Previous Page: Marcion's Gospel and the Synoptic Problem
Several people have connected Marcion with Mark because of the similarity of their names. For example, in his blog Stephan Huller wrote:
… So it is that when we are faced with all these silly stories about Marcion seducing a virgin or Marcion being rejected by Polycarp (there really are surprisingly few stories out there actually) some scholars - the idiots in the pack - simply regurgitate the whole inherited 'package' and literally develop a story with a beginning middle and end out of it. Others recognize the propagandist elements but don't probe any deeper into even the most basic details about his person ... like the fact that his original name must have been Mark.
The -ion ending is a [sic] affectionate diminutive suffix used at the ends of Greek names. Take the example of the application of this diminutive suffix by Dicaeopolis, the lead character in Aristophanes Achamians. At one point in the play he ridicules Lamachippus, another character in the play by calling him "Lamacivppion" i.e. my little Lamachippus. As Sommerstein explains "Dicaeopolis mockingly adds two incongtruous [sic] elements to Lamachus' name, the sonorous and lofty -hippos 'horse' and the affectionate diminutive suffix -ion."
So this is the explanation of the Marcion [sic] if it was naturally developed among Greek speakers - i.e. that he was someone named Mark who was identified with the nickname 'Marky' by contemporaries.
Whether or not Marcion is the diminutive form of Mark, and whether or not Marcion and the evangelist we know as Mark were the same person (as has also been suggested), what is undeniable are the similarities between Marcion’s gospel of the Lord (The evangelion, below Ev, with author aEv except in quotes or direct references) and the Gospel according to Mark. While the majority opinion is that Ev is a ‘cut down’ edited version of the gospel of Luke, what is usually not stated is that, from everything we know of the text, it is at least equally valid to consider Ev to be an expanded version of Mark.
As is discussed elsewhere (see the Summary of the analysis of Ev), if aEv used Luke as the basis for his gospel then the remarkable thing is how much he seems to have only wanted to remove text from Luke that has no parallel in Mark, and much of which is actually unique to Luke (i.e. it has no parallel in either Mark or Matthew). This is particularly the case for the 214 verses in the first four chapters of Luke, of which (See Marcion's Gospel up to Luke 5:13) only 10% are in Ev. Of the 90% that are not in Ev only 3% have parallels in Mark, and (allowing for some overlap between Matthew 1-2 and Luke 2), 73% are unique to Luke. In other words, assuming that aEv (the author of Ev) edited Luke, why did he make his version of Luke 1-4 in Ev so like Mark by removing so much that we see only in Luke?
The same pattern (although slightly less extreme) exists with regard to the 110 verses in Lk 5-24 that Epiphanius identified as being ‘falsified’ or otherwise not in Ev: 73% have no parallels in Mark, and 66% are unique to Luke. However, there are particular sections of Luke in which this tendency is even more extreme, for example in the Lukan Travel Narrative (Lk 9:51 - 18:14: See Epiphanius: Omissions After Lk 5:13), in which nearly 84% of the material that it is claimed that aEv omitted from Luke we see as unique to Luke. Again, the differences here between Ev and Luke seem to largely focus on making Ev looking much more like Mark than Luke.
As Roth points out, both Knox and Tyson make essentially the same observation, although their numbers differ somewhat due to different assumptions regarding which verses of Luke were not in Ev:
Knox observed, “Of the verses which [sic] there is positive evidence to show did not belong to Marcion, 79.7 per cent are peculiar to Luke. Of all the verses of Luke which are peculiar to Luke, 39 per cent are known to be missing in Marcion, whereas of verses of Luke paralleled in Matthew or Mark or both, only 10 per cent are known to be missing from Marcion.”
Tyson also employed this argument in his recent attempt to revitalize Knox’s views, once again relying on Harnack’s reconstruction of Marcion’s Gospel and creating tables like those of Knox. Although there are slight differences in the numbers due to Tyson’s occasionally different evaluation about certain Lukan pericopes, the results are quite similar. Tyson states, “About 12 percent of Lukan material with synoptic parallels is probably absent from Marcion’s Gospel. But 41-43 percent of Lukan Sondergut material is omitted.”
We therefore have a very strange situation. With the exception of the text of the Great Omission (which both Luke and Ev lack), Ev basically contains the great majority of the text of Mark, together with only some of the non-Markan text that we also see in Luke. The material in Luke that is not in Ev is mainly Sondergut (or Special) Luke material, with a little Double Tradition, and therefore is not in Mark either. According to Epiphanius, overall Marcion deliberately removed 266.5 verses from Luke, 88% of which had no parallel in Mark, and 70% of which contained text unique to Luke. To put this another way, the great majority of the text that Epiphanius states Marcion cut out of Luke is text that had no parallel in Mark, and is also primarily Lukan Sondergut, i.e. text that (assuming Markan priority) had been added by aLuke (the author of Luke) before Marcion later removed it!
Again on the assumption that Marcion edited Luke, why did he apparently try so hard to preserve material originating in Mark, while cutting out so much of the material that originated in Luke, the gospel on which he supposedly based his own? Although we can hypothesize reasons why Marcion might have removed primarily non-Markan text from Luke when creating Ev, such editing (so making Ev look much more like an expanded version of Mark than a shorter version of Luke) is very hard to defend. Instead, if Ev preceded both Matthew and Luke then the lack of this text in Ev is easily and naturally explained as later additions by either Matthew or Luke. However, there are also a few passages in Mark that Tertullian and/or Epiphanius indicated were not in Ev. If Ev is essentially an expanded version of Mark, then why did aEv not include these passages from his primary source?
The idea of multiple versions of Mark is not new, with various theories around the end of the 19th century positing either a Proto-Mark (or Ur-Marcus) earlier than canonical Mark, or a later Deutero-Mark. In Textual Criticism and the Synoptic Problem, Peter Head explains that it is the minor agreements between Matthew and Luke against Mark (those triple tradition passages in which Matthew and Luke have the same wording but where Mark is different) that indicate that there may have been different versions of Mark:
… Secondly, however, 2DH [Two Document Hypothesis] advocates have maintained other types of solution to the “problem” of the minor agreements that are not as simply explained by independent redaction, including the ongoing impact of additional oral tradition (on Matthew and Luke), and the possibility that Matthew and Luke had access to an earlier or at least different form of Mark than the canonical form. This position was already adopted by Sanday in the Oxford Seminar volume: “I believe that by far the greater number of coincidences of Matthew and Luke against Mark are due to the use by Matthew and Luke – not of an Ur-Marcus or older form of the Gospel, but of a recension of the text of Mark different from that from which all the extant MSS. of the Gospel are descended.”
Attempts to explain this with some precision in terms of a particular text-critical theory have been made at various points. T.F. Glasson proposed that Matthew and Luke may have used a “Western” text of Mark. J.P. Brown proposed that the Caesarean text of Mark (containing numerous agreements of Matthew and Luke against Mark) originated in an early revision of Mark, a Deutero-Mark. This may serve to bring an important reminder of the realities of ancient textual transmission into the discussion, and mean that reasonable allowance should be made for the fact that Matthew and Luke could not have had access to identical copies of Mark.
The last point above made by Head is often overlooked: If aMatthew and aLuke knew Mark then it is highly unlikely that they both used the same physical copy of Mark, and hence it must be allowed that there were at least some differences between what the two of them knew as Mark. It is therefore conceivable that the document seen by aLuke that included the text of Mark also included material that the document seen by aMatthew did not, and that we know as at least some of Sondergut Luke.
If the above is the case then it is not unreasonable to consider that Ev could be (or could be closely related to) this ‘other’ version of Mark, from which the Matthew/Luke text of at least some of the minor agreements originates. However, as Streeter points out, many of these ‘agreements’ (e.g. the many uses of the historic present in Mark that are in neither Matthew nor Luke, and the use of δέ instead of καί) can be explained as independent linguistic choices by aMatthew and aLuke rather than indicating a different source, so we would not expect aEv to be the source of all of them. In addition, as we do not have evidence of the exact text of Ev in several places, in these places we will not be able to tell whether Marcion agrees with Mark or Matthew/Luke, and so the text used in Ev at the location of the minor agreements cannot be determined in all cases.
As previously indicated, it is clear that Tertullian did not see in Ev some of the text that we see in Luke (mainly in the early chapters, see Marcion's Gospel up to Luke 5:13), but he makes very few claims regarding aEv actually removing or changing any text. Instead, as he reports very little actual difference between Ev and Luke after the early chapters (see Tertullian: Omissions and Differences), a reasonable inference is that in these places the text of Ev was close to that in Tertullian's copy of Luke.
Despite Tertullian not quoting or even alluding to all the text in Ev, so that there are several places where we do not know specific differences between Ev and Luke, and Epiphanius lack of specifics regarding Lk 4:1-5:13, we do know that there were some passages in Mark that were not in Ev. In cases where these passages are also not in Luke (for example the Great Omission), then if Ev follows Luke this is completely unexceptional, while if Luke came after Ev then aEv also did not include the passages (for whatever reason) and aLuke then followed the content of Mark.
There are a few passages in Mark that are not in Ev but are in Luke, and for these the situation is more complex. If Ev is an edited version of Luke then these are easily explainable as deletions by aEv, but if Ev precedes Luke, and possibly depends on Mark, then we could have the odd situation in which aEv apparently omitted text he saw in Mark, but aLuke later re-instated the same text, possibly by taking it from Mark. However, in all cases their existence in Luke does not depend on aLuke having seen them in Mark, and instead they are all explainable as being due to one of the following:
1. They are triple tradition passages, and Luke took them from Matthew;
2. The passage in Mark is either an interpolation, or an assimilation to Luke;
3. Because of difficult variants we are unsure of the original contents of Luke.
Note that point 1 above assumes that aLuke could have known Matthew, a position that is denied by supporters of the Mark-Q synoptic hypothesis. However, as this is just a hypothesis this analysis allows for the possibility that aLuke did know Matthew. Also, although the Mark-Q hypothesis is based on the assumption that aLuke and aMatthew did not used each others' gospel, that does not prevent a similar document (e.g. Ev) having existed even if one author did know the others' gospel. Instead, it would simply mean that the Mark-Q hypothesis (which requires independence) was false.
If Ev preceded Matthew then a couple of otherwise difficult textual problems are easily explained: a doublet in Mk 1:12-13; and the text of the Old Latin in Lk 19:30-34. However, the most important point is that in all cases in which text in Mark has a parallel in Luke, but where according to either Tertulian or Epiphanius Ev lacked the text, the parallel in Luke is also obtainable from Matthew, i.e. it is Triple Tradition text. In other words, if Ev preceded Matthew then everything in Luke that is not unique to Luke (i.e. is not Sondergut Luke material) was obtainable from the combination of Ev and Matthew, and aLuke may never have even needed to have seen a copy of Mark. Of course, it is worth noting here that aLuke knew of many sources of material on Jesus, as he tells us in Lk 1:1-2
Forasmuch as many have taken in hand to set forth in order a declaration of those things which are most surely believed among us, Even as they delivered them unto us, which from the beginning were eyewitnesses, and ministers of the word;
It is important to note that there is nothing that requires aEv to have included everything from Mark in Ev, in the same way that aMatthew and aLuke did not include everything from Mark in their respective gospels, and on the Farrer / MwQH aLuke did not include everything from Matthew in Luke. Nevertheless, on the assumption that both Matthew and Luke have Ev as a source it is necessary to examine the triple tradition text that does not have a parallel in Ev on a case-by-case basis, in case a pattern emerges regarding text for which Ev either does or does not contain a parallel. For example, where Luke appears to follow text in Ev, aMatthew would qite likely have also seen this text, and if Matthew also followed Ev to at least some extent then we would expect that Matthew would be closer to Luke than in places where Ev contained no text. This can (in part) be tested by examining the triple tradition passages (shown below in Markan order) where Ev either does or does not contain any text:
Epiphanius begins his analysis of Ev as follows:
At the very beginning he [Marcion] excised all of Luke’s original discussion – his “inasmuch as many have taken in hand” and so forth, and the material about Elizabeth and the angel’s annunciation to the Virgin Mary, John and Zacharias and the birth at Bethlehem; the genealogy and the subject of the baptism.
Although "the subject of the baptism" might be referring to just Lk 3:21-22, it is more likely that it covers the whole of Lk 3:2b-22, since the first place in Ev at which Tertullian refers to John the Baptist is in reference to Ev 5:33, where he writes:
Whence, too, does John come upon the scene? Christ, suddenly; and just as suddenly, John!
Tertullian is pointing out that just as Jesus is given no introduction in Ev, the same applies to John, as he is not mentioned in Ev prior to v. 5:33 (using the location in Luke to identify equivalent text in Ev). Therefore, if Ev preceded Luke, but Ev did not contain any parallel to Lk 3:2b-22, what was the source for these verses in Luke?
Part of this question is easy to answer, as Lk 3:3a, 5-6, 10-16a, 18-21a have no parallels in either Mark or Matthew, and so they are most likely to be additions by aLuke based on some other other oral or literary source, and because these verses are in neither Mark nor Matthew there is no reason to expect them to have been in Ev. In addition, Mk 1:2b, 5-6, 9b, 10a do not have parallels in Luke, and so again there is no reason to expect them to have parallels in Ev either, because aMatthew could have obtained his parallels to Mk 1:5-6, 9b, 10a directly from Mark.
What is harder to explain is the text common to Mark, Matthew, and Luke (the triple tradition text) that is not in Ev, which in this area includes Mk 1:2a, 3-4, 7-9, 10bc, 11, 14b. If we see text in Mark that is not in Ev, but is in both Matthew and Luke, then:
If aEv edited Luke, why would he remove the baptism, a passage in all three synoptic gospels?
If Ev preceded Luke, why would aEv not have included at least a form of this text similar to that in Mark, and why would aMatthew and aLuke include their own versions rather than following Mark more closely?
The first alternative is easy to explain by suggesting that it was the connection to the Old Testament that aEv wanted to exclude, while the second alternative ignores the fact that the parallel passages in Matthew and Luke each contain significant differences from the version of the baptism in Mark. So, it is not just a question of why Ev differs from Mark, but why do Ev, Matthew, and Luke all differ significantly from Mark:
In both Matthew and Luke the parallels to Mk 1:4 (Mt 3:1b-2 and Lk 3:2b-3 respectively) come before the parallels to Mk 1:2a;
Only Luke has a parallel to Mk 1:2b, and this is located at Lk 7:27b;
Mk 1:5-6 do not have any parallels in Luke;
Mark has no parallels to Mt 3:7-10 and Lk 3:7-9;
Lk 3:10-16a has no parallel in either Mark or Matthew;
In both Matthew and Luke the parallels to Mk 1:7 (Mt 3:11c and 3:16d respectively) come after the parallels to Mk 1:8a;
The parallels to Mk 1:8 include references to repentance and fire (that are not present in Mark) in some mss;
The parallels Mt 3:12 and Lk 3:17 have no parallel in Mark;
Lk 3:18, 21a have no parallels in either Mark or Matthew;
Mk 1:9b and 1:10a have no parallels in Luke;
Mt 3:14-15 has no parallel in either Mark or Luke;
Mk 3:11 has a close parallel at Lk 3:22, but the parallel at Mt 3:17 is different.
Another item to consider is the version of the baptism that is contained in the Gospel of the Ebionites, which contains parallels to Mk 1:4a, 4c-6, 9b-11, and Lk 3:21a (and so conceivably being a source for these verses independent of the synoptic tradition), but nothing corresponding to Mk 1:2-3, 7-9a.
As discussed in Mark 1:1-3 - The Short Beginning there is a possibility that Mk 1:1-3 were not original, and if so then it is possible that Ev did not contain any parallel to these verses for the reason that, along with the Great Omission and the long ending of Mark, aEv did not see them in Mark. If the Gospel of the Ebionites is later than Mark then this could also explain why it did not contain parallels to these verses, while if Ebionites is earlier than Mark then this would lend weight to the suggestion that Mk 1:1-3 are not original, and also perhaps explain why Mk 1:2-3 are ‘swapped’ with Mk 1:4 in comparison with the parallels in Ebionites, Matthew, and Luke.
The above explanations provide possible reasons as to why the majority of the story of the baptism has no parallel in Ev, and also why the versions in Matthew and Luke differ so much. However, we are still left with the fact that Mk 1:4, 7-9a, 9c, and 10b-11 have close parallels in Matthew and Luke but not Ev, and that these verses on their own present a coherent baptism account:
John did baptize in the wilderness, and preach the baptism of repentance for the remission of sins [Mk 1:4]
[And preached,] saying, There cometh one mightier than I after me, the latchet of whose shoes I am not worthy to stoop down and unloose. [Mk 1:7]
I indeed have baptized you with water: but he shall baptize you with the Holy Ghost. [Mk 1:8]
And it came to pass in those days, that Jesus came [Mk 1:9a]
and was baptized of John in Jordan. [Mk 1:9c]
And … he saw the heavens opened, and the Spirit like a dove descending upon him: [Mk 1:10b]
And there came a voice from heaven, saying, Thou art my beloved Son, in whom I am well pleased. [Mk 1:11]
On the assumption of Markan priority both aMatthew and aLuke saw these verses in Mark and used them in their own gospels. However, they both added considerably to what they saw in Mark, not only with double tradition material, but also each with their own unique text. aLuke then would have been able to see that both Mark and Matthew included versions of Jesus’ baptism by John, and chosen to include his own version even though it was not in Ev.
Although this does not show that Ev might have acted as a ‘moderating’ factor, it does provide an example of the opposite: how different the versions of a passage common to all three synoptics sometimes are when there is no parallel in Ev.
Neither Tertullian nor Epiphanius make any reference to any version of the temptation (Lk 4:1-13) in Ev. Tertullian’s first mention of tempting occurs in regard to Lk 10:25, and although he knew of the temptation (as he quotes Mt 4:3 at the beginning of Against Praxeas), there is no evidence that he knew the Lukan version. As previously indicated (See Luke 3:1 – 4:15, 4:31a), after having noted the lack of Luke 1-2 and "the genealogy and the subject of the baptism" all Epiphanius says about the beginning of Ev is:
All this he cut out and turned his back on, and made this the beginning of the Gospel, “In the fifteenth year of Tiberius Caesar,” and so on. He starts from there then [Lk 3:1a] and yet, again, does not go on in order. He falsifies some things, as I said, he adds others helter-skelter, not going straight on but disingenuously wandering all over the material.
Epiphanius’ next comment is on Lk 5:14, so we do not know from this whether the “some things” that he states that Marcion “falsified” included any or all of the temptation. It is worth pointing out at this point that after referring to 'Nazara' at Q 4:16, the IQP's Q has has nothing until The Beatitudes at Q 6:20-21.
Matthew and Luke parallel each other closely in the Temptation, while the version in Mark is much shorter, with only Lk 4:1b-2a having a parallel in Mark (at Mk 1:12-13). Even in these few verses there are some significant differences among the synoptics. As described in Fatigue in Mark – or Damage to Mark ? - The Temptation, these differences center around the two references to the wilderness in Mark, only the first of which is in Matthew, and only the second of which is in Luke. However, instead of aLuke having here added the second wilderness reference from Mark, variants in Mark indicate that the opposite is the case, with the second reference in Mark being a later assimilation to Luke. As neither Tertullian nor Epiphanius refer to the Temptation with regard to Ev it is possible that it was present in some form, but as "and was with the wild beasts; and the angels ministered unto him" is not present in Luke then at most it was just what we see as Mk 1:12-13d, but could have been the even shorter version suggested above.
After the temptation, in Mark and Matthew Jesus first preaches in Galilee. He then performs healings and exorcisms (Matthew), and tells the people to repent because the kingdom of God (Mk 1:15) or heaven (Mt 4:17) is coming. In Luke Jesus also enters Galilee, and although there is no mention of any preaching, healing, etc., news/word about him nevertheless spreads, and he then teaches in the synagogues (Lk 4:14-15). However, none of this introduction is present in Ev, and instead it begins with Jesus entering Capernaum. The only text here that Luke has in common with Mark (but not Ev) is the reference to entering Galilee in Mk 1:14b, and it is quite possible for aLuke to have obtained this from the parallel at Mt 4:12b instead of from Mark.
In Mk 1:29-31 Jesus leaves the synagogue and enters Simon’s house. He heals Simon’s mother-in-law’s fever, and she then ministers to them. This passage has close parallels at Mt 8:14-15 and Lk 4:38-39, except that Matthew does not have Jesus leaving the synagogue first, nor is there any mention of anyone entering the house with him. There is nothing in Luke that requires aLuke to have seen this passage in Ev, nor anything in Mark that requires this passage to have been present in Ev. As with the choices made by aMatthew and aLuke regarding material in Mark, aEv could have not known this passage, or simply have chosen to it.
In the majority of mss the leper in Mk 1:40 does not call Jesus ‘Lord,’ whereas in both Mt 8:2 and Lk 5:12 he does, and the lack of ‘Lord’ results in a minor agreement of Matthew and Luke against Mark. In The Four Gospels: A Study of Origins, Streeter comments:
But the word [Κύριε] occurs in Mark also in B C L 579 Sah., W c e ff2, Θ 700. It is omitted by א D b, Syr. S. Boh. Byz. Hort for once deserts B, thinking B here assimilates (κύριε only once in Mk.). But the combination of the three distinct traditions, Egyptian B C L Sah., "African" W c e, and Caesarean Θ 700, is a very strong one. Either, then, B is right and there is no agreement of Matthew and Luke against Mark; or we have, not only a clear case of B L convicted of assimilation, but evidence of such an orgy of assimilation in these small details that no text can be relied on, and it is just as likely that the presence of κύριε in either Matthew or Luke may be due to the same cause.
Tertullian refers to this passage in Ev in general terms, mentioning only “the cure of leprosy,” and that “The Lord, therefore … touched the leper.” Assuming that Ev precedes Matthew and Luke this suggests that it contained ‘Lord,’ so possibly being the source of this minor agreement of Matthew and Luke against Mark, and also that, if earlier than Luke, Ev could be the cause of Streeter’s apparent “orgy of assimilation in these small details."
Epiphanius states that: “He [Marcion] did not have, “His mother and his brethren,” but simply “Thy mother and thy brethren,”” so indicating that Ev 8:19 did not exist (For details see Lk 8:19-21 in Luke Chapter 8). It is supposed that aEv removed this verse from Luke because if Jesus had a mother and brothers then he must have been corporeal, but if Ev preceded both Matthew and Luke then v. 8:19 was simply not in Ev, perhaps because aEv deemed it redundant in the light of Mk 3:32b. This also seems likely to be the case with Mk 3:31b, 33b-34a, and 35b, none of which have parallels in Luke.
It is instead possible that Mk 3:31 is a later addition to this passage, as there are unusual variants that involve swapping mother and brothers. As Mk 3:31, Mt 12:46, and Lk 8:19 are all slightly different it is possible for Luke to here be derived from Matthew, not Mark, and so the shorter text in Ev is not a problem. It is also possible that aEv knew and used the version we see in Thomas 99, which has parallels to Lk 8:20-21, but not Lk 8:19.
According to Epiphanius, the whole of the 3rd passion prediction in Luke (vv. 18:31-33, and probably also 34) was not present in Ev. Lk 18:31c (the reason for going to Jerusalem) and 34 are unique to Luke, while almost all of the rest of the Lukan version of this passage could be derived from Matthew instead of Mark. The exception is Jesus being spat on in Mk 10:34c, and in Luke this phrase is either swapped with the scourging or is missing. D (The Greek side of Codex Bezae), it, and various other mss omit Jesus being “spitefully entreated,” and some omit him being spat on, neither of which are mentioned by Epiphanius. This and the fact that being spat on is not mentioned in Lk 22-23 or in Ev suggest that Jesus being “spitefully entreated and spitted on” could be later additions.
All three synoptics contain passages describing Jesus’ entry into Jerusalem, at Mk 1:1-17, Mt 21:1-13, and Lk 19:29-46 respectively. However, Epiphanius is clear that either the majority or all of this passage was not present in Ev:
He falsified the passage about the ass and Bethphage, and the one about the city and the temple, because of the scripture, “My house shall be called an house of prayer, but ye make it a den of thieves.”
From this we know that Ev did not contain Lk 19:29-36, 41-46, and possibly also vv. 37-40. Given that all three synoptics contain versions of this passage, how could Ev not contain it if it preceded Luke, or why would aEv exclude it if Ev followed Luke? Although the parallels in Mark, Matthew, and Luke are superficially quite similar, there are some significant differences in their accounts:
Mk 11:11, 16 and some of 11:6 are unique to Mark;
Some of Mk 11:2, 3; 11:14b-16a; and some of 11:13-14, have no parallels in Matthew;
Some of Mk 11:3, 8-9; 12-14; and some of 11:15, have no parallels in Luke;
Mt 21:4-5, 10b-11 are unique to Matthew;
Lk 19:37, 39-40, 41b-44 and some of Lk 19:32-34, 38 are unique to Luke.
Of the three synoptics, Luke contains the greatest amount of unique material, although less so in Bezae, in which a significant portion of Lk 19:30-33 is not present. Even within verses that have parallels in one or both of the other synoptics there are significant wording differences, perhaps the most well known of which is that in Matthew Jesus enters Jerusalem on the backs of two animals rather than one! It should be clear from the above that it is not obvious that the narrative in Luke depends on Mark, as Brent Kinman suggests:
In turning now to the Gospels, it will be observed that the contexts of the entry narrative in Mark and Luke differ. The possibility of a drastic change to the interpretation of the story in Luke’s presentation should not be underestimated. The observation commends such widespread assent that it hardly bears repeating except in the current discussion where it is important to remember that Luke might not simply be taking over Markan theology when he assumes the language of a Markan narrative.
As noted above, some of the verses are unique to Luke, while others (Lk 19:38, 41a, 45-46) could be derived from Matthew rather than Mark, and there are details (e.g. Mk 11:8b-9, 16) that are not present in Luke that may have simply been an editorial decision by aLuke. There are also some details in Lk 19:29-31 that are not in Matthew, and so might have originated in Mark: the fact that the colt was tied; that no-one had sat on it; and the loosing of the colt. However, none of these details regarding the colt are in Bezae (see Luke Chapter 19), and as it is hard to explain why they would be removed, it is more likely that these minor details were originally not in Luke, but were later added as an assimilation to Mark.
The mention of Bethany is perhaps different, as it is not present in Mt 21:1. However, Mk 11:1 refers to “two of his disciples,” Mt 21:1 to just “two disciples,” while Lk 19:29 has two variants: “two of the disciples,” and “two of his disciples.” Willker comments regarding “his” in Luke:
Externally the omission is favored. Internally it could be a harmonization to Mt. On the other hand the longer reading could be a harmonization to Mk, but this is normally less likely. If the pronoun is wrong here, it is interesting that Mk has the pronoun safe, but both Mt and Lk omit it.
If Lk 19:29 originally did not have “his” then it is likely to be based on Mt 21:1, and the addition of “his” is an assimilation to Mk 11:1, with the addition of “Bethany” possibly occurring at the same time. Assuming this to be the case, then there is no need for an early Luke to have included Mk 11:1, or any of the verses in Mark that contain parallels to the omissions from Ev mentioned by Epiphanius.
This parable exists in all three synoptics, at Mk 12:1-11, Mt 21:33-44, and Lk 20:9-18. However, Epiphanius states that none of it was in Ev:
Again, he excised the material about the vineyard which was let out to husbandmen, and the verse, “What is this, then, The stone which the builders rejected?”
As Epiphanius did not know our verse divisions this would suggest that Ev did not contain any of what we see as Lk 20:9-18. Because Matthew contains parallels to every verse of the Lukan version of this parable it appears that aLuke could have used the version in Matthew as the basis of his version. However, as with the entry into Jerusalem, this is one of a few places in the synoptics in which the numbers of people or animals varies. In this case it is the number of servants sent to the husbandmen before the son is sent:
In Mark three servants are initially sent, followed by many more;
In Matthew an unspecified number are initially sent, three of whom are killed, followed by another unspecified number;
In Luke only three are sent.
It is conceivable that aEv knew that the description of the vineyard in Mk 12:1 was derived from Isa 5:1-2, and so did not want to include this allusion to the Old Testament, while aLuke wanted to include it (perhaps because he saw it in both Mark and Matthew), but chose to simplify the description so that it did not refer back to Isaiah. aLuke may also have seen what we know as Thomas 65 (which also has no description of the vineyard, and in which only two servants are sent) and also wanted to have a specific number of servants, but chose three because that was the number used in both Mark and Matthew.
Epiphanius is clear that Lk 20:37-38 had no equivalent in Ev, as he writes:
He excised, “Now that the dead are raised, even Moses showed at the bush, in calling the Lord the God of Abraham and Isaac and Jacob. But he is a God of the living, not of the dead.”
Although Mk 12:26-27 has a close parallel at Mt 21:31-32, the parallel at Lk 20:37-38 contains a significant difference: While in Mark and Matthew God tells Moses who he is, in Luke Moses proclaims who God is. If Ev preceded Luke and had contained a close parallel to Mark/Matthew here, then it seems likely that we would also see a similar close parallel in Luke. Instead, it is perhaps because of the lack of this text in Ev that aLuke felt free to use the framework of the passage from Mark/Matthew, but to alter its main point.
In Mk 10:17 Jesus is asked; “Good Master, what shall I do that I may inherit eternal life?” This same question is asked in the parallels at Mt 9:16 and Lk 18:18, and also in Ev. The same question is also asked in Lk 10:25, but in the (loose) parallels to this verse at Mk 12:28 and Mt 22:35-36 Jesus is asked a completely different question regarding the commandments, while Tertullian gives the question in the parallel in Ev as simply: “What shall I do to inherit life?”
While it is conceivable that aEv might by mistake have removed “eternal” from his copy of Lk 10:25 but not Lk 18:18 when he created Ev, there is no obvious reason why aLuke, seeing questions regarding the commandments in both Mk 12:28 and Mt 22:35, would decide to change it to a question regarding eternal life. On the other hand, although it is not obvious why aEv might change the question regarding the first commandment in Mk 12:28 to one about “life,” if this was the case then it is easier to see aMatthew following Mark, and aLuke following Ev, but then adding ‘eternal” to the parallel in Lk 10:25 to match Lk 18:18.
Here is another example of aLuke apparently taking the framework of a passage from Mark/Matthew but choosing to ignore the meaning of the passage. Although Mark 13:14-19 and Mt 24:15-21 are almost identical, Lk 21:20-23 has little in common with either. For example, Mark and Matthew both refer to “the abomination of desolation,” while Luke mentions “Jerusalem compassed with armies” instead. Also, in Mk 13:15-19 and Mt 24:16-21 Jesus’ statements regarding what the people should do are almost identical, while in Lk 21:21b-22 they are significantly different, as is the description of what will then take place. Luke then relocates Mk 13:17 / Mt 17-19 and adds to it, as Lk 21:23-24.
Epiphanius writes that: “He [Marcion] falsified the following: “Then let them which are in Judaea flee to the mountains” and so on, because of the words subjoined in the text, “until all things that are written be fulfilled.”” This indicates that Ev did not contain any parallel to Lk 21:21-22. However, as Tertullian refers to Lk 21:20, 25 but not Lk 21:21-24, it is possible that Ev did not contain any parallel to Lk 21:23-24 either. Epiphanius then points out that aEv’s actions in removing these words were ineffective, because he left in place other words that made the same point, that: “What the prophets and Moses said came from God the Father, from the Lord himself the Son of God, and the Holy Spirit; and once written they had to be fulfilled.”
However, if Ev was written before Luke then Lk 21:21-22 did not yet exist, and so could not have been “falsified” by aEv. Instead, on this scenario aEv chose not to include a parallel to Mk 13:14-19, which contains no mention of fulfillment, but instead refers to the prophecy of the destruction of Jerusalem in Dan 9:24-27. Why aEv would choose to do so is not known, although a reasonable speculation is that he wanted to avoid having a reference to the Old Testament prophecy. Then, with the lack of a parallel in Ev, aLuke felt free to change the Mark/Matthew words to suit his own purpose.
According to Epiphanius, none of Lk 22:35-37 (and most likely Lk 22:38 as well) was present in Ev. These verses have no parallel in Matthew, and only Lk 22:37 has a parallel in Mark, at Mk 15:28. It is worth noting that the Lukan parallels to Mk 15:26-28 (Lk 23:38, 23:33b, 22:37 respectively) are all ‘out of place’ in comparison to Mark, as Eusebius of Emesa indicates in the case of Lk 22:37:
Peter drew his sword, which the Lord bade him take with him in order to fulfil every particular dispensation connected with His humanity; according to what S. Luke declares, that this saying should be fulfilled, "He was reckoned among the transgressors” a saying which the other Evangelists apply to Christ on the Cross.
Mk 15:26-28 are split up in Luke, so it is possible that aLuke did not see them as consecutive verses in Mark. It is also possible that aLuke did not actually see Mk 15:28 in Mark at all, as this verse is not present in many mss: 01, A, B, C, D, X, Y*, Ψ, 047, 157, pm173, d, k, Sy-S, sa, and bopt. As Lk 22:37 is part of what is otherwise a purely Sondergut Luke passage (Lk 22:35-38), it seems more likely that Mk 15:28 is an interpolation in Mark that aEv did not see.
This approximately 75-verse (the ends of the omitted portion are not ‘clean cut’) section of Mark does not fit the same pattern as the passages above, as although it is present in Mark (and approximately 50% has parallels in Matthew), it has no parallels in Luke. Neither Tertullian nor Epiphanius refer to any parallel to these verses in Ev, and the possibility that equivalent text was in Ev but both Tertullian and Epiphanius somehow failed to mention this fact seems remote in the extreme. Therefore, we must conclude that these verses had no parallel in Ev either.
If Ev was a ‘cut down’ version of Luke then this is of course what would be expected. However, if Ev instead preceded Luke then the question of why aLuke omitted these verses of Mark is no longer relevant. Instead, on this scenario aLuke saw and used Ev (which did not include these verses), and so the relevant question instead becomes: Why did aEv not include these verses? Several reasons have been proposed as to why aLuke might have omitted these verses, and most are generally applicable to aEv as well: using a ‘Proto Mark;’ homeolteuton; having a defective copy of Mark; and just choosing to omit various different pieces of text from Mark. However, a few reasons are not applicable to Marcion:
The Griesbach hypothesis (that Mark was intended to conflate Matthew and Luke) does not apply;
There would be no reason for Marcion to want to omit any discussion of clean and unclean, because he had no connection with Acts.
From what we can deduce about aLuke we can give reasons why he might choose to omit particular sections of text from what we know as the Great Omission, but it is not so easy to do the same for aEv. Although we know a good deal about Marcion’s theology from his Antithesis, if Ev preceded 'his' gospel then Marcion is not aEv, and so we know nothing about aEv's theology. There is also the fact that, as seen above, there is little else of Mark that is not present in Ev, so it is hard to believe that aEv would omit 75 contiguous verses of Mark at this point. This leaves us with the suggestion that aEv had a defective copy of Mark.
On the scenario that Ev followed Luke, one common objection to the idea that aLuke had a copy of Mark that was defective here (see The Great Omission) is that, seeing that his copy of Mark was damaged he could have obtained another copy, or, even if not, could have ‘filled in the gaps’ from Matthew. However, if it was aEv who saw the defective copy of Mark then he may have had no way of obtaining another copy (we know nothing about his circumstances), and, as shown in Marcion's Gospel and Matthew, Ev appears to have preceded Matthew, so it is also highly unlikely that aEv would have been able to fill in the gap from another source, in the way that aLuke could have done from Matthew.
Unless something other than Markan priority is assumed, then (by definition) Luke depends in some way on Mark. Ev may also depend on Mark, and the most common view is that Ev depends on Mark through Luke (i.e. Mark => Luke => Ev). This analysis suggests instead that it is Luke that depends on Mark via Ev (i.e. Ev => Luke), but is it possible that Ev and Luke are independent of each other?
The answer is a very clear no. Virtually all of the content of Ev is present in Luke, and the language of the two (as far as we are able to determine it) is almost identical. Even where we can identify differences, in the majority of cases the language used in Ev corresponds to a known (often, although not always, Old Latin) variant in Luke. However, the single most telling issue is the Great Omission. As noted above, neither Ev nor Luke contain this text. If either (or both) aEv and aLuke had access to a copy of Mark that contained this text then the probability that they could independently omit exactly the same text, with the same oddities at the ‘join’ (e.g. ‘Bethsaida,’ see The Great Omission) is virtually zero. Even if aEv and aLuke saw the same copy of Mark (or copies with identical missing text), then, again, the probability of them independently dealing with the omission in exactly the same way is virtually zero. In other words, it is not possible for Ev and Luke to have independently depended on Mark.
Although Epiphanius identifies in detail the approximately 20% of the text of Luke that he states was omitted by Marcion, he makes no reference to the fact that the great majority of this omitted text has no parallel in Mark (see Tertullian vs. Epiphanius). Either he did not realize this, or he did but for some reason chose not to mention it. At first sight it appears that, by failing to mention (for example) that, in common with Ev, Mark has nothing corresponding to the first two chapters of Luke, he is deliberately concealing the possibility that Ev could be anything other than a version of Luke ‘cut-down’ by Marcion. However, it depends on how well Epiphanius knew the other gospels.
It is conceivable that, even though he was a bishop, Epiphanius did not have a complete bible. He obviously knew Luke, and in his Panarion he quotes phrases that we see in Mark, Matthew, and John, and also refers to: “the Law, the prophets, the Gospels and the [acts of the] Apostles,” so it seems certain that he knew all the canonical gospels, even if they were not gathered together in one volume. That being the case, how is it possible that Epiphanius did not recognize that, in terms of both length and content, Ev could potentially be an expanded version of Mark rather than a cut-down version of Luke?
It is quite likely that Epiphanius considered that the four canonical gospels were created independently, with what we see today as a clear literary relationship between the synoptics being explained instead as the result of divine inspiration giving the same text to all three authors. If that were the case that it may have been almost impossible for him to even conceive the idea of Luke depending on something (i.e. Ev) that itself depended on Mark, and even more so when the dominant position of the Church in which Epiphanius held high office was that Ev had been written by a heretic. Under these circumstances it is not really surprising that Epiphanius only ever saw one answer to the questions arising from the differences between Ev and Luke: that Marcion edited Luke.
The above analysis leads to some obvious questions: Why would the author of Ev use as his 'base text' a gospel (Luke) from which he felt the need to remove so much text, when there already existed a shorter gospel (Mark) in which the great majority of the text that he wanted to remove simply did not exist? Then, assuming that aEv did in fact edit Luke to create Ev, how was he able to arrange that nearly 90% of the text that he did remove was actually text that had no parallel in Mark? And finally, having gone to all that trouble, why also did aEv, according to the evidence of both Tertullian and Epiphanius, still leave in Ev so much of the text of Luke that went against his position that both were able to refute him just from what he left? Why did he (apparently) remove the wrong text from Luke?
Even allowing for the possibility that aEv might have wanted to simply exclude from Luke everything prior to the start of Jesus’ preaching, we are still left with the fact that nearly three-quarters of the other 110 verses that Epiphanius states that Marcion had removed from Luke have no parallel in Mark. To have achieved this by selectively cutting text out of Luke, as Marcion is accused of doing, would have required him to have had a very detailed synopsis comparing Mark with Luke, as the chances of him being able to achieve the result described by Tertullian and Epiphanius without careful selection of the Lukan verses not in Ev based on their presence in or absence from Mark, are very slim indeed.
Return to: Marcion's Gospel and the Synoptic Problem
If you have any comments, questions, suggestions, etc. regarding Marcion or my analysis please email me at davidinglis2@comcast.net