Richard Carrier and Rank-Raglan Heroes, Part 1

Richard Carrier and Rank-Raglan Heroes

The core of Richard Carrier’s argument for the a-historicity of Jesus of Nazareth is his use of the Rank-Raglan mythotype to establish a low prior probability of Jesus’ existence.  For something so key to his argument, however, Carrier plays very fast and loose in his use of the mythotype. Specifically,

1. He modifies key criteria to better fit the story of Jesus of Nazareth;
2. He biases his scoring in favour of finding Jesus of Nazareth a member of the mythotype;
3. He biases his scoring against finding historical figures to be members of the mythotype; and
4. He uses a biased sample of the mythotype in establishing his prior probabilities.

The first three of these points have been pointed out before, notably by James McGrath (“Rankled by Wrangling over Rank-Raglan Rankings”) and Johan Rönnblom (“Is Jesus a Rank Raglan Hero”).  Daniel Gullota has also made similar points in his review of “On the Historicity of Jesus”.  That review, however, is behind a paywall and inaccessible to me.  I only know of its contents through Carrier’s response to that review.  Unlike the first three points, I am not aware of anybody previously making the fourth point.
 

In this blogpost, I will restrict my discussion only to Carrier’s treatment of the Rank-Raglan mythotype, and specifically how he modified the criteria.  I will not discuss how he scores results, and how he selected his sample for comparison. Nor will I concern myself with his use of Bayes theorem on the data to generate a prior probability of Jesus of Nazareth being historical.

Comparing Lord Raglan’s twenty-two criteria for the mythotype (pages 178 & 179 of “The Hero: A Study in Tradition, Myth and Drama” with those of Carrier (Locations 9469-9501 of “On the Historicity of Jesus”) it is easy to see that there are differences.  In five out of twenty two of those criteria, the differences are substantial such that on a strict scoring, Jesus meets those criteria as stated by Carrier, but would not meet those criteria as stated by Lord Raglan.  There are other, inconsequential differences, including the order in which the criteria are presented. The five substantial differences are as follows:

	Lord Raglan’s Criteria		Richard Carrier’s Criteria
1	“The hero’s mother is a royal virgin”.	1	“The hero’s mother is a virgin”.
2	“His father is a king”.	2	“His father is a king or the heir of a king”.
13	“[He] becomes king”.	10	“He is crowned, hailed, or becomes king.”
17	“[He] is driven from the throne and city”.	14	“He is driven from the throne or city.”
21	“His body is not buried”.	18	“His body turns up missing.”

Table 1: Differences in Lord Raglan’s and Richard Carrier’s stated criteria.

Because Gullota (among others) has already raised the issue of Carrier’s changes, we have the good fortune of already knowing Carrier’s defense of his modifications.  He says:

“Gullotta says “it is clear that Carrier has modified Raglan’s qualifications in order to make this archetypal hero model better fit the Jesus tradition,” but in fact I modified it to combine the two into one test, and actually fit the data of the other heroes as claimed by Rank and Raglan, since their counts didn’t actually match the exact wording of their own criteria (and even after I improved the wording, I got different counts than they did, as I note in : pp. 230-31, n. 191). And once again Gullotta fails at logic here. I explain (on p. 231) that by making the criteria even broader than Gullotta thinks Rank and Raglan had applied them, this should have increased the number of historical persons who score above half. In other words, I set each criteria more general than specified. Thus, it should be easier for someone who really existed to score. That they don’t, actually makes what I did a stronger argument for my conclusion, not a weaker one as Gullotta mistakenly claims.”

Carrier’s defence divides into two parts.  In the first part he claims his modifications meet two desirable criteria; that they combine Otto Rank and Lord Raglan’s distinct criteria into one test, and that they allow the criteria to better fit the data provided by Otto Rank and Lord Raglan.  In the second part, Carrier claims that by “making the criteria even broader” he increases the probability of including a historical figure, and thereby “makes what [he] did a stronger argument”. Both parts of his defence fail!

Considering the first part, it is important to note that Otto Rank only had criteria corresponding to the first three of the criteria which Carrier substantially changes from those of Lord Raglan - ie, criteria 1, 2, and 13.  Because criteria 17 and 21 have no corresponding criteria in Otto Rank’s “standard saga” (p 65 of The Myth of the Birth of the Hero), no change to Lord Raglan’s criteria can bring them into closer alignment with those of Rank.

Note: Readers who do not want to go into every detail may want to skip ahead to the section entitled "Summary".

Born of a Virgin

Moving on to those that do have corresponding criteria in Rank’s “standard saga”, we find (matching the first two criteria in Lord Raglan’s list), “The hero is the child of most distinguished parents, usually the son of a king.”  Clearly, with regard to the first criterion, Carrier has removed the distinction - that of prestigious parentage - which is shared by Otto Rank’s and Lord Raglan’s criteria to focus on that which they do not share, ie, virginity.  

At the same time, with regard to Lord Raglan’s first criterion, Carrier’s modification makes the criterion fit less well with Lord Raglan’s practise.  Lord Raglan (and Carrier) count Joseph as being a Rank-Raglan hero in part because his mother was a “royal virgin” (Raglan) or a “virgin” (Carrier) despite the fact that his mother was the favourite wife of his father during a period in which her sister (and fellow wife) conceived and gave birth to six sons and a daughter in successive pregnancies (Genesis 29:31 to 30:21), and the fact that Joseph’s father is said to have “slept with” Rachel (Genesis 29:30).  Similarly Raglan (and Carrier) count Moses as meeting the first criterion even though he was his mother’s third child, with his sister being old enough to watch him in the river as a baby (Exodus 2:4) and his brother also being three years older (Exodus 7:7).

Clearly actual virginity was not an important part in Lord Raglan’s criterion.  What is important, and is the sole corresponding criterion by Otto Rank, is the social status of the mother.  In this case, Carrier’s modified criterion fits worse with Rank’s equivalent, and does not match Raglan’s practise, but includes Jesus where both Rank’s and Raglan’s first criterion would have excluded him.  That alone is sufficient to falsify the first part of Carrier’s defence of his modification of the criterion.

The Son of a King

Carrier fairs marginally better on the second criterion.  Otto Rank’s criterion that the father of the hero be “most distinguished” and “usually the son of a king” is certainly broader than Lord Raglan’s stated criterion that “His father is a king”.  Carrier’s criterion, ie, that the hero’s “...father is a king or the heir of a king”, is broader than Raglan’s stated criterion, if not as broad as that of Rank.

It does not fair so well compared to Lord Raglan’s practise.  Raglan counted as heroes as satisfying his second criterion if they were the son of “a patriarch” (Joseph), or of “the Duke of Cornwall” (Arthur), or of “a Holy Man” (Watu Gunung), or “was of the principle family of the Levites” (Moses).  It may be objected that two of these do not meet Carrier’s arbitrary cut of as to who actually counts as a Rank-Raglan hero, but that is irrelevant. What is relevant is what properties led Lord Raglan to count a potential hero as meeting a specific criterion.  That means Carrier’s criterion is more restrictive than either Rank’s criterion, or Raglan’s criterion as applied. Given that Carrier purportedly modified the criterion to both to make both sets of criteria into one test, and to reflect Rank and Raglan’s practise, his partial success in the former and clear failure in the later again is sufficient to falsify the first part of Carrier’s defence as applied to this criterion.

Once again, however, while failing at its stated purpose, Carrier’s modified criterion does give a blush of plausibility to his claim that Jesus passes this test.  It is clear that Joseph was not a king, and that nor was he a “most distinguished” parent. Indeed, his occupation was used as evidence of Jesus’ lack of distinction by the people of Nazareth according to Matthew (13:55), and hence according to the tale on which Carrier is basing his assessment.  Ergo, Joseph meets neither Rank’s nor Raglan’s second criterion.

At first blush Matthew’s genealogy of Joseph (Matthew 1:6-17), and his being addressed by the “angel of the Lord” as “Joseph son of David” (Matthew 1:20) seem to support the claim.  Evidently Carrier thinks it does, for he says, “Jesus scores twenty out of twenty-two, according to Matthew’s Gospel … . The first nineteen [by his numeration] hardly require defense (e.g. his father is the heir of King David …)” (OHJ location 9561).  Those first appearances, however, are deceiving. Not all descendants are heirs. The author of Matthew has theological reasons to ‘prove’ that Jesus was a descendant of David, but he had no need to prove he was a literal heir. His claim to be King of the Jews was based on his being the Messiah, the anointed and therefore equalled or exceeded David’s claim.  It was theologically necessary that he be a descendant of David to satisfy prophecy, but for that any patrilineal descendant would do. Further, based on 1 Chronicles, chapter 3, the line of inheritance descended from Zerubbabel through his second son, Hananiah. In contrast Matthew has Joseph’s line descend through Abiud. Ergo anybody taking both books at face value could not reasonably conclude that Joseph was David’s heir.  Though Carrier’s modification of this criterion does make it more plausible that Joseph satisfies it, ultimately he fails to do so even on Matthew’s account.

Becomes King

As with the second criterion, Carrier’s version of Lord Raglan’s 13th criterion is broader than that of Raglan, but not so broad as either Otto Rank’s equivalent (“Finally he achieves rank and honors”) or as Raglan’s practise.  In practise, Raglan counts this criterion as being satisfied if the person becomes a ruler of any sort, even (as in the case of Joseph) if subject to a higher authority. He also counts “becomes a man of power” (Asclepios), “becomes a sort of dictator” (Elijah), and even “becomes King of May and ruler of the forest” (Robin Hood).  

One feature of Raglan’s practise is that they actually need to achieve the position, however inconsequential it is.  Robin Hood is “ruler of the forest” because it is he, not the Sheriff of Nottingham who can enforce his will there. That means that Carrier’s criterion that the hero “... is crowned, hailed, or becomes king” is both broader and narrower than Raglan’s practise.  It is broader because mere acclamation as king is sufficient to satisfy the condition, whereas for Raglan’s practise the authority must actually be exercised. It is narrower because the position must be that of king, whereas Raglan in practise allows a whole host of positions to satisfy the condition including one (“King of May”) which is effectively equivalent to being a Prom King.

It is because Carrier’s criterion is both broader and narrower than Raglan’s practise that he can count Jesus as satisfying it.  Specifically, Jesus is acclaimed King in Matthew’s gospel at least twice - once by the wise men (Luke 2:1-12) and again, in mockery, by Pilate’s soldiers (Matthew 27: 27-31).  Arguably he is also acclaimed by the crowds during his triumphal entry (Matthew 21:1-11), although in that episode he is called the “Son of David”, and “he who is coming in the name of the Lord”, but when asked for clarification, “the crowds” said he was a prophet, not a king.  Further, Jesus is also crowned in mockery by Pilate’s soldiers. Being acclaimed twice, and crowned once (even if in mockery), Jesus therefore satisfies Carrier’s criterion, but he does not satisfy Lord Raglan’s - either his stricter criterion as stated, or the less strict criterion based on his practise.  Yet again Carrier’s modifications make it easier to count Jesus as a Rank-Raglan hero.

Driven From Throne and City

Lord Raglan’s 17th criterion has no equivalent among Otto Rank’s criteria.  It follows that Carrier’s modification of it, according to his defence, is solely to match Raglan’s practise.  In this it fails given that Raglan’s practise is much broader than that allowed by Carrier’s modification. For instance, Raglan counts as the hero “... being driven from throne and city” invading heaven (Watu Gunang), being induced by stealth to leave his castle (Llew Llaw Gyffes), and invading France to seek revenge on Lancelot for adultery and betrayal (King Arthur).  Clearly for Raglan the hero merely needs to leave their seat of power, not be driven from it. Carrier retains the idea of involuntary departure, thereby excluding cases that Raglan counted. So, once again, Carrier’s criterion is looser than Raglan’s stated criterion, but more restrictive than Raglan’s practise.

It should also be noted that while Jesus arguably satisfies this criterion in Raglan’s loose practise, it does not satisfy it by his stated criterion.  Specifically, Jesus did not have a throne and therefore could not be driven from both throne and city. By turning the conjunction (and) into a disjunction (or), Carrier changes it so that only one of the two clauses need by satisfied, not both.  That in turn allows Jesus to satisfy Carrier’s stated criterion in this case, where he does not satisfy Raglan’s.

His Body is Not Buried

Finally, with regard to the twenty first criterion, there is again no equivalent criterion by Otto Rank.  Raglan’s practise is again, uncommonly loose. He counts as satisfying this criterion people whose burial place is “uncertain” (Oedipus) or “unknown” (Perseus, Jason, Bellerophon, and Asclepios), whose body “is not found” (Romulus, Heracles), who have “no burial place” (Dionysos), or whose body “is not buried” (Joseph, Moses, Nyikang, and King Arthur), or when “there is no mention of his burial” (Watu Gunung), who has “no real death or burial” (Llew Llaw Gyffes), or even when the “place of his death and burial are variously given” (Robin Hood).  Given the broad range of examples, Carrier’s vague criterion that “His body turns up missing” appears to match Raglan’s practise. As such, in this case alone is the change justified by his stated defence of the alterations.

In this case it is transparent how Carrier’s altered criterion changes Raglan’s criterion so that Jesus satisfies it.  According to all four gospel accounts, Jesus was buried, thereby contradicting Raglan’s stated account. According to three of the Gospel’s (and interpolations in the third) Jesus’ body leaves the grave and hence “turns up missing”.

Summary

To summarize, none of the five substantive changes in criteria by Carrier adjust the criterion to be the same as Otto Rank’s.  For the first criterion, his criterion is less like Rank’s than is Raglan’s. For the second and thirteenth criteria it is more like Rank’s but still not as broad.  For the seventeenth and twenty first criteria, Rank has not equivalent criterion. At the same time, only the twenty first criterion was actually modified to match Raglan’s practise.  The first criterion is much more restrictive than Raglan’s practise. The second, thirteenth, and seventeenth first criteria are modified to be less restrictive than Raglan’s stated criteria, but more restrictive than Raglan’s practise, counting as satisfying it people who would not satisfy a criteria determined on Raglan’s actual practise (and equivalent to Rank’s stated criteria).  Finally, all substantive changes make it easier to count Jesus as satisfying that criterion, although careful analysis shows that Jesus does not satisfy criterion two by Carrier’s definition.

Do Carrier’s Modifications Make His Argument Stronger?

As noted previously, the second part of Carrier’s defence of his modifications is that it makes his argument stronger.  Specifically, he says:

“...I set each criteria more general than specified. Thus, it should be easier for someone who really existed to score. That they don’t, actually makes what I did a stronger argument for my conclusion, not a weaker one …”.

It should be noted, however, that he does not simply make the criteria broader.  Specifically, as noted above, he makes criterion four of the five criterion more restrictive than Raglan’s practise, not less.  Further, for one of the criteria (the first), the set of people satisfying it is not even a subset of those satisfying either Raglan’s stated criteria or practise.  Thus the initial premise of the second part of Carrier’s defence is simply false.

I cannot make much comment on the second premise of the argument without commenting on Carrier’s practise, which would draw me into issues of bias of application that I am reserving for a following post.  I will note, however, that in his scoring of various Rank-Raglan heroes, he scores Theseus, Perseus and Jason one point less than does Raglan, and Bellerophon two points less. He scores no hero that Raglan scored higher than did Raglan.  That shows his practise to have been more restrictive than Raglan’s, again falsifying the first premise of the second part of his justification.

Conclusion

As we have seen Richard Carrier makes key modifications to five of Lord Raglan’s stated criteria.  The modifications would in four out of five cases, as stated, result potential Rank-Raglan heroes not receiving a score on that criterion relative to Lord Raglan’s practise, and in two cases would also in theory result in potential heroes receiving a score on that criterion when they would not based on either Raglan’s stated criteria or practise.  In no case does the modification match Otto Rank’s equivalent criterion. Although in two cases it does shift Raglan’s stated criteria closer, on one case it shifts it further away. Finally, both as stated and in Carrier’s practise, Carrier’s modified criteria are more restrictive than are Lord Raglan’s, contrary to the first premise of Carrier’s defence of his modification.  Therefore, no part of Carrier’s defense stands. This does not mean he cannot justify the changes. It merely means that he has not.