McGrath goes on to explain why further argument is necessary:
Historians are interested in literature that doesn't actually relate historical stories. A historian of early Christianity will be interested in the Infancy Gospel of Thomas, or the Acts of Paul and Thecla, even if they are persuaded that they do not contain any genuine historical information about Jesus or Paul or Thecla. Because when we situate them in their historical context, they tell us about the time in which they were written, the views and beliefs that Christians had in that period. And that contributes to our understanding of history.
If the Gospels were pure fabrication, I would still want to understand when they were written, and what sorts of communities produced them. Historians don't just look at texts to get information from the details of the story about the time in which the story is set. Even when historians feel there is little or no valuable historical information in a story, they still study it carefully to learn about the author (even if we don't know his or her name) and the time in which it was written.
I will admit that I do not hold a doctoral degree in either history or New Testament studies, but it seems to me that if the evidence is inconclusive, there should be no shame in saying that the evidence is inconclusive. In fact, no matter how much one wants to resolve a question, intellectual and academic integrity might demand agnosticism.
Rick Sumner at The Dilettante Exegete made what I thought were some excellent points about the unwarranted surety that New Testament scholars are inclined to express:
Even if we agreed there was an historical Romulus, for example, if you started telling me a specific act he did in the founding of Rome, with details about his consciousness while he did it, I'd laugh in your face. It seems preposterous everywhere but here. And even if we let you get away with it, your statement would be a lot more provisional. If someone replied that they doubt Romulus even existed, you would doubtlessly allow for the possibility, and acknowledge that your later conjectures were based on an earlier conjecture: That Romulus was real.It has always struck me the way that Christian apologists (and I do not put McGrath in that category) claim to be applying the same techniques as other scholars who study ancient sources, and yet think they can know details about what Jesus said and thought and did at precise moments with much greater specificity than any historian would ever claim about Julius Caesar or any other figure from the ancient world.
When we deal strictly with textual evidence elsewhere, we recognize the limits of our conjectures. But here we have none . . . . Not only do we not have to be provisional, we can suggest that all the gospels fundamentally misunderstood Jesus' message, and we know even better than they do what it was. While many critics would disagree with those sorts of efforts, we don't deprive them of dialogue, while virtually anywhere else we wouldn't deem to treat such speculation with a grain of seriousness.
It seems to me that expressing warranted agnosticism in an academic inquiry is a positive good. In the field of textual criticism, scholars like Bart Ehrman have noted the futility of talking about the "original manuscripts" of the New Testament given the paucity of early manuscript evidence. They choose instead to talk about the manuscripts that survive and what they can tell us about the communities that produced them and used them. They profess agnosticism about the original texts because they are effectively irretrievable, however this does not stop them from reaching meaningful conclusions where data exists. I don't see how admitting agnosticism on historicity would do anything but enhance the exploration of those further arguments that McGrath thinks are so important.