Production Note
Cornell University Library pro-
duced this volume to replace the
irreparably deteriorated original.
It was scanned using Xerox soft-
ware and equipment at 600 dots
per inch resolution and com-
pressed prior to storage using
CCITT Group 4 compression. The
digital data were used to create
Cornell's replacement volume on
paper that meets the ANSI Stand-
ard Z39.48-1984. The production
of this volume was supported in
part by the Commission on Prés-
ervation and Access and the Xerox
Corporation. Digital file copy-
right by Cornell University
Library 1991.THE UNIVERSITY OF CHICAGO
FOUNDED BY JOHN D. ROCKEFELLER
The Decennial Publications
SOME PRINCIPLES OE LITERARY CR1TIC1SM AND THE1R
APPLICATION TO THE SYNOPTIC PROBLEM
BY
ERNEST DEWITT BURTON
PROFESSOR AND HEAD OF THE DEPARTMENT OF BIBLICAL
AND PATRISTIC GREEK
PRINTED FROM VOLUME V
CHICAGO
THE UNIVERSITY OF CHICAGO PRESS
1904Copyright 1904
BY THE UNIVERSITY OF CHICAGO
PRINTED DECEMBER 15, 1904SOME PRINCIPLES OF LIÏERARY CRITICISM AMD THEIR
APPLICATION TO THE SYNOPT1C PROBLEM
Ernest DeWitt Bürton
Ever since the days of Augustine and Jerome, not to say earlier, the mutual
resemblances of the first three gospels hâve been observed, and the problem thus
created for the biblical scholar has been discussed. Since 1794, when Eichhorn pro-
posed his theory accounting for the resemblances and divergences of the synoptic gos-
pels by deriving them ail from a common document existing in various recensions, the
question has been vigorously discussed, and almost numberless théories hâve been
proposed for its solution.
It may be doubted, however, whether amid the multitude of these théories
sufficient thought has been given to the formulation of the principles in accordance
with which any solution of the problem must of necessity proceed. In textual criticism
this phase of the matter has received most earnest attention, and principles hâve been
formulated in respect to which there is general, even if not universal, agreement
among textual critics. There seems to be no good reason why a course should not be
pursued, in the matter of the relation of our présent synoptic gospels to one another,
similar to that which has been so efficient in promoting the solution of the problem of
the interrelation of the manuscripts of these gospels and of the other New Testament
books.
Indeed, the work done in formulating the task of textual criticism may well
furnish the starting-point for the effort to formulate corresponding principles applicable
to the problem of the relation of the gospels to one another, if not also the framework
for a provisional statement of such principles. For, in fact, the relation of our several
synoptic gospels to one another, and to the documents or traditions which may hâve
lain behind them, is closely analogous to the relation of the several manuscripts (or
other witnesses to the text) of any New Testament book to one another, and to the
manuscripts which in this case undoubtedly constituted the sources from which our
existing manuscripts were produced. The analogy, of course, must not be pressed
where it does not hold. The methods of scribes and editors, though similar, are not
identical. The analogy must serve mainly to suggest principles whose validity shall
afterward be tested by their applicability to the actual conditions of the class of
problems to which the synoptic problem belongs. But so used it seems likely to lead
to valuable results.
The principles so reached would, if valid, be applicable not solely to the synoptic
problem ; for this problem is simply a spécifie instance of the general problem with
which the historian always has to deal when he finds among his sources documents
which, though similar, are nevertheless not duplicates. Bernheim has discussed this
1954 Principles of Literary Critioism and the Synoptic Problem
problem with spécial reference to the sources for mediæval history in his work, Lehr-
buch der historischen Méthode (Leipzig, 1889), pp. 272-99, and the conclusions which
he reaches, if Sound, are applicable in general to the whole class of cases in which
there exist documents which are similar without being identical, and so in particular to
the synoptic problem.
If, then, we avail ourselves of the work already done in textual criticism and in
the study of the use of sources in historical work in general, continually checking the
suggestions derived from these sources by unquestioned facts, we ought to arrive
at certain principles on the basis of which it would be possible to move with some
degree of firmness and certainty to an assured solution of the synoptic problem. The
aim of this paper is not to review the history of the efforts to solve the problem of
the synoptic gospels, or to examine any of the solutions that hâve been proposed, but,
first, to formulate principles applicable to such problems ; secondly, to state the main facts
respecting the relation of the synoptic gospels to one another ; and, thirdly, to indicate
the conclusions to which the corrélation of these principles and facts seems to lead.
I. PRINCIPLES APPLICABLE TO THE SOLUTION OF THE SYNOPTIC PROBLEM
1. As between some sort of literary relationship and total independence from a
literary point of view, the decision is to be based upon the extent and nature of the
resemblance of documents to one another. Between the two extremes, represented on
the one side by two manuscripts or printed papers word for word identical throughout,
and on the other side by two manuscripts or printed works having nothing in common
save that they are written in the same language, ail grades and degrees of resemblance
are possible. In the former case, relation between documents ceases to exist because
there are not two documents, but two exemplars of the same document. In the latter
extreme, relationship vanishes because of total diversity. Between these extremes
there may exist, for example, the case of two documents which, resembling one another,
but not to the extent of identity, owe their resemblance to the fact that two authors
wholly independently give an account of the same general event. In this case the
two narratives traced to their origin meet, not in any common literary ancestor, but
only in the event narrated. Again, two documents may resemble one another by reason
of the fact that behind both of them there lies a literary work by which both hâve been
influenced. This literary work may be a speech to which both hâve listened, a writing
which both hâve read and which they reproduce from memory, or a writing from which
both transcribe with more or less freedom of reproduction. From the point of view
of the principle we are now endeavoring to state, these différences do not corne into
account. The question is simply this: When we trace the lines of heredity of the two
writings, similar but not identical, to that which accounts for their similarity, is this
common source something which had already assumed literary form, or an event or
group of facts which each writer was endeavoring in his own way to record ? This
fundamental question must, as has been stated, be solved by the extent and nature of
196Ebnest DeWitt Burton
5
the resemblances. How great a degree of resemblance will compel the supposition of
a common literary ancestor, how great a degree of divergence would exclude this sup-
position, can be determined only by expérience and observation. It is important, if
possible, that such observation should pertain to writings the nature of whose relation-
ship is known, and which corne from a period not too remote from that to which the
writings belong whose relationship we are endeavoring to détermine.
2.	It being determined that the relation of the documents in question is literary
in its character, the decision between documentary and oral relationship—that is, a
relationship mediated by written documents and one mediated by oral report or tradi-
tion—must likewise rest upon the nature and extent of the resemblance, account being
taken of order of paragraphs or other like divisions, content of narrative, and verbal
agreement.
The extent and character of the resemblances which would be produced respect-
ively by oral tradition and documentary relationship must be judged, as respects the
synoptic gospels, not by modem custom, but by the usage of the times from which
these books corne. It must be recognized that among the ancient Jews greater
resemblance might be produced by oral relationship than is usual in such cases today,
and that greater variation would usually appear in the case of documentary relation
than would usually occur now. It is obviously impossible to détermine with accuracy
to what extent a written reproduction of an oral tradition would differ from this tradi-
tion itself, since by the nature of the case the original is not in existence, and cannot
therefore be compared with the transcription of it. Only in case we possessed two
writings, both vouched for as independent attempts to reproduce what had previously
existed only orally, could we by actual test détermine how great a degree of resemblance
and diversity would be produced by such independent reproduction of oral material.
On the other hand, the degree of diversity between the source of a document
and the document itself, that source being in written form, can be more accurately
determined, since instances exist in which both the source and the dérivative document
are still extant. In the case of the synoptic gospels we hâve several very valuable
bases of judgment upon this point, concerning which something will be said in a later
paragraph. See II, 8, p. 20.
3.	It being determined that there is between two or more extant documents a
literary relationship of a documentary character, the general principle for determining
the direction of descent—that is, which of the documents is antécédent to the others—
is, that that one is to be accepted as, relatively speaking, the original which will explain
the origin of the others, but cannot itself be explained as the product of the others.
In dealing with our synoptic gospels it must, however, be borne in mind that the
order of dependence is not necessarily the same throughout the whole extent of the
books. In textual criticism we are wont to recognize four parts of the New Testament
which are, as respects their textual history, to a considérable degree independent. These
four are: the Gospels, the Acts and Catholic Epistles, the Pauline Epistles, and the
1976 Principles of Litebaby Cbiticism and the Synoptic Pboblem
Apocalypse. In like manner it must be recognized as possible that in the history of the
gospels which lies back of the textual history, strictly so called, the several portions of
the synoptic material may hâve had a measurably independent history. There may,
for example, hâve been separate narratives of the infancy, the ministry of John, the
Gralilean ministry of Jésus, the Perean ministry, the passion and résurrection history.
If so, it is possible that our présent gospels sustain different relations to one another
in these different parts.
4. When the documents or portions of documents are two in number, and the
previously indicated tests hâve led to the conclusion that they sustain a literary rela-
tion to one another, if the influence of sources no longer extant be excluded from
considération, there remains evidently but one possibility, viz., that one of the docu-
ments is derived from the other. This may be represented by the diagram :
I
a
b
Which of the two documents is primary, and which is dépendent upon the other, must
be decided on the basis of the evidence as to which contains features of a secondary
character. The foliowing may be regarded as évidences of such secondary character:
(1) manifest misunderstanding of what stands in one document on the part of the writer
of the other; (2) insertion by one writer of material not in the other, and clearly
interrupting the course of thought or symmetry of plan in the other; (3) clear
omission from one document of matter which was in the other, the omission of which
destroys the connection; (4) insertion of matter the motive for which can be
clearly seen in the light of the author’s general aim, while no motive can be discovered
for its omission by the other author if he had had it in his source; (5) vice versa,
omission of matter traceable to the motive natural to the writer when the insertion
could not thus be accounted for; (6) alterations of other kinds which conform the
matter to the general method or tendency of the author.
In textual criticism it is regarded as a general, but not invariable, rule that the
longer reading is the later. It cannot be assumed that this rule holds also in our
présent field until the usage of compilers in that period has been somewhat more
carefully studied. Concerning the light which Tatian’s Diatessaron throws upon the
question, a word will be said below. The tendency of a particular writer, if it can be
determined, would in any case be the safest criterion; but one must, of course, be on
his guard against reasoning in a vicious circle in such a case.
5. But the possibility that the similarity of two documents is, in fact, due to the
influence of non-extant sources must also be taken into account. And this fact adds
to the possible relationships, indicated above, still other possibilités which for con-
198Ernest DeWitt Burton
7
venience we number consecutively with those already named. Thus a and b may be
independently derived from a common oral tradition, or a common lost document,
as represented in the foliowing diagram:
II
X
a	b
or a may be derived from a lost document or from oral tradition, and b may be
derived from this non-extant source and a, it being remembered that a may represent
either documont, b denoting the other. This may be represented in a diagram thus:
III1
a---------)b
The distinguishing of these several cases from one another and from cases I and II
must evidently be accomplished by the application of the tests for secondary character,
as indicated above. Thus in case II the marks of secondary character may be expected
to appear now in a, now in b, without clear evidence of originality in either through-
out. Whether the documents in hand be related through common oral tradition or a
written document would be determined by the nature of the resemblance—whether it
approximated that which we actually find in documents known to be derived from
extant sources, with which therefore they can be compared, ôr whether there is a free-
dom of reproduction which seems to exclude written sources. In case III the docu-
ment which drew from its companion, as well as from the source lying back of both,
would be expected to show stronger marks of secondary character than the others. But
since the dérivative document might in some cases displace the secondary readings of
its companion and restore those of the common source, and since in case II one docu-
ment might départ more freely from the original than the other, mere prépondérance
of marks of secondary character would not enable us either to identify the documents
under case III, or to distinguish case III from case IL Only when either a or b
should show clear indications of having combined two readings, one of which was
found in the other extant document, and the other of which was clearly related to it as
its source, would it be possible with any confidence to détermine which of the several
relations covered by cases II and III was the true one.
1 Case III can evidently be resolved into four cases, non-extant source. It would, however, contribute not to
according as the non-extant source is oral or written, and clearness, but to confusion, to enumerate under separate
according as a or b is the middle term between the non- notation ail possible variations of relation. It must suffice
extant source and the other existing document. Still other to represent classes of cases,
cases might also arise through the use of more than one
1998 Principles of Literaby Critioism and the Synoptic Problem
The problem presented by this group of cases is then to be solved, as far as it is
soluble, by the tests for direction of descent, with some possible help from evidence of
conflation.
6. When the related documents are three in number, various relations are possible.
If for the présent those be excluded that involve the hypothesis of a non-extant source,
oral or documentary, the possibilities may be most simply indicated by diagrams as
follows:
IV	V	VI	VII
That is, b and c may be derived from a; a may be derived from b and c; b may be
derived from a, and c from b; b may be derived from a, and c from a and b. But it
must also be noted, as in the previous examples, that diagrams IV, V, VI, and VII
each represent several possibilities. Thus, to apply the diagrams to the problem of
the gospels, diagram IV may represent the dérivation of Mark and Luke from Matt.,
or of Matt. and Luke from Mark, or of Matt. and Mark from Luke. The case is
similar also in respect to each of the other diagrams ; so that, in fact, they represent
eighteen possibilities.
It is obvious that ail the tests indicated in 4 above can be applied also in the
présent group of cases for the purpose of identifying the documents. Thus, if one of
the documents constantly shows marks of originality as compared with each of the other
two, and these two bear the signs of secondary character as compared with it, these
facts would be consistent with the solution suggested in diagram IV, the document
first mentioned occupying the position a; and, if not modified by other facts, would
point to this as the solution of the problem. If, however, to such a constant origi-
nality of a there be added also an originality of b, as compared with c, this would
suggest a diagram of the form VI, with the order, a—b—c. It is unnecessary to
follow out this phase of the problem in detail. An attentive inspection of the dia-
grams will suggest the various ways of applying to the problems of three documents
the tests which are applicable also to two documents.
It is of more significance to observe that the possession of three related documents
enables us to apply a more objective test than is possible when the documents are but
two in number.
Thus if two of the documents, say b and c, are so entirely distinct from one
anotlier as to suggest no interdependence, while the third, a, is a combination of élé-
ments drawn from b and c, such evidence will clearly point to hypothesis V as against
any of the other hypothèses enumerated. For under any other there would certainly
200Ernest DeWitt Burton
9
be material common to b and c as well as to a and b and to a and c. Of the eighteen
possibilities seventeen are thus excluded by a definite objective test.
If, on the other hand, a and b sometimes agréé against c, a and c sometimes agréé
against 6, and b and c sometimes agréé against a, and if ail these three forms of binary
agreement occur with approximately equal frequency, this fact will make strongly for
some form of hypothesis VII as against any of the others named above, since it alone
fnrnishes the conditions nnder which any two of the three documents may agréé against
the third. In such case the eighteen possibilities are by a definite objective test
reduced to four. An occurrence of ail three forms of binary agreement might indeed
arise under a modified form of hypothesis V, viz., if b and c, from which a is derived,
themselves hâve a common source x. But in this case we either hâve four documents,
or are no longer dealing with cases which exclude the influence of a non-extant source
and hâve passed into a new group, to be considered below.
For the further récognition of cases falling under hypothesis VII (or V in its
modified form), and for the identification of the documents, i. e., for determining which
is the dérivative document, there is again an objective test, long ago pointed out as val-
uable in textual criticism. Thus, if one document is based on two others, it may reason-
ably be expected that the compiler will sometimes combine the statements of his two
authorities, producing what is known as conflation. This, of course, will not occur
when the two are separately derived from one, though the appearance of it might
arise if the two dérivative documents should by chance divide between them a phrase
or sentence of the source. A classic ex ample of real or apparent conflation in the
gospels is Mark 1:32, o\jr(a<; 8e 7evotiévrjs, ore e8vaev 6 rçAxoç ; Matt. reading at the
corresponding point, 8:16, oyjrtas 8è yevo/jiévrjs, and Luke, 4:40, Svvovtos 8è tov rfKiov.
Now, if one of the three documents under considération should exhibit numerous and
clear cases of conflation, this would be practical démonstration of its dérivation from
the other two, i. e., that it is c in diagram VII (or a in V). To the test of conflation
may, of course, also be added those for direction of descent as set forth under 5.
If, once more, two of the three documents, let us say a and b, often agréé against
c, and a and c often agréé against 6, but b and c, while having much matter in common
with one another and with a, never agréé against a, it is certain that a is in some sense
the middle term between b and c, and we are pointed to hypothesis IV or VI: to IV in
the form indicated in the diagram, or to VI in the form b—a—c or c—a—6. Hypoth-
esis V is excluded by the existence of material common to b and c. Hypothesis VII is
excluded by the fact that the connection of b with c makes ail three forms of binary
combination possible and probable.2 Thus again by an objective test, based not upon
2 This general statement requires modification only by
the récognition of the fact tliat a writer in the position
indicated by c in diagram VII might conceivably départ
from either or both of his sources, when they were not in
agreement, but avoid departing from them in that to
which they bore concurrent testimony, and thus might
create a situation under VII very similar to that which we
hâve interpreted as pointing to IV or VI. The récognition
of this possibility would increase the uneliminated possi-
bilities from three to five, and require the application of
other tests, as, e. g., of conflation, for determining whether
the case really fell under VII on the one hand, or IV or VI
on the other.
20110 Principles of Literaby Critioism and the Synoptic Problem
délicate considération of the kind of changes that a first century author would prob-
able be disposed to make, but upon the obvious fact that of the three possible kinds of
binary combination two are frequent, while instances of the third are lacking, it is pos-
sible to exclude fifteen of the eighteen possibilités, and reduce the problem to discov-
ering which of the remaining three we actually hâve in hand. If, instead of an entire
absence of the third form of binary agreement, there are of this form only such
instances as may be reasonably ascribed to the coincident action of a common motive
affecting two of the authors, say b and c, and leading them to make the same change
in their source a, such a situation would point strongly to hypothesis IV, and at the
same time indicate which of the three documents was the source of the other two.
The problem of distinguishing between case IV on the one side, and any possible
form of VI on the other, is less simple. It can be solved, generally speaking, by
adding to the tests applicable to three documents those for direction of descent
applicable to two documents. Thus, if by the former test a has been shown to be the
middle term between the other two, a probable verdict on the question whether we
hâve case IV or a form of VI may be reached by observing which of the documents
shows marks of secondary character as compared with the others. Thus, if these
marks of secondary character appear in both b and c, as compared with a, this would
point to IV. If they appear in a as compared with 6, and in c as compared with a,
this would point to VI in the form b—a—c. If they appear in a as compared with
c, and in b as compared with a, this would suggest VI in the form c—a—6.
To discuss in further detail ail the problems suggested by the hypothèses named
is unnecessary. The methods employed will be similar to those already suggested.
The problem of three related documents, non-extant sources excluded, is then to
be solved by the application of the principle that two documents derived from a third
will each agréé with that source, but not with one another against the source, supple-
mented by the tests for direction of descent, and the possible evidence of conflation.
7. Thus far in discussing the problem of three documents we hâve ignored the
possibility that sources no longer extant hâve entered in as factors of the process from
which the three existing documents resulted. But this possibility must, of course, be
VIII	IX	X
taken into account. To represent or enumerate ail the many ways in which a non-
extant document or documents might hâve contributed to the existing resuit is neither
expédient nor necessary. It will suffice to consider a few of the many possible cases.
In case VIII the three extant documents are produced directly from a non-extant
202Ernest DeWitt Burton
11
source. In case IX, a is derived from the non-extant document æ, b from x and a,
and c from a and 6. In case X, b and c are each derived from a non-extant document
and the extant document a. In case VIII ail three classes of binary combinations
would appear as in case VII, and this case (VIII) can be distinguished from VII by
the fact that the marks of secondary character appear in ail three documents with
approximately equal frequency. In case X there would be material common to a, b,
and c, but also material common to b and c not found in a. In the former a and b
would sometimes agréé against c, and a and c against b, but agreements of b and c
against a would not occur, or would be rare and easily explicable. Marks of secondary
character would appear in b and c, but not in a. In case IX there would be material
common to a and b, but probably also material peculiar to each, and the marks of
secondary character in material common to them would sometimes appear in a, some-
times in b. Ail three forms of binary combination would be possible, but, perhaps as
in case VII, agreements of a and b against c would be less frequent than either of
the other forms, since the former could arise only through c departing from both
its sources when these were in agreement; c would be likely to bear the marks of
secondary character now in reference to a, now in reference to b, and very likely show
conflation of a and b. Total absence of such conflation, or the total disregard of mate-
rial found in a or b and germane to the purpose of c, especially of material common to
a and b (hence in both the sources of c) and likewise germane to the purpose of c,
would make against this hypothesis, and suggest some form of X, or some other theory
not included in our bnef illustrative list.
The methods applicable to this group of cases are therefore in general those
indicated under the preceding section. But the possible relations are indefinite in
number, and the particular method to be employed will vary with every practical case.
8.	Account must also be taken of the fact that the significance of agreement or
disagreement may be quite different according as it pertains to matter en bloc or to
details of threefold narrative. Thus, if in threefold narrative documents a and b often
agréé against c, and a and c often agréé against b, while b and c never agréé against a,
there is a strong probability that b and c are derived from a. But if in the same
documents b and c agréé—it may be exactly—in whole paragraphs not found in a, this
does not overthrow the conclusion previously reached, but rather points to the posses-
sion by b and c of a source additional to a. In other words, if to agreements of a and
b against c, and of a and c against &, there be added agreements of b and c against a
in details of threefold narrative, this points to hypothesis VII or VIII. But if this
latter agreement, b and c against a, be not in details, but only through the addition
of matter en bloc, X is the hypothesis indicated.
9.	It must also be évident that an agreement in omission is of quite different
significance from an agreement in addition. Thus in threefold material the common
possession by two documents of any considérable amount of material not found in the
third either shows that that third wTas not, at least in its extant form, the source of the
20312 Principles of Literary Critioism and the Synoptic Problem
other two (excludes IV), or nécessitâtes the supposition of an additional common
source (X). But the same amount of agreement in omission, or even a much larger
amount, might, if the evidence were otherwise clearly in favor of regarding the third
document as the source of the other two (as in IV), be explicable as due to coincidence
or the influence of the same motive upon two minds. The same principle would hold
respecting larger portions of material, except that the agreement of two documents
against the third in the common possession of paragraphs or sections capable of indepen-
dent transmission more obviously than in the other case suggests an additional source.
In brief, two writers might both hâve the same reason or different reasons for omitting
matter found in a third; but the addition of matter verbally the same and of any
considérable extent, whether of phrases in a threefold text, or of entire paragraphs or
sections, would require some other explanation.
To these principles, based on the relations of documents, may be added another,
respecting the value of ancient testimony, too obvious to require defense, but worthy
to be borne in mind.
10. Tradition cannot control the clear evidence of the documents themselves.
But a theory which accords with ancient tradition, especially uncontradicted tradition,
is more probable, other things being equal, than one which contradicts tradition. Of
very spécial significance is the testimony of an author himself respecting the sources
used by him or at his command. Such testimony can be set aside only when contra-
dicted by clear internai evidence. The most probable theory is that which conforms
alike to the internai evidence of the documents, to the testimony of the author, if such
testimony exists, and to ancient tradition.
II. FACTS RESPECTING THE RELATION OF THE SYNOPTIC GOSPELS TO-ONE ANOTHER,
AND THE BEARING OF THESE FACTS ON POSSIBLE THEORIES
To exhibit in detail ail the facts respecting the relation of the synoptic gospels to
one another would require the printing of the Greek text of these gospels in parallel
columns, together with some device for indicating to the eye the extent and nature of
their parallelism. Several attempts hâve been made to do this,3 none of them wholly
successful. lt is not the purpose of this essay thus to show the details, but to state
those general facts which a careful and detailed comparison of the gospels discloses,
and which, in the author’s judgment, point the way to a solution of the problem of
their mutual relation.
1.	It needs barely be mentioned that the synoptic gospels differ one from another
to such an extent as to make them clearly not exemplars of the same work, but
different, and to a certain extent independent, Works.
2.	There is in these synoptic gospels, on the other hand, a large amount of similar
material—a large number of sections which appear in two or in ail threeof the gospels
in closely similar form. In any numerical statement of the facts at this point there
3See Rushbrooke, Synopticon; Wright, Sy nopsis of gelien; Heineke, Synopsederdrei ersten kanonischen Evan-
the Gospels in Greek; Huck, Synopse der drei ersten Evan- gelien; Veit, Die synoptischen Parallelen, Vol. I.
204Ernest DeWitt Burton
13
must be a certain element of arbitrariness, since the length of a section is to a certain
extent a matter of personal judgment. General facts, however, will appear in the
following statements:	In the table appended to this essay there are 120 sections. In
49 sections there are three accounts so closely resembling one another as to indicate
some kind of literary relationship. In 1 section (64) there are three accounts, but the
Luke account is quite independent of the other two. In 15 sections closely parallel
accounts are found in Matt. and Mark ; in the case of four of these Luke has an inde-
pendent but more or less similar account, usually differently located. In 5 sections
parallel accounts are found in Mark and Luke (this number including three in which
the Mark passage is in 16:9-20). In 5 sections parallel accounts are found in Matt.
and Luke, if in this list we may include the généalogies. In 6 Matt. is the only
source; in 1 Mark is the only source; in 37 Luke is the only source, including the
four independent accounts mentioned above. In 1 Luke (13 :18-21) contains material
found also in Matt. in a threefold section.
3. In a large proportion of these cases the resemblance between the parallel
accounts is very close, extending to ideas, words, order of words, and even to the
insertion of parenthetical clauses. One or two examples will suffice to show the
character of this resemblance.
Matt. 3:7-10
Luke 3:7-9
*l8toi/ 8c 7roAAovç tu>v «ËapicratW Kai SaSSov/caitov
èp^o/ievouç «ri to pdimo-fm ciircv avtoÎç
Tewyfiara i^iSvtov, tlç vireSeiiev v/jlîv
<f>vyeîv a7ro rrjç p.eXXovo’rjç opyrjç ;
iroirjuaTt ovv Kapirov a£iov rrjs /xeravoiaç •
Kai piT] 8o£rjre Àcyciv cv cairroîs
Harépa €\opi€v tov ’Afipaap., A.cyto yàp vpXv on
Svmrai 6 Oeos ck tü>v XlBuïv tovtiûv
cycîpai TCKva tü> *Aftpaap.. rfàt] 8e
d^LVTj ITpOÇ TYJV pltflv T(ûv 8evSp(i)V K€ÎTCU •
irâv ovv SévSpov p.rj iroiovv Kapirov KaXov
€KKonrT€Tat Kal ctç nvp /SdXXerai.
*EAcyev ovv roîç €Kirop€vop.evois o^Aoiç fiairTKT0r}vai
vit avrov •
TewrjpLara i8va>v, nç vireSet^ev vplv
rfavyeîv (X7rô rrjç p,eXXovcrr)Ç opyrjç ;
irocrjarare ovv Kapirovs àitovs rfjs /icravoiaç •
Kai /irj dp$rj<rBe Xiytiv iv iavroîç
Ilarepa e^op.€V tov ’AfipaapL, Xkya) yàp vpüv on
Svvarai ô 0€os ck Ttbv XtBu>v rovrtov
eyetpai T€kva t<£ 'Afipadp.. rjbyj 8è
Kal rj à$Lvrj irpos tyjv pl£av rûv 8év8pu)v KCÎTai •
irdv ovv 8iv8pov p*r} iroiovv Kapirov ^/caÀ.oi'J
€KKOirT€TaL Kal cîç irvp ftdXXerai.
Matt. 9:14-17
totc irpocrép^ovrai aura) ol p.aBrj-
rai *Icodvov ÀcyovTeç
Aià ri ijficîç Kai ol &apiaaîoi
vr)<TTevop.ev,
ol 8è p.aOr)Tai <rov ov vrjtTrvoov-
Mark 2:18-22
Kai rjŒav ol p.aBrjral ’ltadvov Kal
ol &apiaaîoi vrj(TT£vovT€s.
Kal tpxovrai Kal Xéyovaiv avrco
Aià ri 61 p.aOrjTal ’lcoavov Kai ol
pLaOrjTal twv ^apiaaiwv vrjcrTev-
ovcriVf
ol 8c o-oi [padrjTaï] ov vrjcrTevov-
triv;
Lüke 5:33-39
oî 8c ci7rav irpos avrov
Oî pLaBrjral ’ltoavov vrjŒTevovcriv
irvKvà Kal SerjcreLÇ iroiowrai,
op.0i<oç Kai oi t<ûv <Papi(TaL(i)V
ol 8c aol icrOtovtnv Kai irwov&w
aiv;
20514 Principles of Literary Critioism and the Synoptic Problem
Kal ct7rcv aÛTOÎç 6 *1 rjvovç
M77 Sûvavrax ot viol tou vup<£üivoç
rrtvBtiv è<f)’ oo’ov pcT’ avra>v
€crrtv 6 vvp(f>ioç ;
èXcucrovrai Se rj pipai orav dirapBrj
a7r’ airojv 6 vvp<f>LOÇ, Kal totc
V7)(TTZV(TOV<nV.
ouSetç Se èniflaXXei è7rt/3Xrjpa
paKOVÇ âyva(f>ov ènl t/xarta)
7raAatco.
aipa yàp to 7r\rjp(opia avrov àno
TOV IpXLTLOVf Kat \€Îpov (T\i(rpa
ytverat.
oû8e fidWov&iv olvov viov eiç dcr-
kouç TraXaioûç *
et Se pijyc, prjyvvvrai ot do-Kot,
Kat 6 otvoç eK^etrat Kat ot acrKOt
a7roXXuvTat •
àXXà /3dWov<riv olvov viov etç
acrKOvs Katvouç,
Kat âp<j>oT€pot crvvrrjpovvrai
Kat €L7T€V aVTOLÇ 6 *1 rjfTOVÇ
Mr/ SuvavTat ot utot tou vup<£a>voç
ev <5 6 vvp<j>ioç pcT* auTtov eortv
VT/oreuctv ;
O (TOV xpovov €\OV<riV TOV VVfJL<j)LOV
/xer’ auTtov, ou SuvavTai vtjo’tcvclv '
êXeuo-ovrat Se rfpipai oTav dnapBrj
an auTtov o vvp<f>ioç} Kal totc
V7J(TT €V(T OVO’LV	€V €KCtVÿ Tÿ
WW-
ovSctç inif3\ripia paKovç âyva<j>ov
ènipanrei ènl ipdnov nakaiov *
et oe piÿ,
atpet ro nXrfpaipua 0^’ avrov ro
Kaivov tov naXaiov, Kal ^etpov
(r^tcr/xa ytverat.
Kat oûSetç fidXXei olvov viov et?
do"Koùç 7raXatouç *
et Se p77, piyfet o otvoç tous d<r-
KOUÇ,
Kat 6 otvoç ânoXXvrai Kal ot acr-
KOt.
[dXXà otvov vcov etç aa-Koùç Kat-
vouç]
ô Se ’lîycroûç curcv 7rpoç avrovç
M77 Swao’^e roùç uîoùç tou vup-
<j>S>voç èv <S 6 vupc^toç per’ aû-
tü>v etrrtv Trottât vi/oreûcrat ;
èXeuo’ovrat Se i^pepat, Kat orav
ànapBrj an auTtov o vvp<f>ioç
totc vyo’Tevo’ovaiv èv ckc tvatç
ratç rjpipaiç.
*EXcyev Se Kat napa/3oXr]V rrpoç
au roùç ort
OûSetç èni/3Xr)pa àno Ipuxrtov
Kaivov cr^to’aç inif3d\Xei €7rt
ipariov 7raXatov •
> /
et oe piyye,
Kat to Kaivov cr^tcrei Kat ra> 7ra-
Xatu> ou o’vpÿiavrjo'ei ro €7rt-
fi\rjpa ro a7ro tou Kaivov.
Kal oûSetç )8aXXet otvov véov etç
âaKoùç 7raXatouç •
et Se p^yc, prj$€ 1 b otvoç 6 veoç
TOUÇ (XOTCOUÇ,
Kat auToç èK^u^aeTat Kat ot acr-
KOt anoXovvTai
aXXà otvov veov etç do’Koùç Kaivovç
j3\yTeov.
[OûSetç 7rtcov nakaiov Bikei viov ■
Xéyet yàp O TraXatoç Xprj&TOç
eo-Ttv.]
The first of these examples shows this resemblance at its greatest. Of the sixty-
three consecutive words in Matt. beginning with yewrjpara and ending with /3aXXerat,
sixty are found in Luke in the same order; for fcapirov âÇiov Luke has Kapirovç àÇi'ovç,
for SoÇrjTe he has âpÇrjaOe, and after ^£77 Se he adds /cat; for /caXoV in Luke the evidence
is not quite conclusive. In the second example the resemblance, especially between
Matt. and the other accounts, is less close, y et still very striking. The student of the
subject does not need to be told that instances of similarly close parallelism are very
numerous. They may be studied in any Greek harmony of the gospels, especially
Rushbrooke, Huck, or Wright. The resemblance is at many points doser than is
ordinarily found between the quotations of the New Testament writers and their Old
Testament source, and closely resembles that whichexists between Tatianand his sources,
the latter, of course, well known to be written and in our possession (see under 8).
206Ebnest DeWitt Bürton
15
4. In order to observe more closely the facts respecting the relation of the
synoptic gospels, the contents of these gospels may be classified into seven classes
(somewhat after the manner of the canons of Eusebius), according as they are
(a) threefold, being found in ail three gospels; (b) twofold, being common to
Matt. and Mark; (c) twofold, being common to Mark and Luke; (d) twofold, being
common to Matt. and Luke; (e) peculiar to Matt., i. e., omitted by Mark and Luke;
(f) peculiar to Mark, i. e., omitted by Matt. and Luke; (g) peculiar to Luke, i. c.,
omitted by Matt. and Mark. This classification may be applied first to the material
en bloc—i. e., to sections, paragraphs, or portions of paragraphs—such that they might
not improbably be transmitted independently;4 and secondly to the several portions,
even to single words or terminations, of what in the previous classification is reckoned
as threefold narrative.5
The general facts regarding the agreement of the synoptic gospels in respect to
whole sections, or paragraphs, or considérable portions of paragraphs are as follows:6
a.	The threefold material consists of the following passages of Mark with the
parallel passages in the other gospels : 1:2-4, 7-15, 29-34, 39-45; 2:1—3:12;
3:16-19; 3:22-4:25; 4:30-32; 4:35—5:43; 6:6b-16, 30-44; 8:27—9:8; 9:14-37,
42; 10:13-34, 46-52; 11:1-11, 15-19, 27-33; 12:1-39; 13:1-20, 24-32; 14:1, 2,
10-25, 29-50,53-72; 15:1-15, 21-47; 16:1-8.
b.	Mark and Matt. agréé in including the following passages not in Luke : Mark
1:5, 6, 16-20; 4:33, 34; 6:l-6a, 17-29; 6:45—7:31; 8:1-21; 9:9-13, 43-48;
10:1-12, 35-45; 11:12-14, 20-25; 13:21-23; 14:3-9, 26-28; 15:16-20 = Matt.
3:4-6; 4:18-22, 24, 25; 5:29, 30; 9:35, 36; 13:34, 54-58; 14:3-12, 22-27;
14:32—15:11; 15:15—16:11; 17:9-13; 18:8, 9; 19:1-9; 20:20-28; 21:18-22;
24:23-25; 26:6-13, 30-32; 27:27-31.
c.	Mark and Luke agréé in including the following not in Matt.: Mark 1:21-28,
35-38; 3 :13-15; 9:38-41; 12: 40-44 = Luke 4: 31-37, 42, 43; 6:12,13; 9:49, 50;
20:47—21:4.
d.	Matt. and Luke agréé in including the following matter not in Mark: Matt.
3:7-10, 12; 4:3-11; 5:1-3, 5, 6, 11, 12, 25, 26, 32, 39-42, 44-48; 6:8-13, 19-33;
7:1-5, 7-12, 16-27;7 8:5-13, 18-22; 9:37,38; 10:15, 16,24-40; 11:2-19, 21-27;
12: 27, 28, 30, 33-35, 38-45; 13:16, 17, 33; 18:12-15, 21, 22; 23:4, 12, 13, 23-39;
24:26-28, 37-51; 25:24-29 = Luke 3:7-9, 17; 4:3-13; 6:20-23, 27-49; 7:1-10,
* The length of the portion which might be trans-
mitted independently would dépend somewhat upon its
character. Even a comparatively brief saying might be
handed down without connection ; but a detail of a narra-
tive if transmitted must hâve corne down as a part of a
story.
5 On the differing significance of agreements on the
one hand in details of threefold narrative, and on the other
in matter en bloc, see I, 8, p. 11.
sThere is necessarily some room for différence of
opinion as to precisely how much should be reckoned as
threefold material, and precisely of what nature and
extent a twofold agreement must be to exclude it from the
list. The above list of passages could doubtless be criti-
cised both as including too much and as excluding too
much. It is believed to be at least approximately correct.
7 Concerning the parallel matter in Matt., chaps. 5-7,
and Luke 6:20-49, see more fully under 10 below. Some
verses of Matt. are included in the above list which are
only partially paralleled in Luke,
20716 Pbinciples of Litebary Cbitioism and the Synoptic Pboblem
18-28, 81-35; 9:57-60; 10:2-15, 21-24; 11:2-4, 9-13, 19, 20, 23-26, 29-32, 34,
35, 39-42, 44-52; 12:1-9, 22-34, 39-46, 51-53, 58, 59; 13:20, 21, 28, 29, 34, 35;
14:25-27, 34, 85; 15:4-7; 16:13, 16; 17:3, 4, 23-27, 34^37; 19:20-27. This
material which Matt. and Luke possess in common, but do not share with Mark,
is partly narrative in character, partly discourse material. The resemblance of form,
words, sentences, order of sentences, is for the most part very close. An example of
it may be seen in Matt. 3:7-10 and Luke 3:7-9 printed on page 13. In location,
on the other hand, there is very little agreement, the only instances in which corre-
sponding material can be strictly said to be correspondingly placed being the
section just referred to, the account of the preaching of John the Baptist, where
Matt. 3:7-10, 12, and Luke 3:7-9, 17 contain the same material and are similarly
placed, and the temptation story, in which Matt. and Luke closely agréé in their
additions to Mark except in the order of the second and thiud temptations.
e.	The matter peculiar to Matt., which therefore Mark and Luke agréé in
omitting (i. e., in not containing), is the following:	Matt., chaps. 1; 2; 4:13-16;
5:4, 7-10, 13-24 (but with partial parallels in Luke), 27, 28, 33-37; 6:1-7, 14-18
(with partial parallels of 14,15 in Mark and Luke), 34; 7:6, 13-15 (with partial paral-
lels in Luke); 8:17; 10:5, 6, 23, 41; 11:28-30; 12:17-21, 36, 37; 13:24-30, 35-53;
14:28-31; 15:12-14; 16:12, 17-19; 17:24-27; 18:10, 16-20, 23-35; 19:10-12;
20:1-16; 21:4, 5, 14-16, 28-32; 22:1-14; 23:2, 3, 5, 8-10, 15-22; 24:10-12;
25:1-23, 30-46, 52-54; 27:3-10, 19, 24, 25, 51b-53, 62-66; 28:2-4, 9-20.
f.	The matter peculiar to Mark, which therefore Matt. and Luke agréé in omitting
(i. e., in not containing), is the following: Mark 3:20, 21; 4:26-29; 7:32-37; 8:22-
26; 9:49, 50; 13:33-37; 14:51, 52.
g.	The matter peculiar to Luke, which therefore Matt. and Mark agréé in omitting
(i. e.. in not containing), is the following: chaps. 1 ; 2; 3:5, 6,10-15,18-20; 4:16-30;
5:1-11; 6:24-26; 7:11-17,29, 30, 36-50; 8:1-3; 9:51-56, 61, 62; 10:1, 16-20,
25-42; 11:1, 5-8, 27, 28, 36-38, 53, 54; 12:13-21, 35-38, 47-50, [54-57]; 13:1-17,
22-27 (with partial parallels in Matt.) 31-33; 14:1-24, 28-33; 15:1-3, 8-32; 16:1-12,
14, 15, 19-31; 17:5-22, 28-32; 18:1-14; 19:1-19, 39-44; 21:34-38; 22:15-17;
24-32 (but with partial parallels in Matt.), 35-38; 23:4-16, 27-32, 40-43; 24:7-53.
5. When there are three parallel accounts—i. e., in the matter referred to in 4a—
Matt. and Luke resemble each other much less closely than either Matt. and Mark, or
Mark and Luke. In a very large prépondérance of the agreements of Matt. and Luke
they resemble one another only in so far as both agréé with Mark. Beyond this their
agreements consist only in the occasional omission of matter found in Mark, and the
occasional agreement in a single word or brief phrase not found in Mark. The facts,
reduced to numerical statement, are as follows:
a.	Matt. and Mark agréé against Luke by addition or substitution in approxi-
mately 1,600 words.
b.	Mark and Luke agréé against Matt. in approximately 860 words.
208Ernest DeWitt Burton
17
c.	Matt. and Lufee agréé against Mark in approximately 2758 words.
A more careful scrutiny of this third and smaller group of agreements discloses
several facts which tend still further to emphasize the disparity of these numbers.
The 275 words of agreement are distributed in about 175 instances, from which it
appears that they average less than two words each. Of these 175 instances, 15 con-
sist in the substitution of ehrov for Xefy», about 20 of hé for tcai., 2 of /ca( for hé.
A comparison of Mark with each of the other gospels shows that in many instances
the substitution of ehrov for Xéyo> and of tcai for hé appears also as between Mark and
each of the others singly. This fact indicates that in these 35 instances we hâve simply
the coincident effect of causes which affected both Math and Luke alike. In about 20
of the 175 instances in which Matt. and Luke agréé (wholly or in part) against Mark,
the three agréé in that they use words of the same root, Matt. and Luke employing a
form diifering from Mark’s in prefix or termination. Many instances of change in
which Matt. and Luke agréé are explicable as due to a common impulse of Matt. and
Luke to improve Mark’s Grreek, as, e. g., by the substitution of an aorist for an historical
présent (Mark 4:38; 5:15; 11:1; 11:7—cf. 15:20—and parallels), or the participle
for a finite verb with kcl( (Mark 1:41; 4:38; 5:38; 6:7; 11:2; and parallels), or to
conform the statement more exactly to the facts as understood by them (cf. Mark 6:14;
8:31; 10:34 with the parallels). There remain, however, a considérable number of
additions and substitutions which are of a different character. But of these instances
scarcely more than one in ten causes a différence in meaning between the several
accounts, and these affect only the merest details. The common additions of Matt.
and Luke to Mark which affect the sense of the passage further than by the change
of tense or an unimportant exchange of prépositions are as follows:	Mark 1:5 =
Matt. 3:5 = Luke 3:3,5 words; Mark l:8 = Matt. 3:ll = Luke 3:16, 2 words;
Mark 2:12 = Matt. 9:7 = Luke 5:25, 4 words ; Mark 3:1= Matt. 12:9 = Luke
6:6, 1 word; Mark 3:18 = Matt. 9:2 = Luke 6:14, 3 words; Mark 4:10 = Matt.
13:10 = Luke 8:9, 1 word; Mark 4:41 = Matt. 8:27 = Luke 8: 25, 2 words; Mark
5: 27 = Matt. 9:20 = Luke 8:44, 2 words; Mark 6:7 = Matt. 10:1 = Luke 9:1, 1
word ; Mark 6:34 = Matt. 14:14 = Luke 9:11, 2 words (?) ; Mark 9:2 = Matt.
17:2 = Luke 9:29, 2 words ( ?) ; Mark 13:19 = Matt. 24:21 = Luke 21: 23, 1 word ;
Mark 14:62 = Matt. 26:64 = Luke 22:69, 2 words; Mark 14: 65 = Matt. 26:68 =
Luke 22:64, 5 words; Mark 14: 72 = Matt. 26: 75 = Luke 26:62, 4 words; in ail,
15 instances, 37 words.
d.	Peculiar to Luke in threefold matter—i. c., omitted by Matt. and Mark—are
approximately 1,100 words, for which there is no équivalent or substitute in the
other gospels.
e.	Peculiar to Matt. in threefold matter—i. e., omitted by Mark and Luke—are
approximately 830 words.
8 These figures are based upon the Tischendorf text as cott and Hoet, would probably differ slightly, but could
printed in Huck’s Synapse der drei ersten Evangelien. The hardly do so materially.
results of a count from another text, such as that of West-
20918 Principles of Literary Criticism and the Synoptic Problem
f. Peculiar to Mark in threefold matter—i. e., omitted by Matt. and Luke—are
approximately 1,000 words. Kespecting the matter of omissions in threefold matter
there is, however, much room for différence of opinion, and the figures given under
d, e, f must be regarded as approximate only.
6.	Respecting agreements and disagreements in order the facts are these :
Matt. and Mark agréé against Luke in the placing of two sections in which the
narrative is evidently threefold:	The true kindred of Christ, recorded in Mark 3:31-35
and Matt. 12:46-50 immediately preceding the parables by the sea (Mark 4:1-34;
Matt. 13:1-53), in Luke follows these parables; the imprisonment of John the Baptist,
recorded in Mark 6:17, 18 and Matt. 14:3, 4 in connection with the results of the
missionary journey of the Twelve, is given by Luke at the close of his account of the
preaching of John, Luke 3:19, 20. In the arrangement of paragraphs within a section
Matt. and Mark agréé against Luke in the account of the last supper and in the narrative
of the trial. Besides these instances there are three in which Luke, though recording
an event similar to that of Mark and Matt., evidently gives a wholly independent
account unrelated in a literary way; and one in which Luke’s account is, in the main
at least, independent of Mark, and Matt. is partly parallel to Mark, partly to Luke.
These passages—Luke 4:16-30; 5:1-11; 7:36-50 and 11:14-32—do not concern
us at this point.
Mark and Luke agréé against Matt. in the location of thirteen sections, which lie
between Matt. 4:23 and 13:58. Within these limits there are certain groups of two or
three sections the sections of which succeed one another in the same order as in Mark and
Luke, but the groups themselves are differently located. In respect to the narratives
which précédé and follow these limits, Matt. agréés with Mark in the order of sections
except in the transposition of the conversation between Jésus and his disciples con-
cerning the withered fig tree to a place in immédiate connection with the cursing of
the tree. But as Luke omits both of these sections, the transposition does not resuit
in a disagreement of Matt. with both Mark and Luke.
Matt. and Luke never agréé against Mark in order of sections or paragraphs.
7.	Of the ancient testimonies bearing upon the origin of the synoptic gospels, it
must sufiice to présent a few which are of spécial significance and importance.
First among these is the préfacé of Luke’s gospel:
*E7T€t87y7T€/) 7roAÀoi è7re\etpr](Tav OLvardèatrOat Sirjyrjcnv irepl twv 7re7r\rjpo<j>oprjpiév(i)v ev ffpuv irpayp.a-
tù)v, Ka0à)ç irapéSocrav Tfp.lv ot chr’ àpxrjs olvtotttou kcll V7rr)p€Tcu yevopevot rov \6yov, e8o£e Kapol
7raprjKoXov0rjKOTL avuiOev iràcriv (XKpif3u)<s KaOetjrjs cr oi ypaij/ai, k par terre ®eo<f> iÀ.e, ira imyvws irepl wv
Karrjxqôris Xoywv rrjv âcr<f>d\etav.
This passage bears for us the important testimony of the author of this gospel
that when he wrote there were already in existence several narratives of the life of
Jésus, more or less complété, and that these narratives were based, in the intention of
their writers at least, on the oral narratives of the life of Jésus which proceeded
from the personal companions of Jésus, men who had witnessed the events from the
210Ernest DeWitt Burton
19
beginning, and from the beginning had been ministers of the word. The author
thus implies that his writing was subséquent both to the formation of an oral tradition
and to the putting forth of not a few written gospels based upon this oral testimony of
the eyewitnesses. He does not definitely state of which of these sources he had made
use in his work, but he expressly affirms the existence of both the oral tradition and
the written gospels, and implies that both were accessible to him. It is instructive to
observe that while the author includes in his own gospel a story of the infancy, he yet
implies that the oral gospel and the writings that preceded his were coincident in scope
with the public life of Jésus; in other words, were of the same general extent as the
gospel of Mark. They who from the beginning were eyewitnesses and ministers of
the word are one class not two. “From the beginning” must therefore mean from
the beginning, not of the life of Jésus, but of his ministry, since only from the latter
time were there ministers of the word. Their testimony, therefore, and the gospels
based thereon, covered only the public life of Jésus.
Second among the important ancient testimonies is that of Papias, transmitted by
Eusebius in the foliowing passage (H. E., iii, 39:)
Kat [ïlamaç] aAXaç 8c rfj 18ta ypacjrrj irapahlhuxriv ’ApioriWoç tov 7rp6cr6ev 8e8r}X<i>flévov twv tov
Kvptov Xoy<ov 8irjyy(T€L<s, Kal tov irptafivrépov ’lcoavvov 7ra/oa8o<rciç • c<£* aç tovç <£iAo/xa0etç àra7r€/xi//avTeç,
<zvayKcuü>ç vvv irpoa,Brj<Topt€V ratç 7rpocKTc0€icraiç avrov <f>wvaTç 7rapd&oo-iv, y irepl Map*ov tov to evayyc-
Xiov ycypa<£oTO$ c#CT€0ciTai 8ià tovtwv *
Kat tovO’ 6 7rp£ar/3vTcpo<; cAcyc. Mci/okoç p.\v €pp.r}vevTr)S Ilcrpov ya/o/xcvoç, ocra ipvrjpiovevo’tv,
aK/atjSok cypcuj/ev, ov piévroi rd£u Ta vtto tov xpicrov rj Xe^Bévra rj rrpayBivTa. Ovtc y dp yKOvat tov
Kvptov, ovtc rraprjKoXovBrjaev avt<o • varepov 8c, wç cÏIct/xü, oç 7rpos Tas xpeiaç cttoicÎto Taç 8i8acr#ca-
Atiaç, àAA’ ov^ cüo’7rep (rvura^cv twv KvptXLKiàv 7roiov/xevoç Xoyaiv • tvore ov8cv rjpapTt Map/coç, ovtcdç Ivia
ypaxf/as aïs âir€p.vrjpi6v€vcr€.v. *Evoç yàp CTroo/craTO rrpovouiv, tov prjSkv <ov tjkov(T€ wapaXureîv, rj i/'evcrao'-
Bat ti cv avToîç.
TavTa piev ovv ifTToprjTai t<o IIa7rta 7rcpt tov MapKOV. Ilcpt 8c tov MaT^atov tœvt eiprjTai •
MaT0aîoç pikv ovv *E/?pat8i ScoAckto) Ta Aoyta o’vvcypai/'aTO. ‘Hp/xrçvcvo'c 8’ avVa, tvç rjv Svrnroç,
€Ka(TTOS.
The assertions of this passage which are of particular significance for our présent
purpose are two. First, Papias affirms, on the authority of the Elder, that Mark
wrote a book containing an account of the things that were said and done by the
Lord, upon the basis of the testimony of Peter, the latter of course being, as Mark
was not, an eyewitness of the events of Jésus’ life. There can be but little doubt that
Eusebius, in identifying the gospel thus written with the gospel of Mark current under
that name in his day, and, we may add, in our own day, was doing what Papias did
before him. We cannot, indeed, assume without proof that the second and fourth
centuries were right in this identification. But it is évident at the least that the
statement of Papias that Mark wrote a gospel based directly on eyewitness testimony
reflects the belief of the second century.
But secondly, Papias also testifies that Matthew wrote a book in the Hebrew
21120 Principles of Literary Criticism and the Synoptio Problem
(meaning either Hebrew or Aramaic) language, consisting of the “sayings” (\071a).
There can be no reasonable doubt that rà Xoyia means in this connection the ntterances
of Jésus. That it was written in the Hebrew language was apparently known to
Papias through tradition, not from his own possession of it in that language ; for the
time when it was necessary for each (Greek) reader to translate it for himself is
spoken of as past. Papias therefore intimâtes a distinction between the gospel that
Matthew wrote and the one that was commonly current in his day and circle, though
not suggesting that the two differed otherwise than in language. It would be
instructive, but would unduly prolong this portion of the paper, to examine the testi-
mony of later writers to the existence even at a later time of Hebrew gospels with
which was associated the name of Matthew, and from such testimony to deduce a conclu-
sion as to how mue h an ancient writer meant when he seemingly identified two Works.
8. Of kindred significance with the ancient testimonies concerning the rise of the
gospels is the evidence respecting the literary methods of the period in which the
gospels arose. Here again it is impossible to marshal ail the evidence. It must
sufïice to refer briefly to three items of spécial significance.
The first of these is contained in the New Testament itself and in no small part
in the gospels, being furnished by the quotations which New Testament writers made
from the Old Testament. These quotations are certainly in a large number of cases
made from the Septuagint version; and, despite ail uncertainties respecting the text
of that version and of the New Testament, it is possible by comparison of the two to
détermine with approximate certainty how accurately a New Testament writer felt con-
strained to quote the text of sacred scripture, which existed in written form and was
presumably accessible to him if he saw fit to refer to it. Such comparison will show
that, while the Septuagint text is often reproduced with almost literal exaetness (see,
e. g., Matt. 13 :14, 15, quoted from Isa. 6:9, 10; Acts, 2:17-21, quoted from Joël
2 :28-32; Rom. 4:7, 8, quoted from Ps. 32:1, 2), yet, on the other hand, the New Testa-
ment writer often allowed himself no little liberty of quotation. Sometimes, indeed, the
variation from the Septuagint is due to the author’s independent use of the Hebrew,
but in other cases he départs alike from the Hebrew and the Septuagint (see, e. g.,
Matt. 2:6, quoted from Mic. 5:2; Rom. 3:14, from Ps. 10:7; Rom. 3:15-17, from
Isa. 59:7, 8; Matt. 2:23, from Isa. 11:1 [?]); sometimes when employing nearly the
same words, he transposes the phrases (as in Matt. 21:33 and Mark 12:1, quoted from
Isa. 5:1, 2, in which the agreement of order between Matt. and Mark against the
Septuagint is suggestive, but not exceptional), or combines quotations from different
passages (as in Rom. 3:10-18, and more notably in 1 Pet. 2:7, 8; Rom. 9:33).
A second instructive illustration is furnished by the appendix to the gospel of
Mark. It is the almost unanimous verdict of textual critics that Mark 16:9-20 is an
addition by a later hand than that which composed the body of the gospel.9 Whether
9 See Conybeare, in Expositor, Vol. IV, No. 8, p. 241; Schluss des Markusevangeliums ; Zahn, Einleitung in das
Vol. IV, No. 10, p. 219; Vol. V, No. 2, p. 401; Rohrbach, Der Neue Testament, 2d ed., Vol. II, pp. 227-35.
212Ernest DeWitt Bubton
21
it is from the hand of Aristion is not significant for our présent purpose, since it is at
any rate later than the rest of the gospel. It is scarcely less évident that it makes use
of the concluding portions of the other gospels, being in part produced from them by
ver y free condensation, in part enlarged by additions from other sources. The situa-
tion is of course exceptional ; but the verses furnish us none the less an example of the
freedom with which a writer, whose work eventually found a place in our gospels,
employed sources that beyond ail reasonable doubt were in his hands in written form.
But a still more instructive illustration of the literary method of that âge is
furnished in the Diatessaron of Tatian prepared by an Assyrian Christian about 175
A. D. From our four gospels substantially as we now hâve them, Tatian with scissors
and paste constructed a new composite gospel, which came into common use in the
churches of Syria, till in the fifth century Rabbula, bishop of Edessa, and Theodoret,
bishop of Cyrrhus, removed this composite work from the churches, putting in
its place the separate gospels. In this case, we are dealing, not with scattered
quotations occurring in the midst of what is in the main an independent work,
nor with an appendix added to complété what was evidently felt to be an incomplète
gospel, but with a work which, on the one hand, was composed wholly of material
already extant, and, on the other, was intended to be a complété work in itself. The
production of the Diatessaron from the extant canonical gospels was, moreover, simply
a further step in the process the earlier stages of which are reflected in the préfacé of
Luke. The sources of the composite book we not only know to hâve been written,
but we hâve them in our hands, and are able therefore, despite the fact that our
only complété witness to Tatian’s text is an Arabie translation, and that the con-
firmatory testimony cornes to us through the medium of other languages than the
original Greek, to compare the résultant work with the sources. It is, of course, a
problem in textual criticism to détermine with accuracy the précisé text of this work
as it left Tatian’s hands, as well as to ascertain the précisé form of the text of the
several gospels which he used. But any uncertainty upon these matters affects only
the details of the problem, leaving its main lines clear and distinct. The exhibition
in detail of the relation of the Diatessaron to its sources is itself a subject for an
extended paper, and cannot be undertaken here. But the student who will make the
necessary examination will find that, while Tatian clearly possessed and used our four
gospels as his sources, and had no others of importance, and while he evidently
intended in general to include every event, discourse, and saying found in the four
gospels,10 and in general to record each such event and discourse but once, omitting
duplicate accounts, and to retain the paragraphs in substantially their original form,
yet there are examples of almost every possible kind of modification of his sources,
including addition, duplication, omission, conflation, modification of grammatical
10 The only considérable omissions are the préfacé of used by Tatian. For a full exhibit of the facts respecting
Luke and, according to the prépondérance of evidence, the	Tatian which bear upon the problem discussed in this
généalogies. The story of the woman taken in adultery	paper see Hobson, The Diatessaron of Tatian and the
and Mark 16:9-20 were without doubt not in the sources Synoptic Problem (Chicago, 1904).
21322 Principles of Literary Criticism and the Synoptic Problem
form, substitution of équivalent terms, and especially of transposition, extending to
words, phrases, sentences, paragraphe and events. It is évident, indeed, that Tatian
framed for himself a scheme of the life of Jésus, and fitted into it the material of the
four gospels, freely changing, in order to accomplish this, the location of events even
when the gospels definitely marked the time, and building up new discourses out of
those which he found in the gospels. He shows no marked tendency either to
abbreviate or to expand the individual paragraphs which he dérivés from his several
sources.
Tatian’s work, not to go further into detail, makes it wholly clear that in the
second century, in which tliere is every reason to believe that the four gospels were
held in even greater reverence, and had more nearly attained a position of finality,
than was the case in the first century respecting the many gospels of which the préfacé
of Luke speaks, it was still possible both that the method of constructing books out of
other existing books, which we know from such examples as the Book of Enoch and
the Teaching of the Apostles to hâve been common in that âge, might be applied to
the gospels, and that a gospel so constructed might even displace, in certain régions
at least, the older and more directly apostolic sources of such a composite gospel. It
cannot, of course, prove that Luke and Matthew were produced by the same process,
but it does forbid us to deny the possibility of their production by such a process on
the ground that it was contrary to the literary method of the âge, or that reverence
for the sources would hâve prevented it.
These facts enable us to test a considérable number of hypothèses respecting the
relation of the gospels to one another. Thus, to represent the matter graphically,
they tend to show the correctness or incorrectness of ail the following views :
G
Matt
A
Mark
Y
Matt.
Luke
B
Mark
Luke
Matt.
c
Matt.
Luke
Mark
F
Luke
Mark
Matt.
Mark Luke
Mark
Luke
Matt	Luke
Mark
Matt.
Matt.
214Ernest DeWitt Burton
23
KLM	N
Hypothesis M, standing apart in some respects from ail the rest, may be con-
sidered first. It is in favor of it that it recognizes the existence of an oral tradition
antedating ail written gospels, which is rendered probable by the préfacé of Luke, and
that it accounts in general for the variations of the gospels from one another in three-
fold material and for the différences of order. But it is against it that it fails to
account for that large amount of close verbal agreement which exists between the
gospels (c/. 3 above, p. 13) ; that it very unsatisfactorily accounts for the remarkable
agreement between the gospels, especially between Mark and Luke, in order of sections
(c/. 6, above, p. 18); and that it furnishes no explanation of the relatively small
amount of agreement of Matt. and Luke against Mark in threefold matter (c/. 5 c, above,
p. 17). According to any hypothesis which predicates a common source for ail three
gospels, the three kinds of binary agreements against the third document should be
not conspicuously unequal. It is, furthermore, an objection to this theory that it
ignores the existence of gospel writings older at least than the latest of our synoptists,
which is attested by the préfacé of Luke, and confirmed by the statement of Papias
concerning the \oyia-collection made by Matt. (c/. 7, above, p. 19), and the strong
probability, in view of the literary methods of the time (c/. 8, pp. 20 ff.), that such earlier
writings would be employed by our evangelists. Still other objections to this view
will appear at a later point. But those already stated, even if they are not themselves
fatal, are sufficient to lead us to inquire whether there is not another hypothesis which
is less open to objection.
Of the other hypothèses, A, B, C, E, G, and I are excluded by the facts mentioned
under 5a, b; A, E, and G, because they exclude any large number of significant
agreements of Mark and Luke against Matt. (c/. 5 b, p. 16); and B, C, and I, because
they exclude any considérable number of such agreements of Mark and Matt. against
Luke (c/. 5 a, p. 16), whereas such agreements are, in fact, very numerous throughout
threefold material. Hypothèses D and F are shown to be inadéquate explanations
because they fail to account for the large amount of material occurring in whole sec-
tions or paragraphs found in both Matt. and Luke, but not in Mark. J, K, and L are
shown to be inadéquate by failing to account for any of the material common to ail
three of the gospels; for any of that which is common to the two documents from
which the third is ex hypothesi derived ; and for material peculiar to the dérivative docu-
ment. J and K are still further open to the objection that the résultant gospel (in J,
Matt. ; in K, Luke) must in a large number of instances hâve departed, in detail of
21524 Pbinciples of Literary Criticism and the Synoptic Problem
threefold narrative, from both their respective sources, even when these were in agree-
ment. Against N is the significant fact (already urged against M) of the relatively
small amount of agreement of Matt. and Luke against Mark in threefold narratives, as
compared with the agreement of Matt. and Mark against Luke, and Mark and Luke
against Matt. Before deciding, however, whether hypothesis N is to be excluded, it
will be well to compare it with certain other hypothèses which are obviously suggested
by the fact that the evidence already examined seems to suggest for Mark a position
as in some sense the middle term between Mark and Luke.
Dismissing for the présent ail other hypothèses, let us compare hypothèses H and
N. The question as between these hypothèses may be put in this form:	Is Mark the
source of Matt. and Luke so far as concerns the threefold narratives, or hâve these
narratives a common source different from any one of them ? If the latter is the case,
wherein did this common source differ from Mark? To answer these questions let us
refer again to the facts stated under 4 and 5 respecting the nature of the agreements
of Matt. and Luke against Mark. It has been pointed out there that the positive
agreements of Matt. and Luke against Mark in threefold narrative are about one-
third those of Mark and Luke against Matt., and one-sixth of those of Matt. and Mark
against Luke; that the agreements of Matt. and Luke against Mark in omissions,
amounting to about 1,000 words, a little less than one-tenth of the gospel, are a little
less than those of Mark and Matt. against Luke, and a little more than those of
Mark and Luke against Matt. ; and that there are no agreements of Matt. and Luke
against Mark in order of sections or paragraphs. It has also appeared that an impor-
tant fraction of the positive agreements of Matt. and Luke against Mark are explicable
as due to a motive for changing Mark by which Matt. and Luke might both be affected,
and that of the total number of such agreements only fifteen cause a material différence
between the several accounts. We hâve already seen that the list of entire passages in
the omission of which Matt. and Luke agréé against Mark is a brief one. See 4 f.
From these facts it will appear that if we choose hypothesis N in preference to H,
the document x which we shall then construct as the common source of Matt., Mark,
and Luke will differ in no important respect from the présent gospel of Mark. Upon
the supposition that Matt. and Luke could not independently agréé in the omission of
occasional sentences and phrases, and five or six short passages, the latter aggregating
about twenty verses, and that the text of Matt. and Luke, as we possess them in our
best critical éditions, has suffered no harmonistic corruption, we should indeed be
compelled, in order to restore x, to add to the présent Mark some 15 phrases aggre-
gating 37 words; to subtract about 1,000 words occurring here and there throughout
the gospel, and some twenty verses occurring in five detached passages, and in 240
words to make slight changes, largely inflectional and not affecting the sense. In
no instance should we transpose any of the material from its position in the présent
gospel of Mark.
But, in fact, neither of the above assumptions is justified. The omission by
216Ernest DeWitt Bürton
25
Matt. and Luke of 1,000 words found in Mark in'sections which, as sections, appear
in ail three gospels, seems at first a weighty fact. But an examination of this material
shows that in a relatively small proportion of cases are the facts best explained by
assuming the absence of this material from the source. In very few cases is the
departure of Matt. and Luke from the Mark narrative the resuit of simple omission of
the same phrase or sentence ; the concurrent omission is only the common résultant of
processes of modification—condensation or free reproduction—which are in them-
selves quite different in the two cases. Thus for example in Mark 11:15-17 and
parallels,' Matt. and Luke both hâve a shorter account than Mark, Matt., however,
passing over Mark’s vs. 16, Luke omitting 15b (longer than 16) and 16. Mark, chap.
5, and parallels, furnish a still better illustration. Matt. and Luke both contain a
shorter account than Mark and coincide in the omission of over 100 words, yet never
effect the omission in the same way. The changes thus made by the two evangelists,
which resuit in a long list of words omitted in common, are moreover entirely similar
to those which appear in passages in which, there being but two accounts, of which
Mark is one, the other gospel, Matt. or Luke, gives a shorter account than Mark ; or
in cases in which, there being three accounts, only one, either Matt. or Luke, is abbre-
viated. See for example Mark 8:30-33; 9:9, 10, 38-41; 10:32-34; 13:4-23, 28-32,
and the parallels. In short the Matt.-Luke coincidences in omission of material
found in Mark are for the most part explicable as the resuit of causes which evidently
affected Matt. and Luke alike, but independently. There remain indeed a few
instances which are not naturally so explicable. Among these are Mark 1:2b, in
which the agreement of the form of the quotation with that in Matt. 11:10 and Luke
7:27, and that too precisely in those particulars in which Matt. and Luke départ from
both the Hebrew original and the Septuagint version, together with the improbability
that if Matt. and Luke derived this quotation from Mark rather than the converse
they would hâve agreed in inserting the quotation at precisely the same point in their
respective narratives, créâtes a probability that Mark is at this point secondary to
Matt. and Luke, or to the document from which Matt. and Luke derived their narra-
tive of the message from John the Baptist. But when this one instance has been
admitted it becomes less improbable that there are others, among which may perhaps
be included Mark 1:13 b, /cal rjv fiera rûv Qr\pi<ùv, 2:27 to adfifiarov èià rov avQpwirov
èyevero, /cal ov% 6 âvOpœiro^ ïkà to ad^^arov^ an expression which one cannot doubt
came from Jésus’ lips, but for the omission of which by Matt. and Luke if they had it
before them in Mark’s gospel it is difficult to assign a reason ; 3:5, fier’ opyr)ç, avv-
\virovfievos èrrl rrf nrcopcoaei tt)ç /capSuiç avrwv, though it is easy to suppose that both
evangelists should hâve had a reason for omitting fier’ opyrfs. Other possible examples
might be cited, but for none of them can a very strong case be made out.
The presence in Matt. and Luke of common additions to the Mark text of three-
fold narratives présents of course a different problem. In so far as it is not explicable
as due to an influence affecting bofh evangelists alike it must be explained, if Mark is
21726 Principles of Literary Criticism and the Synoptic Problem
the source of the other two gospels, as due either to the early disappearance from
Mark of words originally contained in his text, or to early assimilation of the text of
Matt. and Luke. The first of these causes cannot be excluded, but definite evidence
of its operation is of course absent ; the second, however, viz. : assimilation of the text
of Matt. and Luke at so early a point in the history of these gospels that it is
impossible to detect and correct it from our existing evidence, though of course like
the preceding incapable of direct proof in particular cases, is rendered probable by
the general character of the evidence for the text of the gospels. So far from it being
reasonable to assume that we are able with our existing evidence to discover ail
instances in which the two gospels hâve been assimilated to one another, it is prac-
tically certain that the tendencies which we discover by the examination of the
existing evidence were already at work in a period antedating extant evidence.
Nevertheless it would be an unwarranted proceeding to affirm that ail the
agreements of Matt. and Luke in omission are due to accidentai coincidence, or to
the operation of similar motives affecting both minds; or, on the other hand,
that ail positive agreement of Matt. and Luke against Mark is due either to these
causes, or to harmonistic assimilation of Matt. and Luke, antedating existing evidence.
We must remain content apparently with that which the evidence seems clearly to
establish, namely, that the common source of the threefold narrative of Matt., Mark,
and Luke must hâve contained substantially the material which we now hâve in our
présent gospel of Mark, arranged in the same order in which it now stands in that
gospel; that there is little reason to suspect a greater divergence of the présent
gospel of Mark from what we may call the original Mark than is indicated in the
additions and omissions suggested above ; and it may even be the case that a considér-
able part of this apparent divergence is only apparent, being due either to a similar
modification of their source by Matt. and Luke, or to assimilation of Matt. and Luke
to one another subséquent to the construction of their respective gospels, but anté-
cédent to the date of our oldest textual evidence. It is évident that at this point
documentary criticism and textual criticism meet and merge. It is possible that
the divergence of our présent Mark from the form in which it was apparently used
as a source by Matt. and Luke is due almost wholly to causes which, if we were able
definitely to discover them, would be found to belong to the field of textual criticism.
It appears therefore that, in view of the facts, hypothesis N must involve the view
that x is not very différent from our présent Mark, and that hypothesis H must recog-
nize that Mark may hâve undergone some slight change, éditorial or scribal, between
the time of its use by Matt. and Luke and the time to which our textual evidence carries
us back; and that so modified either of these hypothèses will account for the triple
narrative of the synoptic gospels. For convenience we may then dismiss hypothesis
N, and speak hereafter in terms of H, it being understood that this hypothesis is
subject to the qualifications already indicated.
But while this hypothesis H accounts for the threefold material, it as evidently
218Ernest DeWitt Burton
27
fails to account for that large amount of material common to Matt. and Luke not
found in Mark, to say nothing of parts peculiar to each of the gospels. This hypothesis
must therefore be classed with D, F, J, K, L, as inadéquate.
Our next task must therefore be to consider possible modifications of these
inadéquate hypothèses, with a view to accounting for those facts which they fail to
explain. Thus, to account for the facts, D may be supplemented by the hypothesis that
Matt. was used by Luke as well as by Mark; F, by the hypothesis that Luke was used
by Matt. as well as by Mark; H, by the hypothesis that Luke used Matt. also as well
as Mark, or that Matt. used Luke also as well as Mark; J, by that of a non-extant
document used by Mark and Luke; K, by that of a non-extant document used by
Matt. and Mark; and L and H may each be supplemented by the hypothesis of a
non-extant document used by Matt. or Luke.
But we hâve also to recognize that certain modifications of the rejected
hypothèses, A, B, C, E, G, I, by creating new lines of connection between the gospels
and thus accounting for resemblances not accounted for by the simple hypothèses,
relieve these latter from some at least of the objections on the basis of which they
were rejected. Thus A may be supplemented by the hypothesis that Mark was used
by Luke as well as by Matt. ; B, by the hypothesis that Mark was used by Matt. as well
as by Luke; C, by the hypothesis that Matt. was used by Mark as well as by Luke; E,
by the hypothesis that Luke was used by Mark as well as by Matt; G, by the
hypothesis that Luke used Mark also, as well as Matt. ; and I, by the hypothesis that
Matt. used Mark also, as well as Luke.
In other words, these facts lead us to consider the hypothèses indicated by the
foliowing diagrams:
Matt.
Luke28 Principles of Litebary Criticism and the Synoptic Problem
For the testing of these hypothèses we need to take into account certain further
facts concerning the material common to Matt. and Luke, but not found in Mark.
9.	Matt. and Luke each hâve a story of the infancy. In several particulars these
stories agréé; as, e. g., in the names of the parents, Joseph and Mary; the place of the
birth, Bethlehem; the place of subséquent résidence, Nazareth; the supernatural con-
ception of the child, and his name, Jésus. But though each account occupies some
two chapters, the only verbal similarity of the two accounts is in the announcement of
the angel to Mary in Luke 1:31, TeÇy viov, /cal /caXeaeis to ovofia avrov ’lrjaovv, and the
answering words, è/cXrjdr) to ovofia ai/Tov ’Irjaoiky in 2:21 ; and the words of the angel
to Joseph in Matt. 1:21, T€%€Tat, 8è viovj /cal /caXeaeis to ovofia avTOv ’lijaovv, and the
corresponding words, /cal ê/caXecrev to ovofia avTov *lr)<rovvy in 1:23. If one compare
these facts with the facts concerning the relation of Matt. and Luke in other periods
of the life of Jésus—as, e. g., with the close verbal parallelism of the two similarly
located accounts of the preaching of John the Baptist (Matt. 3:7-9, 12; Luke 3:7-
9, 17) and of the temptation of Jésus (Matt. 4:1-11; Luke 4:1-13), or again with
the scarcely less clearly marked verbal parallelism of passages which are very differ-
ently located in the two evangelists (see, e. g., Matt. 11:25-27 and Luke 10:21, 22 ; Matt.
6:25-33 and Luke 12:22-31), it will be évident that the same explanation cannot be
applied to the two classes of facts.
10.	The facts respecting the distribution of the material common to Matt. and
Luke referred to in 4d above, p. 15, are peculiar and noteworthy.
In the infancy narrative there is, as already stated, no common material save one
or two sentences, one of which in Matt. is addressed to Joseph and in Luke to Mary;
and the other in each case is the evangelist’s statement that the instructions were
carried out.
In the narrative of John the Baptist and the associated account of the baptism
and temptation of Jésus, the two narratives hâve, in addition to that which they share
with Mark, two sections of closely parallel material similarly located (Matt. 3:7-9,
12; 4:1-11; Luke 3:7-9, 17; 4:1-13).
In the Galilean ministry there are three sections in which Matt. and Luke
hâve an évident literary relationship and there is no Mark account. These three
sections are the sermon on the mount, Luke 6:20-49, with its parallels in Matt., chaps.
5-7; the centurion’s servant, Matt. 8:5-13 and Luke 7:1-10; John the Baptist’s last
message, Matt., 11:2-19 and Luke 7:18-35. The relation of the two discourses, or
two versions of the one discourse, in Matt., chaps. 5, 6, 7, and Luke 6:20-49, so similar
in certain respects, so different in certain others, will call for fuller considération later.
The other two sections exhibit in the two accounts a verbal similarity only less marked
than in the narratives of John the Baptist and the temptation. As respects location, it
is to be noted that ail three of these sections are assigned by both gospels to the same
general period of Jésus’ ministry, and stand in the same order; but are differently
placed by reason of the intervention of different sections between them.
220Ernest DeWitt Burton
29
A distinctly different situation confronta us when we corne to examine that portion
of the material assigned by Luke’s gospel to the Perean period which is paralleled
in Matt. Of the 379 verses contained in Luke’s Perean section (excluding Luke
18:15-43, which is parallel to Mark’s account of this same period), 127 verses are
found also in Matt. Of this number, 97 stand in Matt. in the Galilean ministry, and
30 in his account of the passion week;11 not a verse is to be found in his account of
the Perean period. There are parallelisms between the Matt. and Luke accounts of
this period, but they are confined to that portion which both share with Mark. Such
a phenomenon is certainly not the resuit of accident, but must find its explanation in
the process by which the gospels were produced. When we turn to the passion week
and résurrection story, we find no material common to Matt. and Luke only, exceeding
the limits of a brief phrase or sentence,12 and assigned by both to this period. Each
has some material not found in Mark, and there is, as already indicated, considérable
material in Matt. at this point that is found in Luke’s Perean section; but the
parallelism of material common to Matt. and Luke only, such as is found in the early
period of the Galilean ministry, wholly disappears in this portion.
These facts enable us to eliminate certain of the hypothèses last enumerated.
The objections to the different hypothèses are naturally to a considérable extent the
same. Yet it will probably be most satisfactory to treat them seriatim, notwithstand-
ing the répétition which this method will involve.
Against hypothesis O are the foliowing considérations:
a. It fails to account for the absence of an infancy story in Mark. In itself this is,
of course, not a weighty objection, since it is entirely possible that a later writer should
for some reason prefer to begin the gospel with the public work of John the Baptist
and Jésus ; but the considération is somewhat strengthened by the evidence that in the
omission of the infancy narrative Mark represents the older conception of the limits
of the gospel narrative. Appeal may be made on this point to the idea of the limits
of the gospel expressed in Acts 1:22, “Beginning from the baptism of John unto the
day in which he was taken up from us.” But more décisive is the testimony of the
préfacé to the third gospel, the author of which, though he includes in his own gospel
an infancy narrative, yet by his phrase, “ they who from the beginning were eye-
witnesses and ministers of the word,” reflects the conception that the beginning of the
gospel dates from the time when there were those who were both eyewitnesses and
ministers of the word. To this may be added the testimony of the gospel of Mark
itself, the first line of which, taken according to its most probable meaning, introduces
the preaching of John the Baptist as the beginning of the gospel; not, of course, the
beginning of this book, but of that gospel which lay back of the written book. When
u Mark may hâve furnished 13 verses of those which
belong to the Galilean ministry, and 1 of those in the
passion week. He contains practically équivalent mate-
rial, but Matt. agréés verbally with Luke more closely
than with Mark.
12 Matt. 26:50 a and Luke 22:48 are alike in substance,
not in words ; Matt. 26:68 and Luke 22:64 hâve five words
in common which are not in Mark ; Matt. 26:75 and Luke
22:62 hâve a phrase of five words in common, of which
two appear in Mark, one of the latter in a different tense.
22130 Principles of Literary Critioism and the Synoptic Problem
due weight is given to these facts, it appears distinctly improbable that a gospel which
contains a narrative of the infancy is the source of a gospel which begins with the
preaching of John the Baptist. In other words, this hypothesis reverses the order of
descent which is rendered probable by the facts respecting the infancy narratives.
b.	This hypothesis fails to account for the total dissimilarity of the infancy stories
as found in Matt. and Luke. That Luke should, as this hypothesis assumes, reproduce
in other portions of his gospel, and in many cases almost verbally, the material which
he had before him in the gospels of Mark and Matt. (combining with it, as we must
assume, material derived from other sources), yet should in the infancy period
entirely ignore Matthew’s narrative of this period, is certainly very improbable. In
other words, this hypothesis contravenes the probability respecting Luke’s method as
shown by the body of the book.
c.	This hypothesis fails to account for the relatively small amount of agreement of
Matt. and Luke against Mark in threefold narrative. On the simpler hypothesis D, of
which hypothesis O is a modification, there is, of course, nothing to produce such
agreement of Matt. and Luke against Mark. But when Luke is supposed to hâve
used, not only Mark, but also Matt., which Mark had already used, there is then
established a direct connection between Matt. and Luke, from which it would naturally
resuit that Luke would sometimes follow Matt. and sometimes Mark in material which
both possess, Mark having derived it from Matt. That Luke, following Matt. at times
as closely as on this hypothesis he must hâve done, and sometimes displacing Mark
for another source, should hâve so constantly preferred Mark to Matt. in details of
material common to them both is not impossible but improbable. The full force of
this objection can be felt only by a detailed comparison of the Greek text of the three
narratives.
d.	The narratives common to Matt. and Mark bear on the whole the marks of
secondary character, not in Mark, but in Matt. Cf. 16, p. 46.
e.	The theory fails to account for the totally different location of the material
found in Luke in the Perean section, and in Matt. in the Galilean ministry or passion
week. If Luke had before him the two sources, Matt. and Mark, it is impossible to
explain why in using material from Mark he should hâve foliowed the order of that
evangelist so closely as he evidently does, and, on the other hand, in using material
derived from Matt. not found in Mark, should so scrupulously assign the major portion
of it to an entirely different period of the ministry from that to which Matt. had
assigned it. This is ail the more improbable in view of the fact that the material which
on this hypothesis Luke must hâve derived from Matt. is in Matt. imbedded for the
most part in extended discourses assigned, as above remarked, to the Galilean ministry
or the passion week. Now, since Luke places this material in the Perean ministry,
and assigns to the several fragments of it distributed through his record of this
ministry spécifie historical occasions indicated by brief historical introductions, these
must hâve been arbitrary inventions of the evangelist. But there is nothing in this
222Ernest DeWitt Burton
31
evangelist’s methods of dealing with his other sources that would lead us to ascribe to
him such a method of procedure.
f. It is inadéquate in failing to account for the large amount of material peculiar
to Luke. This objection belongs, however, in a different class from the preceding,
tending to show not that the hypothesis is in itself incorrect, but that it is inadéquate.
On the whole this hypothesis labors under insuperable difficulties.
The objections to hypothèses P, Q, R, S, T are for the most part similar to those
already urged against O, and may be somewhat more briefly stated.
The objections to hypothesis P are:
a.	It reverses the order of descent rendered probable by the facts respecting the
infancy narratives, The facts already adduced from Acts 1:21, 22, the préfacé of
Luke, and Mark 1:1 make even more strongly against this hypothesis than against O.
b.	A total ignoring of the infancy story of Matt., such as is by this hypothesis
ascribed to Luke, is unaccountable in view of the use which by the hypothesis Luke
must hâve made of the other portions of Matt.
c.	It fails to account for Mark’s omission of that large amount of material which
was furnished him in common by both his sources but is not found in his gospel.
d.	It inadequately accounts for the relatively small amount of agreement between
Matt. and Luke against Mark in threefold narrative.
e.	It fails entirely to account for the totally different location of the material
found in Luke in the Perean section, in Matt. in the Galilean or passion week. Even
aside from any comparison of it with his treatment of another source, such as we were
able to make in considering hypothesis O, the treatment of the different portions of
Matt. which this hypothesis ascribes to Luke is, as pointed out in discussing O, dis-
tinctly improbable.
f.	It is inadéquate in failing to account for the material peculiar to Luke.
The objections to hypothesis Q are:
a.	Like O and P it fails to account for the absence of any influence of the infancy
story of Matt. upon that in Luke.
b.	It inadequately accounts for the relatively small amount of agreement between
Matt. and Luke against Mark in threefold narrative.
c.	It fails to account for the treatment which Luke accorded to that portion of
Matt. which he gathered together into his Perean ministry, differing so markedly as
it does from that which, on this hypothesis, he accorded to other portions of Matt.
and to Mark. The objection urged against O holds with full force against Q.
d.	It is inadéquate in failing to account for the material peculiar to Luke.
To hypothesis R it is to be objected :
a.	It fails to account for the absence of any influence of Luke’s infancy narrative
upon that contained in Matt.
b.	It fails to account for the relatively small amount of agreement between Matt.
and Luke against Mark in material found also in Mark,
22332 Principles of Literary Criticism and the Synoptio Problem
c.	It fails to account for the totally different location of the material found in
Luke in the Perean section, in Matt. in the Galilean ministry and passion week.
The relation between Matt. and Luke is on this hypothesis, and on the two fol-
lowing, the reverse of that supposed in hypothèses O, P, and Q. But the improba-
bility of the hypothesis at this point is only less than in the preceding three. It is
true that if Matt. used Mark, he freely departed from Mark’s order so far as concerns
arrangement of the material in the first half of the Galilean ministry, that is, from
Matt. 4:12 to 13:58. But in no case does he assign discourse material exceeding
the limits of a verse or two and derived from Mark to a different period of the life of
Jésus. Except in the first half of the Galilean ministry, as already indicated, his
departures from Mark’s order are extremely slight. The material contained in Luke
in the pre-Galilean period he also assigns to the position in which it appears in Luke,
and the material found in Luke in the Galilean ministry he likewise assigns to that
period, though exercising the same freedom that he has used in respect to Mark. But
according to this hypothesis Matt. must hâve dealt in an entirely different way with
the material which he desired to use from Luke’s Perean period, carefully placing
every verse so employed in a different location, even in a different period of the
ministry, from that to which Luke had assigned it. This procedure, while indicating
that Matt. must hâve possessed much of the same material which Luke also contains,
is quite inexplicable on the supposition that he had it before him in the précisé form
and position in which it stands in the gospel of Luke.
d.	It is inadéquate in failing to account for matter in Luke not found in Mark,
or for the important material peculiar to Matt.
The objections to hypothesis S are :
a.	By making Mark dérivative it reverses the probable order of descent as indi-
cated by the absence of an infancy story from Mark, and its presence in the source Luke.
b.	It fails to account for the absence of any influence of Luke’s infancy narrative
upon Matt.
c.	It inadequately accounts for the relatively small amount of agreement of Matt.
and Luke against Mark in threefold material.
d.	It ascribes to Matt. the same improbable way of treating the material of
Luke’s Perean section that is involved in hypothesis R.
e.	It is inadéquate in failing to account for the matter peculiar to Matt.
Hypothesis T is open to nearly the same objections as P, but is like R and S in
that Matt. is made secondary to Luke. Thus
a.	It fails to account for the absence of an infancy story in Mark.
b.	It fails to account for Matt.’s independence of Luke in the infancy narratives.
c.	It ascribes to Matt. the same treatment of the material of Luke’s Perean sec-
tion, the improbability of which has been pointed out in discussing R and S, altliough
in this case we are without a basis of comparison, since by this hypothesis Matt. did
not possess Mark.
224Ernest DeWitt Burton
33
d.	It fails to account for Mark’s omission of that large amount of material which
was possessed in common by both his sources but is not found in his gospel.
e.	It is inadéquate in failing to account for the matter peculiar to Matt., or for
the rejection by Mark of this addition to his material found in Luke and of the
peculiar material of Luke. But neither this objection nor the immediately preceding
one can be strongly urged, since we are without basis of comparison by which to
décidé by what motives Mark might hâve been influenced.
Concerning hypothesis U it is sufficient to say that x, being the source of Mark
and Luke, must approximate either the one or the other of these gospels, or contain
the material now found in both or either of them. In proportion as x approximates
Mark the hypothesis approaches B, and is open to the objections already urged against
that theory. In proportion as x approximates Luke the hypothesis approaches S and
is open to the objections urged against that theory. Nor will the difficulty of the
hypothesis be obviated by supposing that x contained the sum of the material now
found in Mark and Luke, for in that case not only will most of the objections
already urged hold, but it will be necessary to suppose further that x contained a con-
sidérable number of duplicate narratives of which the two evangelists in each case
selected a different one.
Hypothesis V in turn approaches Q if x be supposed to approximate Mark, and
is open to the objections urged against Q, or approaches O if a? be supposed to approxi-
mate Matt. Or, if x be supposed to include ail the material now in Mark and Matt.
it would be open to substantially the same objections as O.
Hypothesis W approaches P if x be supposed to approximate Matt., or T if
x be supposed to approximate Luke. If x be supposed to include only the material
common to Matt. and Luke, then it is entirely inadéquate by reason of its failure to
account for the material which each of them possesses in addition to that which it
shares with the other. If x be supposed to include the material now found in both
or either of them, not only must that document liave been a strange assemblage of
duplicates, but the two evangelists which drew from it must, in so large a number of
cases as to be explained only by private arrangement between them, hâve chosen that
form of narative, and those details of narrative, which the other evangelist omitted.
Thus, the infancy narrative must hâve included both the Matt. and the Luke story,
and the two evangelists must hâve so divided the material between them that each
omitted what the other used, except, indeed, that both retained certain words of the
angel concerning the birth of Jésus and his name, but took pains in this case to
ascribe them to different occasions.
Conrady has, indeed, undertaken to defend the hypothesis that these infancy
stories are derived from a common source, which he endeavors to relate in some way
to the protevangelium of James; but such a hypothesis can only excite surprise that
one who has carefully studied the way in which documents were used by ancient writers
could hâve had sufficient confidence in this theory to propose it. Whatever view be
22534 Principles of Literary Criticism and the Synoptic Problem
taken of x the theory is open to two objections, one of which also holds against several
previous hypothèses, the other of which is peculiar to this theory.
a.	By making Mark dérivative it contravenes the probable order of descent as
indicated by the facts concerning the infancy narratives. Considérations similar to
those already adduced against hypothèses O and P on this point hold also against the
présent hypothesis.
b.	It fails to account for the different methods of treatment which on this hypoth-
esis both Matt. and Luke mnst hâve adopted toward different portions of the same
source. The material which by hypothesis was afterward taken up by Mark, both
Matt. and Luke reproduce in closely similar form, and in the main in the same order,
the chief exception being in respect to the first half of the Galilean ministry in Matt.,
as already referred to. But the material which Mark did not afterward take up they
treat differently from the way in which they treat Mark, diflferently from one another,
and differently in different parts of the gospel. This fact must be accounted as
almost conclusive evidence that we hâve in some way by this hypothesis reversed the
actual order of proceeding. If we endeavor to avoid this difficulty by the obvious
suggestion that x is in reality not one document but two, then it immediately appears
that one of the two éléments into which x is thus resolved must hâve been substantially
the gospel of Mark; in other words, this modification of the hypothesis résolves
hypothesis W into hypothesis Y.
If then we turn to consider hypothesis Y, we observe that it is consistent with
most of the facts thus far adduced. The récognition of Mark as a common source
of Matt. and Luke accounts for that portion of the gospel narrative which is common
to ail three. It is also consistent with the fact of agreements of Mark and Matt.
against Luke, and of Mark and Luke against Matt., and of the much smaller amount of
agreement of Matt. and Luke against Mark. The récognition of a second non-extant
source of Matt. and Luke accounts in general for the existence in Matt. and Luke of
material not found in Mark. The hypothesis is, however, inadéquate. Thus—
a.	It fails to account for the existence of entirely independent infancy stories in
Matt. and Luke.
b.	One document additional to Mark is inadéquate to account for the great dif-
férences in arrangement and content of the post-infancy portions of Matt. and Luke.
For it is évident that a document containing only the material common to Matt. and
Luke fails to account for the important matter peculiar to the first and third gospels
respectively. On the other hand a document containing both the non-Markan material
common to Matt. and Luke and the material peculiar to each excludes the explanation
of the great différences between Matt. and Luke in content and arrangement of non-
Markan material by their possession of sources only in part the same, or by a différence
in their method of combining the sources (such as would be natural or inévitable in
case of a multiplicity of sources used by the two evangelists), and compels the suppo-
sition of a rearrangement, by one or the other of the evangelists, of the non-Markan
226Ernest DeWitt Burton
35
document possessed by them both, and a variant treatment of different parts of the
same document by the same evangelist, for neither of whic-h can any motive be dis-
covered. If, for example, the non-Markan source be supposed to hâve contained the
material common to Matt. and Luke in the setting in which it appears in Matt. we
hâve nearly the same situation supposed in hypothesis O, and objection e urged against
that hypothesis holds here also with slight modification. Or, if the non-Markan
material existed in the form and connection in which Luke now has it, the situation is
nearly the same as in hypothesis R, and objection c urged against that view holds here
with little change. The facts still to be considered will set in still clearer light the
inadequacy of one non-Markan source to account for ail the facts.
The récognition of these inadequacies of hypothesis Y calls for its modification
by the récognition of the fact that x is not in reality one document, but two, or even
more.13 In order to discover into how many parts x is to be resolved, and what the
constituents of each part were, it will be necessary now to examine the gospels Matt.
and Luke once more, and more closely, and to compare them with reference both
to the material which they hâve in common and to that which is peculiar to each.
11. It is one of the long-ago observed peculiarities of the first gospel that it con-
tains certain extended discourses of Jésus, somewhat regularly distributed through the
book. There are in ail six of these, occupying nine chapters of the gospel: the
sermon on the mount in chaps. 5-7, the missionary discourse in chap. 10, the parables
of the growth of the kingdom in chap. 13, the discourse on the personal relations of
the disciples in chap. 18, the invective against the Pharisees in chap. 23, and the dis-
course on the end of the nation in chaps. 24 and 25. If now we examine these dis-
courses with reference to the extent to which they are paralleled in the other gospels, we
discover two facts:	(1) In every case there is at an approximately corresponding place
in one or both of the other synoptic gospels, a discourse, or at least a few sentences,
on the theme of the discourse as given in Matt. In the case of the sermon on the
mount this parallel discourse is in Luke only ; in the case of ail the rest it is found in
both Mark and Luke.14 (2) In every case the discourse in Matt. contains, in addition
J3 Hypothesis Y might also be modified by the theory
that Mark also possessed x; or this might be combined
with the resolution of x, in which case Mark might be sup-
posed to hâve possessed one or more of the component
éléments of x. But the only advantage gained by this
dependence of Mark on a; is that it would explain that rela-
tively small amount of agreement which exists between
Matt. and Luke against Mark. But against it are the ob-
jections (a) that it fails to account for the fact that this
agreement is so small in amount ; if Matt. and Luke both
possessed a source of Mark, it would be reasonable to
14 This
following list:
The sermon on the mount :
The mission of the Twelve :
The parables by the sea :
Personal relations of disciples :
Discourse against the Pharisees:
Eschatological discourse :
expect that they would oftener adopt readings of that
source which Mark had modified; (b) that in the form
which leaves x unresolved it supposes Mark to hâve omit-
ted from his sources more material than he used ; in the
form which résolves x into éléments, the only part that
could on the basis of evidence be supposed to be in Mark’s
hands would be a slightly different recension of what we
now possess in Mark ; this amounts simply to saying that
the présent Mark is a slightly different édition from that
which Matt. and Luke used; (c) that it unnecessarily com-
plicates the hypothesis.
14 This general parallelism of the discourses of Matt. with material found in Mark and Luke will appear in the
Matt., chaps. 5-7
Matt. 10:1-11:1
Matt. 13:1-53
Matt., chap. 18
Matt., chap. 23
Matt., chaps. 24, 25.
227
Mark 6:7-13
Mark 4:1-34
Mark 9:33-50
Mark 12:38-40
Mark, chap. 13
Luke 6:20-49
Luke 9:1-6
Luke 8:4-18
Luke 9:46-50
Luke 20:45-47
Luke 21:5-3836 Principles of Literary Critigism and the Synoptic Problem
to the material which is parallel to that of the corresponding section in Mark or Luke,
certain other material. If now we examine this additional material, looking for paral-
lels to it in other parts of the other gospels, we shall discover that it is of three kinds:
a.	There is in each of the discourses a certain portion peculiar to Matt.
b.	There is a large amount of material paralleled only in Luke, and at an entirely
different place, though usually closely resembling it in language. The Luke parallels
of this last material (save that Matt. 10:24 a=Luke 6:40 a) are invariably in his Perean
section, 9:51—18:14; 19:1-28; Le., in that portion of his record of the Perean min-
istry of which there are no parallels in Mark’s record of this period.
c.	There is in some of these discourses a small amount of material found also in
Mark, but in a different position, much of this being also paralleled in Luke in a
position corresponding to that of Mark. It is worthy of notice that in no instance
do the Matt. discourses contain material parallel to that in Luke, except such as is
either in the six parallel discourses (in every instance except the sermon on the
mount the Luke discourse being closely parallel to that found in the corresponding
place in Mark) or in the Perean section.
If now, dismissing these results from mind altogether, we undertake a study of
these discourses of Jésus as reported in Matt. with reference to their content and
structure, we discover that there are certain portions of them which give us difficulty,
either because they interrupt what seems to be the évident logical structure of the dis-
course, or because they are as evidently inappropriate to the occasion described in the
introduction to the discourse. Thus in the sermon on the mount we find that there
are certain portions which, though not so distinctly foreign to the general course of
thought as to arrest the attention of the casual reader, do yet disclose themselves as
such to careful study. For example, chap. 6 opens with the words: “Take heed that
ye do not your righteousness before men to be seen of them ; else ye hâve no reward
with your Father who is in heaven.” It is évident that Jésus is here passing to con-
trast the righteousness of the kingdom of heaven with the conduct of the Pharisees, as
in the preceding chapter he had contrasted it with their teaching, and specifically now
in respect to ostentatiousness. Positively stated, he is insisting upon righteousness of
the heart and before God, who knows the heart, in contrast with outward conduct per-
formed to win the praise of men, just as in chap. 5 he had contrasted righteousness of
the heart with a literalism that kept the letter of the statute at the sacrifice of real
morality. It is scarcely less clear that he here gives three illustrations of this general
teaching, viz., almsgiving, prayer, and fasting. If now the first and third of these illus-
trations be compared, the paragraphs severally devoted to them will be seen to be exactly
symmetrical in structure. They begin with similar words, “When thou doest alms,”
“ When ye fast,” and each ends with the words, “and thy Father which seeth in secret
shall reward thee.” The intervening verses, moreover, are exactly parallel in thought
and form, differing only in that which forms the subject of the illustration. If the
second illustration be examined, it will be found that the first two verses constitute a
228Eknest DeWitt Burton
37
perfect parallel to the first and third illustrations. To these verses, however, there are
added in the case of the second illustration nine more, which not only exceed the
parallelism and destroy the symmetry of structure, but, what is much more signifi-
cant, carry the thought away from righteousness before God, illustrated in the case
of prayer, to prayer in general, first to simplicity in prayer, and then to the forgiving
spirit as the condition of securing the answer to our prayer for forgiveness.
If we extend our study to the sermon on the mount as a whole, we shall find that
similar phenomena characterize the whole discourse. Studied throughout with refer-
ence to its structure and course of thought, and wholly without reference to the bearing
of the results on the synoptic problem, it will be found to contain two somewhat distinct
strata of material. There is first a discourse on a clearly defined theme, the righteous-
ness of the kingdom of heaven, especially in relation to the current ethics which were
professedly derived from the Old Testament. This discourse has a well-defined
structure, and présents no difficulties of analysis that are at ail serious. Overlying this,
or, to drop the figure, interpolated into it, is a sériés of additions, each of which is eitlier
attached, as the words about prayer are, to a paragraph dealing with the same theme,
from which they are nevertheless a digression, or constitutes an entirely new para-
graph more or less closely related in thought to that paragraph of the first stratum
which it follows.
If, furthermore, the two strata themselves be examined with reference to their
parallels in other parts of the gospels, it will be found that the facts respecting them
are different. The first one constitutes a discourse which has a manifest relationship
with that which stands in Luke 6:20-49, and yet is as clearly different from it. The
two are similar in general theme and in certain striking sentences, and, what is most
remarkable, in the order of the similar parts, there being but three or four instances of
parallel verses differently placed, and but one (Matt. 7:12; Luke 6:31) in which the
différence of position exceeds the limits of a paragraph. On the other hand, the two
discourses are very unequal in extent, that of Matt. being three and one-half times as
long as Luke’s, but Luke’s also containing some material not in Matt. They are very
different in point of view, Luke’s form wholly lacking the Jewish atmosphère which is
so characteristic of Matthew’s. It is impossible to doubt that these discourses are in
some way related; it is impossible to suppose that either can hâve been derived wholly
from the other, difficult to suppose that either can hâve constituted a direct and chief
source of the other. The resemblance is widely different in degree, if not in kind,
from that which marks those portions of the gospels which we are constrained to refer
to a common source, and the resemblance between the basal stratum of the Matt. dis-
course and Luke 6:20-49 is widely different from that which appears in comparing
those portions of this discourse which are paralleled in Luke’s Perean section with the
Lukan version of them.
Since neither Luke nor Mark can hâve been the sole source of the basal stratum
in Matt., we are compelled to posit a source no longer extant in separate form, and
22938 Principles of Literary Critioism and the Synoptic Problem
presumably distinct from any document possessed by Luke, since much of the mate-
rial of which we are speaking does not appear in his gospel.
The interjected material, on the other hand, has in general no parallel in the
corresponding discourse in Luke, but almost ail of it is paralleled in other portions of
the gospels. It is of two kinds as respects the location of its parallels. The largest
portion of it is found in Luke’s Perean section.15 A small portion is paralleled in
Mark.16 A few verses are paralleled more or less completely both in Luke’s Perean
section and in Mark.17
Now, these facts, combined with those which we hâve already considered with
reference to the relation of the gospels, point strongly to the conclusion that the
sermon on the mount in its présent form is drawn from at least three sources; first, a
basal discourse—a discourse more or less closely resembling that which is contained in
Luke 6:20-49; second, a document which contained Luke, chaps. 11 and 12, and
probably chap. 16, but which was not our présent Luke; and, third, the gospel of
Mark. The few sentences in this discourse which are paralleled in Mark would not
perhaps require this addition, but, in view of the other evidence that Matt. possessed
Mark, this is the simplest way to account for them.18
The discourse which with its introduction is contained in Matt. 9:36—11:1 is
not less instructive, though the facts are somewhat different. This discourse is said
to hâve been spoken to the Twelve on the occasion of sending them out on a mission-
ary journey. In the early part of it Jésus instructs them to go only to the lost sheep
!5Matt. 5:25, 26 has its parallel in Luke 12:58, 59;
Matt. 5:32 in Luke 16:18; Matt. 6:8 in Luke 12:30; Matt.
6:9-13 in Luke 11:2-4; Matt. 6:19-34 in Luke 12:22-34;
11:34,35; 16:13; Matt. 7:7-11 in Luke 11:9-13. Besides
these seven passages, there are several which do not so
clearly interrupt the course of thought in Matt. Matt.
5:13b is paralleled in Luke 14:34; Matt. 5:15 in Luke
11:33; Matt. 5:18 in Luke 16:17; Matt. 7:13, 14 in Luke
13:24.
16	Matt. 5:29, 30 is paralleled in Mark 9:43, 47 (and Matt.
18:8,9); Matt. 5:32 in Mark 10:11 (and Matt. 19:9) ; Matt.
6:14 in Mark 11:25.
17	Thus with Matt. 5:13 b cf. Luke 14:34, 35 and Mark
9:50; with Matt. 5:15 cf. Luke 11: 33 and Mark 4:21 (also
Luke 8:16) ; with Matt. 5: 32 cf. Luke 16:18 and Mark 10:11
(also Matt. 19:9).
78 If we inquire more closely respecting the origin of
the sermon as given by Matt., two hypothèses suggest them-
selves as possibly accounting for the facts. (a) It may be
that we hâve in Luke’s discourse and in that portion of
Matt. which we hâve designated as its first or basal stratum
two different recensions of the same material. These two
recensions hâve evidently diverged widely from one another
under the influence of the intention to adapt them to dif-
ferent readers or hearers, and hâve exercised little influ-
ence upon one another since their divergence, yet hâve
retained, especially in order of topics treated, clearly
recognizable evidence of their relationship. (b) The dis-
course in the Lukan form may be the basis upon which,
with the use of material drawn from an independent source,
the discourse in Matt. has been built up. The second of
these hypothèses is favored by the close verbal resemblance
of a small portion of the material which Matt. has in com-
mon with Luke (see especially Matt. 7:1-5; Luke 6:37 a,
38 b, 41, 42), and by the close resemblance in order of
parallel material. The first hypothesis is favored by the
independent treatment of the subject which characterizes
most of the Matt. discourse which is parallel to Luke 6:20-
49, by the clearly marked plan and clearly defined and
independent point of view of the basal discourse in Matt.,
and by the fact that the second hypothesis compels us to
suppose that the same writer first built up out of two sources
a discourse having marked unity and definiteness of pur-
pose, and then destroyed that unity by additions from a
third source ; for if it be said that these two steps were
taken by different authors, then the final author of the
gospel already possessed the discourse, not in its éléments,
Luke 6:20-49 and a second source, but with these two com-
bined intowhat wehave called the basal discourse in Matt.
The diverse indications of the evidence suggest modifica-
tions of the first hypothesis in the direction of the second.
Thus the close verbal parallelism of Matt. 7:1-5 with Luke
6:37 a, 38 b, 41, 42, combined with the close resemblance in
order of the sentences common to the discourses, may lead
to the conclusion that, while Matt. possessed a discourse
having substantially the characteristics of tbe basal dis-
course in this gospel, he also possessed Luke 6:20-49 and
made some use of it. This view is somewhat favored again
by observing that, while Matt. 7:1-5 is not distinctly foreign
to the basal discourse, it is less clearly germane to it than
those portions in which the verbal resemblance to Luke
6:20-49 is less clearly marked.
230Ernest DeWitt Burton
39
of the house o£ Israël, but in the next paragraph, 10:16 ff., there is a sudden forward
projection in time, and a sudden widening of the horizon. He speaks of their being
delivered up to councils, of being scourged in the synagogues, of being brought before
governors and kings for a testimony to them and to the gentiles, and finally déclarés:
“ Ye shall not hâve gone over the cities of Judea till the Son of man be corne.” These
words evidently refer to the time after his departure, and their incongruity with this
occasion is évident.
If now we look for the parallels of this discourse that appear in the corresponding
sections in Mark and Luke, we discover that they are confined to six verses (1, 10-14),
and that none of the incongruous material appears in these verses. If we examine
the material outside these six verses, including of course ail the incongruous matter,
as well as some which is not inappropriate to this occasion, we find that (save for
10:24a) it falls into three parts: first, material which has no parallel in the other
gospels ; second, material which is paralleled in the Perean portion of Luke ; and, third,
material which is paralleled in Mark, chiefly in his report of the eschatological dis-
course of Jésus spoken just before his death. Except for half a verse, therefore,
the Matt. discourse contains no material parallel to Luke that is not found either in
the corresponding portion of Mark or in Luke’s Perean section. Thus the general
facts respecting this discourse are the same as respecting the sermon on the mount,
except that, whereas in the sermon on the mount the basal element was a differ-
ent recension of the discourse found in Luke 6:20-49, in this case it is in a passage
of six verses paralleled at the corresponding place in Mark. The facts in detail will be
easily seen by referring to the table appended to this essay.
We may next examine the parables in Matt. 13:1-53. There are seven of them.
Two of them, together with the interprétation of the first and the discussion concerning
the reason for teaching in parables, are closely paralleled in Mark and in the parallel
passage in Luke. The tests for three documents, it may be said in passing, strongly
suggest the originality of Mark and the independent use of Mark by the other two
gospels. Two of the seven are found in Luke’s Perean portion, chap. 13, one of these
being included in the two previously mentioned. The remaining four are peculiar to
Matt. The facts are therefore substantially the same as in the preceding case, but
differ in two minor particulars. There is in this case no material parallel to that which
is found in Mark, except that which appears in the principal and similarly located
parallel passage. One of the parables, that of the mustard seed, appears both in the
corresponding passage of Mark and in Luke’s Perean section. It is instructive to
observe that the parable of the leaven immediately follows the parable of the mustard
seed both in Luke’s Perean section and in Matt. Plurality of sources for Matthew’s
chapter, as also the presence in both sources of the parable of the mustard seed, is
obviously suggested. Of like significance is the double termination of the collection
in Matt.’s chapter, vss. 34, 35, parallel to Mark 4:33, 34, with which he concludes
his collection of parables, forming the first conclusion, and vs. 53, peculiar to this
23140 Principles of Literary Criticism and the Synoptio Problem
gospel, the second. Though we do not possess separately the source from which the
material peculiar to Matt. is drawn, this fact certainly suggests conflation of Mark with
this non-extant document, and the whole chapter has the marks of being a conflation
of éléments drawn from three different sources.
• The facts respecting the other three long discourses in Matt. can be briefly stated.
In the discourse on the personal relations of the members of the kingdom, Matt., chap.
18, the introduction, vss. 1, 2, is similar to the introduction to the parallel passage in
Mark (9:33-36) and Luke (9:46, 47). Of the thirty-three verses which make up the
discourse proper, vss. 5, 6, 8, 9 are closely parallel to Mark’s corresponding section;
vss. 3, 12-14 are paralleled in Luke’s Perean section, though vs. 3 is also paralleled in
a non-parallel section of Mark (10:15); and vss. 12-14 vary sufficiently from the Perean
version (15:4—7) to raise the question whether Matt. drew it from that source or from
some independent source containing substantially the same material. Vss. 10, 15-35
are peculiar to Matt., except for a partial parallelism of vss. 15 and 21 to Luke 17:3, 4.
In the discourse against the Pharisees, Matt., chap. 23, vss. 1, 6, 7a are parallel
to the corresponding passage in Mark (12:38, 39). Vss. 4, 12,13, 23, 25-27, 29-31, 34-
36 are paralleled in thought, and to a certain extent in expression, to passages in the
Perean section of Luke (11:39-52; 14:11; 18:14); vss. 37-39 are paralleled almost
word for word in Luke 13:34, 35; vs. 11 is paralleled in non-parallel sections of Mark
(9 :35; 10:43, 44); and vss. 2, 3, 5, 7b-10, 15-22, 24, 28, 32, 33 are peculiar to Matt.
In the eschatological discourse, Matt., chaps. 24, 25, 24:1-9, 13-25, 29-36,
42 is closely parallel to Mark, with only slight additions and changes; 24:26-28, 37-
41, 43-51 is paralleled in Luke’s Perean section (17:23-27, 34-37; 12:39, 40, 42-
46), the verbal resemblance being for the most part very close; 24:10-12, and
chap. 25 are peculiar to Matt., except for a slight parallelism with Luke’s Perean
section in 25:11, 12 (Luke 13:25) and a somewhat doser resemblance between the
parable of the talents and Luke’s (Perean) parable of the pounds (19:11-27), especially
noticeable in the latter part (Matt. 25:24-29; Luke 19:20-25).
It thus appears that, with individual peculiarities in each case, the general facts
respecting ail these discourses are the same. In each, three éléments appear, with the
addition in some cases of a fourth of less importance. In each one there is a portion,
usually closely connected with the narrative introduction, which is paralleled in the
corresponding section of Mark, except that in the sermon on the mount the parallel is
in Luke ; in each case there is material peculiar to Matt. ; in each case there is material
paralleled in Luke’s Perean section ; in some cases there is a verse or two paralleled in
a non-parallel section of Mark; in no case is any portion paralleled in Luke outside
of 6:20-49, the Perean section, and passages themselves paralleled in Mark.
The portion of these long discourses just referred to of which there is no parallel
in either Mark or Luke consists of approximately the following passages: 5:4, 7-10,
13a, 14, 16, 17, 19-24, 27, 28, 31, 33-39a, 41, 43; 6:1-7, 10b, 13b, 16-18,19
19 Perhaps also 6:19, to which Luke 12:33 is only a substantial parallel.
232Ernest DeWitt Burton
41
34; 7:6, 12b, 15, 22; 10:5, 6, 8b, 16b, 23, 25b, 36, 41; 13:14, 15, 24-30, 35-53;
18:4, 10, 14, 16-20, 23-34; 23:2, 3, 5, 7b-10,15-22, 24, 28, 32; 24:10-12, 30a; 25:1-
11a, 13, 14-30,ao 31-46, being in ail about 160 to 165 verses.
12.	The existence of this material in the longer discourses reported or built up
by the author of the first gospel and peculiar to it, naturally raises the question
whether there are also in the shorter discourses reported by this evangelist sayings of
Jésus not reported in the other gospels. Examination discloses the fact that there are
such and yields the following list : Matt. 3:14, 15 ; 9:13 a ; 11:28—30 ; 12:5—7,11, 12 a,
34; 16:17-19; 19:28; 20:1-15; 21:28-32, 43; 22:1-14; 26:52, 53, to which should
perhaps be added certain passages in which sayings of Jésus are introduced by a brief
narrative, viz., 15:12-14, 23, 24; 17:24-27; 19:10-12; 21:14-16, containing eight
more verses of sayings of Jésus. There is, of course, no reason why these sayings of
Jésus should be separated from those named above. If, then, we combine the two lists,
we find that the following passages of Matt. contain reports of sayings attributed
to Jésus and reported in the first gospel only: 3:14, 15; 5:4, 7—10, 13a, 14, 16, 17,
19-24, 27, 28, 31, 33-39a, 41, 43; 6:1-7, 10b, 13b, 16-18, 34; 7:6, 12b, 15, 22;
9:13a; 10:5, 6, 8b, 16b, 23, 25b, 36, 41; 11:28-30; 12:5-7, 11, 12a, 34; 13:14, 15,
24-30, 35-53; 15:12-14, 23, 24; 16:17-19; 17:24-27; 18:4, 10, 14, 16-20, 23-34;
19:10-12, 28; 20:1-15; 21:14-16, 28-32, 43; 22:1-14; 23:2, 3, 5, 7b-10, 15-22, 24,
28, 32; 24:10-12, 30a; 25:l-lla, 13, 14-46; 26:52, 53. Now, the existence of these
verses in the gospel of Matt., about 230 in number, consisting of sayings of Jésus not
reported in the other gospels and constituting a little over one-fifth of the whole gospel,
cannot but remind us of the statement of Papias quoted by Eusebius (.Hist. Eccl.,
iii, 39): Mardaîos fiev ovv fE/3pathi BiaXetcrœ rà \o<yia avveypdyjraro. 'HpfJLqvevae 8’
avra ûjç rjv SvvaTos e/eao-Toç. The comparison of the gospels certainly suggests that
these passages constituted a source of our gospel of Matt. It is in favor of the suppo-
sition that they were in fact contained in, or constituted, the original collection of
sayings of Jésus to which Papias refers, that it conforms to this ancient and undis-
puted tradition, and that it explains, as no theory which makes the Matthean Logia
a source of both Matt. and Luke or of ail three synoptists can explain, how the présent
gospel of Matt. obtained the name. On this view the présent gospel naturally took
the name of that old document which it alone, of our présent gospels at least, repro-
duced and of which it might almost be considered only an enlarged édition. Because
of the relation of this document to our first gospel, and its probable relation to the
apostle Matthew, we may properly call it the Matthean document, and refer to it by
the abbreviation M.
13.	The peculiar facts respecting the use by Matt. and Luke respectively of that
material which the former has assigned to the Perean period, and which the latter has
included in his great discourses, point strongly to the existence of this material as part
20 Luke 19:1-27 has another version of what is evidently latter portion, vss. 26-29 of Matt., 22-26 of Luke, the paral-
the same parable which Matt. reports in 25:14r-30 ; in the lelism is very close.
23342 Principles of Literary Criticism and the Synoptic Problem
of a separate document possessed b y them in common, but very differently employed.
That both evangelists used such a common source, not one of them the completed
gospel of the other, is shown by facts already referred to (see 10, pp. 28 f., and the dis-
cussion of hypothèses O and R, pp. 29, 31). It remains, however, to inquire what was
the extent of this document, and in what form it existed. Obviously our first inquiry
is whether either of its présent existing forms will account for the other; if so, it is
needless to posit some third non-existent form.
Will the form in which it exists in Matt. account for its form in Luke? To this
a négative answer must be returned. (a) As this material exists in Matt., it is dis-
tributed through the Galilean ministry and the passover week. To suppose that Luke
possessed this would be to suppose that he had a considérable part of the présent gospel
of Matt. But in that case it would be difficult to explain why, reproducing the dis-
course material, often with close verbal exactness, there should be so nearly a total
absence of any influence of the narratives of Matt. in the same period upon Luke’s
narrative sections, (b) It would be impossible to explain why Luke should segregate
this particular material, gathering it out of the period in which Matt. had placed it
into a period in which Matt. had placed none of it. (c) It would compel the supposi-
tion that Luke had detached material from the discourses in which it stood in Matt.
and invented for much of it entirely new historical occasions. For it is a fact long
ago and often observed that the discourse material which Luke introduces by short
narratives, Matt. includes in his long discourses, (d) It would leave still unaccounted
for that portion of Luke’s account of the Perean period which is contained neither in
18:15-43, parallel to Mark, chap. 10, nor in Matt., and so would still compel the
positing of another source.
On the other hand, the hypothesis that a once separate document substantially
identical with Luke 9:51—18:14 and 19:1-28 was possessed by both Matt. and Luke,
the latter incorporating it bodily, save for the interjection of 18:15-43 from Mark,
the former making general excerpts from it, chiefly of the sayings of Jésus, accounts
for ail the phenomena, and is itself open to no serious objection. That it was in its
original form designated as a narrative of the Perean ministry is, in view of the first
evangelist’s use of it, not at ail probable. The position assigned to it by Luke is
perhaps sufflciently explained by a lack of any definite désignation of it as the nar-
rative of a particular period, together with the fact that both this document (Luke
18:35; 19:1) and Mark (10:46) recorded an arrivai at Jéricho, which it was obvious
to identify. The transposition of material from its connection in this document to
the location as given by Matt. would be unnatural in a writer who wished to give the
sayings of Jesus as nearly as possible in tlieir original connection, but not improbable
in a writer who, as was evidently the case with the first evangelist, was building up an
argument out of narrative material, and who, in details at least, controlled his arrange-
ment by topical, not chronological, considérations. It would be easier to account for
the absence from Matt. of so large a part of what is found in this document as it
234Ernest DeWitt Burton
43
appears in Luke, if we could suppose that he possessed a shorter recension of it, or that
Luke added to the original common document material derived from tradition or some
other document. But the simplest view is nevertheless that Matt. and Luke both
possessed a document substantially identical with Luke 9:51 —18:14; 19:1-28,
lacking, however, any indication of the précisé period to which the events belonged.
With reference simply to its location in Luke it may be called the Perean document
(p)-
14. Mention has been made (10, p. 28) of five sections in part or wholly parallel in
Matt. and Luke, this material not being found in Mark. One of these, the sermon on
the mount, has already been discussed. The other four sections are the preaching of
John the Baptist, the temptation of Jésus, the centurion’s servant, and the message
from John the Baptist—ail, it will be observed, narrative in character, and ail in the
John the Baptist period, or in the Gralilean ministry. In the first two instances the
material common to Matt. and Luke appears in close combination with material found
also in Mark. In the last two there is no related matter in Mark. The existence of
these sections, suggesting that if the first and third evangelists possessed Mark as a
common source, they had also some other source in common, naturally raises two
inquiries: first, whether there is in either of the gospels any material that by its
similarity or its connection betrays itself as probably from the same source; and,
second, whether there is any due to the method of either evangelist in using the two
sources. In the attempt to answer the first of these questions our attention is directed
to two groups of material. In the first of these Luke relates events which as events
are either identical with, or very similar to, events also narrated by Mark, but the
accounts of which are evidently not the same. These narratives are the call of the
four (Mark 1:16-20; c/. Luke 5:1-11), the rejection at Nazareth (Mark 6:1-6; cf.
Luke 4:16-30), and the anointing of Jésus in the house of Simon (Mark 14:3-9; cf
Luke 7:36-50). The last named of these, it is to be noted, stands in immédiate con-
nection with the fourth of the narratives spoken of above as in whole or in part com-
mon to Matt. and Luke only. The second group includes passages peculiar to Luke
standing in close connection with material common to Matt. and Luke. These passages
are the words of John to the various classes of people (Luke 3:10-15, 19), the raising
of the widow’s son (Luke 6:11-17), and the women who accompanied Jésus (Luke
8:1-3). Thus the whole of Luke 7:1 — 8:3 is made up of material either common to
Matt. and Luke or found only in Luke ; in other words, of material not found in Mark.
If now we examine ail the material of these three groups—that which is common to
Matt. and Luke, that which apparently displaces similar material in Mark, and that
which is peculiar to Luke but stands in immédiate connection with the sections first
named—we discover a marked uniformity in general literary character. The narra-
tives are ail vividly told, surpassing in this respect even the vivid narratives of Mark,
and in literary style reaching the high-water mark of this gospel. With this material it
is natural to associate the other passage in Luke’s Galilean ministry which is not
23544 Principles of Literary Critioism and the Synoptio Problem
paralleled in Mark, viz., the sermon on the plain, or, if we may so call it, the Lukan
version of the sermon on the mount. Ail the facts taken together suggest that in
these passages, Luke 3:7—15, 17,18; 4:2b-13 [14,15] 16-30; 5:1-11 ; 6:20-49; 7:1—
8:3, we hâve a gospel behind and within our présent gospel.21 The only alternative
supposition is that we are dealing here with fragments of tradition, or brief separate
documents, and against this view may be urged alike economy of hypothesis and the
general literary resemblance of the whole material.22 Having reference to the prevail-
ing character of the content of this document, we may perhaps appropriately call it
the Galilean Document, and refer to it by abbreviation as G.
15. Among the passages just discussed are, as already mentioned, two that take
the place of similar narratives in Mark, not in the sense that they occupy corresponding
positions, but that Luke, who alone inserts them, does not contain the different and
differently placed accounts of similar events contained in Mark (and Matt.). This
suggests the inquiry whether there are other instances of a similar proceeding on
Luke’s part. Examination discloses a number of such.
Thus, in addition to the two instances already named, it is to be observed that
Luke, who tells in 7:36-50 the anointing of Jésus in the house of Simon the Pharisee,
omits the story of the anointing of Jésus in the house of Simon the leper told in Mark
14:3-9 and Matt. 26:6-13; relating in 22:24—27 Jésus’ rebuke of the ambition of the
disciples, he omits the partly similar account which Mark has in 10:35-45; and giving
in 11:14-36 an account of the casting out of a démon and the charge of the Pharisees
that Jésus cast out démons by Beelzebub, he omits the account which Mark gives in
3:22-29 ; narrating in 10:25-37 the answer of Jésus to the question of the lawyer,
he omits the similar conversation with a lawyer recorded in Mark 12:28-34
and Matt. 22:34—40, though retaining certain phrases of Mark’s account. In
the two cases first mentioned Luke and Mark both assign their respective nar-
ratives to the Galilean ministry (though not to identical places in that period); in
the last three cases the two similar events or narratives are assigned to distinctly
different periods. The first of the three Luke places in the Galilean ministry, Mark
in the passion week ; the second Luke places in the passion week, Mark in the Perean
21	Mr. Badham was, I believe, the first to call attention
to the relation of these narratives to one another. He
believed them to constitute the original Petrine gospel of
Mark, no part of which he held is contained in our présent
Mark. See his Formation of the Gospels, first édition.
22	There is not lacking a hint that there was in this
lost book still other material additional to that now found
in Luke. Returning to the narratives which apparently
displace similar accounts in Mark, we may notice that,
while the narratives common to Mark and Luke are almost
without exception in the same order in the two gospels,
the relation of Mark and Luke being in this respect in
marked contrast with the relation of Mark and Matt., yet
in those instances in which Luke gives a distinct account
of what is apparently the same event as that narrated by
Mark, he gives it also a different position. This fact is not
only an indication that in these cases Luke is not simply
expanding and embellishing the narrative of Mark, but is
strong evidence that the narratives did not corne to him as
detached pièces. Their position, different from that of the
sections which they displace, can only be accounted for on
the assumption that they had a place in the source from
which they were taken. But since, of course, place in this
case is not absolute but relative, i. e., consists in being after
or before some other event —it is again strongly suggested
that this lost source had some of the Mark events in rela-
tion to which they are set by Luke. If this be so, then it
follows that Luke has not reproduced the whole of this
book, but that, as is in itself intrinsically probable, he has in
some cases preferred Mark to his other source, just as he
has in these cases preferred the other source to Mark. But
it will be évident that for order and general structure he
has built chiefly on the Unes laid down by Mark.
236Ernest DeWitt Burton
45
period; the third Luke places in the Perean period, Mark in the Galilean. It is very
significant that while Matthew’s account of this event occupies approximately the same
position as Mark’s, between the message from John the Baptist and the visit of Jésus’
kindred to him, and is in considérable part verbally almost identical with Mark’s
account, yet it is also in other portions as closely like the account which we possess in
Luke. It is perfectly clear that Matt. and Luke drew from a common source other
than Mark, and scarcely less so that Matt. drew from Mark also. Whether Luke drew
from this common source only (which in that case must hâve had some verbal simi-
larity to Mark, but not enough to imply common literary origin), or, like Matthew,
used Mark also, is not perfectly certain.
But there are still other facts to indicate a disposition on Luke’s part to avoid
including similar narratives even when contained in the same source. Thus the gospel
of Mark, which, we hâve found reason to believe, was his chief source, contains an
account of the feeding of the five thousand, and also of the feeding of the four thou-
sand. But Luke, having given the first of these two similar accounts, omits the second
one. Along with it, indeed, he omits also ail the events which Mark includes between
the two narratives of miraculous feeding, and the section immediately following the
second one, concërning the sign from heaven. Now, the two chief sayings in the
passage just referred to, Mark 8:10-21, Luke has in his Perean section. Mark 8:12
has its parallel in Luke 11:29; 8:14 in Luke 12:1. Does this perhaps account for
the omission of this section at this point? There is, at any rate, one other similar
case that seems to confirm this supposition. Luke, having given in 16:18 in one
sentence the teaching of Jésus concërning divorce, omits the whole passage which
Mark has in 10:2-12. Whether the explanation suggested for these last two cases be
correct or not, and however we may account for the omission of the events which in
Mark lie between the feeding of the five thousand and the feeding of the four thousand,
it must be évident that Luke avoided in general repeating similar narratives, whether
found in the same or different sources, and that in omitting these apparent duplicates
he did not give the preference uniformly to any one source.23
16. These facts, which seem to indicate a peculiarity of Luke’s method of using
his sources, suggest the possibility of gaining some light upon our problem by a dis-
23 The omission of the walking on the water, Mark
6:47-56, may perhaps be due to its similarity to the stilling
of the tempest, which at 8:22-25 he had already taken from
Mark 4:35-41. Of the omission of the material contained
in Mark, chap. 7,1 can suggest no explanation other than
the not wholly satisfactory one that, finding in the feeding
of the five thousand and the feeding of the four thousand
narratives that he regarded as duplicates, and in the mate-
rial immediately following each of these what he also
wished to omit because of its similarity to material already
included, or which he intended to include, he omitted the
whole block, from Mark 6:47 to 8:21. It would still remain
necessary to account for the omission of Mark 8:22-26.
That this is neither in Matt. nor in Luke suggests the
possibility that it was not présent in the copy of Mark
which the other evangelists used; in other words, that the
original Mark was not quite identical with the présent
Mark. And this in turn raises the question whether the
whole of Mark 6:47 (or 45) to and including 8:26 may
possibly hâve been lacking in Luke’s copy of Mark. Con-
ceming the general view of a primitive Mark see above,
pp. 24-26.
There are but three significant instances in which
Luke has retained two similar narratives. These are (a)
the sending out of the Twelve, 9:1-6, and the sending out of
tbe Seventy, 10:1-16 ; (b) the healing of the leper, 5:12-16,
and of the ten lepers, 17:11-19; (c) Jésus’ prédiction of his
death and résurrection, 9:22-27 ; 9:43-45 ; 18:31-34. But in
none of these instances is the similarity of the two narra-
tives such as to suggest identity of the events.
23746 Principles of Literary Critioism and the Synoptic Problem
covery of like peculiarities on the part of the first evangelist. A survey of the gospels
in parallelism discloses two facts that seem to be of importance. The first of these is
the condensed character of Matthew’s narrative at certain points as compared with
the accounts in the other gospels. Examples of shortening or condensation as com-
pared with Mark are found in ail the following sections:
A day in Capernaum :
Healing of a leper :
Healing of a paralytic :
The Gerasene demoniac :
Jairus’s daughter:
Rejection at Nazareth :
Death of John the Baptist :
Feeding of the five thousand :
Eating with unwashen hands :
Mark 1:21-34
Mark 1:40-45
Mark 2:1-12
Mark 5:1-20
Mark 5:21-43
Mark 6:1-6 a
Mark 6:14-29
Mark 6:30-46
Mark 7:1-23
Matt. 8:14-17.
Matt. 8:1-4.
Matt. 9:1-8
Matt. 8:28-34.
Matt. 9:18-26.
Matt. 13:54-58.
Matt. 14:1-12.
Matt. 14:13-23.
Matt. 15:l-20.24
In the few instances in which the narrative of Matt. is longer than Mark’s, this
arises from the addition of some saying of Jésus, or of some comment of the evangelist,
the latter sometimes in the form of a reference to the fulfilment of prophecy.25
In the few narratives that are common to Matt. and Luke only, the relation
of Matthew’s account to Luke’s is in general the same as to Mark’s in the cases above
enumerated. Thus in the story of the centurion’s servant (Matt. 8: 5-13 ; Luke
7 :1-10), the narrative in Matt. is but two-thirds of the length of that in Luke, the
apparent equality of the accounts being due to the inclusion by Matt. (vss. 11, 12) of
certain sayings of Jésus found in Luke’s Perean section (13:28, 29). In the story of
the message from the Baptist (Matt. 11:2—19 ; Luke 7:18-35), the narrative in Matt. is
again shorter than Luke’s by the omission of the latter’s vss. 20, 21, this shortage in
narrative being balanced in quantity by the inclusion of sayings of Jésus in vss. 12-
15, in part parallel to a passage in Luke’s Perean section (16:16), in part parallel to
Mark 9:12, interpreted as Matt. 17:13 shows that this evangelist understood it of
John the Baptist, but perhaps derived from the independent sources of Matt. In the
narrative of Jésus’ answer to the disciples (Matt. 8:19-22; Luke 9:57-61), the nar-
rative of Matt. is shorter by the omission of the story of the third disciple. It is
particularly instructive to compare the relation of the Matt. and Luke narratives in
the story of the centurion’s servant with that of the Matt. and Mark accounts in the
record of the raising of Jairus’s daughter. In both cases the abbreviation of the
Matt. story is due chiefly to his omission of the part taken by the agent of the princi-
pal person, and the ascription directly to the centurion in the one case, the ruler of the
synagogue in the other, of that which in the parallel narrative is said to hâve been
done for him by others. Now, aside from the fact that there is nothing in Matthew’s
narrative in either case to tempt to expansion by the introduction of an agent to act
for the centurion and the ruler, it is certainly far more probable that the mind of one
25 See Mark 1:14, 15 = Matt. 4:12-17; Mark 2: 23-28 =
Matt. 12:1-8 ; Mark 6:47-52 = Matt. 14:24-33.
238
24 In this case Matt. also adds vss. 12-14.Ernest DeWitt Burton
47
writer should act in the same way in respect to both his sources (in both cases, that is,
condensing by omitting the agent) than that two writers (Mark and Lnke) should
independently pursue the same course toward different narratives in the one source
(Matt.). And this probability becomes still greater when we take into account the
whole list of sections in which the narrative of Matt. is shorter than that of Mark, on
the one hand, or of Luke, on the other. Economy of hypothesis, and the principle
that that form of narrative is secondary which can be accounted for by the observed
tendency of the writer, are decidedly in favor of attributing to Matt. condensation of
both his sources, rather than reversing it and attributing to both Mark and Luke
expansion. And this again is still furtlier strengthened by observing that Luke’s
tendency, as seen in narratives common to Mark and Luke, is to condensation (less
marked than in the case of Matt.) rather than to expansion.
The second characteristic of Matt. as compared with Mark and Luke has already
been referred to, viz., his tendency not only to combine discourse material together
into large discourses, but to include in narrative sections, parallel to Mark or Luke,
sayings of Jésus not found in the parallel accounts, sometimes with brief narrative
introductions also peculiar to him. Instances of this (most of which hâve already been
mentioned) may be seen in 3:14, 15; 8:11, 12; 9:13a; 11:12-15, 28-30; 12:5-7,
11, 12a, 33-37; 15:12-14, 23, 24; 16:17-19; 19:10-12, 28; 20:1-16; 21:14-16,
28-32, 43 ; 22 :1-14. Of these passages, eighteen in number, nine are in the midst
of narratives given by ail three evangelists. That neither Mark nor Luke contains
them points, of course, in the direction of the explanation of them as additions by
Matt. rather than as omissions by Mark and Luke from sources possessed by them.
The other nine cases, though not directly supporting this view, are in no way opposed
to it.
It remains to consider briefly those éléments of Matthew and Luke that hâve not
been discussed.
17. Matt. has a comparatively small number of narrative passages, besides those
in the infancy period, which are not paralleled in either of the other gospels. Some of
them are brief complété narratives, others add details to narratives found in one or
both of the other gospels. The list is as foliows: removal to Capernaum, 4:13;26
Peter’s attempt to walk on the water, 14:28-31; details of the transfiguration, 17:6,
7 ; the repentance and death of Judas, 27 : 3-10; Pilate’s washing of his hands, 27 : 24,
25; the résurrection of the saints, 27 :51b-53; the sealing of the tomb, 27:62-66; the
report of the watch, 28:11-15.
To these it might seem necessary to add the introduction to the sermon on the
mount, 4: 23-25 ; the introduction to the missionary discourse, 9 : 35-37 ; and the heal-
ing of two blind men and the dumb demoniac, 9 : 27-34. But upon examination the
first two of these passages are found to bear ail the marks of being conflations, the
first of material derived from different passages of Mark, the second of material
26 Perhaps derived from Mark 1:21 ; 2:1.
23948 Principles of Literary Critioism and the Synoptio Problem
derived from Mark and the Perean section of Lnke. Thus Matt. 4: 23-25 is substan-
tially paralleled in Mark 1: 39, 28, 32-34; 3:7, 8; and Matt. 9 :35-37 in Mark 6: 6,
12, 13, 34. Such conflation points to the dérivative character of this gospel. The
expression ScScta/ccov èv raZç <rvvay(oyaî<t avrœv /cal /cr}pv<raœv to evayyeXiov tî)ç fiaaikeias
/cal Oepairevwv Trâaav voaov /cal iràaav pa\a/ctavy occurring in both these summaries,
is apparently the writer’s general phrase to describe the work of Jésus, its language
being his own, the idea being furnished by the passages above referred to as his
sources. It is not impossible that in the light of these passages 15: 30, 31 should be
regarded simply as a summary composed by the evangelist. As respects the two
narratives composing 9: 27-34, these are probably to be regarded as duplicates of the
accounts appearing in 12: 22 ff., and 20: 29-34.
The éditorial remarks of the first evangelist, including his references to the
fulfilment of prophecy, call for no comment.
18. Of the passages peculiar to Luke ail hâve already been considered except
those that are found in the passion and résurrection history. It will suffice simply to
point out that in this portion of Luke there are the following passages of which there
are no parallels in the other gospels:
Additions to the story of the triumphal entry:
Additions to the story of the last supper:
Additions to Christ’s last conversation with the disciples :
Gethsemane, mainly an independent account : most of
Variant account of the condemnation of Jésus by the Jews:
Additions to the story of the trial before Pilate:
Additional incidents of the crucifixion:
Independent story of the résurrection moming:
Walk to Emmaus:
Appearance in Jérusalem:
Ascension :
19:37,39-44;
22:15-17, 24-30;
22:31-33, 35-38;
22:39-46;
22:66, 67;
23:2,4-16;
23:27-32,36, 37,39-43;
24:3-5a, 6 b, 7-11;
24:13-35;
24:36-43;
24:44-53.
It is a very notable fact that, while Matt. and Luke each hâve in their Jérusalem
period considérable material not found in Mark, they hâve no such material in
common.27 This fact clearly indicates that the sources additional to Mark from which
Matt. and Luke drew were not, as in the John the Baptist period and the Galilean
ministry, the same, but quite distinct.
The préfacé and other éditorial remarks of Luke require no spécial discussion.
If now we re-enumerate the separate sources or strata of material which we hâve
27 To this statement there are but two exceptions,
these perhaps only apparently so. Matt. 21:44 is clearly
parallel to Luke 20:18. There is, however, sufficient docu-
mentary evidence for the omission of the verse in Matt. (D.
33, a, b, e, ff. 1 2 Or. omit; fcÇBCXZAn une 10 etc. include it),
to lead Tischendorf, in view of it and the internai evidence
furnished by the évident appearance of harmonistic cor-
ruption, to omit the verse, and Treg. and WH. to bracket
it. Weiss argues for its genuineness. Luke 22:30 is
parallel to Matt. 19:28, the usual relation being in this case
reversed, a passage which Matt. has in the Perean period
being found in Luke’s passion week. Were there many
such instances, they would be fatal to some of the positions
maintained in this paper. But the comparison of these
verses with their respective contexts renders it probable
that in this case as in several others (c/. Mark 3:22ff. with
Luke ll:15ff. ; and Mark 4:30-32 with Luke 13:18,19; Mark
9:42 with Luke 17:2 ; and Mark 12:39 with Luke 11: 43) ;
substantially the same saying of Jésus was contained in
two independent documents.
240Ebnest DeWitt Burton
49
seemed to discover lying behind Matt. and Luke—that is, of one or both of them —
and for convenience of reference indicate again the contents of each, we hâve the
foliowing list:
MAJOR SOURCES
1.	The gospel of Mark, substantially as we now possess it.
2.	The spécial Matthean source, probably the Logia of Matthew spoken of by
Papias, consisting of substantially the following portions of Matt.: 3:14, 15; 5:4, 7-
10, 13a, 14,16, 17, 19-24, 27, 28, 31, 33-39a, 41, 43; 6:1-7, 10b, 13b, 15, 16-18,
34; 7:6, 12b, 15, 22; 9:13a; 10:5, 6, 8b, 23, 25b, 36, 41; 11:28-30; 12:5-7, 11,
12 a, 34; 13:14, 15, 24-30, 35-53; 15:12-14, 23, 24; 16:17-19; 17:24-27 ; 18:
4, 10, 14, 16-20, 23-34; 19:10-12, 28; 20:1-15; 21:14-16, 28-32, 43; 22:1-14;
23:2, 3, 5, 7b-10, 15-22, 24, 28, 32; 24:10-12, 30a; 25:l-lla, 13, 14-46;
26:52, 53 (ilf).28
3. The Perean document, consisting of Luke 9:57—18:14; 19:1-28, portions of
it being contained also in the record of the Galilean ministry and passion week of
Matt. (P).
4.	The Galilean document found in Luke 3:7-15, 17,18; 4:2b-13 [14, 15], 16-
30; 5:1-11; 6:20-49; 7:1—8:3 (G).
MINOR SOURCES
5. The infancy story of Matt.—Matt., chaps. 1, 2.
6. The infancy story of Luke—Luke, chaps. 1, 2 (except 1:1-4).
7.	Other narratives peculiar to Matt.
8.	The narratives peculiar to Luke, and not included in 2, 3, or 6, viz., his addi-
tions to Mark’s narrative of the passion week and résurrection story.
Ail the major sources were beyond doubt written. Of the minor sources, 6 gives
internai evidence, which it is not necessary to state here at length, of having existed
in writing before being taken up into the gospel. The other minor sources may
perhaps not hâve been reduced to writing before their incorporation in our gospels.
It is a legitimate question, and one that demands considération, whether each of
the eight sources is really distinct, or whether two or more may not hâve been parts of
one document. From this discussion Mark is, of course, excluded by the fact that we
actually possess it and know its extent. That G and P were parts of one is an obvious
suggestion, but one which is not sustained by the evidence. Against it is the fact
that Matt. has treated these two sources so differently, using G with only that degree
of freedom with which he also used Mark, i. e., condensing accounts, and transposing
events to some extent within the same general period, y et not transgressing the limits
of the period, but employing P almost exclusively as a source from which to draw
discourse material to enrich the great discourses, and always placing this material both
in a different connection and in a different period from that in which Luke has given it
28 On purely internai grounds one would perhaps refer source 7 ; but for such reference there is no objective
16:17-19; 17:24-27; 19:28; 22:7; 26:52, 53 to the minor ground.
24150 Principles of Literary Criticism and the Synoptic Problem
to us. This renders it probable that the documents were not one, but two ; G having
some marks by which to indicate to what period its events belonged, P being without
such marks. The name Perean therefore is correct only as a term of convenience
derived from the position in which the document as we hâve it stands in Luke.
Against the combination of P and M are several reasons: (a) It would be difficult,
not to say impossible, to explain why Luke should hâve omitted so large a part of that
which the document must in that case hâve contained, viz., the whole of that discourse
material which is peculiar to Matt. and which on this hypothesis was contained in a
document which Luke had in his hands. For the combined document must hâve been
either P with M scattered through it, or P with M prefixed or added to it. In the
former case, Luke must simply hâve torn out ail these sections; in the latter case,
the document would hâve been M + P or P -f- Af, and Luke must hâve eut off
M; in either case with no motive that can be discovered. For Matthew’s omis-
sions from P there is a natural explanation, in that his interest was in discourse
material, and that, having taken Mark as his basal narrative, he used both M and P to
build up the discourses, the starting-point of which he found in Mark or in G. But for
Luke’s omission of the whole of the M material there is no explanation consistent with
his general method of employing his sources, (b) Since P and M were ex hypothesi
one, Luke must hâve had both in one, and this supposition deprives us of that expla-
nation of the traditional name of our first gospel which is furnished by the supposition
that they were separate and that M was taken up into the first gospel only. (c) M
alone corresponds much better to the description given by Papias of the (original)
gospel of Matt. than does the supposed combination of M and P. (d) It is difficult
on this hypothesis to explain the way in which Matt. has built up his version of the
sermon on the mount. If he possessed G and another source which combined P and
Af, two methods of procedure are possible. First, G may hâve contained the discourse
as we now hâve it in Luke 6:20-49; on the basis of this, Matt. may hâve built up
that discourse which we find in Matt., chaps. 5-7, when we subtract the passages
found in Luke’s Perean section and in Mark, and then hâve added this latter material.
In that case the same writer has from the same sources first constructed an orderly,
well-planned discourse, and then has marred its structure by additions from the same
source. Secondly, he may hâve found in his Perean-Matthean document a discourse
like that just supposed, Matt., chaps. 5-7, minus the material paralleled in Luke’s
Perean portion and in Mark. In this case he must hâve marred the unity of this
discourse by transferring to it material from another part of the same document. This
is not impossible but it is distinctly more complicated and less probable than that the
partially incongruous material came from different sources.
The hypothesis that G and M were in reality one is open to substantially the same
objections as those just stated against the view of the unity of P and Af. It fails to
account for Luke’s omission of ail that large portion of the combined document which
is peculiar to Matt.; it fails to account for the name of the first gospel; it is not
242Ernest DeWitt Burton
51
accordant with the testimony of Papias, and it complicates the explanation of the
sermon on the mount as found in Matt. This fourth objection assumes now, however,
a slightly different form. The hypothesis that G and M were one involves the
supposition either that this document contained two versions of the sermon on the
mount, which is certainly improbable, or that it contained a version on which both the
existing versions were based, modified in Matthew’s case, of course, in part by the
addition of material from the Perean document, and in part from Mark. We hâve
already had occasion to observe the improbability of this hypothesis.
A further reason might perhaps be urged against the combination of G and P,
but clearly holds against the union of G and M, viz., the clear différence of literary
character between the two éléments. G is chiefly made up of narratives, distinguished
for their vividness and fulness of detail. M is composed of sayings of Jésus without
narrative introduction, so far as the use of them in Matt. indicates, many of them
aphoristic and poetic in form. It is certainly more probable that material so different
in character and so differently treated by the evangelists constituted separate docu-
ments than that they were parts of one and the same document.
On the whole, therefore, the hypothesis of the separateness of these three docu-
ments is more probable than that any two of them were constituent parts of a single
document.
Respecting possible combination of the minor sources with one another, or with
any of the major sources, a few remarks will suffice.
The combination of the infancy narrative of Matt. or of the other narratives
peculiar to him with any source used by Luke is rendered improbable, especially as
respects the infancy period and the passion week and the résurrection story, by the
absence of any influence of these portions of Matthew’s sources upon Luke. The
possibility of simple non-use can never be disproved, but ail the evidence is
against it.
For like reasons the combination of Luke’s narrative of the infancy, or of his
spécial narratives of the passion week and the résurrection story, with any source used
by Matt., is improbable.
A combination of the infancy narrative of Matt. with the narratives peculiar to
him in other parts of the gospel can neither be proved nor disproved. Considérations
of style and purpose are somewhat against it.
A combination of Luke’s infancy narrative with his spécial material at the end of
the gospel is practically excluded by the long interval between them and the marked
différence in style.
The most serious question is whether the narratives peculiar to Matt. are from
the same source as the discourse material peculiar to him. Are M and the minor
sources 5 and 7 parts of the same source? Objective tests fail. Ail three groups of
material are alike in that they are absent from Mark and Luke. The argument from
style would certainly favor diversity of source, so far at least as to distinguish the dis-
24352 Principles of Literary Critioism and the Synoptio Problem
course material from the narrative material. If to these be added the wide différence
in intrinsic character and spirit between the discourses and the narratives, the former
generally conceded to constitute probably the oldest stratum of evangelic material, the
latter bearing marks of being the very latest, the infancy story having exerted no
influence on the Pauline literature, or indeed on any extant document that can be
dated earlier than the gospels of Matt. and Luke themselves, and such narratives as
the résurrection of the saints (27:52-53)29 strongly suggesting remoteness from the
events, there can scarcely remain room to doubt that we hâve in the discourses (M) a
document distinct from the narratives peculiar to Matt. and of older date than these.
The resuit, therefore, of the attempt to combine the different sources enumerated
on p. 49 is négative. The prépondérance of probability is on the side of leaving the
list as it was first given, and of resolving the x of our hypothesis Y (p. 34) into seven
éléments.
But it must now be observed that the facts on the basis of which we hâve reached
this resuit are of value also for the strengthening of the argument by which we excluded
other hypothèses.
Thus the facts stated under 11 and 18, by showing the different relation of Matt.
to different portions of Luke, and the mutual exclusiveness of their additions to Mark
in the latter part of the gospel, are against any hypothesis that makes Matt. dépendent
on Luke as it now stands; i. e., against B, E, I, J, R, S, T, U. The facts stated under 11,
13, 14, and 18 make against any hypothesis which makes Luke dépendent on Matt. and
in favor of the dependence of Matt. on Mark and some source of Luke; i. e., against
A, C, G, K, O, P, Q, V. The facts stated under 11, 13, 14, and 18 are strongly against
N. The complexity and at the same time uniformity of the relations of the several gos-
pels could never be explained by assuming a single document as their common source ;
and if x be resolved into several éléments, the theory is no longer the same. These
facts are for the same reason opposed to M ; and if x be resolved into éléments, as has,
e. g., been done by Arthur Wright, then, in view of the close resemblances of the gos-
pels to one another both in order of events and in language, each of the éléments of x
becomes in effect a document, only ex hypothesi committed to memory instead of
written down. The facts stated under 15 somewhat strengthen the argument in favor
of making Mark a source of Luke, and those under 16 are in favor of making Mark a
source of Matt.; i. e., against C, D, E, F, G, I, J, K, L, M, N, O, P, Q, S, T, and in
favor of H ; or, since that has been shown to be inadéquate, in favor of Y.
CONCLUSIONS
The conclusions to which our whole study has led may then be summarized as
f ollows :
1. Our Mark, or a document in large part identical with it, was employed as a
source of both our first and third gospels.
29 Concerning Matt. 17:24-27 and other narratives introducing sayings of Jésus, see note 28.
244Ernest DeWitt Burton
53
2.	Matt. and Lnke also possessed in common a docnment which contained sub-
stantially the material standing in Luke 3:7-15, 17, 18; 4:2b-13 [14, 15], 16-30;
5:1-11; 6:20-49; 7:1—8:3; herein referred to as the Galilean document (G).
3.	Matt. and Luke also had a document in whole or in part identical with Luke
9:51 —18:14 and 19:1-28, which, however, they used in very different ways; herein
referred to as the Perean document (P).
4.	Matt. also had a document not employed by Luke, chiefly or wholly made up
of discourse material. This is presumably the Logia of Matt. spoken of by Papias (M).
5.	Additional minor sources there must also hâve been, the first and third
evangelists having in the main different ones, as is illustrated in the case of the infancy
narratives and the almost wholly independent additions to Mark’s account in the passion
and résurrection history.
6.	Thus the sources of Matt. are the Matthean Logia, Mark, the Galilean docu-
ment, and the Perean document, besides certain minor sources. In his employment
of these sources the first evangelist gave the chief place to Mark and the Matthean
Logia, employing the Galilean document for illustrative purposes, and the Perean
document for enrichment of the discourses the basis of which was found in the Logia
or in Mark.
7.	Luke has the same chief sources as Matt., with the exception of the Matthean
Logia. In his use of them he made Mark the basis, interpolated material from the
Galilean document, omitting Mark’s similar narratives when they seemed to him less
full and vivid; added the Perean document in two solid sections, making the junction
with Mark in such way that the arrivai at Jéricho indicated in this document should
synchronize with that recorded by Mark.
Each of the two later evangelists pursued a consistent and easily intelligible
method in the use of the sources, but each his own method.
8.	The agreements of Matt. and Luke against Mark in triple narrative which
are scattered through the gospels are an unexplained remainder. To make them
determinative for the whole theory is, however, to set the pyramid on its apex. The
line between éditorial révision and scribal corruption was probably not so sharply drawn
in fact as our rigid discrimination between higher and lower criticism would suggest.
There are many indications that the one class of phenomena shade into the other.
This unexplained remainder probably owes its origin to causes that belong to the
border line between éditorial révision and scribal corruption, or else to some slight
influence of one of these gospels in its final form on the mind of the writer of the
other.
9.	The question of sources behind the above enumerated sources, in particular
the sources behind Mark and the Perean document, remains to be investigated. For
lack of documents to compare, the methods by which this problem must be solved
must be to a considérable extent different from those by which the results of this paper
hâve been reached.
24554 Principles of Literary Criticism and the Synoptic Problem
APPENDIX
TABLE EXHIBITING THE PAKALLELISM OF THE SYNOPTIC GOSPELS30
The following table is intended to show the parallelism of the synoptic gospels as fully as
can be done without actually printing the text. It aims to do this with the least possible dis-
arrangement of the order of the material as it stands in the gospels. The principles on which
it is constructed are the following:
1.	A distinction is made between parallel sections and parallel passages in non-parallel
sections.
Parallel sections are those which report what is shown, either by position and similarity
of content, or by close literary similarity of component material, to be kindred reports of the
same event or discourse. Discourses basally the same and similarly located are treated as
parallel, however different in extent. On the other hand, narratives that are at the same time
differently located and so different in form or content as to indicate that they are of diverse
literary origin are not treated as constituting parallel sections, even though the general simi-
larity of the accounts renders it probable that the events recorded are really the same. Thus
Luke 4:16-30 is distinguished from Mark 6:1-6, and Luke 11:14-36 from Mark 3:22-30,because
in each case what were evidently in the sources two different accounts are by the different
evangelists differently located. On the other hand, Matt. 12:22-45 and Mark 3:22-30 are treated
as parallel sections, because the two are basally the same, and similarly treated; and Matt.
8:19-22 and Luke 9:57-60 are identified because though differently placed the two accounts are
practically identical.
Parallel passages are those which express the same thought. As here employed, the term
covers passages resembling one another in ail degrees from identity of thought and words to
similarity of thought with almost entire dissimilarity of language.
2.	The length of the sections has been determined by the unity of the event or discourse.
In this mat ter there is, of course, room for différence of judgment.
3.	In order of sections Mark has been followed for ail sections that this gospel contains.
This procedure seems required by the fact that the other two gospels, each agreeing with Mark
in a large majority of instances and each sometimes departing from his order, never agréé
against him in such departure.
4.	Sections standing in Luke only and between sections paralleled in Mark, are placed in
the position assigned them by Luke.
5.	Sections standing in Matt. only, and between sections paralleled in Mark, are placed in
the position assigned them by Matt.
6.	Sections common to Matt. and Luke and not found in Mark are placed in the position
given them by Luke, except in the case of the généalogies, which are put in the Matt. position.
This general preference of Luke over Matt. is suggested by the fact that Luke evidently in
general follows his sources in the matter of position, while Matt. as evidently rearranges under
the influence of topical considérations.
7.	When, as in the case of the infancy narratives, there are no parallel sections by which
to détermine the relative position of sections drawn from different gospels, the sections are
located according to such internai indications of relation as the narratives furnish, the order of
each evangelist being in ail cases retained. For the purposes of this table the order of such sec-
tions is a matter of little importance.
30 For the privilège of using in the présent table certain New York. I hâve also to express my thanks to my col-
facts and phraseology already employed in the tables in league, Dr. Edgar J. Goodspeed, and to my friend, Mr.
Stevens and Bürton’s Harmony of the Gospels I desire to John W. Bailey, Ph.D., for valuable assistance in the
express my thanks to my colleague, Professor William préparation and vérification of this table and of other sta-
Arnold Stevens, D.D., of the Rochester Theological Semi- tistical matter in the body of the essay.
nary, and to our publishers, Charles Scribner’s Sons, of
246Ernest DeWitt Burton
55
8.	In the arrangement of the material within the section the same general principles hâve
been followed. But in the story of the temptation the order of Matt. has been preferred to that
of Luke. In the passion week Luke has been conformed to the order of Matt. and Mark, in
accordance with the principle stated in 3.
9.	To many portions of the sections arranged on these principles there are parallel passages
in other sections, sometimes in the same gospel, more frequently in another gospel. Instead of
transposing ail such passages from the position given them by the evangelist, in the présent
table each is retained in the section in which it is included by the evangelist reporting it, and
the fact that it has parallels elsewhere is indicated in the columnsheaded “ParallelPassages in
Non-Parallel Sections.” Over against each such passage, verse, or portion of a verse, of which there
are in non-parallel sections parallels in thought or language exceeding a few words or brief phrase,
there are placed in the right-hand columns references to the passages containing such parallels.
10.	The full extent of each section is indicated by the figures in bold-face type. When-
ever the two or more accounts differ materially in content, or when a portion of the section is
paralleled in a non-parallel section, the passages cited in bold-face type are broken into parts
indicated by the figures in light-face type, similar passages being placed opposite each other,
passages peculiar to any one of the accounts standing without opposite references.
That the exhibit thus made of passages peculiar to each gospel, common to ail three, or to
any two of them, does not in ail details correspond with the statistical statements made under
Part II of the essay is due to several facts : (a) The table shows section divisions (printed in
the table in bold-face type) which the statistics of the essay ignore, (b) The table cannot dis-
tinguish, as the essay does, between agreements en bloc and agreements of detail occurring in
threefold or twofold matter. (e) The table cannot well exhibit peculiarities or agreements, even
if important, which extend only to a very few words, or which are intimately entangled in the
text of the passage. In general the table in its bold-face type exhibits the facts less minutely
than the statistics of the essay, in its light-face type more minutely than the en bloc classifica-
tions of the essay, and less minutely than the statements of the essay respecting verbal agree-
ment or peculiarities occuring in threefold matter.
PART I. THE INFANCY NARRATIVES
Section Titles
Parallel Sections
Matt.
Mark
Luke
Luke’s préfacé...
The généalogies .
Birth of John promised..
Annunciation to Mary —
Annunciation to Joseph..
Mary’s visit to Elizabeth
Birth of John the Baptist
Birth of Jésus..........
The angels and the shep-
herds ..................
The circumcision........
The présentation in the
Temple..................
Wise men from the East..
Flight into Egypt.......
Childhood at Nazareth...
Visit to Jérusalem......
Eighteen yrs. at Nazareth
1:1-17
1:1
U2-Sa‘
1:66-16
1:17
1:18-252
2:1-12
2:13-23
1-4
: 236-38
: 346-38
:316-34a
3:236- 31a
:5-25
: 26-381
: 39-56
: 57-80
: 1-7
: 8-20
: 21
2:22-39
: 40
: 41-50
:51,52
Parallel Material in Non-Parallel Sections		
Matt.	Mark	Luke
		
1	With Luke 1:31 cf. Matt. 1:21.
2	With Matt. 1:21 cf. Luke 1:31. With Matt. 1:25a cf. Luke 2:7a. With Matt. 1:256 cf. Luke 2:21.
24756 Principles of Literaby Cbitioism and the Synoptic Pboblem
PART II. THE PERIOD OF PREPARATION
Section Titles	Parallel Sections			Parallel Material in Non-Parallel Sections		
	Matt.	Mark	Luke	Matt.	Mark	Luke
17. Ministry of the Baptist...	3:1-12	1:1-8 1:1	3:1-20 3 :l-2a			
	3:1-3	1:2,3	3:26, 4 3:5, 6			
	3:4 3:5,6 3:7-10	1:4 1:5,6	3:3 3:7-9 3:10-15			
	3:11 3:12	1:7, 8	3:16 3:17 3:18-20	14:3-5	6:17,18	
18. Baptism of Jésus		3:13-17 3:13 3:14,15 3:16,17 4:1-11 4:1, 2 4:3, 4 4:8-10 4:5-7 4:11	1:911 1:9 1:10,11 1: là, 13 1:12,13a 1:136 1:13c	3:21,22 3:21, 22 4:1-13 4:1,2 4:3,4 4:5-8 4:9-12 4:13			
19. The temptation							
						
PART III. THE GALILEAN HINISTRY
Section Titles	Parallel Sections			Parallel Material in Non-Parallel Sections		
	Matt.	Mark	Luke	Matt.	Mark	Luke
20. Departure to Galilee		4:12-17	1:14, 15	4:14,15			
	4:12	1:14a	4:14a			
	4:13-16					
	4:17	1:146,153				
			4:146,15			
21. Rejection at Nazareth						
(Luke)				4:16-30			
			4:16a	13:54a	6:1a	
			4:166	13:546	6:2a	
			4:17-21			
			4:22a	13:54c	6:26	
			4:226	13:55a	6:3a	
			4:23			
			4:24	13:576	6:4	
			4:25-30			
22. Call of the four (Matt.-						
Mark)		4:18-22	1:16-20				
	4:18	1:16				Cf. 5:2
	4:19, 20	1:17, 18				5:106,11
	4:21, 22	1:19, 20				
23. A day in Capernaum		8:14-17	1:21-34	4:31-41			
		1:21	4:31			
		1:22	4:32	7:286,29		
		1:23-28	4:33-37			
	8:14a	1:29a	4:38a			
		1:296				
	8:146, 15	1:30, 31	4:386,39			
	8:16	1:32	4:40a			
		1:33				
			4:406			
	8:166	1:34a	4:40c, 41a			
			4:416		3:11	
		1:346	4:41c			
24. A preaching tour in Gali-	8:17					
lee 		4:23-25	1:35-39	4:42-44			
		1:35-38	4:42,43			
1	4:23a	1:39	4:44			
s Mark 1:146, 15 contains 10 words not paralleled in Matt. or Luke, but too intricately combined with
paralleled portions to be indicated in this table.
248Ebnest DeWitt Bubton
57
PART III. THE GALILEAN MINISTRY— Continued
Parallee Sections
Section Titl.es
Matt.
A preaching tour in Gali-
lée—Continued..........
4:23b
4:24, 25
Mark
Luke
25.	Call of the four (Luke)...
26.	Healingof a leper......
27.	Healing of a paralytic_
28.	Call of Levi............
29.	Question about fasting...
30.	Plucking grain on a Sab-
bath....................
31.	The withered hand......
32. The famé of Jésus
33. The choosing of the
Twelve..................
34. Sermon on the mount.
8:1-4	1:40-45
8:1	
8:2-4	1:40-44
	1:45
9:1-8	2:1-12
9:1	2:1
	2:2
9:2a	2:3
	2:4
9:2b, 3	2:5-7a
	2:7b
9:4-8	2:8-12
9:9-13	2:13-17
	2:13a
	2:13b
9:9-13	2:14-17
9:14-17	2:18-22
9:14,15	2:18,19a
	2:19b
9:16,17a	2:20-22a
9:17b	(2:22b)
5:1-11
5:1
5:2
5:3-10a
5:10b, 11
5:12-16
5:12-14
5:15,16
5:17a
5:17b
5:18a
5:18b,19
5:20, 21a
5:21b
5:22-26
5:27-32
5:27a
5:27b-32
5:33-39
5:33, 34
5:33-37
5:38
5:39
12:1-8
12:1-4
12:5-7
12:8
12:9-14
12:9, 10
12:11
12:12a
12:12b
12:13, 14
12:15-21
12:15a
12:15b
12:16
12:17-21
2:23-28
2:23-26
2:27
2:28
3:1-6
3:1, 2
3:3
6:1-5
6:1-4
6:5
6:611
6:6, 7
6:8a
6:8b
3:4a
3:4b, 5a
3:5b
3: 5c,6
3:712
3:7
3:8
9 • Q
3:10, lia
3:11b
3:12
6: 9a
6: 9b, 10a
6:10b,11
6:17-19
6:17a
6:17b
6:18,19
3:1319
3:13a
5:1-7:29
5:1, 2
5:3
5:4
5:5, 6
5:7-10
5:11,12
5:13a
5:13b
5:13c
5:14
5:15
5:16
3:13b, 14a
3:14b, 15
3:16-19
6:12-16
6:12a
6:12b
6:13
6:14-16
6:20-49
6:20a
6:20b
6:21
6:22. 23
6:24-26
Parallel Material in Non-Parallel
Sections
Matt.
Mark
Luke
9:35b
Cf. 4:18
4:19,20
l:28;32-34;
3:7, 8
Cf. 1:16
1:17, 18
14:5
14:3b
10:2-4
9:50b
14:34b
14:35 a
4:21
249
11:33, c/.8:1658 Principles of Literary Criticism and the Synoptic Problem
PART III. THE GALILEAN MINISTRY— Continued
Section Titles	Parallel Sections			Parallel Material in Non-Parallel Sections		
	Matt.	Mark	Luke	Matt.	Mark	Luke
Sermon on the mount— Continued		5:17					
	5:18					16:17
	5:19,20 5:21-24 5:25, 26					
						12:58,59
	5:27, 28 5:29			18:9 18:8	9:47 9:43	
						
	5:30					
	5:31					
	5:32			19:9	10:11	16:18
	5:33-37 5:38, 39a					
						
	5:396, 40 5:41		6:29			
	5:42 5:43 5:44a		6:30 6:27a 6 : 276, 28a 6:286			
	5:446					
	5:45		6:356			
	5: 46, 47		6:32-34			
	(cf. 5:44) 5:48 6:1-7 6:8 (cf. 32) 6:9, 10a		6: 35a(c/.27) 6:36			12:30 11:2
						
	6:106 6:ll-13a					
						11:3, 4
	6:136 6:14 6:15			18:35	111:25	
						
						
	6:16-18 6:19-21					12:33, 34 11:34, 35 16:13 12:22-29 12:30 12:31
						
	6:22,23					
	6:24					
	6:25-31					
	6:32(c/.8)					
	6:33					
	6:34 7:1,2a		6:37 6:38a 6:386			
						
	7:26				4:246	
			6:39 6:40 6:41, 42	15:146 10:24, 25a		
						
	7:3-5 7:6 7:7-10					11:9-11 11:13
						
	7:11					
	7:12a 7:126		6:31	22:40		
						
	7:13,14					13:24
	7:15 7:16-18		6:43, 44	12:33		
						
	7:19			3:106		3:96
	7:20 ( = 16a)		(=6: 44a) 6:45 6:46	(=12 : 336) 12:346, 35		
						
	7:21					
	7:22					
	7:23					13:27
	7:24-27 7:28a 7:286. 29		6:47-49 (cf. 7:1a)		1:22	
						
						
35. The centurion’s servant...	8:5-13		7:1-10			
	8:5		7:1			
	8:6 (con-		7:2, 3			
	densed)		7:4, 5 7:6a			
	8:7 8:8a					
			7:66 7:7a 7:76-9			
	8:86-10 8:11					13:29 13:286
						
	8:12a					
-				f 13:42		
	8:126			| 13:50 ■{ 22:136 | 24:51		13:28a
						
			1	L25:30		
250Ernest DeWitt Burton
59
PART III. THE GALILEAN MINISTRY— Continued
Section Titles
The centurion’s son—Con-
tinued ..................
36.	Raising of the widow’s son
37.	Message from John the
Baptist................
38.	Anointingof Jésus (Luke)
39.	A preaching tour, women
accompanying. _________
40. Charges of casting out
démons by Beelzebul.......
41. The kindred of Jésus.
42. Parables by the sea.
Paeallel Sections
Matt.
13a
136
11:2-30
11:2, 3
11:4-6
11:7-11
11:12, ié
11:14
11:15
11:16-19
11:20
11:21-23a
11:236
11:24
11:25, 26
11:27a
11:276
11:28-30
9:32-34
12:22-45
9:32,33a
12:22,23a
9:336
12:236
9:34
12:24
12:25a
12:256,26
12:27,28
12:29
12:30
12:31, 32
12:33
12:34a
12:346
12:34c, 355
12:36,37
12:38
12:39a
12:396
12:40
12:41
12:42
12:43-45a
12:456
Mark
3:20-30
3:20, 21
3:22
3:23a
3:236-26
3:27
3:28, 29
3:30
12:46-50	3:31-35	8:19-21
12:46-48a	3:31-33a	8:19-21a
12:486,49	3:336,34	
12:50	3:35	8:216
13:1-53	4:1-34	8:4-18
13:1, 2	4:1	8:4a
13:3a	4:2	8:46
13:36-8	4:3-8	8: 5-8a
13:9 (=13:436)	4:9 (=4:23)	8:86
13:10,11	4:10,11	8:9,10a
13:13	4:12a	8:106
cf. 13:15 13:14, 15	4:126	
Luke
7:10
7:11-17
7:18-35
7:18,19
7:20,21
7:22, 23
7:24-28
7:29, 30
7:31-35
7:36-50
8:1-3
Paeallel Material in Non-Parallel
Sections
Matt.
cf. 21: 316, 32
13:9
13:436
10:15
28:186*
7:16-18,20
3:76
16:4a
16:46
11:15
Mark
4:9
4:23
8:12
Luke
16:16
8:86
14:356
15:13-15
10:12
10:21
10:22a
10:226
11:14
11:15
11:17a
11:176,18
11:19, 20
11:21, 224
11:23
12:10
6:43, 44
3:76
6:455
11:16
11:29a
11:296
11:306
11:32
11:31
11:24-26
14:356
4	The verbal resemblance here is between Matt. and Mark only; Luke though parallel in thought is not so in
language.
5	In transposed order : Matt. 12: 34c = Luke 6:456 ; Matt. 12: 35 = Luke 6:45a.
6	But Matt. contains reference to the three days and nights, not found in Luke.
25160 Principles of Literary Criticism and the Synoptio Problem
PAET III. THE GALILEAN MINISTRY— Continued
Section Titles
Parables by the sea—Con-
tinued ...........
43. The stilling of the tempest
44. The Gerasene demoniac
45. The daughter of Jairus...
46. Rejection at
(Matt.-Mark.).
Nazareth
47. Mission of the Twelve__
Paballel Sections		
Matt.	Mark	Luke
13:16,17		
13:18-21a	4:13-17a	8:11-13
13:216	4:176	
13:22, 23	4:18-20	8:14,15
	4:21	8:16
	4:22	8:17
	4:23 (see 4:9)	
	4:24a	8:18a
	4:246	
13:12	4:25	8:186
	4:26-29	
13:24-30		
13:31, 32	4 : 30-327	
13:33		
13:34	4: 33 , 34	
13:35-41		
13:42 (=13:50)		
13:43a		
13:436(=13:9)		
13:41-49		
13:50( =13:42)		
13:51-53		
b 8:18,23-27	4:35-41	8:22-25
8:18, 23	4:35,36a	8:22
	4:366	
8 : 24	4:37	8:23
	4:38a	
8:25-27	4:386-41	8:24,25
8:28-34	5:1-20	8:26-39
8:28	5 :l-3a	8:26, 27
	5:36-5	8:296
	5:6	8:28a
8:29	5:7	8:286
	5:8	8:29a
	5:9,10	8:30,31
8:30-34a	5: ll-15a	8:32-35a
	5:156,16	8:356,36
8:356	5:17	8:37 a
	5:18-20	8:376-39
9:18-26	5:21-43	8:40-56
	5:21	8:40
9:18-20a	5:22-25	8:41-43a
	5:26a	8:436
	5:266,27a	
9:206	5:276	8:44a
9:21	5:28	
	5:29-33	8:446-47
9:22	5:34	8:48
	5:35-37	8:49-51
9:23^25	5:38-42a	8:52-54
	5:426,43	8:55,56
9:26		
13:64-58	6:l-6a	
13:54a	6:1a	
	6:16	
13:546	6:2a	
13:54c	6:26	
13:55a	6:3a	
13:556-57a	6:36	
13:576	6:4	
13:58	6:5,6	
9:35-11:1	6:66-13	9:1-6
9:35a	6:66	
9:356		
9:36		
9:37, 38		
10:1	6:7	9:1
10:2-4		
Parallel Material in Non-Para llel
Sections
Matt.
5:15
10:26
7:2 6
25:29
Mark
8:12
22:13
24:51
25:90
4:23
6:34
3:16-19
Luke
10:23, 24
11:33
12:2
6:386
19:26
13:18,197
13:20,21
13:28
4:16a
4:166
4:22a
4:226
4:24
10:2
6:14-16
7This parable (the mustard seed) was evidently in both sources, Mark and P. Matthew’s version is confia te.
The parallelism of the two sources leads to the inclusion in Matt. of the parable of the leaven which in P immediately
foliows that of the mustard seed.
252Ebnest DeWitt Bueton
61
PART III. THE GALILEAN MINISTRY— Continued
Section Titles
Mission of the Twelve-
Continued.............
48. Death of John the Baptist
49. Feeding of the five thou-
sand................
50. Walking on the water —
51. Eating with unwashen
hands.............
Paraleel Sections		
Matt.	Mark	Luke
10:5, 6		
10:7, 8a		9:2
10:86		
10:9,10a	6:8,9	9:3
10:106		
10:11	6:10	9:4
10:12,13		
10:14	6:11	9:5
10:15		
10:16a		
10:166		
10:17,18		
10:19, 20		
10:21		
10:22		
10:23		
10:24, 25a		
10:256		
10: 26		
10:27-29		
10:30		
10:31,32		
10:33		
10:34, 35		
10:36		
10: 37		
10: 38		
10:39		
10: 40a		
10:406		
10:41		
10: 42		
	6:12, 13	9:6
11:1		
14:1-12	6:14-29	9:7-9
14:1, 2a	6:14a	9:7
14:26	6:146	
	6:15,16	9:8,9
14:3,4	6:17,18	
	6:19a	
14:5	6:196, 20a	
	6:206	
14:6a	6:21a	
	6:216	
14:66-12a8	6:22-29	
14:126		
14:13 23a	6:30-46	9:10-17
	6:30	9:10a
	6:31	
14:13	6:32, 33	9:106, lia
14:14a	6:34a	
	6*346	
	6:34c	9:116
14:146		9:11c
14:15,16	6:35-37a	9:12,13a
	6:376	9:13c
14:17-21	6:38-43	9:136,14-17
14:22,23a	6:44-46	
14:236-36	6: 47-56	
14:236-27	6:47-50	
14:28-31		
14:32,33a	6:51	
14:336		
	6:52	
14:34,35	6:53-55	
	6:56a	
14:36	6:566	
15:1-20	7:1-23	
15:1	7:1	
	7:2-4	
15:2	7:5	
15:7-9	7:6,7	
PARALLEIi MATERIAL IN NON-PARALLEL
Sections
Matt.
11:24
24:9a
24:96
24:9c, 13
16:24
16:25
18:5
Mark
13:9
13:11
13:12
13:13
4:22
8:38
8:34
8:35
9:37 a
9:376
9:41
9: 36a
9:366
Luke
10:9
10:4a
10:76
10: 7a, c
10:5,6
10:10, 11
10:12
10:3
21:12,13
[12:11, 12
» 21:14, 15
21:16
21:17, 19
6:40
8:17; 12:2
12:8-6
12:7a; 21:18
12:76, 8
9:26; 12:9
12:51, 53
14:26
9:23; 14:27
9:24; 17:33
9:48a; 10:16a
9 : 486; 10:166
3:19, 20
8 Condensed.
25362 Pbinciples of Litebaby Cbitioism and the Synoptic Pboblem
PART III. THE GALILEAN MINISTRY— Continued
Section Titles
Eating with unwashen
hands— Continued .
52. The Syrophœnician wom-
an..................
53. Return to the Sea of Gali-
lée......................
54.	Feeding of the four thou-
sand.................
55.	Demand for a sign from
heaven...............
56.	The blind man near Beth-
saida....................
57.	Peter’s confession.....
58. Jésus foretells his death..
59. The transfiguration .
60. The epileptic boy.
Parallel Sections		
Matt.	Mark	Luke
. 15: 3-6	7:8-13	
15:10,11 15:12-14a 15:146	7:14,15	
15:15-20	7:17-23	
15:21-28	7:24-30	
15:21	7:24a	
	7:246	
15:22 15:23, 24	7: 25a	
15:25	7:256,26	
	7:27a	
15:26-28	7:276-30	
15:29-31	7:31-37	
15; 29, 30	7:31, 32	
	7:33-36	
15:31a 15:316	7:37	
15:32-38	8:1-9	
15:39-16:12	8:10-21	
15:39—16:2a [16:26, 3J	8:10-12a	
16:4a	8:126	
16:46		
16:4c	8:13	
16:5	8:14	
16:6(=116)	8:15	
16: 7-lla 16: 116 (=6) 16:12	8:16-21	
	8:22-26	
16:13-20	8:27-30	9:18-21
16:13-16 16:17-19a 16:196	8:27-29	9:18-20
16:20	8:30	9:21
16:21-28	8:31-9:1	9:22-27
16:21a	8:31a	9:22a
16:216	8:316	9:226
	8:32a	
16:22, 23	8 : 326,33	
16:24	8:34	9:23
16:25	8:35	9:24
16:26a	8:36	9:25
16:266	8:37	
16:27	8:38	9:26
16:28	9:1	9:27
17:1-13	9:2-13	9:28-36
17:1-3	9:2-4	9:28-30 9:31, 32 9:33-35
17: 4, 5 17:6, 7	9:5-7	
17:8	9:8	9:36a
17:9	9:9	
	9:10	9:366
17:10-12 17:13	9:11-13	
17:14-20	9:14-29	9:37-43a
17:14a	9:14a	9:37
	9:146-16	
17:146, 15a	9:17,18a	9:38, 39
17:16, 17	9:186,19	9:40, 41
	9:20	9:42a
	9:21	
17:156	9:22a	
	9 : 226-24	
17:18a	9: 25a	9:426
	9:256	
17:186	9:26a	
	9:k66	
17:18c	9:27	9:42c
Parallel Material in Non-Parallel
Sections
Matt.
Mark
Luke
12:39a
12:396
18:18
10:38
10:39
10:33
6:39
12:54-56
11: 29a
11:296
12:16
17:25
14:27
17:33
12:9
254Ebnest DeWitt Bubton
63
PART III. THE GALILEAN MINISTRY— Continued
Section Titles	Parallel Sections			Parallel Material in Non-Parallel Sections		
	Matt.	Mark	Luke	Matt.	Mark	Luke
The epileptic boy—Con- tinued 				9:43a			
	17:19 17:20	9:28		21:21	11:22, 23	17:6
61. Jésus again foretells his death		17:22, 23 17:22 17:23 17:24-27 Chap. 18 18:1	9:29 9:30-32 9:30,31a 9:316 9:32	9:436-45 9:436, 44 9:45			
62. The temple tax							
63. Discourse on humility			9:33-50 9 : 33-35a	9:46-50 9:46, 47a	(20:26, 27 (23:11		
	18:2 18:3	9:356 9:36	9:48c 9:476		10:43, 44 10:15	22:26 18:17 10:16a 10:166
	18:4 18:5	9:37a 9: 376	9:48a 9: 486	10:40a 10:406		
		9 : 38,39a 9:396 9:40 9:41 9:42	9:49, 50a 9:506	10:42		
	18:6					17:2
	18:7					17:1
	18:8 18:9	9:43, 45 9:47 9:48, 49 9:50a		5:30 5: 29		14:31a 14:346 15: 4, 5, 76
		9:506 9:50c		5:136		
	18:10 18:12, 13					
	18:14 18:15					17:3
	18:16, 17 18:18			16:196		
	18:19, 20 18:21, 22					17:4
	18:23-34 18:35			6:15	11:25	
						
PART IV. THE PEREAN MINISTRY
Section Titles
Matt.
64. Departure from Galilee...
19:1
19:1
65. Answer to disciples
8:19-22
8:19-22
66. Mission of the Seventy....
Parallel Sections
Parallel Material in Non-Paeallel
Sections
Mark
Luke
10:1
10:1
9:57-62
9:51-56
9:51-56
9:57-60a
9:606-62
10:1-24
10:1
10:2
10:3
10:4a
10:46
10:5, 6
10:7a
10:76
10:7c, 8
10:9
10:10,11
10:12
Matt.	Mark	Luke
9:37, 38 10:16a 10:9,10a	6:8, 9	9:3
10:12,13		
10:106		
10:7, 8a 10:14 (10:15 111:24	6:11	9:2 9:5
25564 Pbinciples op Literaby Cbiticism and the Synoptic Pboblem
PART IV. THE PERE AN MINISTRY— Continued
Section Titles
Matt.
Mission of the Seventy—
Continued.............
67.	The good Samaritan...
68.	Yisit to Martha and Mary
69.	Concerning prayer....
70. Discourse against the
Pharisees...............
71. Concerning trustfulness
and watchfulness.............
Parallel, Sections
Mark
Luke
10:13-15
10:16a
10:166
10:17,18
10:19
10:20
10:21
10:22
10:23, 24
10:25-379
10:38-42
11:1-13
11:1
11:2
11:3, 4
11:5-8
11:9-11
11:12
11:13
11:14r54
11:14
11:15
11:16
11:17a
11:176, 18
11:19, 20
11:21,2210
11:23
11:24-26
11:27,28a
11:29a
11:296
11:30 il
11:31
11:32
11:33
11:34, 35
11:36
11:37,38
11:39-41
11:42
11:43
11:44
11:45,46a
11:466
11:47,48
11:49-51
11:52
11:53, 54
Chap. 12
12:1a
12:16
12:2
12:3-6
12:7a
12:76,8
12:9
12:10
12:11,12
12:13-21
12:22-29
Parallel Material in Non-Parallel
Sections
Matt.
11:21-23a
>10:40a
118: 5
10:406
11:25, 26
111:27
) 28:186
13:16,17
6:9,10a
6 :ll-13a
7:7-10
7:11
5 9:32,33a
( 12: 22, 23a
5 9:34
112:24
12:38
12:25a
12:256,26
12:27,28
12:29
12:30
12:43-45a
512:39a
116:4a
512:396
U6:46
12:40H
12:42
12:41
5:15
6:22, 23
23:25,26
23:23
23:6,7
23:27
23:4
23:29,31
23 : 34-36
23:13
16:6
16:116
10:266
10:27-29
10:30
10:31, 32
10:33
12:31,32
10:19,20
6:25-31
Mark
9:37a
9:376
16:18
3:22
3:236,26
3:27
8:126
4:21
12:38,39
8:15
4:22
8:33
3:28,29
13:11
Luke
9:48a
9:486
8:16
20:46
8:17
21:18
9:26
21:14,15
9	On 10:27 cf. Matt. 22 : 37, 38, and Mark 12:30,31.
10	Matthew and Mark, closely resembling one another verbally, are parallel to Luke in substance only.
11	But Matthew contains reference to the three days and three nights not found in Luke.
256Ernest DeWitt Bürton
65
PART IV. THE PEREAN MINISTRY— Continued
Section Titles
Concerning trustfulness
and watchfulness — Con-
tinued ...................
72.	Galileans slain by Pilate..
73.	Woman healed on a sab-
bath....................
74.	Parables of the kingdom..
75. Question whether few are
saved.....................
76. Reply to warning against
Herod....................
77. Discourse at table of a
chief Pharisee.............
78. Discourse on counting the
cost.......................
79. Parables of grâce .
Parallel Sections
Matt.
Mark
Luke
12:30
12:31
12:32
12:33, 34
12:35-38
12:39,40
12:41
12:42-46
12:47-49
12:50a
12:506
12:51
12:52
12:53
12:54-56
12:57
12:58, 59
13:1-9
13:10-17
13:18-21
13:18,19
13:20, 21
13:22-30
13:22, 23
13:24
13:25a
13:256
13:26
13:27
13:28
13:29
13:30
13:31-35
13:31-33
13:34,35
14:1-24
14 :l-3a
14:36
14:4
14:5
14:6-10
14:11
14:12-14
14:15-2412
14:25-35
14:25
14:26
14:27
14:28-33
14:34a
14:346
14:35a
14:356
Ch&p. 15
15:1-3
15:4, 5
15:6
15:7
Parallel Material in Non-Parallel
Sections
Matt.	Mark	Luke
< 6:32 } 6:8 6:33		
6:19-21		
24:43, 44		
24:45-51	10 : 386, 396	
10:34		
10:35,36 [16:2, 3]		
5:25,26		
13:31,32 13:33	4:30-32	
7:13,14		
25:116,12		
7:23 f 8:12 | 13:42 j 13:50 ] 22:13 | 24:51 [25:30 8:11 f 19:30 f 20:16	10:31	
23:37-39		
12:126	3:4	6:9
12:11		
23:12		18:146
10:37 5 10:38 U6 :24	8:34	9:23
5:136 5:13c ( 11:15 •<13:9 ( 13:436	9:50a 9:506 (note) ( 4:9 l 4:23	8:86
18:12, 13a		
18:136		
12Apparently the same parable as Matt. 22:1-10, but a wholly independent version.
25766 Principles of Literary Criticism and the Synoptio Problem
PART IV. THE PEREAN MINISTRY— Continued,
	Parallel. Sections			Parallel Material in Non-Paraelel		
Section Titles					Sections	
	Matt.	Mark	Luke	Matt.	Mark	Luke
Parables of grâce—Con- tinued				15:8-10 15:11-32 16:1-31 16:1-12 16:13			
						
80. Parables of warning							
						
				6:24		
						
			16:14,15 16:16			
				11:12,13		
						
			16:17 16:18 16:19-31	5:18 5 5:32		
					10:11	
				( 19:9		
81. Concerning forgiveness						
and faith				17:1-10			
				18:7 18:6		
			17:1 17:2			
					9:42	
						
			17:3	18:15		
						
			17:4	18:21, 22		
						
			17:5 17:6 17:7-10 17:11-19 17:20-18:8 17:20-22 17:23, 24 17:25 13			
				517:206	11:22,23	
82. The ten lepers					l 21 :21		
						
83. Thecomingof thekingdom						
						
				24:26, 27 16:21a		
					8:31a	9:22a
						
			17:26, 27 17:28 32 17:33	24:37-39		
						
				510:39 { 16:25 24:40, 41	8:35	9:24
			17:34,35 17:37a 17:376 18:1-8			
						
				24:28		
84. The Pharisee and the pub-						
						
lican				18:9-14 18:9-14a 18:146			
						
				23:12		14:11
8d. Concerning divorce		19:3-12	10:2-Ï2				
	19:3 19:7, 8 19:4-6 19:9	10:2 10:3-5 10:6-9 10:10 10:11		5:32		16:18
		10:12				
86. Blessing little children ...	19: 1CP12 19:13-15	10:13-16	18:15-17			
	19:13, 14	10:13, 14	18:15, 16			
	19:15 19:16-20:16	10:15	18:17	18:3		
87. The rich young man			10:16 10:17-31	18:18-30			
	19:16-20	10:17-20 10:21a 10:216-23	18:18-21			
	19:21-23		18:22-24			
		10:24 10:25-28				
	19:24-27a 19:276, 28a 19:286		18:25-28			22:306
						
						
	19:29a	10:29, 30a 10:306 10:30c	18:29,30a			
						
	19:296		18:306			
	19:30 (=20:16)	10:31				13:30
88. Prédiction of crucifixion..	20:1-15 20:16 ( = 19:30) 20:17-19 20:17a	10:32-34 10:32a 10:326 10:32c-34	18:31-34			
						
						
						
						
	20:176-19		18:31-33 18:34			
89. Ambition of James and						
John 		20:20-28	10:35-45				
						
	20:20-22a	10:35-38a 10:386				12:50a
	20:226, 23a	10:39a				
13 Cf. also Matt. 17:22; Mark 9:31; Luke 9:44; and Matt. 20:18; Mark 10: 33; Lukel8:31.
258Ebnest DeWitt Burton
67
PART IV. THE PEREAN MINISTRY—Continued
Parallel Sections
Section Titl.es
Ambition of James and
John—Continued.......
90. Bartimæus healed
Matt.
Mark
Luke
20:236-25»
20:256, 26a
20:266, 27
20:28
10:396
10:40-42a
10:426,43a
10:436,44
10:45
aoifÊii	10:«-52
20:|9-32a	l0:46-49a
18:35-43
18:35-40a
10:496,50
91.	Zacchœus........
92.	Parable of the minæ
20:326, 33
20:34a
9:29a
9:296
20:346
9:30a
9:306, 31
10:51
10:52a
10:526
18:406,41
18:42
18:43a
18:436
19:1-10
19:11-28
19:11
19:12,13
19:14-16
19:17
19:18,19
19:20-25
19:26
19:27, 28
Par allée Material in Non-Parallel
Sections
Matt.
Mark
Luke
23:11
9:356
12:50a
22:25, 26a
22:266; 9:486
22:276
25:14
25:21,23
25:24-28
25:29
13:12
4:25
8:186
PART V. THE PASSION WEEK
Parallel Sections
Section Titles			
	Matt.	Mark	Luke
93. Triumphal entry		21:1-11	11:1-11	19:29-44 19:29-31
	21:1-3	11:1-3	
	21:4, 5 21:6	11:4	19:32
	21:7, 8	11:5, 6 11:7,8	19:33, 34 19:35,36 19:37
	21: 9a	11:9 a	
	21:96	11:96,10	19:38 19:39-44
	21:10a 21:106,11	11:11a	
			
94. Cursing of the fig tree		21 : Ï8, 19	11:116 11:12-14	
	21:18,19a	11:12,13a	
	21:196	11:136 11:14	
95. Cleansing of the temple...	21:12-17	11:1519	19:45-48
	21:12a	11:15a	19:45
	21:126	11:156	
	21:13	11:16 11:17 11:18a 11:186	19:46 19:47
	21:14-16		19:48
			
96 Lesson from the fig tree..	21:17 21:20-22	11:19 11:20-25	
	21:20	11:20 11:21	
	21:21	11:22,23	
Parallel Material in Non-Parallel
Sections
Matt.
Mark
22:33
Luke
21:37a
17:20
17:6
25968 Pbinciples of Litebaky Ceitioism and the Synoptic Pboblem
PART V. THE PASSION WEEK— Continued
Parallee Sections
Section Titles
Matt.
Mark
Lesson f rom the fig tree—
Continued.............
21:22
11:24
11:25
Luke
97.	Jésus’ authority chal-
lenged...............
98.	Para blés of warniog.
21:23-27
21:28-22:14
21:28-32
21:33a
21:336
21:33c
21:34-40
21:41a
21:416
11:27-33
12:1-12
12:1a
12:16
12:1c
12:2-9a
12:96
21:42a
21:426
21:43
[21:44]
21: 45, 46a
21:466
22 : l-13a14
12:10
12:11
12:12
20:1-8
20:9-19
20:9a
20:96
20:10-15
20:16a
20:166
20:17
20:18
20:19
22:136
99. Questionsby Jewishrulers
22:14
22:15-40
22:15,16a
22:166-22
22:23-28
22:30
22:31,32
22:33
22:34
22 : 35, 36
22 : 37,38 1!
22:39
22:40
100. Christ’s unanswerable
question.................
101. Woes against the scribes
and Pharisees..............
22:41-46
22:41-45
22:46
Chap. 23
23:2,3
23:4
23:5
23:6. 7a
23:76-10
23:11
23:12
23:13
23:15-22
23:23
23:24
23:25, 26
23:27
23:28
23:29-31
23:32
23:33
23:34-36
23; 37-39
12:	: 13-34	20:	: 20-40
12:	: 13,14a	20:	:20a
		20:	: 206
12:	: 146-17	20:	: 21-26
12:	: 18-23	20:	: 27-34a
12:	: 24		
		20:	: 346,3oa
12:	: 25	20:	: 356, 36a
		20:	: 366
12:	: 26,27	20:	: 37, 38
12:	:28a	20:	: 39
12:	: 286		
12:	: 29,30		
12:	:31a		
12:	316		
12:	:32-34a		
12:	: 346	20:	40
12:	35-37	20:	41-44
12:	35-37	20:	4144
12:	38-40	20	: 45-47
12:	38a	20:	45
12:	386, 39	20:	46
12:40
20:47
Parallel Material in Non-Parall.ee
Sections
Matt.
{
6:14,15
18:35
Mark
Luke
f 8:12
13:43
-{ 13:50
| 24:51
[25: 30
13:28
11:186
7:12
22:46
12:346
20:40
20:26, 27

9:356
10:43,44
11:466
11:43
I
22:26
9:486
14:11
18:146
11:52
3:76
11:42
11:39-41
11:44
11:47,48
3:76
11:49-51
13:34,35
14	Matt. 22:1-14 is apparently the same parable that is contained in Luke 14:15-24, but an entirely independent version,
15	On Matt. 22:37 ff. cf. Luke 10:27.Ernest DeWitt Burton
69
PART V. THE PASSION WEEK— Continued,
	Parallel, Sections			Parallel Material in Non-Parallel Sections		
Section Titles						
	Matt.	Mark	Luke	Matt.	Mark	Luke
102. The widow’s mite			12:41-44 Ghap. 13	21:1-4 21:5-38			
103. Discourse on last things.	Chaps.24,25					
	24:1,2	13 : lf 2	21:5, 6			
	24:3a	13:3				
	24:36	13:4	21:7			
	24:4-7	13:5-8	21:8-11			
	24:8	13:9a				
	24:9a	13:96	21:12a	10:17		
	24:14	13:9c 13:10	21:126, 13	10:18		
		13:11	21:14,15	10:19, 20; 10:21 10:22a		12:11,12
		13:12	21:16 21:17			
	24:96	13:13a				12:7a
			21:18	10:30		
	24:10-12					
	24; 13	13:136	21:19	10:226		
	24:15-18	13:14-16	21:20, 21 21:22			
	24:19	13:17	21:23a			
	24:20 24:21a 24: 216, 22	13:18 13:19a 13:196, 20	21:236 21:24			
	24:23	13:21				17:23
	24:24, 25 24:26, 27	13:22, 23				
						17:23, 24 17:376
	24:28					
	24:29a	13:24,25a	21:25a 21:256, 26a 21:266			
						
	24:296 24:30a	13:256				
						
	24:306	13:26	21:27			
	24:31	13:27	31:28 21:29-33			
	24:32-35	13:28-31				
	24:36 24:37-39	13:32				17; 26, 27 17:34, 35
	24:40, 41					
		13:33, 34				
	24:42(c/.25:13)	13:35a (c/.33) 13:356-37				
	24:43, 44					12:39, 40
	24:45-51a					12:42-46
				f 8:12 J 13:42 ] 13:50		
	24:516 (=25:306)					13:28
			21:34-36			
				122:136		
						
	25:1-lla 25:116, 12 25:13 (see					13:256
						
						
	24:42) 25:14					19:12, 13
	25:15-20 25:21					
						19:17
	25:22					
	25:23					19:17
	25:24-28					19:20-25 (19:26
	25:29			13:12	4:25	
	25 : 30a 25:306 (see					( 8:186
						
	24:51)					
	25:31-46		21:37a		11:11 11:19	
			21:376	21:17		
			21:38			
104. Conspiracy of the chief	26:1-5					
priests			14:1,2	22: 1, 2			
	26:1					
	26:2a	14:1a	22:1			
	26:26 26:36 26:3a, 4	14:16	22:2a			
	26:5a	14:2a				
	26:56	14:26	22:26			
26170 Pbinciples or Litebaby Cbitioism and the Synoptic Pboblem
PART V. THE PASSION WEEK— Continued
Section Titles
105.	Anointing in Bethany
(Matt.-Mark.).........
106.	Plot of Judas and the
rulers................
107.	Last supper..........
108. Agony in Gethsemane ...
109. Betrayal and arrest.
110. Trial before
authorities.........
Jewish
Paealiæl Sections
Matt.
26:6-13
26:14-16
26:17-35
26:17,18
26:19,20
26:21
26:22
26:23
26:24a
26:246
26:25
26:26
26:27, 28
26:29
26 : 30-32
26 : 33, 34
26:35
26:36-46
26 : 36a
26 : 366
26:37, 38
26:39
26:40a
26:406
26:41
26 : 42-46
26:47-56
26:47a
26:476, 48
26:49
26:50a
26:506
26:51
26:52-54
26:55
26:56 '
36:57-27:10
26:57, 58
26 : 59, 60a
26:606, 61
26: 62, 63a
26 : 636
26 : 64a
26:646
26: 65a
26:656
26:66
26:67, 68
Mark
14:3-9
14:10,11
14:12-31
14:12-14
14:15
14:16,17
14:18
14:19
14:20
14:21a
14:216
14:22
14:23, 24
14:25
14:26-28
14:29 , 30
14:31
14:32-42
14:32a
14:326
14:33, 34
14:35, 36
14:37a
14:376
14:38
14:40-43
14:43-52
14:43a
14:436, 44
14:45
14:46
14 : 47 ‘
14:48,49a
14:496, 50
14:51, 52
14:53-72
14:53, 54
14:55, 56a
14:566
14:57, 58
14:59
14:60,61a
14:616
14:62a
14:626
14:63a
14:636, 64a
14:646
14:65a
14:656
Luke
22:3-6
22: 7-38
22:7-11
22:12
22:13,14
22:15-17
22:21
22:23
22:22
22:19a
22:196
22:20
22:18
22:24
22:25, 26a
22:266
22:27a
22:276
22:28-30a
22:306
22:31, 32
22:33, 34
22 : 35-38
22:39-46
22:39, 40a
22:41, 42
22 : 43, 44
22:45
406=466
22:47-53
22: 47a
22:476
22:48
22:49
22:50
22:51
22:52,53a
22:536
22:54-71
22:54, 55
22 : 666
22 : 66a
22:67, 68
22:69
22:70
22:71
22:63,64
22:65
Parallel Material in Non-Paraelel
Sections
Matt.
20:25, 26a
( 20: 266, 27
} 23:11
20:28
19:286
Mark
10:42,43a
i 10: 436, 44
1 9:356
10:45
Luke
9:486
262Ernest DeWitt Burton
71
PART V. THE PASSION WEEK—Continued
Section Titles	Par ALLEE Sections			Parallel Material in Non-Parallel Sections		
	Matt.	Mark	Luke	Matt.	Mark	Luke
Trial before Jewish						
authorities—Continued..	26:69,70	14:66-68a	22 : 56, 57			
	26:71a	14:686				
		14:68c				
	26:716-74	14:69-72a	22:58-60			
			22:61a			
	26:75	14:726	22:616, 62			
111. Trial before Pilate		27:1-31	15:1-20	23:1-25			
	27:1	15:1a				
	27:2	15:16	23:1			
	27:3-lla					
			23:2			
	27:116	15:2	23:3			
	27:12-14	15:3-5				
			23:4-16			
	27:15,16	15:6, 7a				
		15:76,8	23:19			
	27:17,18	15:9,10				
	27:19					
	27:20-23	15:11-14	23:18, 20-23a			
	27:24, 25					
	27:26	15:15	23:236-25			
	27:27-31	15:16-20				
112. Crucifixion of Jésus		27:32-56	15:21-41	23:26-49			
	27:32	15:21	23:26			
			23:27-32			
	27:33	15:22	23:33a			
	27:34	15:23				
			23:34a			
	27:35	15:24	23:346			
		15:25				
	27:36		23:35a			
	27:37	15:26	23:38			
	27:38	15:27	23:336			
	27:39, 40	15:29, 30				
	27:41,42a	15:31,32a	23:356			
	27:426	15:326				
	27:43					
			23:36, 37			
	27:44	15:32c	23:39			
			23:40-43			
	27:45	15:33	23:44, 45a			
	27:46-49	15:34-36				
	27:50	15:37	23:46a			
			23:466			
	27:51a	15:38	23:456			
	27:516-53					
	27:54	15:39	23:47			
			23:48			
	27:55a	15:40a	23:49a			
	27:56	15:406				
	27:556	15:41	23:496			
113. Burial of Jésus'		27:57-61	15:42-47	23:50-56			
	27:57a	15:42a				
		15:426	23:54			
	27:576	15:43a	23:50, 516			
			23:51a			
	27:58a	15:436	23:52			
		15:44				
	27:586	15:45				
	27:59, 60a	15:46a	23:53			
	27:606	15:466				
	27:61	15:47	23:55			
			23:56a		16:1	
			23:566			
114. Watch at the sepulcher.	. 27:62-66					
26372 Principles of Litebary Cbiticism and the Synoptio Problem
PART VI. THE APPEARANCES AFTER THE RESURRECTION
Section Titles
115. Résurrection morning...
116.	Report of the watch.
117.	Walk to Emmaus......
118.	Appearance in Jérusa-
lem,Thomas being absent
119.	Appearance to the Eleven
in Galilee...........
120.	Final appearance and
ascension............
Parallel Sections
[16:16-18]
[16:19, 20]
[16:19]
[16:20] *
24:44r-53
24:44-50
24:51
24:52
Matt.	Mark	Luke
28:1-10	16:1-8[911] 16:1	24:1-12
28:1	16:2	24:1
28:2-4		
	16:3	
	16:4	24:2 24:3
	16:5	24:4, 5a
27:5, 6a	16:6	24:56, 6a 24:66,7
28:7, 8	16:7, 8	
28:9,10	[cf. 16:9-11]	24:8-12
28:11-15		
	[16:12, 1316]	24:13-35
	[16:141]	24:36-43
28:16-20		
28:16-18a	[16:15-18]	
28:186		
28:19a	[16:15J	
28:196, 20		
Parallel. Material in Non-Parallel
Sections
Matt.
Mark
11:276
16:19,20
Luke
23:56a
10:226
10 The parallelism in these cases is only of the most general character.
264