www.ibresearch.org

Bulletin for Biblical Research (BBR) 1999

The State of the New Testament

Canon in the Second Century

Putting Tatian's Diatessaron

in Perspective

CRAIG D. ALLERT

TRINITY WESTERN UNIVERSITY

LANGLEY, BRITISH COLUMBIA

In contemporary discussions of the NT canon, focus has been on its polem-

ical aspects, that is, when it was closed. By so doing the idea of a canonical

process suffers. In attempting to understand Tatian's Diatessaron in this

process it is argued here that the very existence of the harmony testifies

against a closed fourfold Gospel canon in the mid–second century. A proper

distinction between canon and scripture is foundational in this under-

standing. Discussions about the closed NT canon belong to a day far re-

moved from Tatian's. By placing Tatian's Diatessaron in the perspective of

process we are less tempted to view his use of the four Gospels as proving

their canonicity, a view which is anachronistic and inaccurate.

Key Words: Diatessaron, harmony, Gospels, sources, canon, scripture

W. L. Peterson begins his informative essay on the Diatessaron with

four reasons why the Diatessaron is important.¹ First, the Diatessaron

is the most extensive, earliest collection of second-century Gospel

texts extant. It is much more comprehensive than the other scattered

references of the second century because it incorporated virtually

the entire text of the Gospels of Matthew, Mark, Luke, and John, as

well as material from extracanonical Gospels. Second, it is the earli-

est example of a Gospel harmony yet recovered in extenso. Therefore

it allows us to examine the techniques and concerns of a second-

century harmonist. Third, the Diatessaron reflects the theology and

praxis of its locale. Fourth, it is usually considered the most ancient

of the versions, most likely being the form in which the Gospels first

appeared in Syriac, Latin, Armenian, and Georgian. Because of this

1. W. L. Peterson, "Tatian's Diatessaron," in Ancient Christian Gospels: Their History

and Development (ed. H. Koester; London: SCM / Philadelphia: Trinity, 1990) 403.

2 Bulletin for Biblical Research 9

it occupies a unique position in the history of the dissemination of

the Gospels, for it served as the foundation of four of the major NT

versions, each of which bears the Diatessaron's imprint.

Since the recovery of the Dura Fragment,² study of the Diatessaron

has blossomed and expanded into a field all its own. I do not propose

in this essay to add anything new to the technical study of the Dia-

tessaron—that is, the original language and provenance, examination

of the witnesses, and so on. My purpose is simply to place Tatian's

harmony in the context of the second century's understanding of

Christian writings. In other words I ask, "How does the composition

and circulation of the Diatessaron help us understand the state of the

NT canon in the second century?" From that viewpoint my purpose

best fits under the second and third points in Peterson's list above.

It is a well-known peculiarity that we possess no direct copy of

the Diatessaron. The text, therefore, must be reconstructed from a

number of witnesses.³ In reality, then, we have only one direct copy

of Tatian's works extant, his Oratio ad Graecos. That Tatian did write

other works is not disputed;⁴ unfortunately, these works have not

survived. The main focus of this study on the Diatessaron, rather than

on the Oratio, is necessitated by the fact that the Oratio is valuable

only in a limited sense in regard to Tatian's use and attitude toward

Christian writings.

I. ORATIO AD GRAECOS

Of course, Tatian's Oratio is valuable for putting together a brief

account of his own conversion to Christianity. He sought to discover

truth, and while he was devoting himself to that endeavor, through

Greek philosophy, he happened upon certain "barbaric writings"⁵

which were too old and too divine to be compared with Greek

2. The Dura Fragment was discovered at Dura-Europos in Syria in 1933. It mea-

sures about four inches square and contains fourteen lines of legible Greek from the

passion narrative. Its date is around 254-57 AD. The text is harmonized and establishes

an extremely early date for the circulation of the Diatessaron. See F. C. Burkitt, "The

Dura Fragment of Tatian," JTS 36 (1935) 255-59.

3. For a discussion of the various witness to the text of the Diatessaron, see B. M.

Metzger, The Early Versions of the New Testament: Their Origin, Transmission, and Limi-

tations (Oxford: Clarendon, 1977) 10-25; Peterson, "Tatian's Diatessaron," 408-19.

4. Eusebius (Hist. eccl. 4.29) states that Tatian left a great many writings but does

not name them save the Orat. In Hist. eccl. 5.13 Eusebius explains that Rhodos credits

Tatian with writing A Book on Problems in which he promised to explain the obscure

and hidden parts of scripture. Tatian himself (Orat. 15) states that he wrote "a treatise

concerning animals."

5. Orat. 29.1, grafai=j tisin e)ntuxei=n barbarikai=j. Quotations from Oratio. are from

E. J. Goodspeed, Die ältesten Apologeten: Texte mit kurzen Einleitungen (Göttingen: Van-

denhoeck & Ruprecht, 1915). English translations are my own.

ALLERT: Putting Taitan’s Diatessaron in Perspective 3

philosophy. These barbaric writings were the prophetic word of God.⁶

Not only was Tatian drawn to these writings because of their old age

and divinity, but especially because their divinity was expressed in

the foreknowledge of future events.⁷ But unlike his teacher Justin

Martyr, Tatian does not employ the argument from prophecy using

Christian writings as historical proof for the occurrence of the proph-

ecies.⁸ Tatian explains that he does not use "witnesses from our-

selves,"⁹ that is, Christian writings, because the Greeks do not accept

them. What writings Tatian would have used we do not know. We

may conjecture that Tatian perhaps thought of the Christian writings

in a different sense than Justin, but this would, indeed, only be con-

jecture. The best conclusion is to consider Tatian's audience in his lack

of reference to Christian writings to prove the prophetic predictions.

Because the audience was Greek, Tatian felt that the use of Christian

writings to prove his points would be counterproductive,¹⁰ even though

he calls the Greek writings mythical accounts when compared with

"our narrations."¹¹

The Diatessaron, on the other hand, is relatively valuable in point-

ing out, not only Tatian's probable attitude toward written Gospels,

but also the attitude of the church at large during that time, espe-

cially the Syrian church, which accepted the Diatessaron as its au-

thoritative Gospel into the fifth and sixth centuries.

The existence of the Diatessaron is attested by a number of early

Christian writers. The way these attestations are understood is seen

to contribute to our understanding of the attitude toward the Diates-

saron from the time it was compiled and into the fifth and sixth cen-

turies. Specifically, attention has been paid to the heresy of Tatian,

and perhaps more importantly, to the sources Tatian used in the

Diatessaron's composition.

6. Orat. 20.2-3; 29; 30.1; 40.

7. Orat. 20.2-3; 29.1.

8. On Justin Martyr's use of the argument from prophecy see, e.g., D. E. Aune,

"Justin Martyr's Use of the Old Testament," Bulletin of the Evangelical Theological Society

(1966) 179-97; L. W. Barnard, Justin Martyr: His Life and Thought (Cambridge: Cam-

bridge University Press, 1967); idem, "The Old Testament and Judaism in the Writings

of Justin Martyr," VT 14 (1964) 395– 406; T. W. Manson, "The Proof from Prophecy,"

JTS 46 (1945) 129-30; H. P. Schneider, "Some Reflections on the Dialogue of Justin Mar-

tyr with Trypho," SJT 15 (1962) 164-75; M. Wiles, "The Old Testament in Controversy

with the Jews," SJT 8 (1955) 113-26.

9. Orat. 31.1, ma/rturaj de_ ou) tou_j oi!koi paralh/yomai.

10. Tatian does include several allusions to Christian teaching which may be in a

written form and may even be NT documents. But none of these allusions contains an

introductory formula or statement about whether it is the words of Jesus or was found

in some memoirs or something else. See Orat. 4.1; 13.1; 19.4; 30.1.

11. Orat. 21.1.

4 Bulletin for Biblical Research 9

II. THE HERESY OF TATIAN

Most of the attention that was exerted toward Tatian in the first few

centuries after the composition of the Diatessaron was not because of

his harmony. That Tatian lapsed into heresy¹² is generally the focus

of the second-century heresiologists who mention the harmonist.

Irenaeus never mentions Tatian's Diatessaron but is sure to discuss his

heresy.¹³ From the outset Tatian is not cast in a very good light. The

Encratites (self-controlled), who are led by Tatian, are said to spring

from Saturninus and Marcion. They preach against marriage, thus

setting aside the original creation of God and blaming God for cre-

ating humankind as male and female for procreation. Irenaeus also

criticizes the Encratites for abstaining from meat because this proves

them ungrateful to God who created the animals for our benefit. At

the top of Irenaeus's list, however, appears to be the belief that

Tatian denies the salvation of Adam. In book 3 of his work Against

Heresies,¹⁴ Irenaeus expands on this particular criticism. He finds

fault in the doctrine because in disallowing Adam's salvation Tatian

and the Encratites do not believe that the sheep which had perished

had been found, an obvious allusion to Luke 15:4. Irenaeus reasons

that if this sheep (Adam) has not been found, then the whole human

race is lost in a state of eternal damnation. Irenaeus's verdict on

Tatian is clear, "Thus also do those who disallow Adam's salvation

gain nothing, except this, that they render themselves heretics and

apostates from the truth, and show themselves patrons of the ser-

pent and of death."¹⁵

Briefly mentioned by Irenaeus is the system of AEons invented by

Tatian which was similar to Valentinus. This is confirmed in Hippo-

lytus who expands the explanation by stating Tatian's teaching that

the world had been created by these AEons.¹⁶ Hippolytus also points

out that he moved away from the beliefs of his teacher Justin in the

area of marriage and the salvation of Adam.¹⁷ Clement of Alexandria

mentions only the teaching of Tatian concerning marriage. In dis-

agreeing with his beliefs Clement sets out to prove the sanctity of

marriage in God's eyes.¹⁸ Finally, Eusebius mentions one writer,

12. Whether this lapse was before or after the death of his teacher Justin Martyr

is irrelevant for our purposes. On this question see, e.g., L. W. Barnard, "The Heresy

of Tatian—Once Again," JEH 19 (1968) 1-10; R. M. Grant, "The Heresy of Tatian," JTS

n.s. 5 (1954) 62-68.

13. Irenaeus Haer. 1.28.1.

14. Irenaeus Haer. 3.23.8.

15. Ibid.

16. Hippolytus Haer. 10.14.

17. Hippolytus Haer. 8.9.

18. Clement of Alexandria Str. 13.

ALLERT: Putting Taitan’s Diatessaron in Perspective 5

Musanus, who wrote a discourse (extant to Eusebius but since lost)

against the Encratites, a group "which was at that time just beginning

to sprout and to introduce into life its strange and corrupting false

doctrine. The story goes that Tatian was the author of this error."¹⁹ In

explaining this false doctrine Eusebius simply quotes Irenaeus from

the passages outlined above.²⁰

The beliefs of the Encratites include: the rejection of marriage

as adultery; abstaining from eating meat; denial of the salvation of

Adam. Also among the beliefs of the Encratites is the rejection of the

drinking of wine. In fact, the Encratites even went so far as to sub-

stitute water for wine in the Eucharist service.²¹

If one were to examine the two extant works of Tatian for a com-

plete, or even partial, outline of his heresy one would be hard pressed.

The fact is that neither his Oratio nor his Diatessaron betrays any bla-

tant heretical ideas. Rather than having heretical leanings, Tatian's

Oratio is similar in argument and style to his teacher's Apologies and,

indeed, with other second-century apologies. On the other hand,

Tatian does offer some minor "improvements" in his harmony of the

Gospels which betray a certain leaning in accordance with his more

ascetic views.²² Some examples include the following:

a. The generally accepted Greek text of Matt 1:19 referring to Joseph

as Mary's husband is changed by omitting the definite article and

possessive pronoun and by taking a)nh/r in a general and not a

marital sense meaning, "Joseph, because he was a just man."²³

b. Some witnesses to the Diatessaron reduce the length of time of the

marriage of the prophetess Anna (Luke 2:36) from seven years to

seven days.²⁴

c. The declaration in Matt 19:5, "For this reason a man shall leave his

Father and Mother, and shall cleave to his wife, and the two shall

19. Eusebius Hist. eccl. 4.28-4.29.1, a!rti to/te fu/ein a)rxome/nhn ce/nhn te kai_

fqorimai/an yeudodoci/an ei)sa/gousan tw|= bi/w|: h{j parektroph=j a)rxhgo_n katasth=nai

Tatiano_n lo/goj e!xei. Quotations from Eusebius Hist eccl. are from the Loeb Classic Library

edition. Volume 1 (Books 1-5): K. Lake (ed.) (Cambridge: Harvard University Press / London:

Heinemann, 1975). Volume 2 (Books 6-10): J. E. L. Oulton and H. J. Lawlor (eds.) (Cambridge:

Harvard University Press / London: Heinemann, 1973).

20. Eusebius Hist. eccl. 4.29.1-6.

21. B. M. Metzger, The Canon of the New Testament: Its Origin, Development, and

Significance (Oxford: Clarendon, 1987) 116.

22. J. F. Kelly, Why Is There a New Testament? (London: Chapman, 1986) 146;

Metzger, Early Versions of the New Testament, 34-36.

23. This reading is found in Ephraem and in the Persian and Venetian harmonies.

24. This reading is present in the Commentary of Ephraem and the Sinaitic Syriac

manuscripts. The Persian harmony does not reduce the length of time of the marriage

but claims a state of celibacy for the couple.

6 Bulletin for Biblical Research 9

become one flesh," (NASB) is put in the mouth of Adam rather

than God. In effect, this changes God's intention of marriage to

being only a spiritual union between man and wife, while the

physical union is more an invention by Adam.²⁵

d. Instead of Jesus' claiming, "I am the true vine" (John 15:1), the

Persian Diatessaron has Jesus claiming, "I am the tree of the fruit

of Truth."

e. The allegation of Jesus' being a glutton and a drunkard (Matt

11:19) is absent, as is the statement "when men have drunk freely"

(John 2:10) from the account of the miracle at Cana.²⁶

f. Instead of Jesus' being offered wine mixed with gall in the account

of the crucifixion (Matt 27:34) Jesus is said to have been given vin-

egar mixed with gall.²⁷

g. Some witnesses change John the Baptist's diet (Mark 1:6 and par-

allels) from "locusts and wild honey" to "milk and honey."

These minor alterations and subtractions are by no means indic-

ative of any of the heresies that the above antiheretical writers ad-

dressed. Certainly no one would suggest that the Diatessaron is latent

with such heresy. I affirm, with Westcott that, "the heretical charac-

ter of the book was not evident upon the surface of it."²⁸ Important,

however, in light of these alterations and subtractions is the fact that

Tatian was not beyond changing things in the Gospels which he per-

haps found at odds with his particular understanding of Christianity.

III. THE SOURCES FOR THE DIATESSARON

The more important question for our purposes is the number of

sources which Tatian used in compiling his harmony. No one denies

that Tatian used the Gospels of Matthew, Mark, Luke, and John. The

question surrounding the sources is did he use any more than simply

these four? Of course there are other questions that flow from the

initial question (how many other sources? what were they? were they

written or oral?), but the main question is prominent here. The dis-

cussion of the sources centers on two issues: the attestation of the

Diatessaron, and the text. I will examine each in that order.

25. This reading is found in several of the Medieval harmonies (Liege, Stuttgart,

Gravenhage, and Theodiscum).

26. This reading is found in Ephraem.

27. This reading is found in Ephraem.

28. B. F. Westcott, A General Survey of the History of the Canon of the New Testament

(5th ed.; Cambridge and London: MacMillan, 1881) 324.

ALLERT: Putting Taitan’s Diatessaron in Perspective 7

1. The Attestation

The first mention²⁹ of the Diatessaron in the West is by Eusebius (ca.

260-340), "Tatian composed in some way a combination and collection

of the Gospels, and gave this the name of The Diatessaron, and this is

still extant in some places."³⁰ Rufinus's (ca. 345- 410) Latin translation

of Eusebius's Ecclesiastical History further expanded on the title Dia-

tessaron by stating unum ex quattuor.³¹ Epiphanius (ca. 315- 403) states

that the Gospel Diatessaron was created by Tatian and that some call

it "according to the Hebrews."³² Later, in the fifth century, Theodoret

(ca. 393-ca. 466) states that,

He [Tatian] composed the so-called Diatessaron by cutting out the ge-

nealogies and whatever goes to prove the Lord to have been born of the

seed of David according to the flesh. And this work was in use not only

among his own party but even among those who follow the tradition

of the Apostles, who used it somewhat too innocently as a compen-

dium of the Gospels, without recognizing the craftiness of its compo-

sitions. I myself found more than two hundred copies in reverential

use in the churches of my diocese, all of which I removed, replacing

them by the Gospels of the four Evangelists.³³

In the sixth century Victor, bishop of Capua in Italy (541-54), discov-

ered a manuscript of a Gospel harmony with no title or author's name.

He had a copy made and in his preface to the new copy (which is the

present Codex Fuldensis) explains how he concluded that the work

must be the harmony of Tatian. Interestingly, however, Victor does not

call the work Diatessaron, but a Diapente (= "through five [Gospels]").

There is much evidence to show the early circulation of the Dia-

tessaron in the East. The Old Syriac Gospels (extant in the fourth-

century Codex Sinaiticus and the fifth-century Codex Curetonianus)

show that they have been influenced by the textual variants and the

harmonistic readings of the Diatessaron.³⁴ Many of the Gospel quo-

tations of the Syrian writers Aphrahat (early fourth century) and

Ephraem (ca. 306-73) are from the Diatessaron.³⁵ Ephraem even wrote

a commentary on the Diatessaron.³⁶

29. What follows is dependent on Peterson, "Tatian's Diatessaron," 405-8.

30. Eusebius Hist. eccl. 4.29.6, o( Tatiano_j suna/feia/n tina kai_

sunagwgh_n ou)k oi} d ) o#pwj tw=n eu)aggeli/wn sunqei/j, To_ dia_ tessa/rwn

tou=to proswno/masen, o$ kai_ para/ tisin ei)j e!ti nu=n fe/retai:

31. "One from four."

32. Epiphanius Haer. 46.1.8-9.

33. Theodoret Haer. fab. comp. 1.20.

34. Metzger, Early Versions of the New Testament, 12-17.

35. F. C. Burkitt, Ephraim's Quotations from the Gospel (TextsS 7/2; Cambridge, 1901).

36. The harmony, however, was not known by the name Diatessaron. Rather it was

known as the Gospel of the Mixed (Euangelion da Mehallete). Ephraem mentions neither

Tatian nor the title Diatessaron in this commentary.

8 Bulletin for Biblical Research 9

The title Diatessaron first appears in Syriac in a fourth-century

translation of Eusebius's Ecclesiastical History. The text (4.29.6) is trans-

lated, "Now this same Tatianus their former chief collected and mixed

up and composed a gospel and called it Diatessaron; now this is (the

Gospel) of the mixed, the same that is in the hands of many unto this

day "³⁷ In comparing the Syriac translation with the Greek original it

can be seen that the translator deleted Eusebius's "I know not how"

and modified the last phrase to emphasize the Diatessaron's continu-

ing use. Even though the title Diatessaron was a transliteration, the

translator also felt obliged to include the explanation that the Diates-

saron is known to the readers as the "Gospel of the Mixed."

The Diatessaron is also named in the fifth-century Syrian work

Doctrine of Addai, but the inclusion of the title is most likely an in-

terpolation.³⁸ It is not until the ninth century that the title is used in

Syriac literature again. In order to influence the writings of Aphra-

hat, Ephraem, and the Doctrine of Addai, the Diatessaron must have

been in circulation in Syria from the beginning of Syrian Christianity,

where it was known as the "Gospel of the Mixed." It was only later

that the title Diatessaron became the common designation in Syria.

This explains why the first use of the title in Syriac, in the fourth-

century Syriac translation of Eusebius's Hist. eccl., is a transliteration

which causes the translator to add the standard Syriac name "Gospel

of the Mixed."

The above attestations of Tatian's harmony are often used as proof

that the name Diatessaron was given to the work by Tatian himself.

For example, B. F. Westcott states, "There can be no reasonable doubt

that the name was given to the work by Tatian himself; and if the

Diatessaron was not a compilation of four Gospels, what is the ex-

planation of that number? If again these four Gospels were not those

which we receive, what other four Gospels ever formed a collection

which needed no further description than the four?"³⁹ The assertion

that Tatian himself called the work Diatessaron further leads Westcott

to state that Tatian used only the four canonical Gospels in his har-

mony. Both of these assertions need to be addressed.

Questions have been raised concerning the above attestations

which claim the name Diatessaron for Tatian's work. R. P. C. Hanson

believes that there is no reason to think that this was Tatian's own title

37. W. Wright and N. McLean (eds.), The Ecclesiastical History of Eusebius in Syriac

(Cambridge: Cambridge University Press, 1898) 243.

38. F. C. Burkitt (ed.), Euangelion da-Mepharreshe: The Curetonian Version of the Four

Gospels (2 vols.; Cambridge: Cambridge University Press, 1904) 2.175. Peterson, "Tatian's

Diatessaron," believes that it is an anachronism simply because it is so anachronistic

in Syrian literature of the time.

39. Westcott, The Canon of the New Testament, 322.

ALLERT: Putting Taitan’s Diatessaron in Perspective 9

for the work.⁴⁰ This belief is intimately linked with the number of

sources that Tatian used. Certainly the title Diatessaron (dia_ tessa/rwn=

"through four") suggests the use of only four Gospels. But if this title

was only a later appellation and not used by the compiler himself we

cannot necessarily argue, as Westcott, that the title is entirely descrip-

tive of the work.

In examining the various attestations of the Diatessaron, G. M. Hah-

neman⁴¹ believes that Eusebius's suggestion that Tatian harmonized

only the four canonical Gospels is not based on firsthand knowledge

of the work—Eusebius had probably never seen the document. Fur-

ther, Hahneman points out that Rufinus's Latin translation of the

same passage (4.29.6) further accentuated the possible error. Also

thrown into the mix is the fact that Victor of Capua called the har-

mony a Diapente, thus suggesting Tatian's dependence on five sources

rather than four.

J. H. Charlesworth⁴² asserts that there is no consensus regarding

the number of sources used. Against the use of only four sources is

the possibility that Eusebius erred when he wrote that Tatian's har-

mony was compiled using the four Gospels. Further, the Syriac trans-

lation of the Eusebian passage does call it the Diatessaron, but this

is probably a transliteration of the Greek name. Why would Tatian

place a Greek name on a Syriac work? Against the use of five Gospels

and the appellation of Victor of Capua (Diapente) is the fact that

Diapente may simply be a musical term. In fact, Metzger points out⁴³

that both terms are musical and can apply to Tatian's harmony.

Bolgiani shows that dai_ tessa/rwn and dia_ pe/nte are technical terms used

in ancient musicology, one referring to three intervals of four notes, the

other to four intervals of five notes. He therefore interprets Victor's

comment to mean that Tatian's "harmony" of the four Evangelists in-

volves not merely four individual notes but four fundamental elements

of symphonic harmony, the diapente. Thus both terms, diatessaron and

Victor's metaphorical use of diapente, are appropriate descriptions of

Tatian's Harmony of the Gospels.⁴⁴

Metzger's belief that only four Gospels were used is irrelevant to the

fact that both terms can be applied to describe something in the

harmony other than the number of sources. For Charlesworth, this

40. R. P. C. Hanson, Tradition in the Early Church (London: SCM, 1962) 230.

41. G. M. Hahnernan, The Muratorian Fragment and the Development of the Canon

(Oxford Theological Monographs; Oxford: Clarendon, 1992) 98.

42. J. H. Charlesworth, "Tatian's Dependence upon Apocryphal Traditions," HeyJ.

15 (1974) 5-6.

43. Metzger, Early Versions of the New Testament, 28-29.

44. Ibid., 29.

10 Bulletin for Biblical Research 9

translates to his belief that neither Eusebius's nor Victor's statement

should dictate the answer to the question of Tatian's sources.⁴⁵

2. The Text

If neither Eusebius nor Victor should dictate the number of sources

used in Tatian, the question remains, what should? There certainly is

no consensus on how many sources were used in the compilation of

the harmony. Some admit only the four canonical Gospels,⁴⁶ while

others admit at least one other written or oral source.⁴⁷

The former view is based entirely on the assertion that Tatian

himself gave the harmony the name Diatessaron. Westcott⁴⁸ does just

this when he claims that the only reason for the name Diatessaron is

that Tatian used only the four canonical Gospels. And Metzger⁴⁹ fur-

ther conjectures, based on the same assertion, that Tatian compiled

his harmony because it was his private judgment that the fourfold

harmony was the most convenient way to present the whole gospel

story at once instead of confusing people by offering them four par-

allel and more or less divergent narratives.⁵⁰ As we have seen, how-

ever, the belief that Tatian himself gave the title to his harmony can

easily be called into question.

45. Charlesworth, "Tatian's Dependence upon Apocryphal Traditions," 6.

46. See, e.g., E. J. Goodspeed, The Formation of the New Testament (Chicago: Uni-

versity of Chicago Press, 1926) 58; Kelly, Why Is There a New Testament? 86; Metzger,

Canon of the New Testament, 115-16; idem, Early Versions of the New Testament, 28-30;

A. Souter, The Text and Canon of the New Testament (rev. ed.; C. S. C. Williams; London:

Duckworth, 1960) 155; Westcott, On the Canon of the New Testament, 322.

47. See, e.g., Barnard, "The Heresy of Tatian—Once Again," 5; F. F. Bruce, The

Canon of Scripture (Downers Grove, Ill.: Inter Varsity, 1988) 128; H. von Campenhausen,

The Formation of the Christian Bible (Eng. trans. J. A. Baker; London: Black, 1972) 175;

Charlesworth., "Tatian's Dependence upon Apocryphal Traditions," passim; R. M. Grant,

The Formation of the New Testament (London: Hutchinson University Library, 1965) 138;

Hahneman, The Muratorian Fragment and the Development of the Canon, 98-99; Hanson,

Tradition in the Early Church, 191, 230; W. L. Peterson, "Textual Evidence of Tatian's De-

pendence upon Justin's 'AIIOMNHMONEYMATA," NTS 36 (1990) 512-34; G. Quispel,

"Tatian and the Gospel of Thomas," VC 11 (1957) 189-207; idem, "Some Remarks on

the Gospel of Thomas," NTS 5 (1958) 282-90; idem, "The Gospel of Thomas and the

Gospel of the Hebrews," NTS 12 (1966) 371-82.

48. Westcott, On the Canon of the New Testament, 322.

49. Metzger, Canon of the New Testament, 116.

50. Related to this conjecture by Metzger is the interesting article by T. Baarda,

"DIAFWNIA--SUMFWNIA: Factors in the Harmonization of the Gospels, Especially in

the Diatessaron of Tatian," in Gospel Traditions in the Second Century: Origins, Rescen-

sions, Text, and Transmission (ed. W. L. Peterson; Notre Dame: University of Notre Dame

Press, 1989) 133- 49. Baarda contends that Tatian's motivation in harmonizing the Gos-

pel accounts was found in his conception of the unity of Christianity and the disunity

of the Greeks. He believed unity and harmony were the hallmark of Christianity, but

the apparent discrepancies in the Gospel accounts could not escape his notice. The out-

working of this belief is his Diatessaron.

ALLERT: Putting Taitan’s Diatessaron in Perspective 11

Not very much more evidence can be offered for the latter view

that Tatian used five or more sources in compiling his harmony. This

assertion has been made simply through an examination of the re-

constructed text⁵¹ of the Diatessaron.⁵² Certainly no one would deny

that the majority sources for the Diatessaron are the four Gospels of

Matthew, Mark, Luke, and John. The point at issue here is the possible

use of sources other than these four. The following are the few ex-

amples that can be presented as possible evidence of Tatian's use of

extracanonical sources.

a. The synoptic accounts of the baptism of Jesus by John the Baptist

are similar in that they do not mention a great light shining after

the baptism. Many witnesses to the Diatessaron,⁵³ however, in-

clude the account of a great light shining. We do know that this

same tradition is found in Justin Martyr⁵⁴ and Epiphanius.⁵⁵ Epi-

phanius even states that it was found in the now lost Gospel of the

Ebionites. Therefore, Justin's reading may come from the Gospel of

the Ebionites, which was then used by Tatian—either directly or

indirectly—when he compiled his Diatessaron, or Tatian and the

Gospel of the Ebionites were both dependent on Justin's harmony.⁵⁶

b. The difference between the canonical Matt 8:4⁵⁷ and the Diatessa-

ron⁵⁸ has lead some to assert that the reading must be evidence of

an earlier Judaic-Christian text.⁵⁹

c. One scholar⁶⁰ has shown the possibility of the Persian Diatessaron's

including nativity readings that are present in the Protevangelium

of James.

These are really the only concrete examples that are presented as evi-

dence that Tatian relied on an extracanonical source. This evidence,

51. On the rules on the reconstruction of the text of the Diatessaron, see Peterson,

"Tatian's Diatessaron," 419-22; idem, "Romanos and the Diatessaron: Reading and

Method," NTS 29 (1983) 484-507.

52. What exactly these sources were is not relevant for this essay. A good over-

view of this discussion can be gleaned from the sources listed in n. 46 above.

53. This reading is included in Erphraem's Commentary, Isho^cdad of Merv, the

Pepysian Harmony, the Latin Poem Vita Rhythmica, and two Vetus Latine manuscripts,

a (fourth century), and g¹ (eighth century).

54. Justin Martyr Dial. 88.3.

55. Epiphanius Haer. 30.13.

56. Peterson, "Tatian's Diatessaron," 422-23.

57. "Go, show yourself to the priest, and present the offering that Moses com-

manded, for a testimony to them."

58. "Go, show yourself to the priest and fulfill the Law." This reading is found in

Ephraem, Isho^cdad of Merv, Romanos, the Liège Harmony, the Venetian Harmony.

59. Peterson, "Tatian's Diatessaron," 424.

60. G. Messina, "Lezioni apocrife nel Diatessaron Persian," Bib 30 (1949) 10-27.

12 Bulletin for Biblical Research 9

or lack thereof, causes Metzger to conclude, "the amount of extra-

canonical material that seems to have been present in Tatian's Dia-

tessaron hardly justifies the opinion of some scholars that Tatian

used extensive use of a fifth, apocryphal Gospel when he compiled

his harmony."⁶¹

IV. CONCLUSIONS AND IMPLICATIONS

The field of Diatessaron studies is vast and often times confusing. The

major issues of witnesses, titles, and sources are areas to which

much significant study has been devoted. Many of the conclusions

reached in theses studies, however, have only minimal bearing upon

our topic. As important as many of these issues are for other matters,

they come little into play in understanding the state of the NT canon

of the second century.

Foremost in our minds in placing Tatian's Diatessaron into per-

spective is the issue of definition. In many discussions of canon there

appears to be a lack of application of the true definitions of scripture

and canon. Many, in fact apply the terms as synonyms. That is, many

move from scripture to canon without recognizing the difference

which lies therein. This has led to confusion for those who have

sought to understand the state of the canon in the second century.⁶²

Years ago A. C. Sundberg, Jr. called for a more precise definition

of the terms canon and scripture in order to distinguish some very

important features of the terms.⁶³ Sundberg stated, "it is necessary to

distinguish between the terms 'scripture,' meaning writings which are

held in some sense as authoritative for religion and 'canon,' meaning

a defined collection that is held to be exclusively, i.e., with respect to

all other books, authoritative."⁶⁴ Sundberg did not like the fact that

many discussions concerning canonicity employed the terms canon

and canonical in a rather loose sense. In other words, Sundberg ob-

jected to the fact that these terms were used to designate documents

that were widely held in authority (scripture), without regard to a

definitive and exclusive determination of a group of such documents

(canon). He considered the application of canonical to these docu-

ments anachronistic because a canon, by definition, is closed and no

such closed canon existed in the second and third centuries. Thus,

61. Metzger, Early Versions of the New Testament, 36.

62. A. C. Sundberg, Jr., "Towards a Revised History of the New Testament Canon,"

SE (TU 102; 1968) 452-61.

63. A. C. Sundberg, Jr., "The Making of the New Testament Canon," in The Inter-

preter's One Volume Commentary on the Bible (ed. C. M. Laymon; London: Collins, 1972)

1216-24; idem, "Towards a Revised History of the New Testament Canon."

64. Sundberg, "The Making of the New Testament Canon," 1216.

ALLERT: Putting Taitan’s Diatessaron in Perspective 13

canon and scripture must be applied and understood in the proper

sense in order to avoid the above mistakes. More recently the impor-

tance of this distinction has been furthered by L. M. McDonald⁶⁵ and

H. Y. Gamble,⁶⁶ albeit in differing, yet complimentary, terms.

McDonald begins his discussion of the proper distinction be-

tween scripture and canon with an appeal to the findings of G. T. Shep-

pard.⁶⁷ There are two ways of understanding the notion of canon in

the ancient world.⁶⁸ Taking his cue from Sheppard, McDonald ex-

plains these two ways as "canon 1" and "canon 2." Canon 1 is essen-

tially a rule, standard, or guide that functions in an authoritative

manner in a community. It is a fluid or flexible authority that is not

yet fixed. Canon 1 is therefore present wherever there is a respect for

some authority within a community, either in written or oral form.

Canon 2 occurs when these authorities (canon 1) become more fixed in

a given community. Canon 2 authority becomes so well established in

a community of faith that very little doubt arises about the authority

of a text thereafter. Canon 2, therefore, is a more fixed authority for

the community.

Gamble (with more reliance on Sundberg) employs the terms func-

tional and formal in his understanding of scripture and canon. Gamble

begins from his understanding that the NT canon "calls special at-

tention to its form, i.e., a fixed collection of precisely twenty-seven

early Christian documents, and to its function, i.e. literature that is

normative for the faith and life of the Christian community."⁶⁹ Im-

mediately Gamble points out the importance of understanding canon

as a fixed collection. This is what he means by a formal canon. On the

other hand, a functional canon is one that operated in the sense of a

norm or a standard, whether written or oral. Gamble emphasizes that

a scrupulousness about such terminology is necessary so that these

real distinctions are not blurred. The history of the NT canon is not

only concerned with the normative use of Christian writings as au-

thoritative documents (as scripture), but with the delimitation of a

65. L. M. McDonald, The Formation of the Christian Biblical Canon (rev. ed.; Pea-

body: Hendrickson, 1995); idem, "The Integrity of the Biblical Canon in Light of Its

Historical Development," BBR 6 (1996) 95-132.

66. H. Y. Gamble, "Canon—New Testament," ABD (ed. N. D. Freedman; 6 vols.;

New York: Doubleday, 1992) 1.852-61; idem, "The Canon of the New Testament," in

The New Testament and Its Modern Interpreters (ed. E. J. Epp and G. W. MacRae; Atlanta:

Scholars Press, 1989) 201-43; idem, The New Testament Canon: Its Making and Meaning

(Guides to Biblical Scholarship; Philadelphia: Fortress, 1985).

67. G. T. Sheppard, "Canon," The Encyclopedia of Religion (ed. M. Eliade; 10 vols.;

New York: MacMillan, 1987) 3.62-69.

68. McDonald, Formation of the Christian Biblical Canon, 20; idem, "The Integrity of

the Biblical Canon," 101.

69. Gamble, "Canon—New Testament," 852.

14 Bulletin for Biblical Research 9

specific group of such writings and with its meaning and function as

a collection (as a canon).⁷⁰

Both McDonald and Gamble are really saying the same thing as

Sundberg. Canon is not synonymous with scripture. As McDonald

has pointed out,⁷¹ there is considerable overlap. The overlap occurs in

the authority of a document, that is, both scriptural documents and

canonical documents are viewed as authoritative in the life of the

community. The line of demarcation occurs where these documents

are chosen to be placed between two distinct poles.⁷² At one end of

the pole is a rule, standard, ideal, norm, or authoritative office or lit-

erature, whether written or oral. Both Sheppard and McDonald call

this pole "canon 1," while Gamble calls this a "functional canon."

However, it seems that some confusion would be avoided if we main-

tained Sundberg's term "scripture." Whichever term one chooses to

use, the emphasis of this pole is on the internal signs of an elevated

status within the community of faith. At the other end of the pole is

a fixation, standardization, enumeration, listing, chronology, register,

or catalog of exemplary or normative persons, places, or things (al-

though for our purposes here we are concerned only with writings).

Sheppard and McDonald call this pole "canon 2," while Gamble pre-

fers "formal canon." If we adopt Sundberg's terminology above for

scripture, it would then be acceptable also to adopt simply the term

canon for this pole. The emphasis here rests on the precise boundary,

limits, or measure of what, from some preunderstood standard, be-

longs within or falls outside of a specific canon.

When taking this understanding of scripture and canon into the

field of the history of the NT canon its significance can at once be

seen. We must understand that the word canon did not begin to come

into use as an appellation for a collection of Christian writings until

the mid–fourth century with Athanasius.⁷³ In the second century,

however, the word designated what the church acknowledged as

having regulative authority for its faith and life.⁷⁴ It was only after

Athanasius that the term canon came to denote a closed collection of

authoritative writings to which nothing could be added or from which

nothing could be taken away.

70. Gamble, "The Canon of the New Testament," 205.

71. McDonald, Formation of the Christian Biblical Canon, 13.

72. Sheppard, "Canon," 64.

73. Athanasius Decr. 5.18. Here Athanasius describes the Shepherd of Hermas as

"not of the canon." See also Athanasius's famous Festal Letter (Ep. 39) of the year 367,

wherein he describes certain Christian books as "canonical."

74. See, e.g., Irenaeus Haer. 1.9.4; 1.10.1; 5.20.1; Dem 3; Tertullian Praescriptio 13; 27;

De virginibus velandis 1; Clement of Alexandria Str. 7.15.90; Eusebius Hist. eccl. 6.13.3.

ALLERT: Putting Taitan’s Diatessaron in Perspective 15

On the other hand, the term scripture designates writings which

are religiously authoritative. They are thus used as such without re-

gard to their systematic enumeration or limitation. This is, in fact,

how Christian writings were used throughout the second century.

The concept of canon presupposes scripture, but the concept of scrip-

ture does not necessarily entail the notion of a canon.

In canon studies one must be aware of not attributing canonical

status to a document that has only attained scriptural status. This is

particularly important in dealing with the second and third centu-

ries. This is so because of the lack of evidence that the church con-

sciously discussed or contemplated a closed or fixed collection of the

NT Scriptures during this time.⁷⁵

In the second and third centuries, the church fathers did not

answer Marcion, Gnosticism, and Montanism with a closed canon of

scriptures (canon 2 or formal canon). The answer to these challenges

was with a canon of truth (canon 1 or functional canon) which was

defended by an appeal to apostolic writings. This can be seen in

Irenaeus⁷⁶ (ca. 130–ca. 200) and in Tertullian⁷⁷ (ca. 260–ca. 225) where

both use the canon of truth to speak to these challenges. Even as late

as the early third century Serapion shows the same thing.⁷⁸ In writing

to his church, Serapion wished to settle the question of whether the

Gospel of Peter could be read in that church. He had previously al-

lowed it to be read in the church but later reversed this decision on

the basis that it denied the humanity of Jesus. The point here is that

Serapion did not revoke his permission to allow the Gospel of Peter to

be read in the church on the basis of an appeal to a closed collection

of scripture but on the basis of the canon of truth, on the basis that

certain doctrines contained in that Gospel were at variance with what

was handed down through the Apostles to the Bishops of the churches.

The issue was dealt with on the basis of orthodoxy, not canonicity.

It is difficult to place the settling of the NT canon in the second

century when doubt concerning certain documents was occurring in

the fourth century. Eusebius's famous list of recognized, disputed,

and rejected books is illustrative of this point.⁷⁹ One must asked why

discussions of this type were occurring in the early fourth century

if the issue was largely settled in the second. The fact is that there

75. The following two paragraphs acknowledge McDonald ("The Integrity of the

Biblical Canon in Light of Its Historical Development," 118-19) for a succinct expla-

nation of this issue.

76. Irenaeus Haer. 1.8.1; 1.9.1-4.

77. Tertullian Praescriptio 8-9.

78. Eusebius Hist. eccl. 6.12.3-6.

79. Eusebius Hist. eccl. 3.25.

16 Bulletin for Biblical Research 9

were simply no discussions by the second-century fathers concern-

ing canonicity.

The above has implications for placing Tatian's harmony into

proper perspective. Because of its second-century date, the Diatessa-

ron must be understood in light of the necessary distinction between

scripture and canon. Unfortunately, in the past, this distinction has

neither been maintained nor understood by many students of the

scriptural canon.

For example, we must wonder how Souter can state that the har-

mony of the Diatessaron in ca. 170 proves the canonicity of the four

Gospels.⁸⁰ What exactly does Souter mean here? If the concept of a

canon, as we have come to understand it, implies no addition or sub-

traction from its contents, then the Diatessaron is in violation of that

concept by its very existence simply because it added to and sub-

tracted from the four canonical Gospels.

Goodspeed also misunderstands this distinction when he attempts

to place the deliberate creation of the fourfold Gospel at AD 115-25.⁸¹

Goodspeed chooses this date because he believes the Preaching of

Peter, 2 Peter, the Gospel of Peter, Papias, the Epistle of the Apostles, and

Justin Martyr all show acquaintance with or dependence on all four

canonical Gospels. This attempt, however, confuses aquaintance with,

and even dependence upon, the four with a conscious decision to ar-

range the four into a fixed collection. The former does not necessarily

mean the latter.

We must also take issue with Goodspeed when he states that

Tatian's use of the four Gospels in this bold rearrangement shows

that, even though these four stood on a higher level than any others,

they were still not considered scripture.⁸² It is quite apparent that

Goodspeed is using canon and scripture synonymously since he later

claims that Tatian's harmony proves that he had no canonical rever-

ence for the four Gospels.⁸³ I would agree with the statement con-

cerning the lack of canonical reverence but not with Goodspeed's

comment that Tatian did not consider them scripture. The very fact

that Tatian used primarily these four Gospels shows, at the very

least, that they were authoritative and thus considered scripture.

B. Lindars also misunderstands the concept of canon when he

discusses the Diatessaron.⁸⁴ Lindars claims that the concept of canon

80. Souter, The Text and Canon of the New Testament, 155.

81. Goodspeed, The Formation of the New Testament, 37-38.

82. Ibid., 58.

83. Ibid., 115.

84. J. Rogerson, C. Rowland, and B. Lindars, The History of Christian Theology,

Vol. 2: The Study and Use of the Bible (Basingstoke: Pickering / Grand Rapids: Eerdmans,

1988) 243.

ALLERT: Putting Taitan’s Diatessaron in Perspective 17

was not so rigid so as to prevent some freedom in handling the text.

Apparently Lindars ignores the fact that the concept of canon was not

associated with writings during the second century. As was pointed

out above, it was associated with the rule of faith. In this light it is

difficult to imagine Tatian having a concept of canon that was appli-

cable to the sources used in his compilation.

I assert that the appearance of a harmony, regardless of the

sources used, shows that a fourfold Gospel canon was not in exist-

ence, practically or even conceptually. Only a handful of scholars

have come to recognize the significance of the above definitions with

respect to the position of the Diatessaron in the second century.⁸⁵ With

them I recognize the importance of the Diatessaron in establishing

authority for the four Gospels. It is, in fact, the first document avail-

able to us that overtly recognizes the authority of the four Gospels.

But it is a mistake to assume that the use of these four shows that

they were already canonical in Tatian's day. Perhaps the lines have

become blurred because the main sources used by Tatian are known

to us as canonical. But, as the above scholars have pointed out, Tatian

used the four Gospels in much the same way that the Gospel writers

employed their sources. But no one has ever argued that, simply

because Matthew and Luke used Mark, Mark was canonical.

Tatian used these four because they were scripture and probably

the most influential Gospels, not because they were canonical. If they

were canonical it seems as though Tatian's compilation would have

met with some sort of opposition from the very start because of his

handling (mishandling?) of the canonical text. But this was not the

case. In fact, the Diatessaron enjoyed great popularity for centuries in

Syria and was probably the first Gospel to reach Syria. The Diatessa-

ron was only dislodged there as the authoritative Gospel under the

influence of Rabulla, bishop of Edessa, in the fifth century. Appar-

ently this dislodging was not a result of a disagreement with the

handling of the texts but because of Tatian's reputation as a heretic.⁸⁶

The church in Syria most likely became embarrassed by the use of a

Gospel text that was compiled by a heretic.

In placing Tatian's Diatessaron in perspective we must understand

that the idea of Tatian's thinking of these four Gospels in a canonical

sense is, in reality, anachronistic—it belongs to a time removed from

his own. The documents were authoritative and they were scripture,

85. Gamble, "The Canon of the New Testament," 210; idem, "Canon—New Testa-

ment," 854; Hahneman, The Muratorian Fragment and the Devlopment of the Canon, 98-99;

McDonald, The Formation of the Christian Biblical Canon, 10, 152, 225; Sundberg, "Towards

a Revised History of the New Testament Canon," 457-58; idem, "The Making of the

New Testament Canon," 1219.

86. Metzger, The Canon of the New Testament, 218.

18 Bulletin for Biblical Research 9

but they were not canonical. The perspective this opens for us is that

the state of the NT canon during Tatian's day was still fluid. Tatian,

in concert with Justin Martyr, Clement of Rome, Ignatius of Antioch,

and Polycarp, betrays neither a concept of nor a desire to form a NT

canon. The possibility, therefore, of a written or oral "extracanonical"

source in the Diatessaron is irrelevant because we are not dealing with

a canon.

The discussions surrounding the significance of Tatian's harmony

on the NT canon point to the need for a better understanding of the

formation of the canon. By focusing on the polemical aspects of the

canon—that is, when it was closed—the idea of a canonical process

suffers.⁸⁷ We are better served by placing Tatian's Diatessaron in this

perspective of process. By viewing this fluid era of Christian writings

from this perspective, we are less tempted to view Christian writings

of the second century from the anachronistic perspective of twentieth-

century Christianity, which already possesses a fixed canon and as-

sumes the same situation at the appearance of certain writings in

early Christianity.

87. A. C. Outler, "The 'Logic' of Canon-Making and the Tasks of Canon-Criticism,"

in Texts and Testaments: Critical Essays on the Bible and the Early Church Fathers (ed. W. E.

March; S. D. Curries Festschrift; San Antonio: Trinity University Press, 1980) 263-76.