C.S. Lewis, writing about New Testament criticism, says, "[R]eflection on the extreme improbability of his own life--by historical standards--seems to me a profitable exercise for everyone. It encourages a due agnosticism." ("Modern Theology and Biblical Criticism," in Christian Reflections, p. 164)
Lewis was talking about the over-confidence with which NT scholars make up the history of a biblical narrative and the unwarranted conclusions that they draw. His example from his own life was this: Suppose that someone in the future learned that he abandoned Christianity in his teens and also that he had an atheist tutor in his teens who had a great influence on him. It would seem indubitable to such an historian that the tutor influenced Lewis to abandon Christianity and therefore that any earlier texts that seemed to show that he was already an atheist before going to the tutor must be "backwards projections." But, as Lewis points out based on his own knowledge, the conjectural historian would be wrong, as in fact he did abandon Christianity before going to the atheist tutor.
Michael Licona quotes Jonathan Pennington as saying, concerning the stories of Jesus' infancy,
Despite our conflation of all these events at the annual church Christmas pageant, these stories do not in fact overlap at all. If Jesus did not appear as the named figure in both of these accounts, one would never suspect they were stories about the same person.
This sort of consideration appears to be the reason why Licona says repeatedly that he doesn't "know what's going on with the infancy narratives," as if the sheer differences between the narratives create some kind of deep problem for the New Testament scholar. Apparently he considers Pennington's comment to summarize a central part of the alleged problem.
But Pennington's comment should, frankly, be regarded as quite pointless. Such failure of "overlap" (an interesting word) between different statements about a given person's life is entirely common. And concluding that two texts complement one another and tell different parts of the same person's story is not "conflation." (Insert pedantic acknowledgement that, contrary to the portrayal in many Christmas pageants, the wise men didn't come to the stable and the shepherds didn't follow the star. Let's face it: If that were all Pennington were saying, any fundamentalist would say the same, and nobody would be out there agonizing over "what is going on with the infancy narratives.")
Real history is always improbable in its particulars. That all of these different things should have happened to one and the same person can always be made to sound incredible, and one could present different instances from the majority of people's lives that one wouldn't "suspect were stories about the same person" if one were not told. This can even be done with stories about approximately the same period of the same person's life.
Following Lewis's example, I present the following infancy stories:
1) She was born to an unwed mother.
2) Her mother was poor at the time of her birth.
3) The child was born with a misshapen skull, causing doctors to suspect mental disability.
Contrast that with this:
1') She had a father and mother who loved her dearly from infancy.
2') She was a child prodigy and was quoting passages of Scripture before the age of two.
3') Her parents had great aspirations for her and worked hard to send her to college.
These stories do not overlap at all! If one were not told that both of these were about the same named person, one would never suspect such a thing!
Yet they are. Both sets of facts are true of me. The apparent "contradiction" concerning the single mother and the loving mother and father is resolved by the fact that I was placed for adoption at birth and adopted in early infancy. The skull anomaly ended up being a minor matter; the doctors who fretted at my birth were just wrong to think there was anything wrong with my brain.
As Lewis says, reflection upon the extreme improbability of one's own life encourages a due agnosticism when it comes to making pronouncements about the alleged inaccuracy of an historical record of someone else's life. Or even when it comes to thinking that there is something "going on" with narratives other than the statement of plain historical truth.
If we want to know the truth about historical events, including the life of Jesus, we need to have the kind of robust approach that does not create problems where no problem exists. New Testament criticism has become adept at borrowing trouble. Reflection on the extreme improbability of one's own life should be a healthy corrective to that tendency.
[Update]: In fairness to Pennington (though to my mind this reflects rather unfavorably upon Licona's use of Pennington), I should say that I have now been able to get a look at more of Pennington's book in which the above quotation occurs, and he explicitly states that "reasoned harmonization" takes care of the differences between Luke's and Matthew's birth narratives! Pennington's original comment is just setting up an alleged problem that he intends to address later. It is extremely odd that Licona quotes Pennington as he does (with the added emphasis of the phrase "my friend Jonathan Pennington") while apparently rejecting Pennington's unfazed solution to the alleged "problem" and also not making it clear that Pennington, unlike Licona himself, doesn't really think that this is an issue. He's just temporarily setting it up that way on a particular page! I'm glad myself to be able to place Pennington's comment in context and avoid attributing to him the artificial creation of a problem where none exists, but apparently his friend Dr. Licona isn't unduly troubled about any such impression that might be given.