The first question anyone has to answer when answering the question “How likely is it that Jesus was a mythical and not a historical person?” is “How often, at that time, were people like Jesus mythical and not historical?” And that requires answering the question, “What does it mean to be like Jesus?” What category of person did Jesus fall into? This was the occupation of Chapter 6 of my 2014 academic study On the Historicity of Jesus.  This still only gets us a “prior probability” that he was or wasn’t historical; that can readily be overcome by evidence. So even if people like Jesus tended not to be historical, there could still be enough evidence to prove Jesus one of the exceptions. But we have to start with the prior, before we can get to that concluding question of whether there is still also enough evidence to believe Jesus existed. So the prior comes first. How often did people like Jesus back then turn out to be at all historical, or entirely mythical?

Where I Got To

In my study I surveyed various approaches to that question, but settled on one for having two virtues over all the others: since Jesus undeniably belongs to the superset of highly mythologized persons, we want to find any subset of ancient persons who were definitely highly mythologized that has the most members of any such set, and whose belonging to that set was difficult to achieve, requiring the largest conjunction of peculiar features. These two conditions typically work at odds: as you have to meet more peculiar features to fit a set, the fewer people who are going to fit it. 

I surveyed examples of that phenomenon by looking at other highly dramatized sets Jesus belongs to, like the Socrates-Aesop “countercultural hero” set (Element 46) and the “ascending-hero” set (Element 48), whose most famous member is Romulus, mythical founder of Rome (see Richard Miller’s peer-reviewed analysis in Resurrection and Reception in Early Christianity, which wasn’t yet published in 2014 but I did cite an early article of his on the subject). These sets are highly particular but have very few pertinent members, creating a statistical risk of anomaly too great to work with (see my analysis of this point, and of the mathematical point to follow, in my response to Kamil Gregor). I also looked at suffering savior lords (Element 31), none of whom were historical (yet all of whom were set in history and given de facto historical biographies), but even those are only a handful in number. Jesus, it turns out, unusually belongs to a lot of heavily mythologized sets. Which simply reflects the general fact that he was highly mythologized. But it’s hard to identify the total membership of such a superset, since what qualified as “highly” mythologized is hard to nail down so precisely.

But lo, we have a set that does meet both the conditions needed for doing that: the Rank-Raglan Hero set (Element 48), which, counting Jesus, has fifteen members, and whose membership requires a significant number of peculiar conjunctions, too many for chance accident (or even gerrymandering) to produce membership to any likelihood worth worrying about. The significance of this is not that Jesus was molded to be one of these heroes. That’s as incidental as that he was molded to fit the Socrates-Aesop set and the Ascending Hero set, and to look like yet another suffering personal savior Lord, and to be a new Moses and Elijah, and so on. Any of that can be done to a historical person as well as a purely mythical one. Rather, what’s significant is that we know Jesus was highly mythologized, and yet here we have a usable proxy to that set to measure with: being substantially rendered into a Rank-Raglan hero is an instance of high mythologization; and we have fifteen people from that same era also being that mythologized; so now we have a definable and countable reference class representative of the larger (if more vague) superset of all “highly mythologized persons.” 

If you want to know how typical it was for a highly mythologized person to turn out to ever have been historical—since obviously both real people and fake ones could receive this treatment, so what we want to know is how often—you can look at the Rank-Raglan set and see how many members of that set were historical. That will proxy to the entire category of highly mythologized persons, because it is one instance of being highly mythologized, has a lot of members, and is hard to belong to. It’s unlikely Rank-Raglan heroes were “more often” historical or mythical than any other kind of highly mythologized person. If few (or indeed none) of them were historical, then it is unlikely any other kind of comparably mythologized person is going to be any more often historical than that, much less a person who is both mythologized into that set and also other highly mythical sets (there is no plausible way that fact is going to increase the prior probability of Jesus being historical).

Certainly, if you want to claim something else, you have to empirically prove it; otherwise, once we see Jesus is in this set and none of its other members were historical, to maintain Jesus nevertheless existed we must admit that Jesus has to be a rare exception to one empirically observed rule. Which makes the historicity of Jesus a remarkable claim, requiring remarkable evidence. Just as for other members of that set, it would be weird for Jesus to be the lone historical example. And weird is just another way of saying improbable. (All the objections one might have to this, such as regarding how quickly this happened to Jesus, or other sets Jesus might also belong to that have more historical members, and so on, I fully address already in OHJ.)

This can’t be artificial. There is no statistically credible way to gerrymander a set of so many mythical persons with so many particular criteria having to be met. This can be done with small sets (hence I discuss the infamous Lincoln-Kennedy convergences in OHJ as an example of a gerrymandered set) or with easy-to-meet conditions for membership (like “religious founders” with just one or two cherry-picked things in common); but hard-to-meet conditions for membership cannot possibly “conveniently” rope in over a dozen members none of whom happen to be historical. Any such attempt would inevitably rope in a lot of historical members, too—unless it was objectively unlikely for historical members to meet those conditions. Which is precisely what we want to know.

Even so, a set as small as fifteen still requires large margins of error to account for potential statistical anomalies, and so I set in my study a large error margin of 4 members. That means, of the fourteen besides Jesus, in actual fact we have no reason to believe any were historical, for a count of 0 historical members (leading to a Laplacean probability any new member would be historical of 1 in 16); but on the other side of my error margin I assume maybe, unbeknownst to us, as many as 4 were historical, leading to a Laplacean probability any new member would be historical of near enough to 1 in 3 that I just set that as the upper bound.

So anyone else we might dig up in historical records who fits the Rank-Raglan class we can say has at best a 1 in 3 chance of being historical—until we look at any evidence particular to them, which could establish they are indeed in that 1 in 3 (or 1 in 16), and therefore one of those exceptions. Jesus just so happens to be someone else in historical records who fits the class. So we cannot privilege Jesus. He cannot be any more likely to be historical, on prior considerations alone, than anyone else in the set (from Moses to Hercules). To act otherwise would be to substitute prejudice for objective historical fact. If you want to argue for a different frequency of ancient historical persons in that set, you have to do the work and prove it. Until then, I’ve met a sufficient burden of evidence: I found no other members in antiquity; and all the members I found were ahistorical; it’s hard to qualify as a member; yet there were a lot of members; and Jesus is a member. You have to simply deal with this fact.

I also increased the a fortiori strength of that case by being more stringent on membership requirements than Rank or Raglan were, such that in some cases I got lower (and in no case higher) scores for the heroes they graded as belonging to the set. This was due to some of the details claimed for them to make them fit one or another criterion being a bit too vague to be certain. For example, Raglan counts Oedipus as “reputed to be the son of a God” based on an inference that maybe he implied it at one point in his mythic cycle, but I find that inference too vague to count. Whereas I grant Raglan counting “the circumstances of his conception are unusual” because, indeed, central to Oedipus’s entire myth is a mystery over who his parents were. Who conceived him was not only in doubt, but indeed turns out so unusual as to result in him unknowingly killing his father and marying his mother. [Update: It has been pointed out that I mistook Raglan as scoring Oedipus at 22, so my subtraction put him at 21, when it should be 20.]  

Note a key fact here: this does not mean miracles or the supernatural are required to meet any of the criteria for Rank-Raglan heroes. That can be a mechanism chosen in any given story, but there are also entirely realistic ways to get the same result. For example, that the hero’s mother “is a virgin” does not mean “virgin at his birth” (though of course that, too, would qualify) but simply “virgin at his conception,” or at least at consummation with the hero’s father. In other words, it just means his mother was a maiden when she married his father—as opposed to having had previous husbands or lovers. Perfectly ordinary.

As such, many historical people would have met that singular criterion, for example. It only starts to match a mythic pattern when found in conjunction with a larger number of other convenient facts. Raglan settled on twenty-two, and counted as members anyone meeting more than a handful. Being more stringent again, I count a member only if they score more than half of Raglan’s twenty-two criteria, which means a minimum score of 12. Any one of those twelve coincidences could be historically common (some maybe less so, but nevertheless, they are things that “could” happen, and not just impossible miracles). But to have at least twelve of them (out of a potential twenty-two) is simply not historically possible by chance accident—it would not be expected to occur historically even once, much less fifteen times! 

This is what makes this evidence of mythologization. All these heroes have definitely been mythologized; indeed even conformed to what was evidently a quite popular mythotype. This should not be controversial. It’s as obvious as that Jesus was conformed mythically to Moses and Elijah (or even, as I also document, Socrates and Aesop). But what we can discern from this discovery is this: we can now look in that set to see how often historical people were that heavily mythologized; and when we do, we find that none were, so far as we know. So evidently it was not common for historical people to become this heavily mythologized. They still could have been (hence my lower and upper error margins of 1 in 16 and 1 in 3). But we cannot claim they typically were. Which means we should not expect Jesus to have been—without evidence. This makes Jesus different from just any random person, like an ordinary military general or a government administrator or a politician’s wife or any other mundane person: such people typically do turn out to be historical (as I explain, for example, with respect to Hannibal, and more generally in an entire chapter on this difference in Jesus from Outer Space). So all we are doing here is empirically differentiating between mundane people, and people like Jesus—people merely mentioned in historical accounts, and people heavily mythologized.

Where to Go from There

Okay. So. All that.

If you need further exposition, I have written on this many times before, from many different angles:

• In Response to M. David Litwa
• In Response to Kamil Gregor
• In Response to Nicholas Covington
• In Response to Daniel Gulotta
• In Response to James McGrath
• In Response to David Marshall
• In Response to Neal Sendlak
• In Response to Crissy Hansen

See also Jesus and the Problem of the Fraudulent Reference Class and Doing the Math: Historicity of Jesus Edition. 

Common lies you will hear about this are that the Rank-Raglan mythotype has “been widely dismissed by folklorists and mythologists” (in fact, apart from apologists denying it, it remains a respected typology and is treated as established across those fields; see my response to Crissy Hansen for evidence and examples) or that “there are more historical people meeting the criteria than mythical.” In truth no one has demonstrated that even when anachronistic and thus inapplicable eras are included (see, for example, my response to Neal Sendlak); but when we correctly limit the reference class to antiquity, not even one historical person has been found to meet the sufficient criteria.

In OHJ I discuss the two who score closest, Alexander the Great and Mithradates, finding they don’t actually score more than half, without such egregious distortion of what counts as scoring as to be ridiculous, a violation of the basic rules of genre and typology. The game is rather like claiming How to Lose a Guy in Ten Days belongs to the genre of Righteous Mass Killers alongside The Equalizer and John Wick, on the basis that it contained a lot of sick burns and someone cared really a lot about a plant that died. The modern Righteous Mass Killers mythotype is real, and is discerned through Rank-Raglan style criteria: apart from other common elements of the trope that aren’t universally present but reinforce the identification when they are (“the killing is usually done with guns,” “the enemy is usually a criminal syndicate,” etc.), typically a retired or otherwise resigned violent man is moved to the mass murder of a large number of curiously undeniably guilty villains deserving of death, on account of a loved one threatened, harmed, or killed by them or their agents—whether that loved one be a friend, relative, romantic interest, or pet.

The Righteous Mass Killer genre is variable: the “loved one” can be a girlfriend, godson, dog—the “differences” are not relevant to the key feature, which is that they must be an object of sympathy and care for the killer, and threatened, harmed, or killed by the villains. Just as in the Rank-Raglan set, how conceptions and deaths can be “unusual” or “mysterious” is variable, the difference mattering no more than that for John Wick it’s, unusually for the genre, just a dog. It’s still the same genre. So variability within the trope does not deny membership in the set. But you still can’t force members into that set that don’t belong. Nor can you try to remove members from that set by the same tactics (“but in John Wick there’s no fridged girlfriend, so it doesn’t belong to the mythotype” is apologetical rather than legitimate reasoning).

Nor can you make the mythotype disappear by absurdly redefining the criteria either (“there is no Righteous Mass Killer mythotype, because anything can be described as ‘killing’ and as ‘a loved one’,” even “sicks burns” and “a mildly plot-relevant potted plant”). That is just as illegitimate. You therefore can no more make the Rank-Raglan mythotype go away than the Righteous Mass Killer mythotype. Which mythpotype, by the way, is also pretty much mythical. There are no, or hardly any, real historical John Wicks; there are lots of historical mass killers, but none avenging fridged loved ones by mass-murdering crime syndicates. So if we heard a John-Wick-like story someone was claiming to be true, we would be warranted in doubting it until we had good documentation. Because the prior odds favor the entire story, hero and all, being made-up. Sure, those odds do not entail that; evidence could reverse this. But you still need that evidence to reverse it.

Scoring Osiris

My main source for the basic scoring of Rank-Raglan heroes is Raglan himself, as produced in a work with an introduction by one of my mentors, the late Alan Segal, In Quest of the Hero (Princeton University Press 1990), following the advice and methods of interpretation laid out by Alan Dundes in that same volume. In the original edition I credited this to Segal as editor. In the revised edition I corrected that to Dundes. The book has an introduction by Segal, wherein the modern scholar’s work I rely on is Dundes, who is there vetting and endorsing Raglan. I also build on Dundes’ Rank-Raglan scoring of Jesus there, completing fourteen members of the set. 

But I found fifteen: in OHJ I also included Osiris. Neither Dundes nor Raglan mention him. So I had to score Osiris on my own (as I explain in OHJ). So the final question here is: How did I do that? Of course Osiris serves as a peculiarly strong parallel for Jesus (not in all the silly ways cranks claim, but in several serious, well-documented ways, also surveyed in OHJ, and in the most important respects in Jesus from Outer Space), but I also went looking for any remaining heroes in antiquity who might score high enough on the Raglan scale, and indeed found that only Osiris met the challenge. I have yet to find any other ancient personage who does; and no one else has produced one either.

To score Osiris I relied on a single treatise: Plutarch’s On Isis and Osiris, which he wrote for a peer of his who was an actual officiant, a priestess of the Isis and Osiris cult—so it is unlikely to be full of errors or his own fabrications (apart from where he outright says it is). It is possible Osiris could score higher in some other source, or across all sources; I make no claim to my result being “his score,” only his minimum score, since it arises even from a single author’s account. This is also Osiris as worshipped in the Hellenized mysteries most comparable to early Christianity; not Osiris as understood by pre-Hellenistic Egyptians hundreds of years earlier. The religion as it evolved into the form Plutarch is writing about has a lot of “Greek” features. But it still constitutes a version of the myth, one clearly officially endorsed by that time.

And that does mean this is, indeed, a mythology of Osiris; not the mythology of Osiris. Just as Matthew is “a” mythology of Jesus; not “the” mythology of Jesus. All gods and heroes scored by Rank, Raglan, and Dundes have many conflicting stories told of them. But the existence of any one coherent story classifying them as a Rank-Raglan hero is the one most pertinent to whether that kind of mythologization happened to them. So of all the variant stories Plutarch also relates of Osiris (and Isis), I follow the one that tracks most to that designation. Some other versions might add or subtract features of that mythotype (or even just leave some out for brevity). But that the features were added is what concerns us, because that shows an effort existed to conform the hero to the type; and the more additions that were attempted, the more mythically the god was then being conceived. So our score must follow the story with the most features added, as that then becomes an indicator of how mythologized the god became.

My numeration of criteria differed from Raglan’s (such as to set aside the ones Jesus fails to meet for review last); and my wording of the criteria I revised to match the way Raglan actually applied them, following the methods and advice of Dundes. This changed none of Raglan’s scores (I downscored only for want of facts behind Raglan’s inferences). And for Osiris, the result was as follows:

• 1. The hero’s mother is a virgin.

Osiris is conceived upon Rhea’s first intercourse, per Plutarch, OIAO §12, “the Sun, when he became aware of Rhea’s intercourse with Cronus, invoked a curse upon her that she should not give birth to a child in any month or year.” The moment Rhea has sex, Sun pounces. Few Rank-Raglan heroes are “born” to virgins; but many are conceived by virgins. Cronus was Rhea’s first lover. As I discussed above regarding Oedipus, “virgin” in the Raglan type did not mean miraculous virgin, but just any virgin, even the ordinary kind. By the account given, Rhea qualifies.

• 2. His father is a king or the heir of a king.

Osiris is born to Cronus, “king” of the Titans (and of the universe, after deposing his own father from that position, the god Uranus https://en.wikipedia.org/wiki/Cronus). Of course this integrates Osiris into the Greek pantheon (Diodorus and other authors likewise report similar Hellenizations of Isis-Osiris cult). Plutarch is writing this essay to a priestess of that cult, says this is the most standard account promulgated across the Empire at the time, and it is the only account he gives in which Isis and Osiris are made to be siblings, which was the most popular native Egyptian tradition as well. Hence even in that tradition Osiris similarly scores, as there Isis and Osiris were begotten by Geb, the god of Earth and the first king of men.

• 3. The circumstances of his conception are unusual.

Osiris’s mother is magically cursed upon (or relevantly soon after—and because of) conceiving him (and not just him, but remarkably four other gods sharing the same womb), per OIAO §12 (already quoted above). Hermes then has to engage in elaborate mythological trickery to get that curse removed so Rhea could eventually give birth. This is as much “the circumstances of his conception are unusual” as Oedipus’s conception being merely a mystery. In the full mythotype, weird things attend the conception, whether mundane (Oedipus) or marvelous (Osiris).

• 4. He is reputed to be the son of a god.

Osiris (§12) is born from “Rhea’s intercourse with Cronus.” Both are gods (Titans specifically, but gods all the same).

• 8. We are told nothing of his childhood.

There are no accounts in Plutarch’s summary of Osiris’s childhood. We skip directly from his birth to his adulthood. This is peculiar, because biographies often (even if not routinely) discussed the childhoods of their subjects.

• 10. He is crowned, hailed or becomes king.

Per §12, “a mighty and beneficent king, Osiris, had been born,” §13 “one of the first acts related of Osiris in his reign,” etc. Osiris in life was definitely imagined as having become a king (indeed of Egypt).

• 11. He reigns uneventfully (i.e., without wars or national catastrophes).

Raglan’s application of this criterion did not signify “no significant events at all” occur during the hero’s reign, but that no public catastrophes do (like wars or plagues or floods or famines) until he loses favor with his subjects or the gods. And so it is for Osiris. Per §13, “Later he travelled over the whole earth civilizing it without the slightest need of arms, but most of the peoples he won over to his way by the charm of his persuasive discourse combined with song and all manner of music” and “during his absence the tradition is that Typhon attempted nothing revolutionary because Isis, who was in control, was vigilant and alert.” 

In other accounts outside Plutarch (e.g. Diodorus) this would not score, as a flood (catastrophe) is related as occuring in his reign. But this version of events is not adopted in Plutarch. It is also not clear from the Diodoran account if that catastrophe leads to Osiris losing favor with the people—as in the case of Oedipus a plague does. The scoring element is that nothing of this scale happens during their reign until its conclusion (not that nothing the like happens). But regardless, in Plutarch’s version, nothing the like happens during the reign of Osiris at all. 

• 12. He prescribes laws.

Per §13, “One of the first acts related of Osiris in his reign was to deliver the Egyptians from their destitute and brutish manner of living. This he did by showing them the fruits of cultivation, by giving them laws, and by teaching them to honour the gods.”

• 13. He then loses favor with the gods or his subjects.

Per §13, “During his absence” Typhon was held in check by Isis, yet for some reason this check failed once Osiris returned, such that “when he returned home Typhon contrived a treacherous plot against him and formed a group of conspirators seventy-two in number.” So we have seventy-two of his subjects turning on him; but also after his death, few others are particularly upset about it (Isis alone mourns); and the gods do nothing to protect him. He went from blessed to doomed. That scores.

• 15. He meets with a mysterious death.

Raglan allowed a broad range of mystery to qualify here, as he did with respect to the “unusuality” of the hero’s conception. Anything clearly unusual or mysterious qualifies. Hence OIAO §13 & 15: at Typhon’s guile, “Osiris got into” a chest or coffin provided by Typhon “and lay down, and those who were in the plot ran to it and slammed down the lid, which they fastened by nails from the outside and also by using molten lead. Then they carried the chest to the river and sent it on its way to the sea” where “the waves had gently set it down in the midst of a clump of heather. The heather in a short time ran up into a very beautiful and massive stock, and enfolded and embraced the chest with its growth and concealed it within its trunk.” This is an unmistakably weird death; indeed even being attended by miracles.

• 18. His body turns up missing.

In addition to going missing in the previous way (so that Isis has to go on a quest to find his cadaver), also per §18: “Isis … bestowed the chest” containing the cadaver of Osiris “in a place well out of the way; but Typhon, who was hunting by night in the light of the moon, happened upon it. Recognizing the body he divided it into fourteen parts and scattered them, each in a different place. Isis learned of this and sought for them again, sailing through the swamps in a boat of papyrus.” So Osiris’s body goes missing even twice (Raglan noted repeated scoring for several heroes like this). Vanishing bodies or looking for the missing body is the common theme. Scores.

• 19. Yet he still has one or more holy sepulchers (in fact or fiction).

Per §18, “The traditional result of Osiris’s dismemberment is that there are many so‑called tombs of Osiris in Egypt; for Isis held a funeral for each part when she had found it.” Plutarch even mentions legends that these tombs don’t contain any actual part of Osiris, but in fact replicas, the real parts having been interred somewhere hidden, and thus are even still missing—as I document on OHJ and JFOS, in some accounts, his body is missing because Osiris is still using it, being risen from the dead. Yet there are tombs.

• 21. His parents are related to each other.

Rhea is literally the sister of Cronus. This is the case even in traditional Egyptian religion (there his parents, Earth and Sky, were also siblings).

• 22. He marries a queen or princess related to his predecessor.

Rhea was not only the sister but also the queen of Cronus, Osiris’s predecessor (that Osiris supplants him is per §12, “at the hour of his birth a voice issued forth saying, ‘The Lord of All advances to the light’,” and §2, “Him who is the First, the Lord of All, the Ideal One”). And Queen Rhea and King Cronus also begat his wife, Isis, thus making her a princess of a preceding ruling house. This is also true in native Egyptian tradition: their parents were also the same people, and their father was, again, Geb, god of Earth and first king of men, making Isis, again, a princess. 

Non-Scoring Items for Osiris

Those points Osiris meets. These points he doesn’t:

• 5. An attempt is made to kill him when he is a baby. 

• 6. To escape which he is spirited away from those trying to kill him. 

I didn’t score these because the conditions are vague. Sun’s curse does represent a kind of attempt to kill Osiris as a fetus, rather than as an infant; and hiding the infant Osiris with Pamyles (next) might indicate continuing the evasion of Sun’s wrath. But none of this is clear enough to warrant the score.

Though it is odd to note that this storyline does appear in Plutarch’s narrative—only about one of Osiris’s sons (OIAO §14):

Isis, learning that Osiris in his love had consorted with her sister [Nephthys] through ignorance, in the belief that she was Isis … sought to find the child; for the mother, immediately after its birth, had exposed it because of her fear of Typhon. And when the child had been found … it was brought up and became her guardian and attendant, receiving the name of Anubis, and it is said to protect the gods just as dogs protect men.

Here indeed Anubis is also produced by “unusual conception,” and meets other Rank-Raglan criteria, but too few (he doesn’t die, marry, or rule in any capacity, which alone clears half the available criteria). Though even if he did qualify as a sixteenth Rank-Raglan hero, he also did not exist. At any rate, let us return to Osiris.

• 7. He is reared in a foreign country by one or more foster parents.

I did not score this even though it is half true (as also for his son Anubis), per §12: “a certain Pamyles, while he was drawing water in Thebes, heard a voice issuing from the shrine of Zeus, which bade him proclaim with a loud voice that a mighty and beneficent king, Osiris, had been born; and for this Cronus entrusted to him the child Osiris, which he brought up.” So, he has “one or more foster parents” but is not raised in a foreign country (Thebes is still Egypt). One might quibble about whether the criterion need include the detail of it being in another land. But in this case I let it go.

• 9. On reaching manhood he returns to his future kingdom.

I did not score this because in Plutarch’s summary there is no discussion of Osiris returning from obscurity or exile to claim his kingdom. It is not related how he becomes king at all. One can infer that he might have begun in obscurity or exile, per his rearing by Pamyles, and so must have “returned” in some sense to claim his kingdom, whether earthly, as Pharaoh, or heavenly, as the successor to Cronus. But none of that is clear in this account—much as I believe Oedipus being “reputed a son of god” was not clear either. So it doesn’t go in. Though perhaps in a more complete narrative it would have.

• 14. He is driven from the throne or city.

This does not score because in no account I am familiar with is Osiris driven from the throne or city before his death.

• 16. He dies atop a hill or high place.

I did not score this because it can’t be confirmed. It is not said Osiris’s death occurs atop a hill. Though one might infer it did, given the topography of Egypt’s Nile valley, whereby all settlements had to be located atop hills to avoid its annual fertility flooding, and he seems to be killed in a settlement during a banquet, but it cannot be ascertained from the account in Plutarch that this was understood to be atop a hill or even where exactly it was. Not even the heather marsh that envelops him in his death is said to be on a hill.

• 17. His children, if any, do not succeed him.

This does not score because Osiris’s son Horus the Younger most definitely succeeds him. Note that Raglan scores Oedipus even though his sons briefly succeed him; thus to score only requires that they fail, not that they never try, to rule in his stead. But Horus is not soon deposed, nor ever by any account. Though not explictly claiming literal descent from Horus, the Pharaonic kingship of Egypt rested its legitimacy on his reign being successful, rather than claiming to have usurped and supplanted him. So, no score.

• 20. Before taking a throne or a wife, he battles and defeats a great adversary (such as a king, giant, dragon or wild beast).

This does not score because Plutarch does not mention Osiris having to battle any person or beast, either with violence or wits, before claiming his kingdom (such as Oedipus does, indeed defeating two such foes, one by violence and one by wits). Although Osiris does something alike to gain his final kingdom—to become the Lord of All he must endure the treachery of Typhon and overcome his murderous scheme (by resurrection)—that does not occur before his death. So it doesn’t count for the mythotype, which in this detail relates to the hero’s journey toward death, not the death itself.

Final Score

Osiris therefore scores 14 out of 22, missing 8. Well more than half. He is therefore definitely a Rank-Raglan hero. He also didn’t exist (we have fairly complete records of Egyptian Pharaohs and the development of the Osiris myth over millennia; we’d know if any such ruler were among them). There is also no plausible way multiple historical people could score this high in reality—they could only score this high by being mythologized, indeed quite heavily, as we see for Osiris. So having a high score (more than half the full 22) definitely signals a lot of mythologization has occurred. 

All that remains is to determine how often that scale of mythologization happened to real historical people. If you want to claim that this outcome included more than 1 in 3 such persons, you need to present actual evidence of that. Which means you need to state how many ancient (and indeed it must be ancient) Rank-Raglan heroes were historical out of all there were, and prove it with evidence. Post-antiquity heroes don’t count (as I explain in other articles, the context has changed, rendering that reference class inapplicable). Nor do low-scoring heroes (as they cannot be claimed to be as heavily mythologized as high-scoring ones, in this particular set).

That is the epistemic function of the Rank-Raglan mythotype and Osiris’s belonging to it. Despite a lot of rhetoric, there has been no factual or logically valid challenge to this yet.