I saw none of the other apostles, except James the brother of the Lord. — Galatians 1:19
On this verse some hang their strongest assurance that Jesus himself was an historical figure. Paul says he met James, the brother of the Lord (assumed to be Jesus), so that is absolute proof that Jesus existed. That sounds like a perfectly reasonable conclusion. So reasonable, in fact, that some quickly denounce as perverse cranks any who deny this “obvious meaning”.
But should historians be content with this? Is it being “hyper-sceptical” to question this explanation?
We need to keep in mind some fundamental principles of historical research and explanations from the professional historians themselves. Renowned conservative historian, Sir Geoffrey Elton, warns against deploying such simplistic methods as citing a single piece of evidence to make a case. In this instance, the case is about evidence for the historicity of Jesus.
Historical research does not consist, as beginners in particular often suppose, in the pursuit of some particular evidence that will answer a particular question (G.R. Elton, The Practice of History, p.88)
If that’s what historical research is not, Elton goes on to explain what it is:
it consists of an exhaustive, and exhausting, review of everything that may conceivably be germane to a given investigation. Properly observed, this principle provides a manifest and efficient safeguard against the dangers of personal selection of evidence. (p.88)
Since I am currently reading and reviewing Richard Carrier’s Proving History: Bayes’s Theorem and the Quest for the Historical Jesus I am taking time out in this post to see what happens if I test this “obvious” interpretation of Galatians 1:19 by means of Bayesian principles. Carrier argues that Bayes’ Theorem is nothing more than a mathematical presentation or demonstration of what goes on inside our heads when we are reasoning about any hypothesis correctly. So let’s try it out on the conclusions we draw from Galatians 1:19.
The way it works is like this. (But keep in mind I am a complete novice with Bayes’ theorem. I am trying to learn it by trying to explain what I think I understand so far.) I see a verse in Paul’s letters that appears to have a simple explanation. I think of myself as a geologist looking at strata in a rock face and I think about all I know about strata and the evidence in front of me and with all that in mind I try to work out how that strata came to look the way it does. This verse is like that strata. My task is to test a hypothesis or explanation for how it came to be there and to appear as it does.
So the explanation, or hypothesis, that I decide to test is: That James, whom Paul meets according to this letter, was a sibling of Jesus. That’s my initial explanation for this verse, or in particular this phrase, “James the brother of the Lord”, being there.
It seems pretty straightforward, surely. This should be easy enough to confirm.
So let’s set it out in the theorem structure.
The probability that my explanation for Gal 1:19 is true =
The probability that my explanation for Gal 1:19 is true
(How typical the explanation is) X (How expected the evidence is if the explanation is true)
|(How typical the explanation is) X (How expected the evidence is if the explanation is true)||
(How atypical the explanation is) X (How expected the evidence is if the explanation is not true)
So the first value I need to enter is “How typical the explanation is”.
This means we need to make an assessment of its typicality given all our relevant background knowledge. I put the negatives in red and the positives in green.
According to the Gospels Jesus did have a brother named James.
Now if the phrase said “James the brother of Jesus” then of course we would all agree that such a phrase points to a sibling relationship. But we do have many instances where “brother” is used of Christians and in Hebrews Jesus speaks of having many brethren.
“The Lord” is a religious title here and Jesus is not named. So we have some small space for alternative possibilities.
But against both of the above we might think it is more typical to depart from the spiritual meaning of such a term when one is attempting to clarify the personal identity of one person.
We know of no other instances of people in this context being called the “brother of a spiritual Lord” (or God) so this reduces the typicality of the explanation.
But we have one tradition that Jesus had no siblings at all. We also have other knowledge that James was reputed to have been a renowned leader of the Jerusalem church, and his relationship with God was so close that he was known as old ‘camel-knees’, a repetitive strain injury/side-effect from overmuch praying. (Presumably this is even evidence he wore short tunics otherwise how would this be known, but we can save that little profundity for the theologians.) Our interest is in the likelihood of such a phrase in this context being an indicator that James and Jesus were siblings.
Another circumstance we do know was common enough in ancient times was the tendency for copyists to edit works, usually by adding the odd word or phrase or more. Sometimes this was entered as a gloss in the margin by way of commentary, with a subsequent copyist incorporating that gloss into the main body of the text. That’s a possibility, too, given what we know of both Christian and pagan texts.
Given what we know about the evolution of texts, the alterations to manuscripts and so on, it is by no means sure how secure any wording, especially a slight one, in a New Testament text should be considered which is that far removed from the autograph. How can a decision be made about key questions based on this inherent degree of uncertainty, an uncertainty justified by the general instability of the textual record visible in the manuscripts we do have? And yet arguments are formulated on such slender reeds all the time — and not excepting by mythicists. (Jesus: Neither God Nor Man, pp. 61-62)
On the other side of the ledger we have the likelihood that if Jesus were known as a Son of David then it is reasonable to imagine that his royal heir would be his next-in-line brother.
I would say the typicality of our hypothesis should be “very probable”, let’s say 0.95. Or am I being overly generous here? Will I think the same tomorrow?
By assigning 0.95 as the probability for how typical our explanation is, we by default assign 0.05 as the value of how atypical our explanation is. The two values most add up to 1.
The next value we need to enter is one to indicate how expected the evidence is if the explanation is true.
This includes taking into account a lack of evidence where we have strong reasons to expect to find that evidence if our hypothesis is true.
Well, if our hypothesis were true, yes, we would expect someone who met James to inform readers of his letter that the James he met was indeed the brother of Jesus if that’s what “Lord” refers to. (And certainly Jesus is called “Lord” very often elsewhere. So is God, but Jesus is too.) So to that extent the evidence is just what we would expect.
Against this, however, is the problem that if our hypothesis were true — that James, a leader of the church, really was a sibling of Jesus — we would expect to find supporting claims to this effect in the contemporary or near contemporary literature.
But in the Book of Acts we have what is surely a strange silence about James being related to Jesus despite his prominence in the Jerusalem church. Additionally, we have the unexpected failure to explain how this James acquired this position of pre-eminence. The beginning of the book indicates only twelve apostles and a total of 120 brethren were the original Christian club. James is not singled out. Yet we inexplicably find James leading the Jerusalem conference in Acts 15. It should further be kept in mind that we have no reason to assume that the designation “brother of the Lord” in Galatians was a reference to a “head” of the church as James appears to be in Acts.
The letter attributed to James in the New Testament gives no hint that its author knew that the name and person of James was a blood relation of Jesus. One would have expected some such indication in a letter sent to brethren far and wide (to “the twelve tribes”) to alert readers to the presumed author’s authority. This would be especially so if James were a reasonably common name. Given the often contentious nature of early Christian correspondence it is difficult to explain why any information to enhance the author’s authoritative status would not be made explicit.
The letter attributed to Jude in the New Testament is just as unexpected in the way it identifies its author as the brother of James and not Jesus — if indeed our hypothesis were correct.
The Gospels indicate that James, though a brother of Jesus, was hostile to Jesus. There are no indications anywhere in the Gospels that this hostility was ever resolved. So on the strength of what we know from the Gospels we must suspect that the James Paul met in Jerusalem was not the same as the brother in the Gospels. If he were the same we would expect some hint somewhere that he came to have a change of heart.
Another factor in the Gospel account is the unusual combination of the names assigned to the brothers of Jesus. Any discussion on whether or not Jesus had literal siblings necessarily embraces Mark’s naming four brothers:
Isn’t this Mary’s son and the brother of James (=Jacob), Joseph, Judas (=Judah) and Simon (=Simeon)? (Mark 6:3)
Although the names may have been common, to find these particular names all bracketed together is still striking. Jacob, Joseph and Judah are three of the most prominent of Israelite patriarchs, and Simeon, too, is strongly associated in this status with Judah. As historical Jesus scholar Paul Fredriksen remarks:
It’s a little like naming a string of Olsons Washington, Jefferson, Hamilton, Franklin: the names themselves convey a close identification with the nation’s foundational past. (Jesus of Nazareth, p.240)
Paul in Galatians expresses no interest in learning about Jesus things that only a brother could know. He even scoffs at the idea that James might have anything to teach him.
The context in which the brothers of Jesus appear in the first Gospel (Mark) is the theological message that prophets are not accepted by their own kith and kin. The scene is presented to illustrate this message. It sets Jesus in the tradition of other men of God: Abel, Joseph, Jephthah, Moses, David . . . So the purpose is not to convey historical information but to illustrate a theological message and claim about Jesus. Given the absence of any other evidence clearly supporting historicity, this is a point against the historicity of the relationship between the two persons.
There is no external witness to Galatians 1:19 till the time of Origen (3rd century) despite its apparent potential usefulness in arguments against Marcionites by “orthodox” representatives such as Tertullian (second century).
There is a critical case of some slight cogency against the authenticity of Gal. i, 18, 19, which was absent from Marcion’s Apostolicon; the word “again” in Gal. ii, 1, which presupposes the earlier passage, seems to have been interpolated as it is absent from Irenaeus’s full and accurate citation of this section of the Epistle to the Galatians in his treatise against Heretics. (p. 76 of Jesus Not A Myth by A. D. Howell Smith.)
This record of evidence and “negative evidence” is all very unexpected if our hypothesis were true. I would say it is “very improbable”.
“Very improbable” sounds like a 5% chance of us finding this state of evidence if our hypothesis were correct.
That is, we can assign a probability of 0.05 to the expectedness of the evidence that we do have given our hypothesis is true.
So let’s multiply these two:
[How typical the explanation is — 0.95]
[How expected the evidence is if the explanation is true — 0.05]
Now we need to weigh the alternative hypotheses.
How atypical is the explanation? We have already assigned that the value of 0.05.
How expected is the evidence if our hypothesis were not true. That is, how expected is our evidence if James were not literally in real life the brother of Jesus?
Given the considerations listed above, I would say that the evidence is just what we would expect if James were not a literal sibling of Jesus. It is also just what we would expect (not being attested until the third century despite the anti-Marcionite value of such a concept, and slight hints it did not appear in the text known to Tertullian) if the phrase “brother of the Lord” entered as a gloss.
So how expected is the evidence if our hypothesis were not true? I have to say it is “extremely probable”: that is, 0.99.
So let’s calculate the values in the denominator:
The first one is the same as the numerator: 0.0475
This will have to be added to the following:
How atypical the explanation is: 0.05
How expected the evidence is if the explanation is not true: 0.99
The probability that my explanation for Gal 1:19 is true
This is a very simplified use of Bayes’ Theorem applied to the likelihood that Galatians 1:19 is evidence that James was a sibling of Jesus.
The answer I arrive at is a 50% probability that it is reliable evidence Jesus and James were siblings.
I have only begun to read Richard Carrier’s book and I am sure I have much more yet to understand. I imagine Carrier and others more experienced with this sort of thing will be aghast at mistakes or oversights or oversimplifications I have surely committed.
If I juggle the figures a bit to try to be more accommodating to what evaluations I would expect historicists to make, I can bump up the figure to 66% probability the hypothesis is true.
What this exercise has taught me is the problematic nature of assigning number values. I know Carrier addresses this question later in the book. Part of that difficulty, however, was consciousness of how my own biases might be being quantified and made to stand out like a boil on my nose. It is easy to be a bit too easy on the values I assign simply to avoid any suspicion of letting my biases affect my judgment.
Until I read Carrier’s section on how to assign number values, I suggest that the difficulty must necessarily force one to air the probabilities with peers, both supporters and opponents. There needs to be some agreement for the final results to be acceptable to anyone. And that discussion can only be a good thing by forcing the various factors in an argument to the up-front attention of all stakeholders in the debate. One cannot glibly dismiss an argument on the vague grounds that it is “not persuasive”.
I could well say, with simpler figures, for example, that the numerator should be 0.18. That is, 0.9 times 0.2 (how expected the evidence is given that James was the brother of Jesus). The expectation that the evidence we have is what we would expect if James and Jesus were not siblings and the phrase “brother of the Lord” originated as a gloss: = 0.9
The result then would be 0.67 or two-thirds, 67%.
That’s when I try to bend the results as favourably as possible towards the bias that James and Jesus were siblings.
Even that result demonstrates that the argument that Galatians 1:19 is slam-dunk proof that James and Jesus were siblings is false. The verse can by no means be upheld with absolute confidence that it the guarantor of James being the physical brother of Jesus.
30% doubt leaves a lot of room for further questioning and wider investigation.