Here continues my series on reviews of my book On the Historicity of Jesus. If you know of reviews I haven’t covered, post them in comments (though please also remark on your own estimation of their merits).
In these entries, Covington actually proposes that I under-sell my argument, and that in fact it’s even stronger than I think. He may be right.
Did Demons Kill Jesus?
In part 4, Covington does a good job of explaining why we should concur with several background ideas that my theory rests upon (a more basic, stripped-down Doherty thesis), in this case the theory that Paul probably is referring to demons as the killers of Jesus in 1 Corinthians 2, and that it actually was readily understood in antiquity that things could be planted, built, and buried in outer space (as to them the layered heavens were just other levels of the world, each with their own contents, paralleling everything found on earth).
I need only add a few remarks. When Covington says he’s “no expert in ancient Greek” so “for all [he] know[s] it could be routine for historians to analyze the meanings of Greek words within extremely narrow time-frames [of just fifty years], nonetheless it strikes me as contrived” to do that. Covington is right. It would be extremely bizarre for Classicists to change the definitions of words in half-century time-units. In fact ancient languages were fairly stable, and definitions typically took centuries to shift significantly. Attempting to claim words completely changed their meaning in the course of just one average lifetime is an unusual claim. It requires unusual evidence. It therefore cannot simply be asserted.
Moreover, in respect to 1 Cor. 2:6-8, Gordon Fee’s claim that “‘rulers of this age’ is not used to refer to demons until the second century” is either moot (we have so few documents from the period between Paul and the 2nd century that no claims can be made like this based simply on what extant documents contain) or implausible (I extensively document the evidence of, and scholarship on, multiple different words and phrases for the demonic powers and their use and meaning in OHJ, Element 37). As Covington notes, Paul calls Satan the “god of this age” (2 Cor. 4:4), but I should add the Gospels (even Mark) call Satan the archon, ‘ruler’ (Mk. 3:22). It’s easy to see what people would understand by ‘rulers of this age’…a phrase, incidentally, never used of humans. So Fee’s own methodology, “if we don’t see it, it didn’t happen,” would nix his own argument from the start. Fee is engaging in apologetics, not scholarship.
Certainly, the ambiguity still exists, since the phrase could refer to humans, although I do adduce two strong reasons why in this case that’s unlikely in OHJ, pp. 564-66 and 572-73. But what if we assume it’s still ambiguous and could as easily mean either? Covington concludes with a remarkable point that I actually didn’t consider in my book: since on mythicism the only way Paul could refer to the execution of Jesus is with a phrase that can mean demonic powers, the probability of our seeing such a phrase here is 100%; whereas on historicity, we have no particular reason to expect Paul to have used such a phrase, he could as easily have used one that excludes demonic powers and explicitly identifies earthly powers, and when we have no particular reason to expect either more than the other, the probability of the evidence we then have is 50%. This alone creates a 2:1 factor in favor of mythicism. I did not include that in my math. And it’s not clear how a historicist can escape the logic of doing so.
I had mistakenly assumed that since this phrase can mean either, it is equally expected on both theories (and therefore argues for neither). But I think Covington may be right. If anyone can argue me out of that conclusion, please try. Because it becomes rather damning of historicity if accepted. In OHJ (pp. 563-74; 594) I treat the entire array of “things Jesus did” (including get crucified) as 2:1 in favor of mythicism (3:4 a fortiori). But if this one phrase alone carries that much weight, then my estimates are significantly more generous to historicity than they logically should be.
Covington became less certain after further discussion (see the first section of his part 9 and the comments section of his part 4, although much of that comes from Bernard Muller, who is known to deal in disinformation, false claims, and irrational inferences). But I don’t think the objections raised alter the calculation, because the ambiguity is still not expected on historicity, whereas especially in conjunction with b (our knowledge of the filter), it is expected on myth.
Should the Shift in Who Is Proclaiming the Gospel Count More?
In part 5, Covington proposes:
I’m going to take Carrier to task for missing an important piece of evidence … a piece of evidence that supports his theory. … In the Pauline letters, Jesus is the proclaimed, whereas in the gospels Jesus is the proclaimer. In other words, the gospels portray Jesus as one proclaiming the kingdom of God, whereas in the Pauline letters, the order is reversed: Jesus is what the kingdom of God proclaims.
I actually do mention this in OHJ (e.g., pp. 554-55; cf. pp. 594, 530, 528, 520-21, 516-17, and also in respect to Clement of Rome, p. 314), but Covington does a better job of calling attention to it (and he even cites more scholarship in support). But I zero-out it’s significance, concluding this is equally likely on either theory (minimal historicity or mythicism). Covington disagrees and thinks it should favor mythicism by a factor of 9 to 5.
I’m sympathetic to this. It is certainly a reasonable proposal when arguing a judicantiori (rather than a fortiori), i.e. when modeling what you personally think must be the case even if you are certain stalwart deniers could never be persuaded of it. And I do mention in OHJ from time to time that even my a judicantiori estimates are often over-generous (e.g., pp. 308 and pp. 557, n. 55, and 574-75, n. 82). This may be another instance.
But I was not fully persuaded when I wrote OHJ.
First, although indeed the switch from “Jesus proclaimed” to “Jesus proclaimer” is weird and makes little sense if the Gospel accounts are histories, the Gospel accounts are not histories. And there’s the rub. When we are testing minimal historicity, we are not presuming that what the Gospels say is what happened. To the contrary we are fully allowing that it is not. Consequently, it is entirely possible that Jesus did not do any proclaiming (beyond just prophesy, say), and that it was his followers who started that up after his death. The Gospels then fabricate his “ministry of proclaiming” in order to more effectively allegorize what his followers actually had until then been proclaiming about him. On minimal historicity this is as expected as anything else (because minimal historicity does not predict the contrary).
Although Covington can still capitalize on this. Because this commits the historicist to such a stripped-down historicity as to put the kaibosh on almost every cherished theory of the historical Jesus. But if they want to rescue the baby of historicity, they may be willing to throw out the bathwater of comforting theories about what Jesus actually did. Although that does create even graver problems for them, as I note in OHJ, pp. 557, n. 55, and 574-75, n. 82. So it’s worth pushing them into that corner, even if only to compel them to realize what it entails.
But second, it’s also just as likely Jesus did proclaim, but was just so unknown and unlistened to, that his followers had to switch to proclaiming him (becoming his ad men—in effect, for what was in life a failed product pitch). The Gospels could then have restored the original order, even if with fabricated accounts of it, because they are a different kind of teaching tool than the Epistles. Thus, all historicist attempts to “explain away” the silence of Paul’s letters (with respect to Jesus having a ministry) apply here as well: maybe Paul never did have an occasion to clearly say that’s where any Christian teachings came from. Personally, I do think that’s unlikely. But a fortiori?
So I think Covington is at least right that my a judicantiori estimate for Jesus’ sayings is over-generous to historicity and should not be 1:1, on the same reasoning as previous: the Epistles display the only way Paul could talk about the gospel on mythicism (or almost, see below), but that is not the only way he could have spoken on historicity. One might counter that he could also have spoken more explicitly of the cosmic origin of the gospel (so what we have is not the only way he could have written about it), but b (our background evidence) counters that in turn, because we know the Epistles passed through a very corrosive filter (OHJ, Element 21) that would have destroyed any such thing (to a very high probability, e.g., if Paul wrote any letter saying something explicitly anti-historicist, it simply would not have been preserved for us to know if it: OHJ, pp. 552, 593-94, and ch. 8, §12). Whereas that filter would have had the opposite effect on historicity: had Paul written anything more explicitly historicist, that probably would have been preserved.
So Covington may be right: for the Epistles to speak of proclaiming Jesus rather than Jesus proclaiming anything could be 2:1 more likely on mythicism.
At first look I should have thus replaced my 1:1 (for ‘things Jesus said’) with that 2:1 (OHJ, p. 594) in the ‘worst’ case column. For the ‘best’ case column, it should be 3:4, on the reasoning that the text as we have it is all but 100% expected on mythicism, but on minimal historicity there is a 50% chance Jesus never did proclaim a gospel, and if he did, there is a 50% chance it was so unknown and unlistened to that it had to be proclaimed by others first, before his fame would be enough to write stories about what he proclaimed (in each case it’s 50% because, I am assuming, we don’t know one way or the other which is more likely—since mininmal mythicism says nothing about that, and there is nothing in b that makes either more likely than the other…if we think a fortiori, although see again p. 557, n. 55). That means even a fortiori the evidence of the Epistles is only 75% likely on minimal historicity, on the most generous framing of options possible (50% + [50% x 50%] = 50% + 25% = 75%). That gives us 75:100, which is 3:4.
But there is one hitch: on mythicism it is actually possible to have preached Jesus as a proclaimer from the start—by saying the heavenly Jesus proclaimed a message (to the apostles, and through their proxy, to the world) which is now being delivered. There may well be texts already that say this. So the sequence ‘proclaimed-proclaimer’ (our e) may be more likely on mythicism, but the sequence ‘proclaimed-proclaimed’ was also genuinely possible on mythicism, so P(~e|m) is not zero, which entails P(e|m) cannot be 100% (Proving History, pp. 230-31, 255-56, and 302, n. 13). When we carry out the math from there I think this may entail a wash a fortiori (1:1), and at best 3:4 against historicity a judicantiori (to reflect the fact that it’s more expected to see our e if historicity is false, given that those rescuing assumptions for historicity are not really 50/50, because in fact our world knowledge in b renders them unusual and thus unlikely).
Although that still means, so far, it looks like Covington’s review of OHJ is making its conclusion stronger.
For a complete list of my responses to critiques of OHJ, see the last section of my List of Responses to Defenders of the Historicity of Jesus.