Updated: Sep 5
The standard answers are not the best Christian response to the question of why the Gospel authors contradict on Jesus' family tree.
Why are the two Synoptic genealogies of Jesus—in Matthew 1 and Luke 3—so wildly contradictory? Although this objection has been raised by Richard Dawkins and other critics of Christianity, many Christian resources on the topic are very poor. A vague answer like “one is the royal line” is not good enough to satisfy a sincere seeker, as it does not explain why Joseph appears to have two different fathers, two different paternal grandfathers, and so on. I intend for this column to serve as a convenient resource that will introduce the problem, review the possible answers to the contradiction, and argue for one particular answer: that there are very good reasons to think Matthew 1 originally contained a genealogy of Mary.
Although I plan to convince you that an interpolation theory is correct, I also want to disabuse you of the other Christian explanations that are sometimes offered for the contradiction, especially the truly ridiculous Africanus-Eusebius theory. Christians have never taken the view that our manuscripts perfectly preserve the original texts in every detail—so it is unnecessary for Christians to resort to bizarre mental gymnastics to defend it.  Finally, I will argue against the theory that Matthew or Luke was lying.
Jesus was a literal blood descendant of David.
Before we contrast the genealogies in Matthew 1 and Luke 3, let’s point out something important that both genealogies share in common—and it’s something that differentiates them from Paul and other writers.
Paul was eager to emphasize that Jesus was a blood descendant of David. For example, Paul wrote to the Romans that Jesus was “descended from David according to the flesh.” Some Christians take other verses out of context to suggest that Paul thought Jesus’ ancestry didn’t matter—and even that we shouldn’t talk about it—but this is clearly incorrect: Paul repeatedly emphasized Jesus’ biological descent from David alongside Jesus’ resurrection from the dead. 
Paul was not alone in emphasizing Jesus’ descent from David.  In Revelation, for instance, an angel refers to Jesus as “the Lion of the tribe of Judah, the Root of David.” In the context of first century Judaism, this statement supports Paul’s assertion that Jesus was a blood descendant of David. 
The formal genealogies in Matthew and Luke, however, could hardly be more different from Paul. Far from emphasizing that Jesus had the "flesh" of David, neither Matthew or Luke seems to have the slightest interest in Jesus’ blood ancestry. In fact, neither Matthew 1 nor Luke 3 appears to tell us anything at all about Jesus’ biological family except that Jesus’ mother was Mary. Both tell us that Mary was Jesus’ mother, and that Joseph was her husband, but then each genealogy only gives us ancestors of Joseph—not Mary—going back at least to David.
How are Matthew and Luke in conflict?
Having noticed this peculiarity, let’s confront the crux of the problem, comparing Joseph’s genealogy in Matthew 1:15-16 to Luke 3:23-24. Matthew states that Joseph was the son of Jacob, the son of Matthan, the son of Eleazor, the son of Eliud. Luke, however, says that Joseph was the son of Heli, the son of Matthat, the son of Levi, the son of Melchi.
Could it be that Joseph was the son of a man named “Jacob Heli,” who sometimes went by one name and sometimes the other? No: Jacob and Heli can’t be the same person, because Jacob is the son of Matthan and Heli is the son of Matthat. Just as Joseph cannot have two fathers, neither can Joseph’s father.
Of course, Matthat and Matthan do sound curiously similar. “Perhaps there’s something to this Jacob Heli theory after all,” you might think. But this increasingly implausible answer only forestalls the problem. Matthan is the son of Eleazor, son of Eliud, while Matthat is the son of Levi, the son of Melchi. Like Jacob and Heli, then, “Matthan” and “Matthat” cannot be the same man.
Now that we can see the contradiction, it’s important that we take a moment to talk about the general concept of New Testament contradictions. Doing so is necessary in order to fully understand what is unusual about this contradiction.
The problem of the genealogies is unlike any other alleged contradiction in the New Testament.
If you haven’t spent much time in the Bible, you may have absorbed the common secular assumption the Bible is full of contradictions just like the one we're discussing here—in which case you might not be surprised that Matthew 1 and Luke 3 are so flatly contradictory. The fact is, though, that there really isn’t anything else—anywhere in the New Testament—quite like this. That makes this problem particularly puzzling.
Let’s demonstrate this fact by comparing our problem to another alleged contradiction: the differing accounts of the death of Judas Iscariot. This is an example that Bart Ehrman—a very sharp, scholarly critic of Christianity—likes to use in order to convince his Christian students that the Bible is unreliable and they should give up their faith. You can rest assured, then, that I haven’t cherry-picked an easy example to make Christianity look good.
In Matthew 27, Judas—consumed by guilt—gives his thirty pieces of silver back to the Pharisees, then hangs himself. Because blood money is regarded as ritually impure, the Pharisees don’t want to put the money into the treasury. Instead, they buy a field with it, which they then use as a burial field. Because the field was bought with blood money, Matthew writes, it is called the “Field of Blood” to this day.
However, in Luke’s account—Acts 1:18-20—Judas himself buys a field with the money that he was paid for betraying Jesus. Eventually, Judas dies in this same field, “falling headlong,” so that his guts gush out on the ground. From then on, Luke writes, this field was called Akeldama, or “Field of Blood.”
Two things about these stories are striking. First, although the stories are different, they share a number of particulars. They agree that:
Judas was paid blood money for betraying Jesus.
A field was purchased, near Jerusalem, with the use of said blood money.
The field was known at the time of writing as the “Field of Blood,” and:
Although other disciples were still alive at the time, Judas had died.
Secondly, the stories are clearly independent of one another. The fact that the stories disagree in some important particulars shows that neither author simply copied his story from the other. Likewise, the authors didn’t collude in order to create a unified story: they didn’t “get their story straight.” Instead, each author seems to have been honestly reporting on the information as he learned of it—and both seem to have had access to what, in jurisprudence, we might call a shared "nucleus of fact." If anything, this means that we can be especially confident that the four facts just listed are historically accurate.
Far from suggesting that the Gospels are essentially unreliable sources about Jesus and the events surrounding him, the Judas discrepancy is just the kind of thing we would expect from honest writers, in proximity to the events being reported on, genuinely recording history as they experienced it, or as it was reported to them. As a lawyer, I have repeatedly seen present-day illustrations of this same principle: if two people's statements are identical in even minor particulars, those statements are generally suspect.
The contradiction between the Synoptic genealogies is different, however: the genealogical contradictions could not plausibly arise from a shared nucleus of fact. If both writers were honestly reporting on the same man’s genealogy, you would not expect to get "Eleazor son of Eliud" in one genealogy and "Levi the son of Melchi" in another. The implication seems clear: if we have the genealogies as Matthew and Luke originally recorded them, then at least one of them was lying. Not speaking allegorically or poetically, but simply inventing false statements of fact in bad faith. If that is the case, then our trust in the reliability of one or both of these authors is going to be seriously undermined.
The Africanus-Eusebius theory: levirate marriage
We’ve now established that there is a contradiction between our manuscripts of Matthew and Luke, and that the contradiction is unique. How then should we approach it? Before arguing for my own answer, I first want to dispose of two competing Christian theories which you might encounter online or in church, respectively.
First, let’s turn to a third-century writer named Africanus—a Roman soldier and Christian thinker who wrote on the genealogies and a range of other topics. In general, Africanus seems to have been an intelligent man. On this issue, unfortunately, he has saddled us with a cockamamie genealogical Twister game that haunts the church to this day. That’s because Africanus’ theory was adopted and quoted by the father of church history, Eusebius.
Africanus’ theory technically reconciles the few generations we’ve just discussed, but it has so many convoluted and implausible layers that you need to contort yourself into a pretzel to accept the whole thing. Even if you do, we’re still left with other contradictions between the genealogies, as well as contradictions between both genealogies and other ancient Jewish sources, like the genealogy in 1 Chronicles. Without miring ourselves in the details of Africanus’ insanity just yet, let’s quickly consider just the key component of that theory: the idea that we have one legal genealogy of Joseph and one biological genealogy of Joseph. Even this narrow proposition makes no sense on its face, as both Matthew’s and Luke’s Gospels are biographies of Jesus, not Joseph.
If Jesus had one biological lineage and one legal lineage, might we want to know both? Sure—and we can certainly see why first century Jews might want that information. But why would we possibly want to know the biological, but non-legal, ancestors of Joseph? Even Jesus’ enemies, by this time, didn’t say that Joseph was his biological father—they said that Mary had been unfaithful.  It follows that, if Joseph had ancestors who weren’t even Jesus’ legal forefathers, those people aren’t anyone to Jesus—not his biological or his legal ancestors.
Why would a writer of a biography of Jesus, like Matthew or Luke, spend a whole chapter giving us a genealogy of people who have nothing to do with Jesus? He wouldn’t. Thus, the Africanus-Eusebius theory makes no sense.
Unfortunately, I feel I must briefly review the details of what Africanus says. Africanus claims that Joseph was the product of a levirate marriage, which is an ancient Jewish custom in which a child becomes the legal heir of his deceased paternal uncle. While this sounds plausible enough by itself, it explains only one of several discrepancies in the genealogies: if your paternal uncle adopted you, after all, this would not change who your paternal grandfather was. Apparently to patch this problem up, Africanus adds in that Joseph's legal father and separate biological father were half-brothers by the same mother.
The mistranslation theory: Luke's genealogy is Marian
Secondly, I have to discard one other theory, which we'll call the mistranslation theory. The idea here is that the genealogy in Luke is actually of Mary. Not the genealogy in some alternate or lost manuscript of Luke, mind you, but the genealogy printed in Luke in your Bible.
How can this be? Luke’s genealogy, after all, doesn’t even mention Mary. Look at Luke 3:23-24; in my Bible—I primarily use an ESV—this verse says that Jesus was “the son (as was supposed) of Joseph, the son of Heli, the son of Matthat” and so on, giving other ancestors of Joseph.
The theory here is that this verse is mistranslated. In the original Greek, the theory goes, this verse does not indicate that Jesus was supposed to be the son of Joseph, Heli’s son. Instead, it indicates that Jesus was supposed to be the son of Joseph, but was really the son of Heli. Therefore, says the mistranslation theorist, “Heli” must refer to Mary’s father—Jesus’ maternal grandfather—and this entire genealogy must be of Mary. If you’ve ever asked about the problem of the Synoptic genealogies in church, the mistranslation theory is likely the answer you were given.
If this is the case, however, why did Luke—a careful historian—not even mention Mary's name in her own genealogy?  Some might suggest that this is because Mary was a woman. This is inconsistent with the fact that Luke is our most Marian Gospel: Mary is the poetic heroine of Luke 1. Luke even ascribes to Mary a beautiful and theologically rich speech, the Magnificat.
More importantly, though, virtually no Christian scholar translates the Greek text to suggest that Jesus’ maternal grandfather was Heli. The only context in which anyone does so, in fact, seems to be when that nervous guy in your Bible study is trying to explain why the genealogies conflict. This, after all, is why your translation of Luke almost certainly says that Heli was the father of Joseph, not of Jesus.
Let’s compare the 28 translations of Luke 3:23 on BibleHub. 25 of those translations state unambiguously that Heli was Joseph’s father. Two translations are ambiguous—the Berean Literal Bible and the New Heart English Bible—but, in context, they still suggest that Joseph was the son of Heli. Only one translation, the so-called Contemporary English Version, actually agrees with that guy from your church—and it’s hard not to think this isn’t the exception that proves the rule. As the CEV has Luke 3:23:
When Jesus began to preach, he was about 30 years old. Everyone thought he was the son of Joseph. But his family went back through Heli...
According to the mistranslation theory, all English translations have gotten this verse wrong except one that, in all likelihood, the mistranslation theorists themselves do not read. As it happens, although the CEV preserves the traditional biblical verse structure, it often takes liberties with the text, engaging in paraphrase in the vein of The Message, as you can see by comparing its rendering of Psalm 127:1 to other translations.
No Christians in the second century believed that there was a contradiction in the genealogies.
I think we’ve disposed of Africanus-Eusebius and the mistranslation theory. As our manuscripts themselves seem unavailing, the next logical step should be to ask how the earliest Christians read the texts. Paul cannot help us here, as he was writing mostly before the Gospels were written down. Instead, let’s turn to the early church fathers—the first generations of Christians after the New Testament. These writers were deeply concerned with who Jesus was, had access to early New Testament manuscripts, and also had access to eyewitness accounts and other information now lost to us.
When we turn to the early fathers for an answer, here’s the first thing we notice: in the second century AD, nobody seems to have noticed that the genealogies contradicted. One New Testament scholar, H.A. Blair, has said “The really surprising thing is that no one in the second century found it surprising, or made any comment on their disagreement.” 
Nor were the second century church fathers uneducated or stupid. On the contrary, the second century AD saw the first generation of Christian intellectuals. These intellectuals included Tertullian, a trained lawyer and a prolific writer who, I think, was of the caliber of Augustine. They also included Clement of Alexandria, an extraordinarily learned man and a philosopher who, moreover, clearly had some interest in Jesus’ genealogy, as he directly cites Matthew’s genealogy of Jesus in his writings.
How did these men not notice that Matthew and Luke had completely different genealogies? I want to suggest the answer to that question is the same as the answer to our overall question about Matthew and Luke.
Let’s start with the writing of Clement of Alexandria, who I just mentioned.  In his work Stromata, written in the second century, Clement of Alexandria included a chapter arguing that the Jewish prophetic tradition is far more ancient than Greco-Roman civilization. In this chapter, Clement mentions Matthew’s genealogy.
Now-lost second-century manuscripts had Matthew giving Jesus' blood ancestry through Mary.
As we quote Clement, note Clement’s use of numerology, as well as the way he mentions Mary. Clement writes:
And in the Gospel according to Matthew, the genealogy which begins with Abraham is continued down to Mary the mother of the Lord. ‘For,’ it is said, ‘from Abraham to David are fourteen generations; and from David to the carrying away into Babylon are fourteen generations; and from the carrying away into Babylon till Christ are likewise other fourteen generations...
First, this numerical pattern is important. Clement here is quoting Matthew 1:17, using the same quotation found in our manuscripts. Matthew says there are 14 generations from Abraham to David, 14 from David to Babylonian exile, and 14 from the Babylonian exile to Christ. In Matthew’s words:
So all the generations from Abraham to David were fourteen generations, and from David to the deportation to Babylon fourteen generations, and from the deportation to Babylon to the Christ fourteen generations.
There is a problem, however. If you open your Bible and count the generations actually listed in Matthew 1, you’ll see there aren’t three sets of 14: the third set is missing a generation, giving us sets of 14, 14, and 13 generations. If our Matthean genealogy is original to Matthew, this means two things: first, Matthew made a mistake and left out a generation. Secondly, Clement and others failed to notice Matthew’s mistake. To a non-Christian especially, this possibility may seem attractive.
However, the larger problem is what Clement says about Mary. Clement says that Matthew’s genealogy “begins with Abraham” and “is continued down to Mary the mother of the Lord.” In our manuscript of Matthew, Mary is mentioned as Joseph’s wife, but it doesn’t continue down “to Mary.” It doesn’t look like she’s a descendant of anyone: instead, it looks as if she’s just stuck in there.
We’ve just mentioned the possibility that Matthew and Clement both made a mistake. Compare that option to another possibility: when Clement was writing, he was looking at a previous version of Matthew 1 that traced Mary’s biological descent from David.
This theory, pioneered by H.A. Blair, has tremendous explanatory scope. It wouldn’t just explain the contradiction between Matthew and Luke—it would also answer three other discrete problems. First, it would explain why nobody seems to have noticed the contradiction between Matthew and Luke in the second century. Secondly, it would explain why our manuscript of Matthew claims to list three sets of 14 but does not do so, and, consequently, why Clement didn’t mention a missing generation in the third set. Thirdly, of course, it would explain why Clement affirmatively described the genealogy as Marian.
Still not convinced just yet? That’s fine, because I’m not done. Let’s look at the writings of another church father—the third century church father Victorinus, who died about 300 AD. In his commentary on Revelation, Victorinus analogized each of the four symbolic “living creatures” in Revelation to one of the four Gospel writers. In Victorinus’ words:
The first living creature was like to a lion, and the second was like to a calf, and the third had a face like to a man… the living creature like to a lion designates Mark, in whom is heard the voice of the lion roaring in the desert. And in the figure of a man, Matthew strives to declare to us the genealogy of Mary, from whom Christ took flesh.
Note here the parallel to Paul’s language in Romans 1: Jesus is descended from David according to the flesh, by way of Mary, from whom he took flesh. Both Clement and Victorinus are reading, in their copies of Matthew, a biological genealogy of Mary.
The theory that Matthew used a biological genealogy also fits nicely with another contextual theme: the fact that the genealogy in Matthew, unlike Luke’s legal genealogy of Joseph, conspicuously highlights several women: Tamar, Rahab, Ruth, and Bathsheba. As H.A. Blair put it, “All these things were, [Matthew] would seem to suggest, the sort of irregularities which should be expected to lead up to the final irregularity of a son whose Davidic descent was traced through his mother.”
If Blair’s theory is correct, when did the interpolation first occur? It must have occurred very early—and certainly no later than around the time Clement was writing—but not so early that there weren’t still Marian manuscripts of Matthew when Victorinus was writing in the third century. This doesn’t answer every question about the genealogies, but it points us in the right direction.
Interpolation in the genealogies does not undermine the reliability of the New Testament's claims.
An interpolation theory is even stronger when we consider the fact that—historically—the genealogies almost seem to have been the part of the New Testament most prone to interpolation. Multiple ancient manuscripts either lack a genealogy where we would expect it—or else contain a different version of our genealogy. 
My favorite example of a copyist messing with a genealogy happened in the High Middle Ages, as described in Bart Ehrman’s book The Text of the New Testament. Luke’s genealogy is supposed to end by saying that Adam was the son of God. But, in the manuscript "Codex 109," a scribe scrambled the genealogy—apparently by lazily copying the names from a table—so that God is inserted into the midst of the genealogy and is listed as the "son of Aram." 
All of this bolsters the probability that Matthew’s genealogy was originally Marian. A genealogy is a complex and precarious thing: slightly tinkering with one is more catastrophic, and perhaps also more tempting, than slightly tinkering with a story or saying.
To conclude our discussion of the interpolation theory, it’s a documented fact that scribes have altered the genealogies, whether to correct perceived mistakes or through error. It’s also highly likely—based especially on the third-party evidence—that what was originally a genealogy of Mary now appears to be a genealogy of Joseph. While I haven’t answered every question about the genealogies in detail, I think I have given the most probable answer to the primary and the general problem.
This all raises a deeper question: if the New Testament was subject to significant and early interpolation, does that pull the rug out from under the essential reliability our text? Absolutely not. Christianity emerged in what G.K. Chesterton called “the full summer of the Roman Empire,” and we have many papyrus and uncial manuscript fragments of the New Testament from the second and third centuries.  Contrary to post-Enlightenment Western assumptions, not everyone who lived more than three centuries before 2019 was an unlettered goatherd in a ramshackle mud hut.
The consensus among scholars, including secular scholars, is therefore that our New Testament mostly reflects its original text. That some interpolation has occurred is to be expected and doesn’t change this fact.
Finally, if you’re averse to Christianity, and have somehow stuck with me so far, you might still have another theory you’d like to raise. A skeptic may want to say that I was on the right track much earlier, when I raised the possibility of a lie. Since, according to many skeptics, Christian writers often lie, we can answer all the issues I’ve raised by simply positing a lot of lying. You could then insist that both of the genealogies were legal genealogies of Joseph, as they now are, and that they differ because Matthew and Luke made them up in bad faith. You would also need to claim, of course, that Clement and Victorinus lied, attributing Matthew’s genealogy to Mary in order to cover up the inconsistency. We’ll call this the fraud theory.
The fraud theory—that Matthew or Luke was lying—
does a poor job of making sense of the data.
Here’s the key problem that the fraud theory can’t explain: when Clement and Victorinus attributed Matthew’s genealogy to Mary, neither of them was talking about the question of contradicting genealogies. Recall that we have no evidence that neither of them even noticed the contradiction.
To illustrate the point, imagine Clement had instead written: “The docetics claim the genealogies contradict, but they are wrong: Matthew’s genealogy is of Mary.” That might cohere with a fraud theory, as it would allow the fraud theorist to claim that Clement was simply covering for Matthew. But that’s not what Clement wrote. Instead, Clement simply happens to mention a genealogy of Mary while arguing for the proposition that Judaism is much older than Western civilization. In the same way, Victorinus mentions the original Marian genealogy in a commentary on Revelation—not while making a point about the genealogy itself. The best explanation for both writers describing the genealogy as Marian is that the Matthean manuscripts they were looking at were genealogies of Mary.
In fact, the fraud theory, which at first might have tempted us, ultimately explains so little—not second century silence on the contradiction, nor Matthew’s numerology, nor testimony about Marian manuscripts, nor the conspicuous women in Matthew—that it has become as much of a Twister game as the Africanus theory that we rejected early on in this discussion.
In summary, although we should be watchful for manuscript interpolation, then, we have no cause to doubt the good faith of Matthew and Luke as authors—and we have no reason to doubt their essential reliability as to Jesus of Nazareth and his life. Ian Huyett is a litigation attorney and a commentator on law, religion, and technology for Staseos and other outlets. You can follow him at @IanHuyett.
 This does not contradict any traditional Christian understanding of inerrancy. See e.g. Chicago Statement on Biblical Inerrancy ("Article X. We affirm that inspiration, strictly speaking, applies only to the autographic text of Scripture, which in the providence of God can be ascertained from available manuscripts with great accuracy. We further affirm that copies and translations of Scripture are the Word of God to the extent that they faithfully represent the original. We deny that any essential element of the Christian faith is affected by the absence of the autographs. We further deny that this absence renders the assertion of biblical inerrancy invalid or irrelevant."); Wayne A. Grudem, Systematic theology: an introduction to biblical doctrine (1994) (“Scripture in the original manuscripts does not affirm anything that is contrary to fact").
 Paul wasn’t one of the disciples, so why is he a good source about Jesus’ genealogy? For one thing, Paul knew some of the disciples, like Peter and John. He also knew James, Jesus' brother. See Galatians 2:9, see also Acts 15. Since Paul had contact with the primary sources, and is our earliest New Testament writer, he’s actually one of our best available sources about Jesus’ ancestry.
 1 Timothy 1:4 is an example of a verse used to allege that we should not talk about Jesus' genealogy. There is no clear reason to think that Paul is talking about Jesus’ genealogy in these verses.
 As Paul wrote to Timothy, Paul suffered for “Jesus Christ, risen from the dead, the offspring of David, as preached in my gospel.”
 People that Jesus encountered sometimes addressed him as “Son of David.” See Matthew 21:9 (“Hosanna to the Son of David! Blessed is he who comes in the name of the Lord!”) Mark 10:48 (“And many rebuked him, telling him to be silent. But he cried out all the more, ‘Son of David, have mercy on me!’”); Luke 18:38 (“And he cried out, ‘Jesus, Son of David, have mercy on me!’”).
 Revelation is attributed to John by the earliest and most reliable extrabiblical Christian writers—men like Ignatius of Antioch and Irenaeus, both of whom personally corresponded with or knew John’s student Polycarp. John, of course, was a disciple of Jesus, and was also charged by Jesus with caring for Jesus' mother Mary, who might have known something or other about Jesus’ genealogy.
 This statement in Revelation is a little bit less clear than Paul—it doesn’t come right out and say “Jesus was a descendant of David according to the flesh.” But, in context, it seems to be referring to the Jewish belief that the messiah would be a blood descendant of David. See, for example, Isaiah chapter 11, which says that a branch from “the roots of Jesse,” who was David’s father, shall bear fruit.  The book True Doctrine, written by the Roman philosopher Celsus, is the first book-length attack on Christianity. According to the church father Origen, it was written about 130 AD, or about 100 years after the events of the Gospels. In the book, Celsus repeats a rumor that Jesus was the illegitimate son of Mary and a Roman soldier named Panthera. The same claim was made hundreds of years later in the Talmud, a holy text of rabbinic Judaism. For example, the Talmud includes stories about a man named Jacob of Kefar Sama, a Christian who the Talmud suggests could perform miraculous healings. At one point the Talmud says “Jacob of Kefar Sama came to heal… in the name of Jesus son of Pantera.” Why do we care at all what the Talmud says? According to Peter Schäfer, the author of Jesus in the Talmud, it is “highly probable that both the Talmud and Celsus draw on common sources (most likely Jewish sources).” These attacks on Mary suggest that she claimed during her lifetime that she was a virgin when Jesus was born. What better explains these sorts of hostile rumors? If Mary claimed all along that Jesus was the biological son of her and her husband, then a rumor that Mary had been unfaithful would’ve been less likely to circulate around the birth of Jesus. A rumor that Mary was unfaithful to Joseph is just the kind of thing we might expect from Mary’s enemies if the New Testament gives us historically accurate accounts of the birth of Jesus.
 On Luke's historical method, see Luke 1:1-4.  Harold A. Blair, Matthew 1,16 and the Matthaean Genealogy, Studia Evangelica II (1964).  Clement of Alexandria is not to be confused with Clement of Rome, a different church father who wrote even earlier.  See e.g. Bruce M. Metzger & Bart D. Ehrman, The Text of the New Testament 268 (2005) ("In Matt. 1.8, Codex Bezae and the Curetonian Syriac insert several additional Old Testament names into Jesus' genealogy.").  Id. at 259.  See e.g. Kurt Aland & Barbara Aland, The Text of the New Testament 57 (1989) (Showing around 50 fragments of the New Testament from before or around 300 AD, with a handful of papyrus fragments from the second century, including P52, depicted above.).