Loading ...
Sorry, an error occurred while loading the content.

Answer to the Diatessaron query (long post)

Expand Messages
  • William L. Petersen
    ... An excellent question, Rod. I ll try to answer you, but first a couple preliminaries to set the stage, and make sure some distinctions and phenomena I ll
    Message 1 of 1 , Jul 16, 1997
    • 0 Attachment
      Rod Mullen asked:

      >The recent diatessaron discussion has reminded me of a question that I had
      >upon first reading Wm Petersen's book on "Tatian's Diatessaron" (Leiden:
      >Brill, 1994). Bill, please permit the question of a novice in matters
      >diatessaronic. On pp.374-75 you discuss Quispel's second criterion for
      >determining the likelihood of a proposed diatessaronic reading being
      >original. If we apply this criterion strictly in reconstructing readings,
      >would it not mean that the value of the diatessaron as an independent
      >witness to the text of the canonical gospels is highly problematic at best,
      >even though it is often cited in the apparatus to UBS4 and NA27? Perhaps we
      >would have attestation of a stage in the gospel tradition, but it seems that
      >hope of reaching an ur-text of the gospels with the help of the diatessaron
      >would elude us. How would you assess this relationship? Interested to hear
      >your reply, thanks-- Rod Mullen

      An excellent question, Rod. I'll try to answer you, but first a couple
      preliminaries to set the stage, and make sure some distinctions and
      phenomena I'll speak of later are clear.

      (1) When Tatian composed the Diatessaron (circa 172), he obviously had to
      use the gospels in the form they then had. He may also have
      incorporated/used/relied upon the synoptic harmony apparently used by his
      teacher, Justin Martyr. If so, that does not change the situation, for
      *Justin's* harmony also had to have been created from the gospels as *he*
      (or whoever created the harmony) knew them, circa 140-160.

      (2) As a broad, general statement, scholarship (regardless of whether on the
      Diatessaron or on a version) has found that (gospel) texts which deviate
      from the *then* "standard" gospel text are bowdlerized (in Diatessaronic
      studies, we say "vulgatized") by having the deviating reading excised and
      replaced with the "standard" reading of the day.

      An example of this phenomenon, independent of the Diatessaron, is in the
      Syriac versions, and alluded to in my review of George Kiraz's parallel line
      edition of the Vetus Syra (Syr-sin, Syr-cur), the Peshitta, and the
      Harclean, just posted in TC. One sees how the earliest Syriac (Syr-sin)
      deviates the furthest from the TR, while each successive step (Syr-cur,
      Pesh, Harc) moves closer and closer to the TR type of text. The process
      reaches its apotheosis in the Harclean, which attempts a literal,
      word-for-word translation of the Greek, even to the point of violating the
      rules of Syriac grammar and syntax. It's text represents (with a wink to
      Harnack) "the acute Hellenization" of the ancient, pre-Pauline, Semitic church.

      The parallel situation exists with the Vetus Latina, where the earlier MSS
      are "wilder" (deviate further from the TR or NA or UBS--whatever base you
      want), and the later MSS move closer to the Greek. And then comes the
      Vulgate, which is a systematic attempt to replace the Vetus Latina with a
      text more in tune with the Greek used elsewhere.

      Within the Diatessaronic tradition, Codex Fuldensis (Vulgate MS F; date 546
      CE) is a magnificent example of "vulgatization," for it can be empirically
      demonstrated from the MS itself that *in the process of copying, the old,
      deviating Diatessaronic readings were excised, and replaced with the
      standard Vulgate reading*. This has been known for over a century (it was
      first noted by Th. Zahn: see my *Tatian's Diatessaron*, pp. 85-86; esp.
      126-129). The telltale signs are the "capitularia" (the "table of
      contents") and the manuscript's *corpus*: while the readings and sequence
      of the *corpus* are Vulgate, the readings and sequence of the "capitularia"
      agree with other Diatessaronic witnesses, and deviate from the Vulgate.
      Conclusion: in the process of "purifying" the text of Codex Fuldensis, only
      the *corpus* was "vulgatized"--through oversight, laziness, or chance, the
      "capitularia" were left untouched. [Comment: This is empirical evidence
      that scribes were "up-dating" and "modernizing" the text of the gospels to
      the standards of their day, and discarding older readings.]

      (3) From a text-critical standpoint, the interesting phenomenon is the
      "distance" from the TR/UBS/NA text, which is found in the Vetus Syra *and
      also* in the Vetus Latina. Since both of these are "versions" and, thus,
      translated from a Greek base, we may presume that they reflect a Greek base
      which *also* had these same deviating readings. This, of couse, suggests
      that the early Greek gospels--upon which these versions were based--were
      different from those current in the fourth and later centuries. (Note:
      Such a suggestion is also supported by the deviating readings found in the
      earliest fathers [Justin, Clem. Al, Origen, Irenaeus, Ephrem, Aphrahat,
      Novatian, Cyprian, etc.], esp. when they *agree* with the Vetus Syra or the
      Vetus Latina and against the TR/NA/UBS or other modern collation base.)

      So far we are hypothesizing, which requires nothing more than imagination.
      One leaves the realm of hypothesis, however, when one finds the *identical*
      deviating reading in *both* the Vetus Latina *and* in the Vetus Syra. The
      only way scholars have been able to explain this is to point to the early
      Greek gospels--which were the base of *both* versions--as having at one time
      contained such a deviating reading. And when an early Father or two (such
      as Justin and Clement) *also* has the same deviating reading *in his Greek
      text*, then we are on pretty firm ground: such a deviating (= non-standard
      reading by *later* points of reference, such as the TR, UBS, NA,
      whatever...) reading must have circulated in the early Christian centuries,
      in a *Greek* gospel. [Comment: This is empirical evidence that the gospel
      text of the second, third and fourth centuries did *not* agree with the
      later gospel text(s), at least in some passages.]

      With this as prologue, we may now turn to Rod's question directly, and
      address the Diatessaron:

      (1) The Diatessaron is regarded as a version of the gospels, as is shown by
      its treatment as such in the handbooks of Voeoebus and Metzger. In fact,
      all handbooks on either the text of the NT or the versions treat the
      Diatessaron, and usually (as Voeoebus and Metzger do) as the *oldest* of
      the versions.

      (2) Since the "autograph" Diatessaron is lost, its text must be
      reconstructed from the various "witnesses" to its text, that is, from
      translations (the Arabic Harmony, the Venetian Harmony, the Tuscan Harmony,
      various Latin Harmonies [Codices Fuldensis, Sangallensis, etc., etc.; more
      than 20 MSS are known], the Middle Dutch family of harmonies [the Liege
      Harmony being the most famous, although the Cambridge, Stuttgart, Haaren,
      and The Hague harmonies are also important], the Middle and Old High German
      Harmonies [the Zurich, Sankt Gallen, and Kassel MSS being important], the
      Middle English Pepysian Harmony, and a miscellany of other MSS in other
      languages), quotations (in Ephrem, Aphrahat, Romanos, Augustine, the Liber
      Graduum, the Acts of Archaleus, the Acts of Thomas, etc., etc., etc.), and
      from texts influenced by the Diatessaron (the Vetus Syra, Vetus Latina, the
      Arabic, Armenian, and Georgian versions). This is only a partial list;
      some of these are better sources, others poorer. Experts know the
      difference on the basis of long and intimate work with them; I provide
      pointers and caveats in *Tatian's Diatessaron.*

      It must be noted that *each* of these sources has undergone "vulgatization."
      In some cases (such as in the case of Codex Fuldensis) it was quite
      thorough; in other cases (as in the case of the Middle Dutch Liege Harmony)
      it was much less thorough.

      Here, Rod, is now the crux of your question and the answer:

      (3) Much of even the original ("autograph") Diatessaron agreed verbatim with
      our current UBS/NA text. "Blessed are the poor" seems to have stood in all
      the versions, in virtually all of the texts. (See p. 369 in *Tatian's

      The *problem* for the Diatessaronic scholar, however, is: Given that all
      Diatessaronic texts have undergone vulgatization, which sought to erase
      deviating readings, and substitute the "standard" Vulgate reading, *how can
      we be sure we have isolated the text of the ancient Diatessaron, and are not
      mistaking a later, vulgatized passage for the text of the original Diatessaron?*

      It follows logically that, in those cases where the original Diatessaron's
      text agreed with our present collation base (TR, UBS, NA, whatever), we
      *cannot* be sure that we have the text of the Diatessaron, for *all* of the
      Diatessaronic witnesses might have been "vulgatized" at this point. Note
      carefully: It is possible that *none* of the Diatessaronic witnesses have
      been "vulgatized," and that they *are* preserving accurately the 2nd century
      text of the Diatessaron (and possibly the 2nd century gospels upon which it
      was based); the problem is that we cannot be *sure* of that. Because of
      our logical inability to distinguish between a genuine Diatessaronic reading
      and a "vulgatized" reading *in and only in cases such as this,* where the
      Diatessaronic witnesses text agrees with the "standard" text of today, we
      must remain silent about the contribution of the Diatessaron at these
      points. Some might say that the Diatessaron agrees with the "aleph-B" text,
      or with the TR--but experts will know and recognize the worthlessness of
      that statement, for it is based on defective logic.

      (4) "Because of this, the text of the Diatessaron can be recovered *with
      certainty* *only* when it *deviates* from the vast majority of gospel
      manuscripts" (i.e., the TR, aleph-B, etc.) (*Tatian's Diatessaron*, p. 369)

      What this means is that we must find places in the Diatessaronic witnesses
      where they have *not* been vulgatized--where the "standard" reading has
      *not* been substituted for the genuine Diatessaronic reading. And the laws
      of logic dictate that that can only be done where we find a "non-standard"
      (= "deviating") reading.

      Finding a deviating reading in one Diatessaronic witness means little (in my
      opinion: Baarda differs, and I agree with him, with qualifications, in
      specific instances: see the discussion on pp. 375-377 in *Tatian's
      Diatessaron*), for each witness has its own transmission history, with
      scribal errors, and its own distinct language, with its own syntax, idioms,
      and grammar. BUT, at those points--and they stick out to the trained eye
      like a sore thumb--where *multiple* Diatessaronic witnesses *all* deviate in
      the *same* precise manner, *then* we have the distinct possibility of having
      found a reading which dates back to the early Diatessaron, probably to the
      second century. In my own research, I always insist (unless there are clear
      reasons for transgressing this rule) that the Diatessaronic witnesses with
      the identical reading be split between "eastern" witnesses and "western"
      witnesses: that is, for example, the Middle Dutch Liege Harmony must agree
      with Ephrem's Syriac Commentary on the Diatessaron.

      The reason I insist upon "bilateral" support (as I call it)--east and west
      agreeing--is to guard against the influence of "local texts." Example: If
      the Old High German Codex Sangallensis agreed with Latin Codex Fuldensis, I
      would be cautious about calling that a Diatessaronic reading. Both MSS were
      copied in the West, in fairly close geographic proximity. Perhaps this
      reading was generated by an early medieval tradition floating around in
      southern Europe during the sixth to ninth centuries. In that case, the
      reading has nothing to do with the Diatessaron.

      Similarly--and here we come to the "second criteria" of your question, which
      is one I, not Quispel, developed (Quispel developed the first criterion,
      that is, requiring eastern and western support [read p. 374 carefully])--the
      deviating reading should *not* be prominent in texts *outside* of the
      Diatessaronic family of texts. This criterion is *specifically* designed to
      help us *solely* in identifying genuine *Diatessaronic* readings. How
      readings which pass these criteria relate to the gospels must be decided
      later. In my experience, the readings which pass muster *sometimes* seem to
      be ancient gospel readings, commended by various internal and external
      factors. (Remember, if we *have* identified a genuine Diatessaronic
      reading, then we are holding in our hands a text of which it can be said--on
      an empirical basis--that it circulated in the final decades of the second
      century....) The question remains whether the deviating reading is a
      "Tatianism"--that is, a reading introduced into the Diatessaron by Tatian
      (contrary to the view of a century ago [e.g., v. Soden], subsequent research
      suggests Tatain contributed little to the Diatessaron [the *only* obvious
      case would seem to be some Encratite readings]; see *Tatian's Diatessaron*,
      pp. 79-82)--or whether it accurately reflects the second-century gospel text
      used by Tatian when creating his Diatessaron.

      The reason for requiring limitation to the Diatessaronic family of texts is
      meant to guard against the possible influence of local texts (it serves as a
      double-check on and back-up to the first criterion, which requires eastern
      and western support). Example: A deviating reading is widespread in the
      Vetus Latina, but absent from the Vulgate, aleph-B, and the TR; yet the
      same deviating reading shows up in both eastern and western Diatessaronic
      witnesses. What does one do? I toss it out, with my very high standard of
      proof. Why? After all, it has bilateral support, doesn't it? The reason
      why I discard it is simple: *In the WEST* one cannot be sure that the
      Diatessaronic witnesses obtained the reading from the *Diatessaron*, for
      they *might* have obtained it from the Vetus Latina. In that case, the
      bilateral support is questionable, as is the Diatessaronic status of the

      (5) The key to your question, Rod, is to remember that (a) the criteria (1,
      2, 3 in the book) are intended to insure that we get to the text of the
      *Diatessaron,* and *not* get hoodwinked by a "vulgatized" reading, or by a
      "local text" in either the east or the west; and (b) *if*, with these
      stringent criteria, we can recover a genuinely Diatessaronic reading, *then*
      we have a versional text from the second century. At that point, depending
      on one's tastes, one can assume that the reading is either from the pen of
      Tatian himself, or that it is--aah!--the text of a gospel known to Tatian
      and accurately appropriated by him at that early date.

      Since, Rod, I assume you are working your way through the book, it may be
      helpful to look back at the two examples given on pp. 14-20, which address
      this very point; also take a look at "F. The Character of the Diatessaron's
      Text" on pp. 437-438 (ahead of where you are now), which also addresses
      exactly this point, but in prose, and with categories (there are 4 types of
      readings in the Diatessaron: (a) textual trivia [probably from its own
      transmission history and the influence of "local texts"], (b)
      glosses/clarifications (perhaps from Tatian or later revisors], (c)
      liturgical /theological changes [perhaps from Tatian or later revisors], and
      (d) ancient gospel readings) and "H. Examination of Tatian's Methods and
      Motives" on pp. 443-444 (which deals with "Tatianisms" and changes perhaps
      due to Tatian). The four types of readings in the Diatessaron (a-d above)
      can be disentangled and categorized, and while all are (because of their
      antiquity and link with Semitic Christianity) highly significant, the NT
      textual critic will probably be most intersted in "d," "ancient gospel

      For the uninitiated, and to illustrate one of these possible "ancient gospel
      readings," an example:

      In the East:

      Ephrem's Commentary on the Diatessaron (4th cent.; composed in Syriac, here
      Arm. [Syr. *hiat*]), Isho'dad of Merv's Commentary (ninth cent.; Syriac),
      Romanos the Melodist (6th cent.; Greek, but knew Syrian traditions as he was
      born in Syria, grew up there, and was bilingual), and

      in the West:

      the Middle English Pepysian Harmony (1400) and the Latin *Vita Rhythmica*
      (c. 1220)

      both report that a "great light" shown in the Jordan River when Jesus was

      This variant is completely absent from the Greek gospel MS tradition as it
      has come down to us. In the Latin, it is found in two MSS of the Vetus
      Latina, MSS *a* (4th cent) and *g1* (6th cent.). It is *possible* that our
      western Diatessaronic witnesses were influenced by the Vetus Latina MSS, and
      do not genuinely reflect the tradition of the Diatessaron. However, we know
      that the Vetus Latina was also influenced by the Diatessaron (which has been
      thought by some scholars [Baumstark and others] to be the first gospel in
      Latin); therefore, it is possible that the reading in the West *does* stem
      from the Diatessaron. Since, however, the only other sources in the West to
      preserve this reading are gospel harmonies (= the Pepysian Harmony) or works
      influenced by the Diatessaron (= the *Vita Rhythmica*), it tips the scales
      in the direction of the Diatessaron. Further, in the East, Isho'dad
      specifically states "And straightway, *as the Diatessaron testifies*, a
      *powerful light* shone, and above the Jordan...", and Ephrem's commentary is
      specifically written on the text of the Diatessaron. Hence, this is
      apparently a Diatessaronic reading.

      The question remains: Did it stand in an ancient (2nd cent) Greek gospel
      text? I suggest it did, for Justin also quotes the passage (Dial. 88.3).
      It is also cited by the Sibylline Oracles, 7.81-84. How does the
      Diatessaron's reading help us here? Three ways: (1) It provides us with a
      *third* fix on this reading in the second century. (2) It prevents one from
      writing off the *Sib. Or.* reading as "poetic licence" and Justin's citation
      as a "memory lapse." (3) Since, upon examination, the Diatessaron appears
      to have been a rather straight-foward harmonization of the text of the
      gospels as known in the 2nd cent., it suggests that this reading was part of
      the gospel text as known in the 2nd cent.--as does the presence of the same
      reading in the two Vetus Latina MSS, which are "canonical" gospels, and
      which insert the passage *ante* Matt 3:16.

      [For full disclosure: later, Epiphanius (late 4th cent.) states that the
      reading stood in "the Hebrew gospel" used by the Ebionites (Pan. 30.13);
      elsewhere, however, he states that "the Diatessaron gospel" was called by
      some "[the gospel] according to the Hebrews" (Pan. 46.1.8-9 ). Further,
      Jerome remarks that a "Gospel according to the Hebrews" was considered by
      "many" to be the "original/autograph" gospel of Matthew--and it is in
      Matthew that Vetus Latina MSS *a* and *g1* place this reading...]

      The value of the Diatessaron's reconstructed text is that it is ancient
      (clearly 2nd cent.) and that, in this instance, it corroborates other
      ancient evidence which might otherwise be ignored or dismissed becuase it is
      not clearly a "text" (Justin's is a "citation/allusion/memory
      lapse/paraphrase/oral tradition he knew," etc.; the Sib. Or. is "poetic/not
      a gospel," etc.). But the Diatessaron is based on the gospels as they
      circulated in the 2nd cent., as multiple examples show.

      For another example (discovered by Baarda and published in his "The Flying
      Jesus: Luke 4:29-30 in the Syriac Diatessaron", in VigChr 40 [1986], pp.
      313-34), see my review in TC 1 of Baarda's *Essays on the Diatessaron*,
      which reprints the article. I summarize it there. Examine his arguments
      and evidence, and I think you'll see once again the immense value of the
      Diatessaron for reconstructing the earliest strata of the gospel tradition.

      Hope that helps, Rod.

      --Petersen, Penn State University.
    Your message has been successfully submitted and would be delivered to recipients shortly.