If Jesus lived should we have heard about him from the numerous writers who lived in and around his time? That’s the topic of Chapter 8 Section 2 through 4 in On the Historicity of Jesus (OHJ). I have a dilemma of sorts when it to comes to this subject. On the one hand it bores me to tears for the simple reason that I’ve heard it discussed so many times before. On the other you can’t reasonably form a conclusion without at least considering it. There are the Josephus passages, after all, and numerous silences in authors who we might expect to have written about Jesus. I’ll just have to find some way to make it interesting to myself.

The silences of pagan authors are covered in Chapter 8 Section 3. Carrier documents several people who had every opportunity and reason to write about Jesus but didn’t (or if they did their writings were not preserved). The section documents many types of missing evidence but I have in mind those authors closest to the time Jesus would have lived, who failed to mention him. Ultimately, Carrier decides the silences as a whole favor neither historicity or mythicism, and thus assigns 1/1 as the likelihood ratio.

I’m inclined to think he is being too generous to historicity. To understand why we’ll need to look at what (and how) it is we are estimating when we state a likelihood ratio. (I’ll note that I haven’t fully worked through this intuition I have, so enjoy watching me fumble in the dark a bit here).

When estimating our likelihood we are given two things; the hypothesis under consideration, and our background knowledge. By given I mean, we are to assume they are true during our analysis. You can think of it like this: given the hypothesis, imagine all the possible worlds where it is true. These worlds will have properties (or states if you will) that are entailed by the hypothesis (i.e. they logically must be the case if our hypothesis is true), and properties that are merely allowed by it (because the hypothesis doesn’t specifically disallow them or have any logical effect on them). It is important to note that we are not to limit the space of possible worlds to the evidence we will be considering. In fact that’s the whole point, to discover what percentage of the total space has our evidence in it.

Now one of the excuses often given for all of the silences is that, unlike his portrayal in the gospels, Jesus wasn’t at all famous. However, this a conclusion based on the very evidence Carrier will have us consider here. Though he discusses this possibility in Section 2, he hasn’t included it in the extensive background knowledge chapters, and neither is it part of our minimal historicity hypothesis. Therefore ‘Jesus was famous’ is not ruled out by either, and therefore must be part of the total space of possible worlds under historicity. Remember, the space of possible worlds will include worlds whose evidence is nothing like ours.

What this means is that a (potentially) large fraction of the possible worlds under historicity will contain lots of early written information about Jesus because he was famous. But that is not what our actual world looks like at all. So the only way the silences could be a wash is if there is an identical large fraction of possible worlds under mythicism where we expect to have such references even though they are false. Otherwise we would expect the entire space for mythicism to contain no references. Obviously there is the possibility of false references (I believe the Testimonium Flavianum is one such reference) but is it enough to balance out the references we should expect if Jesus had indeed been famous? I think, perhaps, not.

This obviously has implications for analyzing the passages in Josephus. So I may have found a way to make this interesting for me, after all. In the future I’ll explore how exactly I want model this intuition and how specific evidence and arguments can effect this more general evaluation.

Though I reserve the right to change my estimate in the future, for now I’m going to say that the silences are at least 10% more common on mythicism, which gives me a ratio of 11/10.