A closer investigation, however, reveals that the nativity story occurs in just two of the four gospels, Matthew and Luke, and that they hardly agree on any of the details. It’s almost as if they are describing two entirely different people being born….
Matthew says that news of Mary’s pregnancy was announced only to Joseph, by an angel in a dream, and only after Jesus had been conceived. Luke, on the other hand, says that Gabriel appeared to Mary while she was awake and explained everything to her before Jesus was conceived. Neither angel cautions silence, so it’s unlikely that one wouldn’t have told the other — and so they can’t both be true.
What I want to do in this post is argue not only that the annunciation accounts are consistent, but also that they're consistent in a way that poses problems for other skeptical claims about the infancy narratives.
Let's consider some examples of how Matthew and Luke might contradict each other in this context:
- The number of annunciations. Was there only one? If more than one, how many?
- The timing of an annunciation. Did it occur ten years before the birth? A year beforehand? Six months? A day?
- Geography. Where did it occur?
- If there were multiple annunciations, who received one first? Joseph? Mary? Did they both receive one at the same time?
- How did they react to it? If Mary received an annunciation, did she tell other people about it just afterward? Or a month later? A year later? Who did she tell?
These are just a few examples illustrating that there's a lot of potential for contradiction. If Matthew, Luke, and/or their sources were independently making up stories, were careless, didn't have much concern for historical accuracy, etc., we'd expect some degree of contradiction in a context like this one. So, if Matthew and Luke are consistent instead, that's significant accordingly.
If there are inconsistencies, we could still accept other material in the accounts as historical, depending on the circumstances. For example, scholars believe that Josephus is inconsistent on some matters, including in what he reports about events he was supposed to have witnessed, but those inconsistencies don't prevent scholars from accepting a large percentage of what Josephus tells us.
And we have to allow some room for ambiguity. Authors, whether they're writing in a historical genre or not, have to be selective in what they include in a document. No two authors are going to have all of the same priorities, information, audiences, etc. Matthew and Luke don't have to report all of the same details in order to be consistent.
I'll start with Luke, since his account begins in an earlier timeframe. When Gabriel comes to Mary, he refers to a future pregnancy, and Mary responds as if she wasn't expecting it. Her question in Luke 1:34 suggests that Gabriel implied a pregnancy that would occur in the near future, and 1:39-45 makes the most sense if Jesus had already been conceived at the time of those verses. (The earliest extrabiblical sources who commented on the subject thought Mary became pregnant just after Gabriel arrived, since they refer to the pregnancies of Mary and Elizabeth overlapping. See here.) Mary is a virgin at the time, the pregnancy is premarital, it occurred through the Holy Spirit, and Mary was given a name for the child by Divine revelation rather than the naming being left up to her and Joseph. Luke tells us that Mary was in a hurry to go to Elizabeth (1:39) and that she stayed with Elizabeth for about three months (1:56). The best explanation I'm aware of for why Mary would be in a hurry and would stay with Elizabeth for so long is that she was concerned about how people would react to her pregnancy. She knew that Elizabeth was also pregnant by Divine intervention (1:36). So, Mary went to somebody she could expect to believe her and to be sympathetic to her situation. The fact that Joseph didn't resolve the matter earlier, so that Mary wouldn't have to stay with Elizabeth for so long, suggests that he hadn't received anything like the supernatural revelation given to Mary. And something happened after three months, so that Mary was willing to return home at that point, but Luke doesn't tell us what it was.
I don't see anything inconsistent with those circumstances in Matthew. To the contrary, the two accounts fit together well. The most substantial charge of contradiction I've come across is the claim that Matthew's geography contradicts Luke's, but I've explained elsewhere why that's a bad argument. Matthew tells us that Mary was "found to be pregnant" (1:18), which suggests that the pregnancy was discovered in a roundabout way. Initially, Mary didn't communicate to Joseph or other relevant individuals (directly or through an intermediary) the fact that she was pregnant. That scenario lines up with what Luke tells us about Mary's circumstances. Matthew 1:18-20 suggests that Joseph didn't find out about the pregnancy until after it occurred, which is at least consistent with Luke. More likely, for reasons I've described above, Matthew isn't just consistent with Luke on this point, but is even corroborated by him. Matthew also seems to agree with Luke that Mary didn't expect the pregnancy well ahead of time. If she had been expecting the pregnancy long before it occurred, it would have been in her interest to have informed Joseph about what to expect. Predicting the pregnancy ahead of time would have given credibility to her claim to have received a Divine revelation. Perhaps the annunciation to Joseph occurred about two to three months after the one to Mary, so that Joseph took initiative to resolve the situation (Matthew 1:24) at that point, leading Mary to return home (Luke 1:56). Both gospels present a scenario in which an annunciation to the other spouse would make sense, even though each gospel only narrates one annunciation. (And since Matthew's entire infancy narrative gives more attention to Joseph than Mary, while Luke's does the opposite, it makes sense that each chose to focus on the annunciation to the spouse to whom he gave more attention.) Matthew also agrees with Luke that Mary is a virgin at the time of the pregnancy, that the pregnancy is premarital, that it occurred through the Holy Spirit, and that a name was given for the child by Divine revelation rather than the naming being left up to the parents.
The annunciation accounts in the two gospels dovetail. They're consistent, they overlap a lot, and each provides details that make the other account more coherent.
Another factor needs to be taken into account with regard to the manner in which the accounts are consistent. Critics frequently tell us that Matthew, Luke, and/or their sources were making up stories in an attempt to parallel Jesus' life to what we see in the Old Testament and other ancient Jewish sources. Since we're discussing annunciation accounts, the most relevant Old Testament figure is Abraham. His son, Isaac, provides the paradigm instance of a miraculous conception in the Old Testament era. And Abraham is a prominent figure in both of the gospels we're discussing. That's true in both infancy narratives, not just the other portions of those gospels. Matthew starts his narrative with a reference to Abraham (1:1) and begins his genealogy with Abraham rather than Adam (1:2). Luke repeatedly mentions Abraham by name in his infancy material (1:55, 1:73, 3:34). The first of those three passages I just cited comes from Mary's comments shortly after the annunciation. So, Abraham is a major figure in the Old Testament, in both gospels, in both infancy narratives, and in close proximity to both annunciations. If critics are right about how much Matthew and Luke's accounts were derived from the Old Testament and related sources, then who would provide a better model for an annunciation than Abraham? Yet, Matthew and Luke not only depart from Abraham's example in some ways, but also depart in the same manner. Abraham found out about Sarah's upcoming pregnancy before Sarah did, and they both were told about the pregnancy long before it occurred (Genesis 12:1-4, 15:4-5, 17:16-21, 18:10-15). By contrast, Matthew and Luke both have Mary finding out about the pregnancy before Joseph does, and both gospels suggest that neither Joseph nor Mary knew about the pregnancy well ahead of time. And there are other differences. There is no infertility because of the advanced age of Joseph and Mary, there is no virgin birth in the Genesis account, etc. Some aspects of what Matthew and Luke agree about not only aren't found in the material about Abraham, but aren't found in the accounts of any other Old Testament figure either.
Think about the significance of Matthew and Luke's agreements. If they and/or their sources were independently making up narratives, were as unconcerned about historical accuracy as critics often suggest, and so forth, they probably wouldn't have come up with scenarios that are so consistent and so unlikely to have been fabricated. How do two sources independently making up stories just happen to hit upon the same timeframe (around the time of the death of Herod the Great), the same names for the parents, a virgin birth, the Holy Spirit's role in bringing about the pregnancy, the same scandalous context (a premarital pregnancy), the child's name being determined by Divine revelation (as opposed to the frequent Old Testament practice of the name being chosen by the parents), neither parent expecting the pregnancy up until just before it happened, Mary finding out about the pregnancy before Joseph does, etc.?
You could agree with much or all that I've said here, yet still reject the historicity of the annunciation accounts. You could acknowledge that the accounts are largely or entirely consistent, acknowledge that there isn't much that can be explained by Old Testament parallels, and acknowledge that there weren't two or more sources independently making up stories, yet maintain that Matthew and Luke were drawing from an earlier tradition that's inaccurate.
But there's a price to pay for adopting that view. You're acknowledging that the accounts are largely or entirely consistent, which undermines the skeptical position I was originally addressing. And you're pushing the Christian claims into an earlier timeframe. The tradition Matthew and Luke would be drawing from would be an earlier source, and it would be one so widely established that two gospel authors accepted it and expected their audiences to do so (which they did, early and across the Christian world). That earlier tradition would be so different from Old Testament annunciations, most significantly the paradigm case of Abraham, that you would have to reject or substantially weaken any objection based on Old Testament parallels. In other words, you would be giving up or significantly weakening some of the most foundational objections to the infancy narratives in general and the annunciation passages in particular (inconsistencies between Matthew and Luke, the lateness of the accounts, Old Testament parallels). And you'd still have to explain why you reject the earlier tradition behind the gospel accounts. That would require that you interact with the arguments for their historicity: arguments for the Divine inspiration of those gospels, evidence for their general historical trustworthiness, aspects of the annunciation passages that are unlikely to have been fabricated, corroboration from other sources, etc.