"Hamilton College prepared me for New Testament studies with four years of Greek and strong training in English Literature. I arrived at Union Theological Seminary in 1963 with a passion to support those who suffered racial injustice. The Civil Rights Movement relied on the Hebrew prophets and Jesus’ concern for the poor, but countless white Christians, especially in the Bible Belt, refused to repent of America’s original sin. I hoped to find the real Jesus beyond conflicted church practices.

Union Seminary introduced students to historical critical study of scriptures: first, sources of the Pentateuch, then the Synoptic Problem. One key text was B.H. Streeter’s seminal 1924 book, The Four Gospels: A Study of Origins. Streeter defined Q — Quelle, or source — by listing material common to Matthew and Luke that did not appear in Mark.

Marcion and the Synoptics

During my first two years at Union, Professor John Knox encouraged me to examine the history and editing of the Synoptic Gospels. He had published Marcion and the New Testament in 1942, but amid the turmoil of World War II, his book got little attention. Knox described deeply human struggles that shaped the New Testament.

Marcion had preached that the harsh God of the Torah could not be the loving Father that Jesus proclaimed. The Church destroyed Marcion’s scriptures, but Tertullian and Epiphanius refuted him in such detail that scholars have used their attacks to recreate most of the lost gospel. Knox thought it contained the bulk of Luke between Jesus’ baptism and his resurrection.[1] Few doubted that Marcion’s version of Luke had once existed, but Q remained a literary ghost.

During a year I took at Cambridge University, 1965-66, Professor C.F.D. “Charlie” Moule guided my study, and I relied on his 1962 volume The Birth of the New Testament. He urged me to examine the Synoptic Problem in its ecclesiastical context, including whether Matthew might have copied from an early version of Luke, rather than Q.

I recognized Matthew as a brilliant editor. It appeared that he used Mark as a framework, then gathered material into a topical gospel that was superbly suited for preaching. It appeared that he condensed Marcan pericopes to save space. If he hoped to have his gospel copied and circulated, he needed to fit his material onto a standard size scroll or codex.

William R. Farmer had offered a contrarian view in his 1964 study, The Synoptic Problem: A Critical Analysis. He argued that Matthew wrote first, that Mark copied from Matthew, and Luke adapted material from both. Farmer built his argument for Matthean priority in sixteen steps,[2] but he never asked why Luke or Mark — if they used Matthew — would have dismembered Matthew’s five magnificent topical discourses, only to scatter the fragments in sometimes jarring catchword groups. Nor did Farmer explain why Mark or Luke might have omitted cogent parables that appear only in Matthew: the Unforgiving Servant (Mt 18:25-35), Workers in the Vineyard (Mt 20:1-16), the Two Sons (Mt 21:28-32), and the Wise and Foolish Wedding Guests (Mt 25:1-13).

I wondered why Farmer dedicated his 1964 book to John Knox but never mentioned Knox’s work on Marcion. Nor did he explore what part Marcion’s version of Luke might have played in gospel editing.

The evidence lay in contested ground between Streeter, Farmer, Knox, Moule, Austin Farrer, B.C. Butler, George Kilpatrick, Rudolf Bultmann, and other scholars.

Questions about Omissions — Women

Streeter listed one condition under which Matthew might have omitted material from Q: “To feel confidently that any at all notable saying in Q was omitted by Matthew one must see clearly that the saying would lend itself to an interpretation by the faithful which he definitely disliked.”[3]
Both Streeter and Bultmann were baffled by the possibility that Matthew had seen and rejected the story of the Widow’s Mite (Mk 12:41-44/Lk 21:1-4). The poor widow had not served Jesus, shown faith in him, or humbled herself before him, yet he singled her out for high praise. Jesus said she had given more than rich men who contributed large sums.

Over the centuries, commentators had noted Luke’s interest in women. In fact, Mark, Luke and John all show Jesus responding empathetically to women and welcoming them among his followers. Such practice would have been extraordinary for a traveling rabbi in Palestine, and it would have shocked male leaders in the early Church.

Under the principle of lectio difficilior potior, I believed it more plausible that Jesus dealt with women in ways described by Mark, Luke and John than that these three gospels all embellished Jesus’ treatment of women.

If Matthew edited Mark, his editing reduced personal detail about women to a minimum. In the stories of Jairus’s daughter and the woman with chronic bleeding (Mt 9:18-26/Mk 5:21-43/Lk 8:40-56), Mark tells readers that the woman had suffered under many doctors and spent all she had, but her condition worsened. In the Marcan story of Jairus’s daughter, the mourners express anguish over the child’s death. Matthew drops these dramatic details. By sheer word count, he trims Mark’s nested narratives by more than half. Similarly, in the empty tomb narratives (Mt 28:1-10/Mk 16:1-8/Luke 24:1-11), Matthew drops the plaintive question the women ask in Mark: “Who will roll away the stone from the entrance of the tomb?”

I could not find any example in Matthew where Jesus heals unless the sufferers, their relatives, or friends approach him humbly and with faith. Both Marcion’s reconstructed gospel and canonical Luke both show Jesus responding empathetically to women, even though they showed no prior faith, devotion, or humility. In the Widow’s Son at Nain (Lk 7:11-17) and the Woman Healed on the Sabbath (Lk 13:10-17), Jesus approaches the women involved and performs miraculous cures.
I wondered how Matthew and his congregation would have reacted to the Lucan story, which appeared in Marcion, where Martha wants Jesus to send Mary to help her prepare dinner (Lk 10:38-42). “Martha, Martha,” Jesus replies, “you are worried and upset about many things. One thing is needed. Mary has chosen what is better, and it will not be taken away from her.” In a similar vein, the parable of the Unjust Judge (Lk 18:1-8) affirms a widow’s relentless demand for justice, while comparing God to a cynical magistrate.

None of this proves that Matthew actually saw these stories in Q or an early version of Luke, or that he decided to exclude such pericopes from his gospel. But if Matthew held traditional views on the place of women, he would probably have found Jesus’ treatment of women disturbing, particularly the stories of the Sinful Woman with Ointment (Lk 7:36-50) and the women who traveled with Jesus (Lk 8:1-3). These pericopes appear in both Marcion’s Gospel and canonical Luke. They satisfy Streeter’s criterion: sayings or stories that lent themselves “to an interpretation by the faithful which [Matthew] definitely disliked.”

Professor Moule welcomed my approach and shared insights he would soon publish in his introduction to a collection of scholarly papers: The Significance of the Message of the Resurrection for Faith in Jesus Christ (1968). In all four gospels the Empty Tomb narratives name the women who went to do the traditional women’s work of washing and anointing a body. Moule argued that this unanimous record of women as first witnesses provided striking proof of the resurrection. He noted that since Jewish and Roman rules of evidence discounted or barred women’s testimony, anyone perpetrating a hoax would have produced male witnesses.

Scholars writing about the Synoptic Problem in the 1960s were all men, and I shared their blind spots about women. During my senior year in college, several of us went to hear a woman preach.[4] One joked afterward by quoting Samuel Johnson’s comment about a female preacher: “Sir, a woman’s preaching is like a dog’s walking on his hind legs. It is not done well; but you are surprised to find it done at all.”

Now my work on the Synoptic Problem forced me to face my own prejudices. I recognized that gender bias had played a decisive but overlooked role in gospel editing.
Streeter made what seemed to be two dubious assumptions: first, that all the material in Q was as useful and as acceptable to Matthew as Mark was; second, that Q had no greater proportion of sayings which Matthew found offensive than was to be found in Mark.
I asked how Matthew might have reacted if his second source resembled what we knew of Marcion’s Luke. It seemed unlikely that Matthew would have used anything like Marcion’s gospel as fully as he used Mark. I wondered what other biases might have affected Matthew’s editing.

Disputes over Law

Streeter and Bultmann were baffled by Matthew’s apparent omission from Mark of the Widow’s Mite. Nor could either explain Matthew discarded Mark’s story of the Strange Exorcist (Mk 9:38-41/Lk 9:49-50). This apparent deletion alerted me to another apparent bias in Matthew’s editing. In contrast to Marcion and others who rejected the Hebrew scriptures, Matthew shows Jesus fulfilling the law and commanding his followers to obey it. In verses peculiar to Matthew’s Gospel, Jesus insists on keeping the Law, down to the tiniest letter or pronunciation mark. He also commands his followers to surpass the scribes and Pharisees in fulfilling it. Only Matthew quotes Jesus warning against those who relax the commandments and teach others to do so (Mt 5:17-20). He calls them “false prophets . . . wolves in the clothing of sheep,” and “evildoers” (Mt 7:15-6, 23). These unique Matthean texts took direct aim at antinomian Christians who shared Marcion’s approach.

I wondered how well-documented disputes between Jewish Christians and gentile converts over keeping the law might have influenced Matthew’s editing. It appeared that Matthew inserted a peculiar comment in Mark’s story of the Canaanite woman. He quotes Jesus telling her he was sent “only to the lost sheep of Israel” (Mt 15:21-28/Mk 7:24-30). This idiosyncratic theme appears again in Matthew’s Mission Discourse where Jesus commands the disciples: “Do not take the gentile road, and do not enter a Samaritan city, but go rather to the lost sheep of the house of Israel” (Mt 10:5-6). Only Matthew explicitly opposes a mission to the Samaritans and Gentiles.

I could not believe that both the anti-Samaritan and pro-Samaritan elements in the Gospels were authentic from the ministry of Jesus. It seemed to me that either Jesus refused to deal with Samaritans and commanded his disciples to do the same, or he dealt with them and allowed his disciples to do so. Although Mark reveals nothing about this, Luke and John are fully aware of the mutual loathing between Jews and Samaritans, yet both show Jesus deliberately establishing contact. Matthew’s opposition to the Samaritans includes only isolated sayings, but Luke and John include vivid narratives of Jesus shattering anti-Samaritan stereotypes.

In one Lucan story, residents of a Samaritan village refuse to receive Jesus and his followers as they walk toward Jerusalem (Lk 9:51-6). James and John offer to call down fire on them — in Marcion’s text, they cite the example of Elijah[5] — but Jesus rebukes them. Luke also shows Jesus telling an unforgettable parable about a Samaritan who fulfills the high command to “love your neighbor as yourself” (Lk 10:29-37). The parable’s torque springs from the fact that the Samaritans had no status under Jewish Law, yet this Samaritan far surpasses religious leaders. This struck me as precisely what Streeter would have considered “an interpretation by the faithful which [Matthew] definitely disliked.”

Wealth and Poverty

A third major difference between Matthew and Luke — or Marcion’s version of Luke — involved questions of wealth and poverty. G. D. Kilpatrick argued in 1950 that Matthew was compiled for use in a wealthy urban Jewish Church.[6] Kilpatrick demonstrated how Matthew modified material he found in Mark and other sources, making it more palatable to rich Christians. Few commentators asked whether the radical blessings and curses in Luke’s Sermon on the Plain (Luke 6:20-26) might represent words of Jesus more accurately than the highly spiritual beatitudes in the Matthew’s Sermon on the Mount (Mt 5:3-11). It seemed to me that the rhetorical power of Matthew’s sermon concealed its tolerance of wealth. This visible editorial bias provided a second reason for Matthew to pass over Mark’s story of Jesus praising the poor widow for giving more than the “many rich people.”

Tension between Luke and Matthew over wealth may also appear in their editing of Mark’s story of the Rich Young Man (Mt 19:16-22/Mk 10:17-22/Lk 18:18-23). Luke shows Jesus telling the devout young man to sell everything he owns and give to the poor. Matthew softens that challenge when he quotes Jesus saying: “If you want to be perfect (teleios), go, sell your possessions and give to the poor…” Matthew apparently inserted this notion of aspiring to perfection through radical sacrifice. That concept took root in the early church and continues both in Roman Catholic and Methodist traditions. Of four questions John Wesley asked candidates for ministry, three were about “going on to perfection . . . [being] made perfect in love in this life,” and “earnestly striving after it.” Matthew admits that Jesus occasionally asked his followers to renounce possessions, but he does not suggest that all Christians must surrender all wealth. He calls for charity but shuns pericopes that equate possessions with evil.

The parable of the Rich Man and Lazarus (Lk 16:19-31) seems, at first glance, as deserving of a place in Matthew’s Gospel as the parable of the Sheep and the Goats (Mt 25:31-46). Both demand charity for the poor, and both threaten hell for those who refuse. But here, as in other Lucan parables, judgement falls harshly on the rich man “who dressed in purple and fine linen and lived in luxury,” while sympathy flows to the poor beggar outside his gate. In Matthew’s parable of the Sheep and the Goats, the judgement has nothing to do with the distinction between the rich on one hand and the poor on the other. The only Matthean criterion is whether they have cared for least of these.

Beyond My Biases
​
Research into what appeared to be Matthew’s editorial bias helped me see the way I and other men routinely undervalued women. Only days before my first term at Cambridge, I married a family friend. On our ocean voyage to England, she practiced signing her new name “Mrs. H. Philip West Jr.” During our year there, she struggled with depression and her identity as a woman.
Back in New York, marriage and motherhood seemed to make her feel worse. At that point, we recognized parallels between the self-loathing of many blacks who lived with the legacy of chattel slavery and women who had long been treated as their husbands’ property. Slaves and women were routinely given their masters’ names, denied educational opportunities, treated as inferiors, and beaten into submission. Many blacks and women internalized a sense of inferiority, even self-hatred.

My wife and I joined the National Organization for Women. As she began advocating for equal treatment under law, her depression lifted, and our marriage improved. She went back to using her birth name, Anne Grant. She worked for girls’ equal education and produced award-winning programs on women’s history. Her graduate thesis explored how Elizabeth Cady Stanton and a committee of women who were barred from voting challenged male supremacy and started publishing The Woman’s Bible in 1895.

In 1971, Professor Leonard Swidler published his persuasive declaration that “Jesus was a Feminist.”[7] He argued that women in Jesus’ time were segregated in an outer court of the Temple, forbidden to pray publicly, and not counted toward a quorum for worship. Swidler quoted one First Century rabbi who declared that “the words of the Torah be burned than entrusted to a woman.” He concluded: “Jesus vigorously promoted the dignity and equality of women in the midst of a very male-dominated society: Jesus was a feminist, and a very radical one.”

I remain grateful that Professor C.F.D. Moule proposed my paper, “A Primitive Version of Luke in the Composition of Matthew,” for publication in New Testament Studies.[8] He and Professor John Knox encouraged me to continue this scholarship, but in the turbulent 1960s I felt an urgent call to urban ministry.

I spent 22 years in church work, then 18 years leading a secular reform group, Common Cause Rhode Island. In retirement, I wrote a first-person account of how scandals shook a state that was notorious for public corruption. Coalitions of angry citizens rallied to demand changes in the system of government, and historic reforms followed.[9]

Recently, I reviewed current research on the Synoptic problem. I was delighted to find scholars challenging the existence of Q and presenting evidence that Matthew had copied from Mark. I applaud the work of Ronald V. Huggins, Evan Powell, Erik Aurelius, Robert MacEwen, and Alan Garrow. 

During more than fifty years since my paper about Matthew’s likely omissions from an early version of Luke appeared in NTS, scholars have wrestled over Q and largely reconstructed Marcion’s gospel. In 2015, Dieter T. Roth published an exhaustive historical recreation of Marcion’s Gospel.[10] It shows Jesus embracing women, people outside the law, and the poor. Even a quick reading shows why Matthew would have found this portrait of Jesus disturbing.

I recently found David Inglis on the Internet. He describes himself as a computer professional, not a scholar. But he has created a comprehensive website on the Synoptic Problem. His research on Marcion includes a chapter-by-chapter parallel layout of Marcion and Luke. Inglis proposes what he calls “MwEL: A New Synoptic Hypothesis,”[11] which goes far beyond what I proposed in 1967.
One dynamic remains clear: those who proclaim Jesus often tailor their message for particular audiences. As the gospel spread across the Roman Empire, evangelists inevitably colored Jesus and edited his words for congregations they cherished. It appears that Matthew reworked his material to soften Jesus’ shocking empathy for women, poor people, antinomians and those outside the fold.
Some may object that my research revealed what I wanted to find. I would counter with a metaphor of digging for fossils. The challenge for paleontologists is to uncover and recognize evidence that has been hidden from view. Only further study will tell whether Matthew edited material from an early version of Luke. For now, it appears that scholars who argue that Matthew edited and compiled material from Mark and other sources will prevail.

I can say with certainty that Jesus’ responses to women helped me break out of cultural assumptions I had absorbed in the 1950s. His example made me a better pastor and advocate. Since our year at Cambridge, Anne Grant and I have become equal partners in a liberating marriage adventure of nearly 55 years.

I regret that many men who govern major Christian denominations still claim scriptural authority for excluding women from power. I long for the day when all church leaders will follow Jesus’ example in breaching historic animosities, and I believe Jesus would rebuke those who twist his words as they preach the Prosperity Gospel."

[1] Knox, John, Marcion and the New Testament: An Essay in the Early History of the Canon, (Chicago: Univ. of Chicago, 1942) 110.

[2] Farmer, William R., The Synoptic Problem: A Critical Analysis, (New York: Macmillan, 1964) 199-232.

[3] B.H. Streeter, The Four Gospels: A Study of Origins, (London: Macmillan, 1924) 290.

[4] Rev. Letty Russell was one of the first women ordained in the United Presbyterian Church (1958) and a founder of the East Harlem Protestant Parish, where she trained church leaders.

[5] 2 Kings 1:10

[6] Kilpatrick, G. D., The Origins of the Gospel According to St. Matthew, (Oxford: Oxford Univ. Press, 1950) 124-26.

[7] Swidler, Leonard, “Jesus was a Feminist,” Catholic World, Jan. 1971.

[8] West, H. Philip, Jr., “A Primitive Version of Luke in the Composition of Matthew,” NTS, Oct. 1967.

[9] West, H. Philip, Jr., Secrets and Scandals: Reforming Rhode Island, 1986-2006, (East Providence: Rhode Island Publications Society) 2014.

[10] Roth, Dieter T., The Text of Marcion’s Gospel, (Leiden: Koninklijke Brill) 2015

[11] Inglis, David, “MwEL: A Mew Synoptic Hypothesis,” https://sites.google.com/site/inglisonmarcion/Home/the-synoptic-problem/mwel-theory (accessed 5/13/2020).

"My article 'Matthean Posteriority: A Preliminary Proposal' (1992), originated as a paper written for the Q Seminar of the late Professor Heinz O. Guenther, which I took during the first semester of my doctoral studies at the University of Toronto/Toronto School of Theology.  My argument was that Q might be dispensed with if we assume that Luke used Mark and that Matthew used both Luke and Mark.  The form in which the article appeared in Novum Testamentum is identical to the copy I turned in to Professor Guenther, indeed I submitted both, albeit quite late with regard to the seminar, on the same day, one copy to Professor Guenther and another to the post office addressed to a journal.  

Guenther, always extremely fastidious when it came to the interests and scholarly development of his students, returned the paper with several pages of comments made in red ink, concluding that it was “beyond the pale” for the course, but nevertheless, to my great relief, assigning it an A.  Not long afterwards, I remember Professor Guenther taking me aside, and, in a fatherly sort of way, urging me that “life was too short,” to strike off on my own looking for a new solution to the Synoptic Problem.  Better, he said, to join one of the established schools and engage in meaningful research along with others there.

In those days the University of Toronto was particularly strong in Synoptic Studies, as I suppose it still probably is.  This is seen in the fact that prominent Q scholar John Kloppenborg came a little before me in the program and Bob Derrenbacker a little after.  As a result, I became cognizant very early of both the elegance and tremendous explanatory power of the dominant two-source theory, a thing I still fully appreciate. I did wonder, however, whether part of the reason it seemed so persuasive to me was that I wasn’t hearing the other side of the arguments.  This led me to feel I needed to explore the potential viability of the alternative solutions as well, including of course the one presented in Austin Farrer’s 1955 article “On Dispensing with Q.”  Farrer attempted to dispense with Q by proposing that Matthew used Mark and Luke used Matthew and Mark.  His most energetic disciple today is Duke University’s Mark Goodacre.

At first Farrer’s hypothesis seemed promising, much more so than other alternative solutions I’d looked at, because unlike them, it affirmed Marcan Priority (the idea that Matthew and Luke used Mark, not the other way around), which by that time seemed to me to be an inescapable fact.

In any case, I set to work trying to make sense of what I saw going on in the Synoptic Gospels (and had been made to see by Heinz Guenther and then afterward by Joseph Plevnik S.J.) through the grid of Farrer’s solution.  I understood that since Synoptic Studies had gained a great deal of precision since Farrer’s article, I might even need to refine the theory a bit myself in order to make it work.   I very quickly found myself liking Farrer’s view and actually hoping he was right.  And yet as I put it to the test point by point, passage by passage, I kept finding myself again and again running up against places where it just didn’t commend itself as a solution, places where Matthew seemed more developed than Luke. 

This was, of course, what I had in a way expected, since I was already aware of the claim by defenders of the two-source theory, that both in order and in form, Luke is generally “more primitive” than Matthew in the double tradition (i.e., in Q).  This is why for example Q passages usually take their chapter and verse references from Luke rather than Matthew.   According to The Critical Edition of Q there are only eleven places where Matthew rather than Luke is thought to preserve the original order of Q.   So, I faced the question rather early whether Matthew was really ever indisputably “more primitive” than Luke when it came to their shared Q material. And I was surprised to find that even as I looked at those eleven passages, which included, for example, Jesus’s temptation in the wilderness, and his teaching about houses built on rock and sand, it was by no means obvious to me that Matthew really could be shown to be “more primitive” than Luke. 

For two-source theorists the idea that sometimes Matthew and sometimes Luke preserves Q in its more original form and/or order, a phenomenon they call “alternating primitivity,” is a key argument given for why Luke and Matthew had to get their shared non-Markan material from Q rather than from one another.  Yet looking at those few places where Matthew was deemed “more primitive,” it was not long before I began to suspect that the two-source defenders found their few examples of Matthew’s being “more primitive,” because their theory led them to expect to find them. In other words, alternating primitivity was being discovered, only because it was already presumed to exist.  I could not see that it was a position they were in any way pressed into embracing by the evidence itself.  This is why I put the matter as definitively as I did in the article: “The solution offered here will ultimately stand or fall on whether it can be demonstrated beyond doubt that Matthew is more primitive than Luke at certain points in the double tradition.”

In any case, the very fact that I was finding myself agreeing that Matthew was anyway most often less primitive than Luke in material designated Q represented a significant blow to Farrer’s thesis.  If Luke really was using Matthew, why and how was he making the material consistently look “more primitive”?      

As I looked more closely at particular passages, the inadequacy of the Farrer position became even more obvious.  The reason was that in order for Farrer to be right it would require Luke taking nicely ordered things from Matthew, picking them apart, and scattering them around in different places in his Gospel for no apparent reason.  A case in point is the famous woes passage of Matthew 23. In Matthew we find the woes all neatly presented one after another in a row, each one starting with an identical stereotyped introduction: “Woe to you Scribes, Pharisees, hypocrites” (23:13,14,15, 23, 25,27,29).  So, if we want to say Luke is getting his woes from this passage in Matthew a number of whys emerge to trouble us.  First, why did Luke drop two of the woes (23:15 and 25)? What was his driving 'redactional' (editorial) principle for doing so?  Why did he break up the passage itself, putting three woes (in different order) in his 11th chapter, and one in his 20th chapter?  But most significantly, why did he get rid of Matthew’s neatly repetitive introductory formula, 'Woe to you Scribes, Pharisees, hypocrites,' and replace it with a range of different introductions: 'Woe to you Pharisees!' (11:42 & 43), 'Woe to you...' (11:44-47), “Woe to you lawyers also!” (11:46), 'Woe to you lawyers!' (11:52), 'Beware of the Scribes' (20:46)? 

Now I could see of course that with a little ingenuity I could invent what might pass for a reason why Luke had done these things.  I could say, for example, that he was interested in being more precise in attributing certain woes to certain groups.  But I already knew that the problem was larger than this. This was only one instance where if Matthew came last his editorial activity would be showing evidence of a propensity for greater organization, whereas if Luke came last, his could only be described as being driven by a tendency toward greater disorganization.   And yet why should that be the case since where we can also check him against Mark we find that he  treated his material there quite differently, quite conservatively.

Two places where this really came home to me in passages shared by Matthew and Luke, were (1) the Salt of the Earth/Light of the World sayings, and (2) the Sign of Jonah saying. 

(1) Salt of the Earth / Light of the World:

​In this first case we also have a Mark/Q overlap which I didn’t show (see Mark 9:49-50).  And Luke does agree with Mark’s version to some extent.  Yet there is enough overlap between Matthew and Luke where they are not following Mark to be able to tell they have shared additional material.  So, if Luke is following Matthew then it represents, in this case, one piece in his larger project of demolishing and redistributing Matthew’s Sermon on the Mount.  First of all, Luke would be breaking the passage in two, putting one part in chapter 14 and the other in chapter 8.  And at the same time jettisoning Matthew’s nicely balanced doublet “You are the salt of the Earth...You are the light of the world.”  But why would he do that? Again, I felt sure that with a little ingenuity I could come up with a reason, yet I knew in my heart of hearts that a better explanation would probably be that Matthew added the nice couplet to his material rather than that Luke took it out. 

(2) The Sign of Jonah (Matthew 11:39-40/Luke 11:29)
 
This was actually for me the straw that broke the camel’s back with regard to my finally rejecting the Farrer hypothesis.  In Matthew 12:39 and 40, Jesus says:

'No sign will be given...except the sign of the prophet Jonah.  For just as Jonah was three days and three nights in the belly of the sea monster, so for three days and three nights the Son of Man will be in the heart of the earth.'

But in Luke 11:29 we read only:

'No sign will be given to it except the sign of Jonah.' 

As I began reflecting on this passage, whatever vestiges of confidence I had in the possibility that Farrer might be salvaged evaporated.  The question was whether I thought that if Luke really had the text of Matthew lying there before him, and actually saw and read in Matthew’s text this very striking prophesy of the death, burial, and resurrection of Jesus, from the mouth of Jesus himself, would he really have been likely to remove it as he transferred the passage over into his own Gospel?  And for me at least the answer had to be no. No, he would not have removed it.  What? Censor a word from the Lord like this?  Surely, he would have kept it, wouldn’t he? Surely.  And so, with that, I simply surrendered. Farrer’s solution wouldn’t work. Truth be told he had the whole thing ass backwards.  And what I was seeing wasn’t Luke crudely pulling things apart but Matthew elegantly pulling them together.

All this meant for me at the time was that the dominant two-source theory—Mark and Q as sources for the independent Matthew and Luke—was probably the most viable one after all. And yet the time I’d spent working through the material from the perspective of the Farrer hypothesis had borne fruit by forcing me to ask myself again and again whether Luke or Matthew looked “more primitive.”  In rejecting Farrer’s view passage by passage, I was tacitly affirming to myself that if there was interdependence going on it would have made much more sense to think that Matthew was getting his material from Luke rather than that Luke getting his from Matthew. 

And yet at it hadn’t yet occurred to me to ask myself whether Q might actually be set aside by arguing that Matthew used Luke.  That would only happen a bit later when I was once more engaged in reading defenses of the two-source theory, especially were they were making their arguments about Matthew and Luke not knowing one another.  First of all, of course, they would appeal to alternating primitivity as a proof, which I had already discounted in my own mind, but then they would often move on to present arguments, standard since Streeter, as to why Luke couldn’t have used Matthew.  Arguments I now regarded as fully valid.   But then they would just jump from there to their conclusion by asserting Luke could not have used Matthew, nor Matthew, Luke.    It was there that I saw it.  Nobody was actually looking at the evidence the other way around, i.e., thinking through how the evidence would look if one supposed not that Luke used Matthew, but that Matthew used Luke.  Instead they all seemed to be tacitly taking for granted that if they’d proved that Luke could not have used Matthew, it followed as well that Matthew could not have used Luke.  So I decided one day to sit down to try and see how things would look from that perspective, really expecting to arrive at a dead end rather quickly.  To my surprise however that didn’t happen, indeed I found the approach illuminating and more satisfying than I had expected.  At the time I had no idea that anyone had attempted this before.  I was aware of H. P. West’s similar thesis, but felt that he had framed it in such a way as to sidestep certain issues that I did not feel I could credibly avoid.  I only heard quite a bit later from Stephen Carlson that Christian Gottlob Wilke had already argued for my view back in the 1830s."

Ron Huggins' article is available in two locations: 
​
"Matthean Posteriority: A Preliminary Proposal." Novum Testamentum 34 (1992): 1-22.
​Reprinted in The Synoptic Problem and Q: Selected Studies from Novum Testamentum, ed. David E. Orton. Brill's Readers in Biblical Studies, vol. 4, 204-25. Leiden: Brill, 1999.

The following is an extract from James R Edwards: The Hebrew Gospel and the Development of the Gospel Tradition (Eerdmans, 2009) pp.245-252

Matthean Posteriority
“Posteriority,” a rarely used antonym of “priority,” needs a word of interpretation.  The historical-critical method endeavored to uncover or recover the oldest and earliest forms of a given tradition.  “Earlier” connoted (more) genuine, whereas “later” was regarded as derivative and less authoritative.  On this scale of values, “posterior” is at least slightly denigrating.  In the current discussion, however, I wish, if possible, to free “posterior” of any pejorative connotations, for the Gospel of Matthew is not a lesser and redundant work.  Throughout the history of the church, the Gospel of Matthew, along with the Gospel of John, has been held in special esteem as one of the two most important gospels.  Readings for Sundays and holy days have traditionally been taken from Matthew, and only from other gospels where Matthew was thought to be deficient. At least until modern times, Matthew has always been regarded as the “primary” gospel among the Synoptics. If Matthew was not primary in terms of chronology, from a historical perspective it rightly claims a primacy in terms of definitiveness.[1]

But exactly how is Matthew definitive? Certainly, it is definitive in terms of structure.  None of the four gospels exhibits the symmetry of the First Gospel.  The other gospels display basic structures, to be sure, even elementary symmetrical structures.  Both Mark and John devote roughly the first halves of their gospels to Jesus’ initial ministry in Galilee, and the second halves to his concluding ministry in Judea and Jerusalem.  The Third Gospel adds a lengthy travel narrative in 9:51-18:14 between the Galilean and Jerusalem ministries of Jesus, creating a three-part structure.  But none of the three gospels exhibits the complex design and proportion of the First Gospel, which begins with two chapters of infancy prologue, and ends with two chapters of passion narrative.  In between, the body of Matthew consists of five major divisions, each consisting of three subdivisions, contain-[p. 246] ing a block of narrative material, followed by a block of didactic material, and a summary refrain in conclusion, “When Jesus had finished all these things . . . . “[2]  The Gospel of Matthew displays greater design, balance, proportion, and order than any of the other three gospels.   

Perhaps the most distinctive aspect of the Gospel of Matthew is the Sermon on the Mount.  The material in Matthew’s Sermon, gathered in chapters 5-7, is also present in Luke, but in Luke the material is dispersed throughout chapters 6, 8, 11, 12, 13, 14, and 16 where it is intermingled with Jesus’ travels and healings and parables.  The same material is decidedly less distinctive and accessible in Luke. Matthew’s Sermon on the Mount, on the contrary, presents readers with an arranged manual of Jesus’ teaching. Delimited by subjects and presented in formulaic refrains (e.g., “You have heard it said of old . . . . but I say to you”), Matthew offers the church an anthology of Jesus’ teaching, primarily on ethical matters.  It is a catechetical handbook that can be memorized, taught, and transmitted.     

To somewhat lesser degrees the blocks of didactic material elsewhere in Matthew follow similar canons of arrangement.  The instructions to disciples in Matthew 10, for example, or the unit on parables in Matthew 13, the litany of woes against Pharisees in Matthew 23, and the eschatological discourse in Matthew 24, all display similar editorial intent and design.  None of the other gospels orders material into the clearly defined categories that are evident in Matthew.[3]

[p. 247] The same attention to symmetry and formulaic harmony is evident in Matthew’s version of the Lord’s Prayer.  Luke’s prayer in 11:1-4 seems truncated and incomplete compared to Matthew’s prayer in 6:9-13.  In Luke, God is addressed simply as “Father,” and two brief ascriptions of praise (“hallowed be your name; your kingdom come”) are followed by three petitions (for bread, forgiveness, and deliverance from temptation). Matthew’s version of the Lord’s Prayer, by contrast, is perfectly crafted and balanced: three formulaic ascriptions of praise to God are complemented by three subsequent petitions, the whole of which is prefaced and concluded (at least in a large portion of the textual tradition) with liturgical formulas.  Few literary critics will doubt that Matthew’s version represents the full flowering of an earlier version of Luke’s prayer.[4]

Another illustration of the same developmental flow occurs in the NT accounts of the Lord’s Supper. All three Synoptics preserve a generally fixed Eucharistic form in which Jesus first institutes the bread and then the cup (Matt 26:26-29; Mark 14:22-25; Luke 22:19-20).  The Eucharistic liturgy of the early church antedated (e.g., 1 Cor 11:23-26) and undoubtedly influenced the earliest gospel accounts of the Lord’s Supper.  But Luke is unique among the canonical gospels, and other early Christian Eucharistic texts preserved in the Didache, for example, in prefacing the words of institution with a preliminary cup attended by eschatological sayings. This produces a Lucan Eucharist sequence of cup—bread—second cup.  The only other early Christian Eucharistic account known to include the Lucan preface to the Lord’s Supper is the Hebrew Gospel quoted by Epiphanius. On form-critical grounds scholars have long suspected that Luke 22:15-18 preserves the earliest authentic Eucharistic words of Jesus.[5]  The agreement of this material with the wording of the Hebrew Gospel not only confirms that suspicion but makes it highly probable that the source of Luke 22:15-18 is the Hebrew Gospel itself.   In contrast to Luke’s longer and complex account of the Lord’s Supper, Matthew (and Mark also) presents a concise account of the words of institution, evenly proportioned between bread and cup.  Matthew’s account is close to Paul’s account of the same in 1 Corinthians 11:23-26, and appears to preserve the ultimate consensus on the Eucharist.

[p. 248] The Greek of the First Gospel must also be mentioned in regard to its dating.  Although the Gospel of Mark exhibits dramatic structural features by sandwiching a defining narrative into the midst of a host narrative,[6]or signaling a relationship between two texts by placing them side-by-side (e.g., Mark 4:35-5:20), the Greek of the Second Gospel can be and often is less than felicitous.  Luke writes a technically superior Greek to Mark, but it is generally more complex, not the least because of its Hebraisms. Matthew’s Greek, on the other hand, is clean and consistent, and his style and wording rarely need to be (or can be) improved.  The Gospel of Matthew appears to have passed through many editorial filters. The result is a gospel that affords memorization and is eminently suitable for public reading.  

Matthew also presents a more developed Christology than does Mark or Luke.  This is especially true of titular Christology, as evinced by Matthew’s use—or frequency of use—of such terms as Son of Abraham, Son of David, Son of God, Emmanuel, Prophet, Christ, Son of Man, and The Coming One.  These titles appear in greater frequency in Matthew than they do in Mark and Luke.  If Matthew were the source of the Synoptic tradition or Double Tradition, it is difficult to explain why such terms would be reduced or eliminated in Mark and/or Luke. Matthew also exhibits a sophisticated implicit Christology, an example of which appears in the use of prose,rcomai, which occurs 52 times in Matthew, but only five times in Mark and 10 times in Luke.  In the Pentateuch, προσέρχομαιbears unmistakable cultic connotations, indicating an approach to God (or a representative of God in priest, altar, tabernacle, etc).  In Matthew, Jesusis overwhelmingly made the object of this verb, thus implicitly transferring divine status to him.[7]  Luke, in particular, knew the LXX too well to have missed the significance of Matthew’s achievement with προσέρχομαι)  It seems inexplicable why Luke, who otherwise showcases Hebraisms in his gospel, would omit such a term if he were drawing on canonical Matthew as one of his written sources.              

The above evidence of Matthean posteriority can be illustrated and augmented by considering, briefly, a number of specific Matthean texts that are best explained as developments of earlier Lucan texts. [p. 249]

• Luke’s first beatitude in 6:20, “Blessed are the poor,” is expanded into a spiritual virtue in Matthew 5:3, “Blessed are the poor in spirit.”
• Jesus’ teaching about turning the other cheek in Luke 6:29 is further specified as the right cheek in Matthew 5:39.
• The metaphor in Luke 6:43 that “a good tree cannot make bad fruit” is developed into a moral maxim (that compromises the metaphor) in Matthew 12:33, “Do the good tree and its fruit . . . .”
• In the story of the father who brings his demon-possessed boy to Jesus for healing, Luke 9:39 ascribes the possession primitively to “a spirit,” whereas Matthew 17:15 provides a rational explanation as “epilepsy.”
• Luke’s  metaphor of Jesus casting out demons by the “finger of God” (11:20) is changed in Matthew 12:28 into the less Hebraic “spirit of God.”
• Jesus’ lament over Jerusalem, which in Luke 13:34-35 is uttered five chapters before he arrives in Jerusalem, is relocated more reasonably in Matthew 23:37-39 to Jerusalem itself.
• The same text clearly signals a Lucan origin of the saying, for Matthew’s spelling of Jerusalem follows the Lucan spelling of the city ( Ἰερουσαλήμ)rather than the normal Matthean spelling ( Ἰεροσόλυμα).  
• Luke omits the controversy with the scribes and Pharisees over clean and unclean in Mark 7:11, whereas Matthew Hellenizes Mark’s “Corban,” leaving the simple Greek “gift” (δῶρον)in its place.
• In Mark 10:17-18 and Luke 18:18-19 the rich man addresses Jesus as “Good teacher,” to which Jesus responds, “Why do you call me good?” Matthew 19:16-17 alleviates the offense of Jesus questioning his own goodness by changing the rich man’s question to, “What good thing must I do to inherit eternal life?” 
• The perplexing parable of Luke 19:11-27, seemingly a composite of two parables (a parable of a king and his rebellious subjects, and another of pounds), has been stripped of the first parable and groomed into a coherent Parable of the Pounds in Matthew 25:14-30.
• In the same parable, Luke 19:13 speaks of ten servants who were summoned, only three of whom were given money; Matthew 25:14 eliminates the superfluous servants and regulates the story to three servants and three loans.
• Luke 23:33 inexplicably omits “Golgotha” from the crucifixion account, whereas Matthew 27:33 restores “Golgotha” on the basis of Mark 15:22. [p. 250]
• Matthew 28:9-15 is unique among the gospels in offering an explanation for the Jewish failure to respond to the proclamation of the gospel.  That explanation indicates a period late in the first century when the breach between church and synagogue appeared to be final, and such an explanation was needed.

These examples—and more could be added—can be satisfactorily explained on the assumption that the Matthean form of the material is a later development of a more primitive Lucan form.  The examples can be less satisfactorily explained, however, on the assumption that Luke has reworked earlier Matthean material.[8]  The comparative data point to the more primitive nature of the Lucan material, which was later developed by Matthew.[9]

This conclusion appears to be corroborated in a study of textual transmission, in which Robert McIver and Marie Carroll argue that “[a]ny sequence of exactly the same 16 or more words that is not an aphorism, poetry, or words to a song is almost certain to have been copied from a written document.”[10]  McIver and Carroll note 46 passages, each consisting of 16 or more words, that two of the Synoptic gospels have in common.  In no case do all three Synoptics share a text of 16 or more words in common.  Luke and Mark have three passages in common,[11]and Luke and Matthew have 11 passages in common.[12]  In addition to the three passages that Mark has in common with Luke, Mark has nine passages in common with Matthew.[13]  Of the 46 passages, 14 appear in Luke, 12 appear in Mark, and 20 [p. 251]appear in Matthew. Of a total of 23 pairs shared by Luke and Mark, or Luke and Matthew, or Mark and Matthew, Matthew shares 20of them. Matthew, in other words, is almost twice as likely to share material in common with Mark, or in common with Luke, as either Luke or Mark is to share material in common with the other. This statistic suggests that Matthew is either the sourceof material for Mark and Luke; or, conversely, that Matthew is the recipientof material from either Mark or Luke.  The theory of Markan priority supplies solid and repeated evidence for the assumption that Matthew has been the recipient of Markan material.  The foregoing arguments, and especially the list of examples on pages 7-8 above, argue the same, that Matthew is the recipient of Lucan material.  The work of McIver and Carroll seems to supply further statistical evidence that Matthew concludes and consummates the Synoptic tradition.

Already in 1838, Christian Gottlob Wilke argued that the similarities between Luke and Matthew could best be explained by Matthew’s use of Luke.[14]  The Gospel of Matthew appears to represent positions on women, sinners, tax-collectors, lawless people, wealth, and possessions that reflect a time period later than the Gospel of Luke.[15]  Detailed literary analyses of the relationship between Matthew and Luke are offered by Ronald Huggins and George Blair.  Huggins argues that the problem of the relationship between Matthew and Luke is reasonably and satisfactorily resolved on the supposition that Matthew used Mark as his main source, which he supplemented by the use of Luke.[16]  In a thorough and minute comparison of Matthew and Luke, Blair sets forth the similar thesis that Mark was the first gospel, Luke’s a revision of Mark’s, with additions, and that Matthew was a revision of both.[17]

[p. 252] The posteriority of Matthew can also be advocated on historical grounds.  Martin Hengel argues that Luke stands closer to the catastrophic fall of Jerusalem in 70 than does Matthew, and that he displays better knowledge of Judaism prior to it.  Hengel further argues that the divide between the Christian movement and the Roman state is less critical in Luke than in later Matthew, and that Matthew presupposes the consolidation of Judaism under later rabbinic programs.  Hengel’s arguments that the gospels named for apostles are laterthan those named for non-apostles are also well-known.[18]  

From both a global and detailed perspective, the Gospel of Matthew looks like the terminus of a long process of kerygmatic incubation in the early church.  If Matthew were prior to the other two Synoptics, it would be difficult to conceive why its symmetry, practical design, topical organization, and structural felicity would be dismembered and parceled into more pedestrian roles in both Mark and Luke.[19] According to virtually every standard of literary creativity and development, the design and content of canonical Matthew suggest a later provenance in the Synoptic birth order.  

------------------------------

As a supplement to the above, James Edwards adds the following:

Matt 26:55 uses the expression καθ᾽ ἡμεραν only here, but this expression is frequently used in Luke and is considered quintessentially Lukan.  The sole appearance of καθ᾽ ἡμεραν in Matt. seems explainable on the basis of its frequent usage in Luke.  Again, Matt 27:8, the temporal delimiter "still to this day" suggests a long time after Jesus' death.  This phrase assumes the writing of the Gospel at a significantly later date.
​
Finally, in my article "The Hermeneutical Significance of Chapter Divisions in Ancient Gospel Manuscripts" NTS 56/3 (2010) 423, I note that the precision of Matthew's Passion Narrative produces a symmetry in the Passion Narrative that rivals the otherwise symmetry in his Gospel as a whole.  No other Gospel achieves the symmetry that Matthew does in notes #63-68 in the Eusebian Canons (= Matt 26:17-28:20).  I attach a typescript of this art., pp. 18-20 discuss Matt's Passion Narrative.
​
-------------------------------
Footnotes to The Hebrew Gospel and the Development of the Gospel Tradition (Eerdmans, 2009)

[1]G. A. Blair, The Synoptic Gospels Compared,  311:“Matthew was called the ‘first’ Gospel . . . because [it] was the most important of the Gospels, the ‘definitive Gospel’ (at least with respect to the three Synoptics).”

[2]Division I: narrative, 3:1-4:25; teaching, 5:1-7:27; summary refrain, 7:27; 
  Division II: narrative, 7:28-9:38; teaching, 10:1-42; summary refrain, 11:1; 
  Division III: narrative, 11:2-12:50; teaching, 13:1-51, summary refrain, 13:53; 
  Division I V: narrative, 13:54-17:27; teaching, 18:1-35; summary refrain, 19:1; 
  Division V: narrative, 19:2-22:46; teaching, 23:1-25:46; summary refrain, 26:1.

[3]M. Hengel, The Four Gospels and the One Gospel of Jesus Christ, 176-177, summarizes the above situation as follows: “[Matthew’s] five great discourse complexes are predominantly fashioned by this [marked theological concern].  Luke has little to set over against this overarching skill in composition. . . . Luke’s ‘Sermon on the Plain’ is itself a mere ‘shadow’ of the Sermon on the Mount.  Therefore, it is utterly improbable that, for example, Luke reshaped a Matthean original. He would not have torn apart discourses which have been worked out so masterfully, but integrated them into his work.  One could make a Sermon on the Mount out of a ‘Sermon on the Plain,’ but not vice versa.  Therefore, Luke cannot be dependent on Matthew, as is consistently asserted.”  Similarly, H. P. West, Jr., “A Primitive Version of Luke in the Composition of Matthew,” 91: “it is incredible to think that the Sermon on the Plain comes from the Sermon on the Mount.  To argue that Luke has so abridged Matthew’s sermon is to accuse him of using Matthew in a way utterly different from his use of Mark.  We remember that Luke nowhere selectively abridges a Marcan block in this way.”

[4]On Luke’s preservation of a more primitive form of the Lord’s Prayer, see S. Hultgren, Narrative Elements in the Double Tradition, 290-309.

[5]Str-B 4.75; J. Jeremias, The Eucharistic Words of Jesus, trans. N. Perrin (New York: Scribners, 1966) 97-100; 164; L. Goppelt, poth,rion( TDNT6.153-154.

[6]See J. Edwards, “Markan Sandwiches: The Significance of Interpolations in Markan Narratives,”NovT31/3 (1989) 193-216.  Reprinted in The Composition of Mark’s Gospel(Brill’s Readers in Biblical Studies 3, compiled by D. Orton; Leiden: Brill, 1999) 192-215.

[7]See J. R. Edwards, “The Use of προσέρχεσθαιin the Gospel of Matthew,” JBL 106/1 (1987) 65-74.

[8]See the discussion on the very considerable difficulties in assuming that Luke followed Matthew in S. Hultgren, Narrative Elements in the Double Tradition, 329-335.

[9]So M. Hengel, The Four Gospels and the One Gospel of Jesus Christ, 69-70: “ . . . as a rule the more original version is attributed to Q-Luke as opposed to Q-Matthew.”

[10]“Experiments to Develop Criteria for Determining the Existence of Written Sources, and Their Potential Implications for the Synoptic Problem,” JBL 121/4 (2002) 680 (italics in original).

[11]Luke 18:15-17//Mark 10:13-16; Luke 4:31-37//Mark 1:21-28; Luke 20:45-47//Mark 12:38-40.

[12]Luke 10:21-24//Matt 11:25-30; Luke 12:41-48//Matt 24:45-51; Luke 16:10-13//Matt 6:24; Luke 3:1-20//Matt 3:1-12; Luke 11:9-13//Matt 7:7-12; Luke 9:57-62//Matt 8:18-22; Luke 11:29-32//Matt 12:38-42; Luke 7:18-35//Matt 11:1-19; Luke 7:1-10//Matt 8:5-13; Luke 5:12-16//Matt 8:1-4; Luke 9:21-27//Matt 16:21-28.

[13]Mark 13:3-13//Matt 10:16-25; Mark 8:31-9:1//Matt 16:21-28; Mark 13:14-23//Matt 24:15-28; Mark 14:12-21//Matt 26:17-25; Mark 7:1-23//Matt15:1-20; Mark 12:35-37//Matt 22:41-46; Mark 10:35-45//Matt 20:20-28; Mark 8:1-10//Matt 15:32-39; Mark 13:24-31//Matt 24:29-35.  For a chart of all 46 passages and their relationships among the Synoptics, see Ibid, 681.

[14]C. G. Wilke, Der Urevangelist oder kritische Untersuchung über das Verwandtschaftsverhältnis der drei ersten Evangelien (Dresden und Leipzig: Gerhard Fleischer Verlag, 1838) 685-693.  On pages 460-462 Wilke specifically argues this case for Matt 3:1-12; 10:1-14; 12:9-14; 12:22-32; and 13:1-35.

[15]H. Ph. West, Jr., “A Primitive Version of Luke in the Composition of Matthew,” 80-88.

[16]R. Huggins, “Matthean Posteriority: A Preliminary Proposal.” Likewise, G. D. Kilpatrick, The Origins of the Gospel According to St. Matthew, (Oxford: University Press, 1946) 7, 140.

[17]G. A. Blair, The Synoptic Gospels Compared, esp. 307-312.

[18]See The Four Gospels and the One Gospel of Jesus Christ,169-179; 254-255; and 303-305 (notes 663-670).  In a personal letter to me dated 24 May 1998, Hengel succinctly summarizes the same: “Dass Lk früher ist als Mt ergibt sich 1. aus seiner grösseren Nähe zur Katastrophe von 70, 2. aus seiner besseren Kenntnis der Verhältnisse vor 70, 3. aus der Tatsache, dass die den Apostelschülern zugeschriebenen Evangelien älter sind als die angeblich von Aposteln verfassten, 4. dass Mt schon die Konsolidierung des Judentums unter rabbinisher Führung voraussetzt und 5. dass Lk noch denkt, dass ein Kompromiss mit dem römischen Staat möglich ist, d.h. er gehört in die frühere Flavierzeit.”

[19]S. Hultgren, Narrative Elements in the Double Tradition, 333: “In general it has been thought more likely that Matthew grouped disparate sayings material into a few great speeches than that Luke broke up Matthew’s well-constructed speeches and scattered the sayings throughout his gospel.  This argument is compelling, and it makes it that much less likely that Luke is dependent on Matthew in the double tradition.”

"My attitude to the Synoptic Problem was influenced, during my undergraduate studies, by RT France. He was highly sceptical of Q and also generally steered clear of favouring any one hypothesis. By the time I finished my BA, therefore, I was definitely a Q skeptic - although one that generally leaned towards the Farrer Hypothesis.
 
A major change occurred in 1998 during my doctoral studies (supervised by the 2DH advocate Christopher Tuckett). While working on the Didache's relationship to Matthew's Gospel I came across the curious triangular relationship between Did. 1.2-5a, Matt 5.38-48 and Luke 6.27-37:

These Didache sayings are remarkable because there are very good reasons for suspecting that Did. 1.3-5a was added to a version of the text that already included Did. 1.2. This is important because, in Luke 6.27-36, elements similar to both Did. 1.2 and 1.3-5a are combined into a seamless, unified whole. This caused me to suspect that Luke's set of sayings had been woven out of the cloth provided by the whole of Didache 1.2-5a (rather than the reverse as was more commonly assumed). Furthermore, I then noticed a very peculiar pattern of similarity and difference between all the three versions: wherever Matthew's version is dissimilar to Luke's, his version is similar to the Didache's. The only reasonable way to explain this, it seemed to me, was to allow that that Matthew had attempted to conflate Luke's version with the sayings in the Didache. (A fuller presentation of this proposal is offered here).

Contemplating this possibility reminded me of a comment made by Michael Goulder in, 'Is Q a Juggernaut' (JBL, 1996): '[I]n a scientific subject a paradigm shift is possible because new and irrefutable evidence may come to light, new evidence in arts subjects is rare, and so are paradigm shifts' (p. 669). Here, it seemed to me, was a tiny piece of new evidence. It was tiny, but it was enough to shake me out of my prior assumptions. Why could not Matthew have used Luke? And why, if Matthew used Luke, could they also not have had access to sources they both shared (in addition to Mark)? 

Once I started to look at the Synoptic relationships with these questions in mind I came increasingly to suspect that Matthew was an author who had sought to harmonise and conflate together a range of earlier traditions, including Mark, Luke, portions of the Didache, and probably several other written and oral sources. This line of approach, it seemed to me, preserved all the strengths of the Farrer and Two Document Hypotheses without reproducing their weaknesses. 

Sometimes friends suggest that my arguments for Matthew's use of Luke would be more appealing if I kept it simpler - i.e. stopped including Did. 1.2-5a. I understand what they mean, which is why I tried to keep the Didache to one side in my initial presentation of the Matthew Conflator Hypothesis. For me, however, Did. 1.2-5a is the jewel of great price that has lain abandoned in a field since its discovery in 1873. New historical evidence of this quality, as Michael Goulder rightly observed, hardly ever comes to light. It seems a shame, therefore, to refuse to allow it to shake our old assumptions. But, of course, such a suggestion is stupidly naive. If we allowed our old assumptions to be disturbed, we would face the massive upheaval of having to wrestle with something disturbingly new."

Erik answered my question as follows:
​ 
"I have read my NTS-article from 2001 again and now I remember.

What put me on the track was the astonishing correspondence between Lk 4:25–27 about the widow in Zarephath in Sidon with a sick child and the pagan officer Naaman on the one hand and on the other
 hand Mt 8:5–13 about the pagan officer and Mt 15:21–28 about the woman in the district of Tyre and Sidon with a sick child. Only Matthew has both these stories and the Mt-redaction has assimilated them to each other, especially through putting in the "great faith" at the end, in 15:28 (not in Mark 7:24–30 – but in Mt 8:10), making these two pagans, one man and one woman, the only two persons whom Jesus praises for their great faith. Matthew obviously wants the stories to be seen as a pair that belongs together, as the two OT counterparts do in Luke 4:25–27. The Mt-redaction also put in "Sidon" at the beginning, in Mt 15:21, not mentioned in Mk 7:24, but in Lk 4:26 – a sign that Mt also wants us to see the connection between the two NT-pagans and their OT counterparts mentioned in Lk 4:25–27. That was and is enough to convince me that Matthew knew the saying Luke 4:25–27. My reasons are elaborated in paragraphs II-IV (pp. 431-436). And my explanation of the fact that Matthew does not render the saying himself, although it seems to be so important to him, is presented in paragraph VI (p. 439f). In short: he did not want to contradict the program as it is expressed in Mt 10:5f (and 15:24): first to Israel. But he also wanted to show how the incalculable reality, i.e. the surprising great faith of two pagans overcame the program.

If this were the only evidence it might, however, also be explained through the assumption that Matthew just knew (the tradition behind) the saying recorded in Luke 4:25–27, not necessarily the whole Gospel of Luke. The tradition could be Syrian, as both examples are connected to Syria – and Mt is often presumed to emanate from Syria (p. 438). But I was more impressed and convinced especially by von Dobschuetz, Hengel and Huggins, and the last years even more by Garrow, that Mt in fact knew the whole of Luke (as well as the Q-tradition[s], p. 439 n. 41)."
 

"During my PhD program at Dallas Seminary in the 2000s, I took a course on the Synoptic Problem taught by Harold Hoehner. Dr. Hoehner was a proponent of the Two Gospel Hypothesis and had been a personal friend of William Farmer, who had also taught in Dallas, at Perkins School of Theology in Southern Methodist University. I began Hoehner’s course favoring the Two Document Hypothesis, but the required readings showed me that there were reasons to question Q. I then started to wonder why it seemed that everyone who disbelieved in Q concluded that Luke must have used Matthew as a source. Why not the reverse? Then I came across the article by Ronald V. Huggins, “Matthean Posteriority: A Preliminary Proposal” (NovT 34 [1992], 1-22). I found it intriguing and persuasive."

Rob went on to publish his PhD dissertation as:
Robert K. MacEwen, Matthean Posteriority: An Exploration of Matthew’s Use of Mark and Luke as a Solution to the Synoptic Problem, LNTS 501; London: Bloomsbury T&T Clark, 2015.

Here is the final paragraph of an RBL review by J Andrew Doole:​

For most of my career I thought the Q hypothesis was the most convincing explanation of the material common to Matthew and Luke, though with the proviso that maybe Q was not a single source. When comparing parallel passages in these two Gospels in detail, I usually found that redactional changes to a common source by the respective evangelists explained satisfactorily their similarities and differences. What changed my mind about Q was the problem of envisaging what the source as a whole could have been like. Q is often said to have been a sayings collection, but in fact the Q material begins with a substantial amount of narrative and then seems to abandon narrative form in favour of sayings alone, sometimes with a brief narrative setting but often not. As a piece of literature, this is incoherent. I doubt if this problem can be solved by dividing Q into more than one source, though I would be open to a sustained argument along those lines if anyone were to develop one.

But the impression one gets from virtually all studies of the Synoptic problem is that the only viable alternative to Q is the Farrer hypothesis (Luke’s use of Matthew). I have never found that hypothesis credible. The major problem I see is the way that Luke would have had to rearrange the material he shares with Matthew. The common objection along these lines has been: Whywould Luke break up Matthew’s well structured and coherent discourses and scatter the material around his Gospel? Advocates of the Farrer hypothesis have devoted their defences to answering that question. But my problem is with the how, not the why. The process by which Luke would have had to select and arrange the material he takes from Matthew (which would have involved very carefully avoiding any Markan material in the context of this Matthean material) would have been extremely complex. Alan Garrow has shown this very convincingly, and it can only be appreciated by people who are willing to sit down with a synopsis and examine the parallel texts in detail. Francis Watson in his book Gospel Writing has now made the best available attempt to explain and account for Luke’s compositional procedure. I doubt if anyone could do better, but I ended reading it with the conclusion: This is unbelievably complex. It is also completely unparalleled in the way that other ancient authors worked with their sources. 

So I hung onto Q for longer than I would otherwise have done because I still found it more credible than Luke’s use of Matthew. But occasionally the question occurred to me: why has hardly anyone even attempted to explore the possibility that Matthew used Luke? It seemed like a solution that had never been tried, rather than one that been tried and found wanting (as one might say of the Griesbach hypothesis). I discovered that it had been proposed (by Christian Gottlob Wilke), back in the nineteenth century just when the Q hypothesis originated and became popular. At that stage most scholars adopted the latter, and the “Wilke hypothesis” (as we might have called it had anyone pursued it then) was forgotten. When eventually an alternative to Q again appeared on the scholarly horizon, it was the Farrer hypothesis.

To me Matthew’s use of Luke had the obvious advantage of avoiding what I found incredible in the Farrer hypothesis. Though Matthew’s compositional procedures, if he took the “Q” material from Luke, would involve often conflating Luke with Mark (in the way that advocates of Q envisage him conflating Q with Mark), his procedures would be much less complex and difficult than those of Luke according to the Farrer hypothesis. A further consideration is that Q scholars, when judging whether Matthew or Luke preserved the more “primitive” form of a saying of Jesus, tend in a large majority of cases to decide for Luke. While the criteria for making these judgments are often not very secure, the overall impression that Luke has the earlier form of sayings he shares with Matthew should surely make it worth considering the hypothesis that Matthew took the “double tradition” material from Luke, rather than vice versa. (Since B. H. Streeter, the case for Q has standardly included the “alternating primitivity” of the “double tradition” material. In other words, sometimes Matthew, sometimes Luke preserves the more primitive form. This assertion seems to have obscured the fact that actually most scholars working with the Q hypothesis judge Luke to be more primitive far more often than they judge Matthew to be.) 

These considerations mean that, if one begins from the position of thinking that the Q hypothesis works rather well in many respects, as I do, but find it in the end unsatisfactory, as I do, then the more obvious of the other two principal possibilities must be Matthew’s use of Luke, not Luke’s use of Matthew. Many of the really good arguments for Q can also support Matthew’s use of Luke. 
Encouraged by the work of Robert MacEwen and Alan Garrow, who have devoted far more time and effort to exploring the “Matthean Posteriority” hypothesis (as MacEwen calls it) than I can spare, I decided I had sufficient reason to adopt it as a working hypothesis. The reason I am 99% convinced of Markan priority is that I have worked with it as a plausible hypothesis throughout my career and have always found it made good sense of the data. I have now made a start to working with Matthew’s use of Luke as a working hypothesis and so far it seems to me to doing well. 
 
The main reason the Synoptic Problem is difficult is that it involves working on both macro- and micro-levels. In other words, a hypothesis must give a plausible account of what an evangelist is doing with his sources overall and why he should be doing it, but it must also meet the test of explaining the precise similarities and differences between the parallel texts of each Gospel in each pericope. Probably this is a reason why so few people have so far taken “the road less travelled” – Matthew’s use of Luke.