to the churches.
14. for this stratification, see e. Bruce Brooks, “The epistle of James,” Alpha (forthcoming).
15. it may be asked, if the caligula prediction did not come off, why was it retained in Mark (and repeated in Matthew)? i reply that authority texts in their
The DiDache: a MissiNG Piece Of The PuzzLe
the Titus destruction of 70 (which is unmistakably reflected in the parallel of Luke 21:20). The final state of Matthew is thus most plausibly dated before 70 and the final state of Luke after 70. The impetus for composing both may well have been the deaths of the major apostles Paul (probably in 60, since the two years of virtual freedom in Rome mentioned in acts 28:30–31 may be an instance of the consistently pro-Roman stance of acts)16 and Peter (probably in 64, the first year of the Neronian persecution). These, not to mention the other turmoil of the time, could have suggested the need for a new account of Jesus and his legacy. The available time window is small and seems to require something like: Luke a (ca. 66) > Matthew (ca. 68) > Luke B (after 70) This would have been during the Jewish War, making a Palestinian place of composition problematic. But Luke has been persistently associated with antioch, and since the two gospels are obviously closely related, syria may be a reasonable hypothesis for Matthew also.17 The difference between Luke (who consistently speaks for the poor) and Matthew (who is comfortable with large numbers, and, as in Matt 23:2 and many like passages, is somewhat in reaction against Jesus’s abridging of the traditional laws) is as likely to be social as geographical.
1.5. Q
This is a conjectural source, supposed to be earlier than Matthew and Luke and used by both. “Q” is thought to account for much of the material common to Matthew and Luke but absent in Mark. Perhaps the strongest evidence for such a source is: (1) the apparent bidirectionality (often called
“alternating primitivity”) displayed by the common material and (2) the nature do not subtract; subtraction is an admission of error in a text already to some degree public. What usually happens instead is that the seeming difficulty is made good by reinterpretation.
16. The final imprisonment of Paul is imagined very differently by whoever wrote 2 Tim 4:6–18. That this describes a second imprisonment has been shown to be untenable; see John Macpherson, “Was There a second imprisonment of Paul in Rome?”
AJT 4 (1900): 23–48.
17. see the careful discussion in W. D. Davies and Dale c. allison, A Critical and Exegetical Commentary on the Gospel according to Saint Matthew, 3 vols., icc (edinburgh: T&T clark, 1988–1997), 1:138–47; in the end they opt for antioch.
BROOKs: BefORe aND afTeR MaTTheW
253
assumption that both Matthew and Luke are integral texts, written at a single date. The latter assumption seems to hold for Matthew (i see no philological warrant for a stratification analysis and do not believe that one has ever been proposed), but as argued above, not for Luke. Michael Goulder sought to undermine the Q hypothesis by demonstrating Matthew > Luke directionality for all the common material. This works for the parable of the talents (Matt 25:14–30 || Luke 19:11–27), which involves a narratively superfluous king in the Lukan version, and thus suggests the directionality Matthew > Luke.18 But it founders on the parable of the feast (Matt 22:1–14 || Luke 14:16–24), which involves a narratively superfluous king in the Matthean version, and thus equally implies the directionality Luke > Matthew. But if the earliest stage of Luke were earlier than Matthew and the second stage of Luke were later than Matthew and aware of Matthew, the above and other facts can be accommodated. a further advantage of that model is that the minor agreements (not addressed in the Q hypothesis) can also be accommodated. so can such elements as the respective birth narratives, which cannot be derived by the usual redaction scenario from a common ancestor (they have almost no details in common), but have a clear Matthew > Luke directionality (in Luke, but not in Matthew, the birth story overrides a perfectly satisfactory beginning in the historical synchronisms of Luke 3:1–2).19 The conjectural “Q” will thus play no part in the following argument.
1.6. Paul
The strongest argument for an early christianity is that Paul began by persecuting it. The second strongest is that he sometimes quotes what have been recognized as pre-Pauline formulas.
Paul was a vehement personality and was just as zealous to assert his apostolic credentials (against those of the “superlative apostles,” 1 cor 18. it also involves other absurdities; see his delightful account in Michael Goulder, Luke: A New Paradigm, 2 vols., JsNTsup 10 (sheffield: JsOT, 1989), 2:680–83.
19. The effect of the transition has often been noticed, e.g.: “at this point, the narrative emerges into the clearer light of day, and joins the main stream of the older evangelical tradition which began not with the birth but with the baptism of Jesus”
(William Manson, The Gospel of Luke, MNTc [London: hodder & stoughton, 1930], 24). for the divinization trajectory that is implied in the divine birth narratives of both Matthew and Luke, see again Brooks, “Gospel Trajectories.”
The DiDache: a MissiNG Piece Of The PuzzLe
11:5 and 12:11) as he had earlier been to exterminate the christians in his vicinity (Gal 1:13–14). Paul is an author to read with care, not least because, in addressing christians whose beliefs differed from his, he was prone to adopt their way of speaking initially as a first step in converting them to his own quite different point of view. an extreme example is Rom 1:2, where he speaks of the Davidic descent of Jesus, seemingly to ingrati-ate himself with those he imagines himself as addressing, at least some of whom evidently took the old Davidic messiah tradition of Jesus seriously.
Paul had no interest in the Davidic or any other descent of Jesus, or in any other aspect of Jesus during his lifetime (see 2 cor 5:16, where that knowledge is superseded by a higher knowledge of Jesus). it was exclusively Jesus as resurrected that interested Paul (emphasized as late as 1 cor 15:14: “if christ has not been raised, then our preaching is in vain, and your faith is in vain”). The joke is that the author of the deutero-Pauline text of 2 Timothy took this passage as genuine rather than rhetorical and included it accordingly (in 2:8) as Paul’s own belief and indeed his teaching, doubtless thinking to strengthen thereby the perceived Pauline character of his work.
for present purposes, there are three major points requiring to be explained about Paul: (1) his persecution of the early christians, (2) his conversion to christianity, and (3) his adoption of the atonement interpretation of the resurrection. for the second, biographies of Paul usually find it sufficient to refer to the experience on the Damascus road, though this is a Lukan fiction not greatly resembling what is probably Paul’s own account of the same experience in 2 cor 12:2–5. in any case, what inner breakthrough prepared, or was expressed by, that experience? The first two questions are perhaps both best answered by Paul himself, with a guarded assist from Luke. immediately after recalling that he “persecuted the church of God violently,” Paul in Gal 1:14 describes himself as “extremely zealous … for the traditions of my fathers” (the zeal and the persecution are even more closely linked in Phil 3:6: “as to zeal a persecutor of the church”). These traditions would have included not only the written commandments, with the Decalogue at their center, but the oral elaborations developed and prized by the Pharisees. Jesus in Mark is frequently at odds with the Pharisees over such traditions as food purity and such Decalogue prescriptions as the one enjoining observance of the sabbath. further, and more specifically, in Mark 10:19 Jesus gives an abridged version of the Decalogue. This is sometimes characterized as the “second table of the Decalogue,” but this is not precisely correct. The actual inventory (with its
BROOKs: BefORe aND afTeR MaTTheW
255
sequence numbers in the Decalogue at Deut 4:7–21, which is also that of exod 20:1–17) is murder (#6), adultery (#7), theft (#8), false witness (#9), fraud (not in the Decalogue),20 and respect for parents (#5). This abridge-ment, plus the Markan Jesus’s frequent and conspicuous offenses against the sabbath commandment and several Pharisaic purity rules, would be well calculated to infuriate the Pharisee Paul.
as to Paul’s motive for converting to christianity, i think the likeli-est possibility is that the christianity to which Paul converted was not the same as the christianity that he had previously persecuted. if at some point there had arisen within the christianity of the time a new theory that emphasized the resurrection of Jesus, this might easily have resonated with Paul’s Pharisaic belief in the resurrection of the dead and produced the sort of about-turning—the sudden recognition—that Paul’s conversion seems to exhibit.21 That this idea was a development in christianity and not original doctrine (Jesus can hardly have taught it to the crowds during his lifetime, and no gospel depicts him as doing so) has been argued above.
in support of the centrality of the resurrection in Paul’s conversion, we may notice the emphasis that Paul, to the end of his writings, places on the resurrection (1 cor 15:14, above cited), and his early visit to Jerusalem to confer with Peter, who had been the first to receive a vision of the risen christ (this gospel tradition is repeated by Paul at 1 cor 15:5: “and that he appeared to cephas, then to the Twelve”). Luke in acts also makes the resurrection the distinguishing point for Paul, saying that he “preached Jesus and the resurrection” (acts 17:18b) and later in Jerusalem defended 20. But compare Mal 3:5, which also mentions other Jesus commandments. The prohibition of fraud is a signature of certain early christian traditions. Besides Mark 10:19, it also occurs at Jas 5:4 (with a probable echo in 1 cor 6:8) and Luke 19:8. There are similar prohibitions in the Two Ways vice list.
21. a point in favor of the two-christianity view suggested here is that Paul as a christian continued to oppose those who did not share his theological views, in particular those who did not place equal emphasis on the death of Jesus. see his curse in the afterthought at the end of 1 corinthians (“if anyone has no love for the Lord, let him be accursed [anathema]. Our Lord, come!”). The latter translates the aramaic prayer Maranatha, given in its original form in Did 10.6. Paul’s meaning is, “Let the Lord indeed come, and condemn these heretics.” at this and similar moments, i suggest that we are seeing an example of the general pattern to which attention was called by Walter Bauer in 1934 (see his Rechtgläubigkeit und Ketzerei im ältesten Christentum, BhT 10 [Tübingen: Mohr siebeck, 1934]): an original belief becoming heretical in terms of a later belief. Paul here represents the later belief.
The DiDache: a MissiNG Piece Of The PuzzLe
himself by saying that he was being persecuted by other Jews who, unlike himself, did not believe in a resurrection (acts 23:6; so also before felix, acts 24:21).
The atonement doctrine is our third difficulty. Rudolf Bultmann, following Wilhelm Bousset, suggested that Paul himself invented it, but the available facts do not seem to support this. We have firsthand testimony for Paul’s theology only during seven years from late in his life: from 1 Thessalonians (ca. 51) to Romans (ca. 57). But even in that short span some development may be observed, and it seems that the atonement, which is so vigorously stressed in Romans, is not yet present in 1 Thessalonians.22 it is present, but as a seemingly late addition, in Mark. Given a latest terminus post quem of circa 45 for Mark, it is probable that the atonement doctrine had been acknowledged by that text not long before, perhaps circa 43. This is eight years before the date usually assigned to 1 Thessalonians. it thus seems that Paul got the atonement doctrine from somewhere else, perhaps most likely from theoreticians at Jerusalem, the zone of John, Mark’s primary acquaintance, and as the site of the temple, also the place where sacrifice-based concepts like the atonement might naturally have arisen.
1.7. Torah
The term “law” in our texts does not invariably invoke the whole of the Torah, whether written or oral. We have seen that Jesus in Mark preaches a reduced Torah (as it happens, precisely those parts of it that do not involve a specifically Jewish form of piety). in this paper, i will not be equating the term “law” with the full Torah. i will also not be using the term “Jewish christian,” which implies that the only significant difference within early christianity is the contrast between Jewish and gentile origin. Though this polarity existed and was consequential, i find that doctrinal differences among christians of whatever origin were also significant.
22. see, for example, herman N. Ridderbos, “The earliest confession of the atonement in Paul” in Reconciliation and Hope: New Testament Essays on Atonement and Eschatology Presented to L. L. Morris on His 60th Birthday, ed. Robert L. Banks (carlisle: Paternoster, 1974), 76–89; and compare Karl Paul Donfried, Paul, Thessalonica, and Early Christianity (London: T&T clark, 2002), 81–83.
BROOKs: BefORe aND afTeR MaTTheW
257
1.8. alpha christianity
There are several extant texts which have in common that (1) they seem to portray a relatively early form of christianity, and (2) they do not mention the resurrection. These include James, the hymn embedded in Phil 223 (which features an exalted and even preexistent christ but not one whose death is said to have salvific significance), and from the preceding argument, the early layers of Mark. among later texts, it is conspicuous that Peter in the Pseudo-clementine Recognitions preaches endlessly, but never mentions the resurrection or any doctrine connected with the resurrection. This seems to represent a vigorous survival (despite what was by then its heretical status) of alpha doctrine.24 in this paper, i will attempt to show the affinity of the Didache with this alpha christianity, whose ritual prescriptions i believe it preserves.
23. as reconstructed by Lohmeyer, it includes an intrusive half-verse added by Paul in quoting it and introducing, if not the resurrection in an explicit form, an allusion to the theologia crucis. see archibald Macbride hunter, Paul and His Predecessors, 2nd ed. (London: scM, 1961), 41; Joseph a. fitzmyer, “The ascension of christ in Pentecost,” in To Advance the Gospel: New Testament Studies, 2nd ed. (Grand Rapids: eerdmans, 1998), 267.
24. i do not wish to imply that preresurrection christianity, here called alpha, was itself uniform. There were undoubtedly many local variants and varieties. for example, a fairly coherent group, the ebionites (seemingly a christian counterpart of the essenes), shared many of the traits here mentioned, including the focus on the poor (whence the name) that is developed in Luke, but they also had some distinctive beliefs and used their own texts, including a gospel based on Matthew (Jean Danielou, The Theology of Jewish Christianity, trans. and ed. John a. Baker [chicago: henry Regnery, 1964], 55–64). That they used Matthew rather than Mark need occasion no surprise. Beginning at some point not long after the destruction of Jerusalem, everybody, including Luke, was increasingly using Matthew; for its prominence, see Édouard Massaux, The Influence of the Gospel of Saint Matthew on the Christian Literature before Saint Irenaeus, trans. Norman J. Belval and suzanne hecht, 3 vols., NGs 5.2 (Macon, Ga: Mercer university Press, 1993), 3:183–89. for another focus, see Ray a. Pritz, Nazarene Jewish Christianity Jewish Christianity: From the End of the New Testament Period Until Its Disappearance in the Fourth Century, stPB 37 (Leiden: Brill; Jerusalem: Magnes, 1988).
The DiDache: a MissiNG Piece Of The PuzzLe
2. The Didache25
The Didache has been an object of interest since the discovery of an almost complete text by Bishop Bryennios in 1873 (first published 1883), but opinions about its place in the scheme of things still differ. One focus of attention is the text’s gospel parallels, the strongest resemblances being chiefly to Matthew.26 The question is whether these are intrinsic, in which case 25. Like nearly all titles of ancient texts, the ones attached to the Bryennios manuscript were most likely labels meant to distinguish the text from others and are thus not original. Of them, “The Teaching [ΔΙΔΑΧΗ] of the Twelve apostles” is physically emphasized in the Bryennios manuscript (see the facsimile in J. Rendel harris, The Teaching of the Apostles [London: c. J. clay; Baltimore: Johns hopkins university Press, 1887], plate i) and is probably earlier; after it, and in the manuscript run together with the text proper, is “The Teaching of the Twelve apostles to the Gentiles.” This may come from a later time and reflect a later perception. i find nothing in the Didache itself that unambiguously suggests a limitation to gentile converts or churches.
26. authorities differ as between reliance on common tradition and the unequivocal use of Matthew. in the end, one must use one’s own judgment. i have consulted chiefly Oxford society of historical Theology, The New Testament in the Apostolic Fathers (Oxford: clarendon, 1905); Massaux, Influence of the Gospel, 1950; Robert a.
Kraft, Barnabas and the Didache, af 3 (Toronto: Nelson, 1965); Wolf-Dietrich Köhler, Die Rezeption des Matthäusevangeliums in der Zeit vor Irenaeus, WuNT 2/24 (Tübingen: Mohr siebeck, 1985) as appended to Massaux, Influence of the Gospel; and William Varner, The Way of the Didache: The First Christian Handbook (Lanham, MD: university Press of america, 2007). Discussion still continues, and the century update of the Oxford committee study (andrew Gregory and christopher M. Tuckett, eds., The New Testament and the Apostolic Fathers: The Reception of the New Testament in the Apostolic Fathers, 2 vols. [Oxford: Oxford university Press, 2005]) still offers room for it. for a survey, with whose conclusion i cannot agree, see christopher M.
Tuckett, “synoptic Tradition in the Didache,” in The Didache in Modern Research, ed.
Jonathan a. Draper, aGJu 37 (Leiden: Brill, 1996), 92–128. i consider two instances to be unequivocal. Didache 8.2, “and do not pray like the hypocrites, but like the Lord commanded in his gospel. Pray this way.” There follows the Lord’s Prayer in its Matthean, not its Lukan, version (see also Jonathan a. Draper, “The Jesus Tradition in the Didache,” in The Jesus Tradition Outside the Gospels, vol. 5 of Gospel Perspectives, ed. David Wenham [sheffield: JsOT, 1985], 279). Didache 15.4, “and do your prayers and alms and all your actions, as you have it in the Gospel of our Lord” (cf. Matt 6:1–4, 5–15) confirms that the author of these passages regarded the Gospel of Matthew (not some more indefinite “gospel” teaching) as authoritative for the Jesus tradition. Then these verses of the Didache are unambiguously aware of Matthew in its Gospel form, not in some preliminary version. That Matthew might have used the earlier Didache, or something having the same tendency, is not out of the question; see further below.
BROOKs: BefORe aND afTeR MaTTheW
259
the Didache is post-Matthean and thus late, or interpolative, in which case the Didache or its earliest textual state is pre-Matthean and thus early.27 i here seek to show, with special attention to the formal evidence, that the Matthean (and Lukan) elements are interpolative, and that the original Didache attests the earlier form of christian belief and practice which i have above called alpha, in contradistinction to the Beta or resurrection christianity preached by Paul.
The principal witness is the Bryennios manuscript, written in 1056.
Other significant witnesses are the P.Oxy. 1782 fragments (late fourth century) and a fifth century coptic fragment (Br. Mus. Or. 9271) containing Did. 10.3b–12.2a, probably an extract rather than part of a copy.28 a version of the Two Ways portion also appears at the end of the epistle of Barnabas, and the entire Didache is included in a comprehensive fourth century Beta document, the apostolic constitutions and canons.
The Didache consists of three readily distinguishable parts. The first is the Two Ways document, Did. 1.1–6.2, originally a separate tract,29 of Didache 1.5, “To everyone asking you for anything, give it and do not ask for it back.”
here the text is closer to Luke 6:30 than to the parallel Matt 5:41, 40 (Varner, Way of the Didache, 20). Then the Didache was aware of Luke as well as Matthew (the source for Did. 1.4, immediately preceding) and thus aware of both the second-tier gospels, though regarding Matthew more highly. a claim of contact with the Gospel of John (at Did. 9.4) has been refuted by arthur Vööbus, “Regarding the Background of the Liturgical Traditions in the Didache,” VC 23 (1969): 81–87. With respect to the Gospels, the Didache as we have it seems to occupy a position not unlike that of the Gospel of Thomas: some echoes of Mark (organic; see below), contact with Matthew and/or Luke, but none with John. The Luke > Thomas directionality has been convincingly argued most recently by Mark Goodacre, Thomas and the Gospels: The Case for Thomas’s Familiarity with the Gospels (Grand Rapids: eerdmans, 2012). The directionality of the Didache/Matthew parallels, and whether some form of the Didache existed before those contacts were made, is precisely the question explored in the present paper.
27. With the third-text option i have dealt in the “Q” section of §1.5, above, by suggesting an alternative to “Q.”
28. for details, see Kurt Niederwimmer, The Didache: A Commentary, trans.
Linda M. Maloney, hermeneia (Minneapolis: fortress, 1998), 19–26. some other relevant manuscripts will be mentioned below.
29. The instructional part of the Two Ways seems to end with the list of evil deeds in 5.2, which concludes “May you be delivered, children, from all of these!” (trans.
Varner, Way of the Didache). Didache 6.1, “see to it that no one leads you astray from this way of teaching, since he is teaching you apart from God,” refers instead to the whole of the Two Ways teaching and, with 6.2, is probably a transition to the following church order section of the Didache. compare the introductory phrase in Barn. 18.1,
The DiDache: a MissiNG Piece Of The PuzzLe
which a version is included in Barn. 18–20. There are partial parallels in the vice lists of Rom 1:19–2:29 and Gal 5:13–6:10, both of which have been challenged as interpolated in the respective Pauline epistles.30 The Two Ways tract functions as the ethical section of the Didache; it is addressed to ordinary believers: members rather than leaders of local churches.
comparison of the Bryennios and Barnabas versions permits recovery of their archetype: the passages common to both.
Next comes Did. 6.3–12.1, with instructions for administering the sacraments and receiving apostolic visitors. This is the church order or liturgical section of the Didache; it is addressed to church leaders. attention to the structural use of the itemization formula περὶ δέ helps to distinguish an original core from later supplements.31
Last is the apocalypse, Did. 16, the doctrinal section. Like the Two Ways, it is addressed to ordinary believers and thus contrasts with the preceding material. Didache 16 itself appears to be an integral production.
The Didache is incorporated whole into apos. con. 7, but with “ortho-doxizing” changes that introduce the resurrection doctrine and thus compromise its original alpha character.32 apostolic constitutions 7.32.5 continues beyond the defective end of the Bryennios manuscript and seems to
“Let us now pass on to another kind of knowledge and instruction,” which seems to link the Two Ways tract to a text that had ended satisfactorily at Barn. 17.2, “for if i should write you concerning things present or yet to come, you would not grasp them, because they are as yet hidden in parables. Let this, then, be enough” (trans. James a. Kleist, The Didache; the Epistle of Barnabas; the Epistles and the Martyrdom of St.
Polycarp; the Fragments of Papias; the Epistle to Diognetus, acW 6 [Westminster, MD: Newman, 1948]). The implication of these bridging passages is that a Two Ways tract was appended to an already complete epistle of Barnabas and separately prefixed to the preexisting liturgical portion of the Didache.
30. for Rom 1:18–2:29, see William O. Walker Jr., Interpolations in the Pauline Letters (London: sheffield, 2001), 166–89; for Gal 5:13–6:10, see John cochrane O’Neill, The Recovery of Paul’s Letter to the Galatians (London: sPcK, 1972), 65–71. i consider these arguments to be definitive.
31. for details of this argument, see §2.2.1 below.
32. The lack of the resurrection and associated doctrines in the Didache has been noticed. The concept of Jesus as redeeming others through his suffering can be connected to the Didache only by its use of the Greek word πᾶς, considered as evoking the
“suffering servant” of isa 59:2–4 (Varner, Way of the Didache, 94). There is no direct warrant in the text for that linkage.
BROOKs: BefORe aND afTeR MaTTheW
261
preserve the material lost from that manuscript. These last verses i number as Did. 16.8–11.33
This is the Didache text with which i will work. i now take up the three sections in turn, for closer examination, paying particular attention to possible interpolations.
2.1. The Two Ways: Did. 1.1–6.2
This is at bottom a list of good and evil deeds, plus the statement that the path of good deeds leads to life and the path of bad deeds leads to death. it is clearly not the “faith” scenario preached by Paul in Romans. it is instead in the same category as the “works” scenario argued by an alpha document, the epistle of James.34
33. i follow Robert e. aldridge, “The Lost ending of the Didache,” VC 53 (1999): 1–15, except that with support from Boniface, which lacks it, and from aldridge himself (p. 10), who suspects that it “may be a later addition,” i regard the next verse, his Did. 16.12, “and they shall rejoice in the Kingdom of God, which is in Christ Jesus”
(p. 15), as a Beta addition in the apostolic constitutions and canons. The preceding Did. 16.11, “to inherit those things which eye hath not seen, nor ear heard, nor have entered into the heart of man, such things as God hath prepared for them that love him,” can be seen as drawing on 1 cor 2:9, but that passage is itself an explicit quote whose exact original is not to be found in the septuagint. Opinions about its source vary. James Moffatt, agreeing with Jerome, suggests a free rendition of isa 64:4. fitzmyer points to further possibilities, including an apocryphon of elijah (so Origen). it suffices for our purpose to say that Paul was quoting some text then current, whether freely or precisely, which may also have been available to the writer of this part of the Didache. it is not proved (but see n. 100 below) that the Didache is here derivative from 1 corinthians. With aldridge, then, but with the above qualification, the Didache originally ended in this way (alpha tenets emphasized): “Then the world will see the Lord coming upon the clouds of heaven, with the angels of his power, in the throne of his Kingdom, to condemn the Devil, the deceiver of the world, and to render to every one according to his deeds. Then shall the wicked go away into everlasting punishment, but the righteous shall enter eternal life, to inherit those things which eye hath not seen, nor ear heard, nor have entered into the heart of man, such things as God has prepared for them that love him” (16.6–11, trans. aldridge, “Lost ending of the Didache”). This would seem to be a fittingly sonorous conclusion to the entire document, in its final form.
34. for the sharp interchange between these two texts, see under “Luke” in §1.3
above.
The DiDache: a MissiNG Piece Of The PuzzLe
2.1.1. Barnabas
The Oxford committee finds that Barnabas quotes or echoes Jewish scriptures, including noncanonical ones,35 inexactly and sometimes in combination; the same applies to its use of canonical New Testament texts.
On that understanding, Barnabas has sometimes striking resemblances to Romans among the genuine Pauline epistles, and of the deutero-Pauline epistles, to ephesians, hebrews, and the Pastorals. The case for knowledge of 1 Peter is considerable, but not for 2 Peter. There are possible links to Matthew among the Gospels.36 This tends to locate Barnabas before 1 clement (ca. 96) and to identify it as aware of several deutero-Pauline writings. Then authorship by Paul’s associate, Mark’s cousin Barnabas, is chronologically improbable, and Barnabas must instead be thought to be borrowing the established name of Barnabas.
Barnabas ends naturally at Barn. 17 (“Let this, then, be enough”). The text is in effect reopened to include the Two Ways at 18.1 (“Let us now pass on to another kind of knowledge and instruction”). The seventeen-chapter text is attested in a ninth century Latin manuscript at st. Petersburg;37 the Two Ways addendum is in place in both sinaiticus (fourth century) and the Bryennios manuscript (copied in 1056). it seems clear that the Two Ways portion was added to an original seventeen-chapter Barnabas. The authorial stance of Barnabas should then be assessed from those seventeen chapters.
Barnabas is structured like a Pauline church letter and certainly has that model in mind. But despite its profession of warm regard, it has no discernible address and was apparently intended as a general letter. The issue for Barnabas is the status of Jewish tradition and its appropriation by the christians, and to this message it gives systematic attention. Whereas hebrews as it were seeks to symbolically reinhabit Jewish sacrificial tradition from the inside, Barnabas rejects and interiorizes sacrifices (ch. 2) and fasts (ch. 3). it claims the covenant for the christians (ch. 4), explains the atonement (chs. 5–8), and interiorizes circumcision (ch. 9) and food laws (ch. 10). it cites scriptural predictions of baptism and explains the 35. Kraft (Barnabas and the Didache, 182–84) also notes many Barnabas quotes from now unidentifiable Jewish writings.
36. Kraft (Barnabas and the Didache, 181–85) sees a quotation (?) of Matt 22:14 in Barn. 4.14b and a “strong allusion” to Matt 26:31 paralleled in Barn. 5.12b.
37. Kirsopp Lake, trans., The Apostolic Fathers, 2 vols, LcL 24–25 (cambridge: harvard university Press, 1912), 1:338.
BROOKs: BefORe aND afTeR MaTTheW
263
cross (chs. 11–12) and again claims the covenant, this time specifically for the gentiles (ch. 13). it allegorizes the sabbath (ch. 14) and rejects other month and day observances (ch. 15). it enjoins the building of a spiritual temple for the Lord (ch. 16)38 and concludes with a brief passage (ch. 17) saying that nothing has been omitted that is “necessary for salvation.” it is thus a complete presentation of the teachings that a gospel might include, but not structured as a life of Jesus. The same might be said of Romans, the only genuine Pauline letter of which Barnabas is unquestionably aware.
Doctrinally, Barnabas is a Beta document. its allegorical style of argument, like that of hebrews, is alexandrian.
2.1.2. The Barnabas Two Ways
The version of the Two Ways that appears as Barn. 18–20 includes one reference to the atonement (19.2, “glorify him who ransomed you in death,”
no parallel in the Didache version), presumably to adjust the alpha stance of the Two Ways to the Beta position of the rest of Barnabas.39 its list of good and bad deeds is very different in sequence from that of the Didache Two Ways. Which is earlier?
here are the two versions, with sequence numbers in each appended to the other:
Didache Two Ways 5.1
Barnabas Two Ways 20a–d
1. murders (φόνoι) (7)
1. idolatry (εἰδωλoλατρεία) (6)
2. adulteries (μoιχεῖαι) (6)
2. audacity (θρασύτης) (20)
38. Barnabas 16.4–5 “owing to the war it [the temple] was destroyed by the enemy” makes plain that Barnabas is post-70, which agrees with the evidence for relative date above cited. On the strength of 16.4b “at present even the servants of the enemy will build it up again” is taken by some to refer to the period under hadrian (ca.
120) when the possibility of actually rebuilding the Jerusalem temple under Roman auspices seemed to exist; Joseph B. Lightfoot (The Apostolic Fathers: Revised Texts with Short Introductions and English Translations, ed. J. R. harmer [London: MacMillan, 1912], 241) argues otherwise. The present tense of the text is probably signaled by the Dan 7:7–8 prophecy about ten kings subdued by one king (Barn. 4.4). One opinion sees the tenth king as Vespasian (ruled 69–79). Prophecy is perhaps by nature dark, to be made clear only in the event, and too much should not be built on these numbers or possible numbers.
39. a similar conclusion is reached in Jonathan a. Draper, “Barnabas and the Riddle of the Didache Revisited,” JSNT 58 (1995): 89–113.
The DiDache: a MissiNG Piece Of The PuzzLe
3. lusts (ἐπιθυμίαι)
3. pride of power (ὕψoς δυνάμεως) ( ~
21)