The two-gospel hypothesis is that the Gospel of Matthew was written before the Gospel of Luke, and that both were written earlier than the Gospel of Mark. It is a proposed solution to the Synoptic Problem, which concerns the pattern of similarities and differences between the three Gospels of Matthew, Mark, and Luke. The hypothesis, following an original proposal by Augustine and expanded by Johann Jakob Griesbach (it was once called the Griesbach hypothesis), was introduced in its current form by William Farmer in 1964. This hypothesis is the most serious alternative to the two-source hypothesis. Its main advantages over the two-source hypothesis include the fact that it relies not just on internal evidence, that it does not require lost sources like the Q document, and that it is supported by the view of the early Church. Unlike the two-source hypothesis, the two-gospel hypothesis concludes that the traditional accounts of the gospels (order and date of publication, as well as authorship) are accurate. A further development of the Augustinian and Griesbach hypotheses is found in the hypothesis of Eta Linnemann, followed by F. David Farnell, that the "two Gospels" were required by the "two witnesses" rule of Deuteronomy.
The hypothesis states that Matthew was written first, while Christianity was still centered in Jerusalem, to calm the hostility between Jews and Christians. After Matthew, as the church expanded beyond the Holy Land, Luke was written as a gospel to the Gentiles. But since neither Luke (nor his patron Paul) were eyewitnesses of Jesus, Peter gave public testimonies that validated Luke’s gospel. These public speeches were transcribed into Mark’s gospel and distributed immediately thereafter, as recorded by the early Church father Irenaeus. Paul then allowed Luke’s gospel to be published.
The proposal suggests that Matthew was written by the apostle Matthew, probably in the 40s AD. At the time, the church had yet to extend outside of Jerusalem. The primary political problem within the church community was caused by the fact that Jewish authorities were outright hostile to Jesus and his followers. Matthew wrote his account in order to show that Jesus was actually the fulfillment of what Jewish scripture had prophesized. It has been long recognized that Matthew is the most “Jewish” of the gospels. It, for example, heavily references Jewish scripture and Jewish history.
When Stephen was martyred, as recorded in the Book of Acts, the disciples scattered beyond Jerusalem into Gentile (mostly Greek but also Syriac) towns. There they began preaching, and a large number of Pagans in Antioch quickly became Christians. By the mid 50s, Paul, who converted and claimed the title of "Apostle to the Gentiles" began to realize the need for a gospel to the Gentiles. This gospel would have to deemphasize the Mosaic Law and recent Jewish history in order to appeal to Greeks and Romans. Paul commissioned his associate, Luke, who used Matthew, as well as other sources. The first verses of Luke’s gospel reference the fact that “many have undertaken to draw up an account” of the testimony of the actual eyewitnesses, and, as such, he has “carefully investigated everything from the beginning” in order to “write an orderly account”. Once the gospel had been written, Paul delayed its publication. He decided that he needed Peter’s public testimony as to its accuracy, since neither Paul nor Luke had known Jesus before his death.
Paul asked Peter, who was the leader of the Apostles, to testify that Luke's account was accurate. According to early church sources, Peter gave a series of speeches to senior Roman army officers. Due to the commonality between Mark and Luke, these speeches would have constituted Peter’s public “seal of approval” upon Luke’s gospel. These church sources suggest that Peter was ambivalent when Mark asked him if he could write down the words of the speeches. However, since the Roman officers who heard the speeches liked them, they asked for copies, and so Mark made fifty copies of Peter’s speeches. These copies began circulating, and became Mark’s gospel. Only after the speeches by Peter were made (and Mark’s transcriptions began circulating) did Paul feel confident enough to publish Luke’s gospel.
The two-gospel hypothesis assumes that Peter made sure that his speeches agreed with both Matthew and (the still unpublished) Luke. Since Matthew was the primary source for Luke, and Matthew’s gospel (the only published gospel at the time) would have been well known to Peter, he mostly would have preached on the contents of Matthew. Knowing Matthew better than Luke, Peter was more likely to mention details found in Matthew and not Luke than vice versa. This would explain why there are more details found in Mark and Matthew but not Luke than there are details found in Mark and Luke but not Matthew. It also explains why Mark is so much shorter than Matthew and Luke, is more anecdotal and emotional, is less polished, and why only it begins immediately with Jesus’ public ministry. Peter was giving public speeches as to what he saw, and never intended his speeches to become a full gospel. This was directly asserted by the early church historians, and explains why there are so few commentaries on Mark (as opposed to Matthew, Luke and John) until a relatively late date. It appears to have been considered the least important gospel in the early church.
Internal and external evidence
Much of the evidence for the two-gospel hypothesis comes from the gospels themselves ("internal evidence"), while some of the evidence is found in the testimony of the early church ("external evidence"). The early church didn't just testify as to who wrote the gospels, in what order, and when they wrote them, it also testified on the specific circumstances surrounding the creation of each gospel. For example, early church documents claim that Mark's Gospel was created after Mark made fifty copies of a series of speeches that Peter had given in Rome. The external evidence (mainly the testimony of the early church) is the main difference between the two hypotheses. The two-gospel hypothesis does not dismiss the views of the early church, and makes assumptions based on both the internal and external evidence. The two-source hypothesis, in contrast, assumes that evidence still in existence (mostly internal evidence such as sentence structure or length) should be used, and makes assumptions using mostly that.
Contrasted with the two-source hypothesis
Approximately 25% of Matthew and 25% of Luke are identical, but are not found in Mark. This has been explained in the two-source hypothesis as coming from the hypothetical Q document, although by the two-gospel hypothesis, this material was copied by Luke from Matthew, but not testified to by Mark because Peter had not seen it. The two-source hypothesis also assumes that the information unique to Matthew (“M”) and Luke (“L”) came from unknown sources. The two-gospel hypothesis, in contrast, assumes “M” to be mostly Matthew’s testimony and “L” to be the eyewitness accounts mentioned in the first verses of Luke’s gospel. In addition, it gives a specific reason for the fact that Mark has more in common with Matthew than it does with Luke.
By the 1960s, scholars considered the two-source hypothesis to be the unquestioned solution to the synoptic problem. By the 1990s, however, the consensus had ended, and some scholars claimed that the two-source hypothesis had even been disproven. Subsequently, the two-gospel hypothesis has emerged as the most serious challenger to the two-source hypothesis.
The two-gospel theory is less of a conjecture than the two-source hypothesis because, unlike that theory, it doesn't assume a priori that the accounts of the early church are unreliable. Since the two-source hypothesis rejects the evidence of the early church, it relies mostly on internal evidence (such as the shortness of Mark) and conjecture (e.g. ‘why would Mark write a shorter version of a gospel in existence?’)
Compared to the Griesbach hypothesis
The Griesbach hypothesis is similar to the two-gospel hypothesis. However, unlike the two-gospel hypothesis, the Griesbach hypothesis is principally a literary hypothesis. What came to be labeled the Griesbach Hypothesis was already anticipated by the British scholar Henry Owen (1716–1795), in a piece he published in 1764, and by Friedrich Andreas Stroth (1750–1785) in an article he published anonymously in 1781. Johann Jakob Griesbach (January 4, 1745 – March 24, 1812), to whom this source hypothesis was first accredited, alluded to his conclusion that Matthew wrote the first of the canonical gospels and that Luke, not Mark, made first use of Matthew in composing the second of the canonical gospels in an address celebrating the Easter season at the University of Jena in 1783. Later, for similar Whitsun programs at Jena (1789–1790), Griesbach published a much more detailed "Demonstration that the Whole Gospel of Mark is Excerpted from the Narratives of Matthew & Luke."
Griesbach's theory was, therefore, one of direct literary dependence between and among the gospels of Matthew, Luke and Mark, or what German scholars came to call a "utilization hypothesis." According to Griesbach, the historical order of the gospels was, first, Matthew; second Luke, making use of Matthew and other non-Matthean tradition; and third, Mark, making use of both Matthew and Luke. In proposing this hypothesis, Griesbach maintained Matthean priority, as had Augustine before him, along with every other scholar in the church prior to the late eighteenth century. Griesbach's main support for his thesis lies in passages where Matthew and Luke agree over and against Mark (e.g. Matthew 26:68; Luke 22:64; Mark 14:65), the so-called Minor Agreements.
Many generic arguments in favor of Markan Priority and/or Two-source hypothesis also work as arguments against the two-gospel hypothesis. While it is impossible to list all arguments in favor and against the theory, some notable arguments are as follows.
- If Luke had access to the final version of Matthew (as opposed to both drawing independently on other sources), why are there so many significant differences between Luke and Matthew on issues such as Jesus' genealogy, circumstances of birth, and events following resurrection? While Luke and Matthew do share a lot of text which is not present in Mark, almost all of it is confined to teachings and parables. Construction of the gospels in accordance with the two-gospel hypothesis would require Luke to rewrite major parts of Matthew's narrative – even though Matthew was presumably an eyewitness who lived in Jerusalem and was surrounded by other eyewitnesses, and Luke was neither.
- "The argument from omission": why would Mark and Peter omit such remarkable and miraculous events as virgin birth of Jesus and particularly his appearance to apostles following resurrection? Both Matthew and Luke explicitly attest that Jesus appeared to the eleven disciples, including Peter, after his resurrection, and it seems incredible that Peter would not testify to that fact in his public speeches. And why is Sermon on the Mount completely omitted? 
- By design of the two-gospel hypothesis, the Gospel of Matthew had to be written originally in Hebrew. No such copies of the Gospel of Matthew exist. (The majority view is, in fact, that Matthew was written in Greek to begin with.) This fact eliminates a major potential advantage of the theory, because it eliminates the need for a hypothetical extinct source (Q) but introduces a different hypothetical extinct source (Hebrew Matthew).
- Many scholars (and particularly most Jewish scholars) hold that the concept of virgin birth in Christianity originated with mistranslation of Isaiah 7:14 into Greek, and therefore that authors of Matthew and Luke were Gentiles. This can be reconciled with the traditional Biblical chronology that dates both gospels no earlier than 75 AD and attributes "the Gospel of Matthew" to an unknown author, but it's in clear contradiction with the position of the two-gospel hypothesis that Matthew was the Saint Matthew, and, therefore, a Jew.
A related theory has Luke drawing not directly from Matthew, but from a common source, seen as a proto-Matthew. This was advanced in the nineteenth century by de Wette and Bleek, and more recently revived by Powers.
Matthaean priority is also a cornerstone of the Augustinian hypothesis, which, however, has Luke drawing from Mark rather than vice versa.
- Thomas. RL., Three Views on the Origins of the Synoptic Gospels, Kregel Academic, p. 10.
- Robert L. Thomas Three views on the origins of the Synoptic Gospels 2002 p255, and p322 "Farnell 's third axiom notes, quoting Linnemann, that the reason for four independent Gospels stems from the legal principle of Deuteronomy 19:15b: "[O]n the evidence of two or three witnesses a matter shall be confirmed.""
- "The Priority of Mark".<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
- "The Synoptic Problem" (PDF).<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
- Bromiley, Geoffrey W. (1959). The International Standard Bible Encyclopedia. p. 281.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
- Asher Norman. Twenty-six reasons why Jews don't believe in Jesus. pp. 91–96.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
- Powers, B. Ward (2010). The Progressive Publication of Matthew: An Explanation of the Writing of the Synoptic Gospels. ISBN 0805448489.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
For Griesbach's life and work, including the full text of the cited work in Latin and in English translation, cf. Bernard Orchard and Thomas R. W. Longstaff (ed.), J. J. Griesbach: Synoptic and Text-Critical Studies 1776–1976, Volume 34 in SNTS Monograph Series (Cambridge University Press, hardback 1978, paperback 2005 ISBN 0-521-02055-7).
- Beck, David (2001). Rethinking the Synoptic Problem. Baker Academic. ISBN 0-8010-2281-9.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
- Black, David (2001). Why Four Gospels?. Kregel Publications. ISBN 0-8254-2070-9.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>