Introduction
Introduction The Didache represents the preserved oral tradition whereby mid-first-century house churches detailed the step-by-step transformation by which gentile converts were to be prepared for full active participation in their assemblies. As an oral tradition, the Didache encapsulated the lived practice by which non-Jews were initiated into the altered habits of perceiving, judging, and acting characteristic of one branch of the Jesus movement during the mid-first century.
This Didache reveals more about how Christians saw themselves and how they lived their everyday lives than any book in the Christian Scriptures. It is not a gospel and, accordingly, it does not attempt to offer guidance by narrating a life of Jesus. In fact, it is older than the canonical gospels and was written in the generation following the death of Jesus when the message of Jesus was not yet encapsulated in stories about Jesus. Nor is the Didache a letter like the writings of Paul. In fact, the Didache was created at the time of Paul's mission to the gentiles, but it shows not the slightest awareness of that mission or of the theology that undergirded it.
The Didache is an anonymous document. Like so many other early Christian books, it did not belong to or originate with a single individual. It belonged to various communities of householders who had received a Way of Life revealed to them by the Father through his servant Jesus. Given the manifest clues of orality within the Didache itself, one can be quite certain that it was originally composed orally and that it circulated on the lips of the members of this community for a good many years before any occasion arose that called for a scribe to prepare a textual version.
The Didache did not originally have any title. When it was used, everyone knew what it was and how it was to be applied. When the written copy did finally get a title it was called "The Training of the Lord Through the Twelve Apostles to the Gentiles." Scholars today have abbreviated this long title as Didache (usually pronounced "Did-ah-Kay")—the Greek word for the systematic training that a mentor (or a master craftsworker) would give to an understudy (apprentice). This was a remarkably fitting working title even though, as the commentary will show, it has not been adequately appreciated or understood.
The Didache represents the first concerted attempt by householders (Crossan 1998) to adapt the way of Jesus to the exigencies of family, occupation, home— the very things that Jesus and his wandering apostles had left behind (Theissen
1977). Paul did this for the communities he founded. The twelve apostles undoubtedly did this for the community at Jerusalem. From Paul, however, we have only occasional letters. From the Twelve we have nothing. The Acts of the Apostles gives only passing details regarding community life in the Jerusalem church and in the churches founded by Paul. The Didache, in contrast, offers a fullblown description of nearly every aspect of community life:
One overhears a candidate being trained from scratch by a mentor who becomes his beloved "father" or "mother." One witnesses the fasting and the solemn rite of baptism, preferably by immersion in flowing water. One overhears the daily prayers and the weekly eucharist—both of which are outlined in full detail.
One learns how visiting prophets were a blessing and a danger at the same time. One comes to understand how manual work, the sharing of resources, and the cultivation of gratitude worked together to provide a mainstay for individual wellbeing within a community. One learns how the confession of failings, the correction of backsliders, and the shunning of recalcitrant members worked to maintain the community's standards of excellence and to insure that their sacrifice was pure. Finally, one discovers how a community poised on the threshold of the Kingdom of God shared the same passionate expectation of God's future Jesus had preached to the Jewish peasants and fishermen of Galilee (Milavec, Didache 2003, ix).
Initially the ordering of the material in the Didache may seem ragged and confusing.
Upon careful examination, however, one can discover the organizational thread that accounts for the flow of topics and reveals the marvelous unity hidden below the surface from beginning to end. This is the same organizational thread that those who originally recited the Didache relied upon for ordering their recitation. The organizational thread is this: the Didache unfolds the comprehensive, step by step program used for the formation of a gentile convert. By following the order of the Didache, mentors training novices were assured of following the progressive, ordered, and psychologically sound path that master trainers had effectively culled from their own successful practice in apprenticing novices. From the vantage point of the novice the ordering of events within the Didache reveals how a candidate came to progressively enlarge those habits of judgment and ritualized experiences required for a full and active participation in the community. Needless to say, this organizational thread will not be entirely evident at first reading. With time, however, one will come away with the growing satisfaction of noticing how, at every point, nothing comes too early, nothing comes too late, everything comes in just its proper place.
Christians have come to regard the books of the Christian Scriptures (New Testament) as including all the authentic writings produced by the Jesus movement during the first century. Sad to say, this judgment is not correct. The Gospel of Thomas and the Q-Gospel represent two instances of first-century productions not found in the canon. Nor, for that matter, do the Christian
Scriptures say anything of the thriving community at Qumran or of the books in the fabulous two-thousand-year-old library nearby that was discovered in 1943 and has come to be known as the Dead Sea Scrolls. But think again of the Pauline letters. While Luke claims, in typical Hellenistic fashion, to have gathered primary sources (Luke 1:1-4), these sources apparently did not include even a single copy of any of Paul's letters, which, at the time of his writing, had been in circulation for over twenty years. The modern reader, consequently, need not be entirely dismayed that the Acts of the Apostles makes no reference to the Didache and that, even after being circulated and used regionally, the Didache failed to have enough clout to gain inclusion in the fourth-century universal canons of books (the New Testament).
The Modern Rediscovery of the Didache Books deteriorate and are destroyed with the ravages of time. Even books handprinted on prepared sheepskin seldom last for twenty generations. Thus it is not so surprising that the revival of classical learning known as the Renaisssance was unable to bring forward even a single known copy of the Didache. Then, to the surprise of everyone, a single complete copy of the Didache was found in 1873. As it happened, Archbishop Philotheos Bryennios was browsing in the library of the Greek Convent of the Holy Sepulchre in Istanbul when, by chance, he noticed the text of the Didache hidden away within a bound collection of early church writings. This discovery was all the more unexpected since various professional catalogers had already systematically combed through the codices on these shelves and had failed to notice the presence of the Didache.
Bryennios was forty at the time and had spent his early years teaching church history before being advanced to various administrative and pastoral posts in the Greek Orthodox Church. In fact, Bryennios had been appointed as bishop a year earlier and was actively committed to church reform. Thus, even with his increased responsibilities, he deliberately made time for his scholarship because he wished to actively foster "a piety built upon a clear understanding of the transformations that modern society requires of the ancient churches" (cited in Sabatier, La Didache 1). Given the wide range of interests and occupations calling for Bryennios' attention, another ten years passed before he fully recognized and finally published his extraordinary find.
Almost overnight, scholars in Europe, England, and America expressed their complete astonishment that such an ancient and important work had finally surfaced. When the first English translation prepared by Hitchcock and Brown was
released on 20 March 1884 in New York bookstores, five thousand copies were sold on the first day (Sabatier, La Didache 5). On the other hand, some scholars regarded Bryennios' find as too good to be true and rejected it as "a modern forgery" (see Hitchcock and Brown, The Teaching v). It was almost as though a document lost for nearly fifteen hundred years and overlooked repeatedly by scholars cataloging the Istanbul library was not allowed to show up so unexpectedly. After a few years, however, the judgment of authenticity prevailed and skeptics were silenced.
The Didache is approximately one-third the length of Mark's Gospel. It now resides in the library of the Orthodox Patriarch of Jerusalem where is has been catalogued as Codex Hierosolymitanus 54. The vocabulary and grammar are typical of popular Greek (koinē) used in the first century. "The style is simple, natural, terse, sententious, and popular" (P. Schaff, Oldest Church Manual 96). As to its vocabulary:
The Didache contains 2190 words. Its vocabulary comprises 552 words … 504 are New Testament words,
497 are classical, and 479 occur in the Septuagint. 15 [words] occur for the first time in the Didache, but are found in later writers. 1 [word, prosexomologein, found in 14:1] occurs only in the Didache [but its meaning can be easily surmised by combining known words] (ibid. 97).
The Unity of the Didache and Its Independence from Known Gospels In the course of the last fifteen years my ideas regarding the Didache have changed many times. During this period, however, two convictions have stood the test of time and have shaped my approach to interpreting the Didache:
1. Unity of the Didache. Up to this point a unified reading of the Didache has been impossible because the prevailing assumption has been that the Didache was created in stages, with the compiler splicing together pre-existing documents with only a minimum of editing. The end result, therefore, was a complex (or even a haphazard) collage that joined together bits and pieces of traditional material coming from unidentified communities and/or unknown authors. The conviction undergirding my commentary, however, is that the Didache has a marvelous unity from beginning to end that, up to this point, has gone unnoticed. Once it is revealed, however, the reader quickly discovers how seemingly incidental clues and abrupt jumps follow a marvelously constructed ordering of the whole.
Accordingly, this present study of the Didache will expend only passing energy on issues of source and redaction criticism and will concentrate on hearing the text as a whole and endeavoring to discern the organizational thread that guided the framers in the ordering of their material.
2. Independence of the Didache from the Gospels. The Didache has been widely understood as citing either Matthew's Gospel or some combination of the Matthean or Lucan traditions. From this vantage point it followed that the date of composition had to be set beyond the 80s and the synoptic material could be used to help interpret and understand the Didache. Thanks to my work with Willy
Rordorf, I came to an early appreciation of the possibility that the Didache might have been created without any dependence upon a known gospel. My extensive study of this issue (Milavec, Didache [2003], Chapter 11) demonstrates that the internal logic, theological orientation, and pastoral practice of the Didache run decisively counter to what one finds within the received gospels.
The repercussions of this conclusion are of decisive importance for the dating and interpretation of the Didache. If one supposes an early-second-century origin for the Didache, for example, one is naturally disposed to find points where the Didache shows dependence upon one or more known gospels then in circulation.
The widespread supposition of gospel dependence, therefore, has blocked most scholars from seriously entertaining the possibility of a mid-first-century origin of the Didache. The supposition of gospel dependence has also encouraged an "inappropriate" interpretation of the text. If one presumes, for example, that the Didache made use of Matthew's Gospel one could justifiably make use of Matthew's theology and church practice as sources for clarifying the intent and background of the Didache. On the other hand, if one supposes that the Didache is independent of Matthew, then it would be unwarranted to use that gospel to clarify obscure segments of a text created outside of its influence. My conclusion that the Didache was composed independently of any known gospel thus means that the gospels can provide studies in contrast and comparison, but they cannot be used to fill in the intent of the framers of the Didache. Within my commentary, consequently, great importance is placed upon allowing the internal evidence of the text to speak for itself free of the influence of what was believed and what was done elsewhere. The case of the Didache is thus comparable to that of the letter to the Hebrews. As soon as it was discovered that Paul was not the author it was likewise required that Hebrews be interpreted based upon its own internal logic and rhetoric, quite independent of the theology of the authentic Pauline letters.
Norms for Establishing a Working Greek Text The Greek manuscript discovered by Bryennios is well preserved, carefully written, and employs a score of tachygraphic signs or abbreviations in common use during the Middle Ages when it was copied. The scribe who made the copy identified himself as "Leon, scribe and sinner" and dates the completion of his
work as 1056.
J. Rendel Harris (The Teaching of the Apostles 1–10) produced a careful transcription of the manuscript in which he expanded the medieval tachygraphic signs. The Greek text found in this book follows the transcription made by Harris as corrected by my own careful examination of a reproduction of the original. In approximately two dozen places Harris made minor transcription errors (details in Milavec, Didache, "How the Greek Manuscript Was Discovered, Transcribed, and Translated"). Furthermore, the Greek manuscript itself contains minor errors: in three places colons were omitted (10:2, 4; 14:3) and in three places short words were omitted (9:4, 10:4, 11:5). These corrections have been widely accepted by current scholars. For the sake of fidelity to the original, however, these minor additions are presented in pointed brackets (< >) to indicate that they were added to the Greek text.
Going beyond this, a long line of scholars has tried to improve the intelligibility of the text by replacing words or phrases with near equivalents. Harris himself gave much attention to a number of substitutions brought forward by Harnack and Hilgenfeld. In the end, however, he preferred not to alter the original text on the following grounds:
Indeed it must be admitted that with slight exceptions the attempts to emend the text have not been very successful. The most difficult passages have yielded to interpretative skill … and this should assure us that any alterations in the text must not be more than moderate if they are to be in any degree acceptable
(Harris, The Teaching of the Apostles 14).
This, then, is the rule guiding this commentary: lectio difficilior potior ("The more difficult reading is preferable"). In my judgment the purpose of the scholar is best served by allowing the singularity of the text to stand out and refraining from harmonizing the text on the basis of what was written or done elsewhere.
Scholars such as Aelred Cody, Willy Rordorf, and Georg Schöllgen have followed this approach to the text very successfully. Jean-Paul Audet, Kurt Niederwimmer, and Klaus Wengst, on the other hand, allow corrections on the basis of comparison with the Latin or Coptic fragments or the Apostolic Constitutions. Wengst, in his
Greek reconstruction, introduces fifty alterations to the received text (see Bordewijn Dehandschutter, "The Text of the Didache"). In the end this creates a hypothetical hybrid that can never be known ever to have existed or been used.
Even seemingly moderate alterations are suspect. Niederwimmer, for instance, suggests that the unusual word klasma ("fragment") in Did. 9:3-4 ought to be replaced by the usual word artos ("loaf"). Erik Peterson ("Über einige Probleme,"
168–69), Arthur Vööbus (Liturgical Traditions 89, 146–48), and Klaus Wengst (Didache 97–98) agree with him. He explains himself as follows:
Verse 3 gives the second benediction, the blessing of the bread, introduced by the rubric peri de tou klasmatos ["And concerning the fragment"]. This suggests that we should understand klasma ["fragment"] to mean the bread broken at the meal celebration. In that case, the plural peri de tōn klasmatōn ["And concerning the fragments"] would seem more appropriate… . The parallels from later liturgies have artos in the analogous location. Peterson has pointed out that klasma is a technical term in the eucharistic language of Egypt; it refers to the particle of the host. The expression could then have entered the text at a secondary stage (Niederwimmer, Didache 148).
If the technical term klasma goes back to the Egyptian liturgy and accordingly hints at a late origin of the Didache (Charles Bigg, "Notes on the Didache" [1905]
414), one might be willing to drop the tainted "late word" in favor of the supposed earlier term, namely, artos. This substitution in Did. 9:3, however, would require that Did. 9:5 be altered such that "fragments of bread," or better, "grains of bread" would be scattered over the hills. It makes no sense to scatter a loaf
(artos). The substitution allowed by Niederwimmer (as well as by Wengst, Peterson, and Vööbus) thus resolves one difficulty by creating another one.
Going deeper, however, what if it could be shown that the term "fragment" goes back to an ancient Jewish usage whereby the loaf was broken prior to being blessed and thus only a "fragment" was held up as representing the whole? In that case klasma would offer a hint of the Hebraic idioms that might once have flourished in the archaic form of the eucharistic liturgy found in the Didache.
Hence by retaining the more difficult reading of the original one retains the necessity of finding a suitable reason for why this (klasma) and not that (artos).
Those who would substitute artos gain a "quick fix" in terms of intelligibility but lose the precious hint that the Didache may contain archaic idioms that were part of its foundation and that, with a little prying, might reveal untold mysteries of its true origins and pastoral genius. I have elected in every case, therefore, to stay with the original.
Norms for Preparing an Analytic, Gender-Inclusive Translation The Greek transcription and English translation on the following pages have been set out so as to facilitate the reading and analysis of the received text. The Greek manuscript prepared by Leon was presented on continuous lines with little or no space left to indicate the end of words, of sentences, and of sections. To make the Greek text and its English translation more accessible, the following pages group and organize the material in such a way that on first reading the internal logic begins to make itself evident. Thus one can identify units of thought, repetitious patterns, and key transitions that ordinarily only a close study of the
text would reveal. This, a technique I learned from Jacob Neusner, is what is meant here by an analytic translation.
The English translation produced here retains the literal Greek meaning without slavishly following the Greek word order and usage. Idioms in the Greek are rendered into English with only small adaptations in order to achieve a close dynamic equivalence without yielding stilted English. Words or phrases placed in brackets serve to clarify the elliptical intent of the Greek while acknowledging that they are absent from the literal Greek text. English words linked together by underlined spaces signal instances where a single Greek word needs to be rendered by a phrase in English. In sum, the English translation in this volume is conservatively constructed so that, in principle, it could be retranslated back into the original Greek.
Since the English language has lost its ability to differentiate between the singular and plural "you," this translation uses "ÿou/ÿour" to signal the plural.
Likewise, when the context fails to make it evident that a verb has a plural subject, an umlaut is used to signal the plural form. Thus, for example, "Präy for ÿour enemies" (Did. 1:3) allows the English reader to know that the Greek has a plural imperative ("Präy") matched by a plural pronoun ("ÿour").
The post-positive particle de is frequently used in Greek to signal a continuation of the foregoing topic. An English-speaking storyteller who links sentences with "and" or "and then" is effectively doing the same thing as a Greek speaker would do. Accordingly, when de appears in the Greek, the English translation will render this de as "(and)." Placing the "and" in parentheses at the beginning of the sentence indicates that current English style normally avoids it. Where kai ("and") is found in the Greek it is consistently rendered as "and" without parentheses. The post-positive particle de can sometimes have an adversative sense. In these cases it is rendered into English as "(but)" or "on the other hand." Where alla ("but") is found, it is consistently rendered as "but" without any parentheses.
Finally, the Greek language assigns gender to nouns and pronouns somewhat differently than does English. Nouns in English designating "things" are neuter, while nouns designating "living beings" are either masculine or feminine. Greek nouns designating things may be masculine, feminine, or neuter (as in most
European languages). When Greek verbs allow for either a male or female subject or a female or male object this will be appropriately signaled by using an inclusive English translation. Overall, the entire Didache is 99% inclusive, yet every modern English translation I have yet examined leaves the mistaken impression that "men" were addressing "men" about "manly things" throughout.
This translation will demonstrate that "women" were also addressing "women" about "womanly things." This commentary will thus rectify a gender distortion that has misrepresented the intent and use of the Didache.