It may be helpful to familiarize yourself with the Conclusions of the analysis of Marcion's Gospel on this site before continuing with this page.
As described in The Synoptic Problem, much of the text of Mk, Mt, and Lk is very similar, and in some places identical, in either two or all three of these gospels, and for this reason they are collectively referred to as the Synoptic (or ‘seeing together’) Gospels, or just the ‘synoptics.’ Marcion's Gospel [Mcg] is obviously related to the synoptics, appearing to be a shorter version of Lk, and the generally accepted explanation for this is that Marcion edited Lk, removing text that did not suit his theology. However, there is no known evidence (save for the acknowledged biased opinions of Tertullian, Epiphanius, and others) that Marcion actually created Mcg, and there is nothing in the content of either Mcg or Lk that could not instead be explained by Lk being an expanded version of Mcg. If so, then it is conceivable that if Mcg is earlier than Lk, then it either is, or is based on something that is, also earlier than Mt. Consequently, if we compare the text of Mcg with that of Lk only, and make decisions regarding its origin on that basis, we are ignoring the parallel (i.e. obviously closely related, but not necessarily identical) text in Mk and Mt, and so it is important to compare Mcg with the other synoptics as well.
There is currently (Feb 2014) no agreed solution to the synoptic problem, and perhaps the most significant issue on which agreement has not yet been reached is that in some places the text of Mt appears to pre-date that in Lk, while in others text in Lk appears to pre-date text in Mt. Of the two possible solutions that can be considered the 'front-runners,' one (the Mark-Q theory) requires the use of an additional hypothetical source, usually known as Q, to solve this and other issues regarding Mt and Lk. However, as Mcg is an additional known (not hypothetical) source that is clearly related to Lk, it is worth investigating whether the text of Mcg could provide all the 'explanatory power' for which Q was originally created, and so effectively replace Q in synoptic solutions.
As we know that Mcg is basically an old, shorter, version of Lk, it is important to know whether or not the text in Mcg is more primitive than (i.e. pre-dates) the parallel text in either Mt or Lk itself. If Lk is more primitive than Mcg then it is likely that Mcg is an edited version of Lk. However, if Mcg is more primitive than Lk then there are a number of possibilities, depending on the relationship between Mcg and Mt. If Mt is more primitive than Mcg then it is unlikely that Mcg can add anything to the synoptic problem. However, if Mt is not more primitive than Mcg we need to look at the agreements (or disagreements) among Mcg, Mt, and Lk: If Mcg is more primitive than Mt we have added to the evidence that Mcg is also earlier than Lk. This is also the case in places where Mcg and Mt are the same, but both are more primitive than Lk. In both these latter cases Mcg could be a source for Mt as well as for Lk.
When investigating the synoptic problem some ‘sections’ of text in the synoptics are given particular names, based on how the synoptics overlap, or do not. Although these names are not universal, they are common enough that they will be used here as well:
In order to investigate the relationships between Mk, Mt, Lk, and Mcg, we need to look at all the Triple and Double Tradition material (everything in both Mt and Lk). Additionally, although neither Sondergut Lk nor the Mk/Lk Double tradition text contain anything from Mt (by definition), it may also turn out that this material can provide extra information. Because neither Tertullian nor Epiphanius mention the complete text of either Lk or Mcg we cannot identify all the Triple and Double Tradition material that they saw, and even in those places where we can, the text they give may not contain direct quotes. In addition, much of the Triple and Double Tradition material does not show any clear signs of directionality, and as a result, we may be very limited in our ability to determine the directionality between text in Mcg and the synoptics. Nevertheless, it is worth trying. For example:
We should also consider the rather stunning fact that, with the exception of one name (Bethany), everything that is generally believed (on the basis of Markan priority) to be material that aLk took from Mk, he could have equally have taken from Mcg instead, and perhaps never have actually seen Mk itself. These points all provide support for the view that not only is Mcg earlier than Lk, but that it is also earlier than Mt, and that both aMt and aLk knew and used, either Mcg, or prehaps a very similar Greek document from which Mcg (perhaps an Old Latin translation) was developed.
The evidence points to Mcg having been created by combining something consisting largely of edited versions (close to the Lukan parallels) of verses in Mk (excluding The Great Omission) with some parts of the text that we refer to as sondergut Lk. It appears that the only synoptic on which Mcg depends is Mk, and that Mcg could be considered to be an ‘intermediate’ form, part way between Mk and what we see as Lk today. Despite the prevailing opinion that Mcg is a later, edited version of Lk, in all respects the known text of Mcg instead meets the criteria for an early version of Lk. This strongly suggests that Mcg is at least very closely related to a pre-Mt version of Luke (Early Luke) as described in The MwEL Theory: A New Synoptic Solution.