DDG - I was responding to a tone I perceived in several early posts from Lib, Bryan Eckers, Sam Stone, and JT.
DDG, not sure what your problem is with the calculation. All we need is one additional assumption, that the probability of having a certain name is independant of the brother’s/father’s names (OK, not very likely, but it’s only a “back-of-the-envelope calculation”). Then the probability of the combination of the 3 names is
0.140.090.02 = .000252
Multiply this by the estimated number of inhabitants to get the likely number:
.000252*65,000 = 16
Nothing wrong here with the logic or use of statistics.
No.
The logical process is sound, but the assumptions are very dodgy.
The population estimate of 65,000 is a bit arbitrary. Earlier in the thread we have a range of 60,000 – 120,000. So why is it 65,000?
Secondly, the names inscribed on ossuaries are not necessarily representative of the population as a whole. I wouldn’t be surprised, for instance, if there were a lot more mens’ names than womens’. If so, the fact that a particula male name appears in 14% of ossuary inscriptions does not mean that 14% of the population had that name. We would need to control for this sex bias, if it exists. And I would kind of assume that higher status individuals were more likely to have ossuary burials or to have their names inscribed, and in many societies prevalence of name is linked to social class. I might be wrong in that assumption, but it needs to be tested. Nor would I be surprised to find that names were not evenly distributed throughout the population, but ran in families, or were associated with people having links to different parts of Judea, or whatever. If Joseph and James were common in the same families, and/or in the same region, the chances of a Joseph/James family connection are greater than would be suggested simply by multiplying the two percentages together.
In other words, we don’t have (or, at any rate, nobody has cited) anything like the data we need to make a useful back-of-the envelope calculation of the likelihood of the Joseph/James/Joshua combination arising.
Oh, and one more thing. The relevant figure is not the average population of Jerusalem at the time when ossuary burials were taking place. It’s the total number of people who lived (or, rather, died) in Jerusalem at the time when ossuary burials were taking place, which would be a larger figure.
Here we go again.
It’s not just me, Abe. The overwhelming majority of historians believe that Jesus of Nazareth actually did exist.
And, as I’ve said before, you insist on imposing standards of proof which go far, far, FAR beyond those used by professional historians when evaluating ancient history. No competent historian, for example, would dismiss the gospel accounts simply because they were written by followers of Jesus. Such an approach to history reflects a desire to prove one’s agenda, rather than dispassionate scholarly investigation.
If you want to keep harping against the claim that Jesus was a historical figure, then I can’t stop you… but once again, the overwhelming majority of scholars who are actually trained in this matter disagree with that conclusion.
UDS, I don’t agree that most of the points you mention would skew the results that much. We’re only talking order of magnitude estimate: if men’s names are 20% more prevalent it won’t make that much of a difference.
Your final point, though, is absolutely correct, although I would modify it to “the number of people who had ossuary burials during the period to which the ossuary can be dated”. That period could be as much as the whole time that ossuaries were used, which would up the population figure considerably. However, only a small fraction of the total population would have been placed in an ossuary. So the validity of the argument would depend on how well we ascertain both of those factors.
After thinking about this a bit, the timing seems very suspicious to me. Just 10 years ago the ossuary of Caiaphas was discovered (in an archaeological dig, not just lying around). Now James’s ossuary shows up. Hmmm…
—Don’t you see the point here? The percentage of certain names written on tombstones in Decatur has absolutely nothing to do with the percentage of those people who may have had siblings or parents named “whatever”.—
Why can’t they just use a conditional probability estimate?
—The main supporters of Josephus as a serious historian are the conservative Christians, who hail him as confirmation of Early Christianity’s validity (“A witness to Christianity!”).—
While obviously there are many problems with the later editing of his works, as well as the fact that “history” as we know it as a discipline didn’t exist yet, what reason do we have to doubt his matter of fact telling of the date of the death of James? Why lie or mislead or embellish about that?
Apos is correct. While Josephus was prone to embellishment and exaggeration, this does not mean that he was utterly irresponsible in matters of history. Despite his failings, modern historians still employ his body of work.
Many laymen are prone to discard Josephus’ accounts, or those of other ancient documents, on the basis of certain errors or alleged exaggerations. Learned historians, however, do not operate that way. They do not claim than an account is fundamentally unreliable, merely on the basis of some inaccuracies (perceived or otherwise).
Did Josephus tend to exaggerate? Perhaps. However, I have yet to see any evidence that he was wont to concoct events or personages out of nothing.
I wasn’t saying I have a problem with Josephus’ dating of James–I’m saying I have a problem with Lemaire using that dating as “proof” that the ossuary belonged to James. Lemaire is saying, in effect, “We know that this ossuary contained the bones of James the brother of Jesus, and we know from Josephus that James died in 62 A.D., so therefore James’ bones would have been put in this box in 63 A.D., so therefore the box can be dated to 63 A.D.” It’s that chain of reasoning that I disagree with.
Josephus is not really the issue here. What I objected to, when I said I was reluctant to accept him as a cite, as “proof” of anything, was Lemaire’s pinning his ossuary’s date to Josephus’ mention of James, especially given that most historians agree that Josephus tended to exaggerate a bit on occasion.
Okay, since you seem to know what it is and how it works, and I don’t, why don’t you go ahead and do a conditional probability estimate of the chances of a man named James in 1st century A.D. Jerusalem having a father named Joseph and a brother named Joshua, using Rachel Hachlili’s figures?
And, show all work, please. 
As mentioned above, the probability estimate rests on several crucial assumptions. In particular, the calculation assumes that
[ul][li] there is no corrolatoins between an individuals name and his father’s or brother’s name,[/li][li] the number of ossuaries “out there” is 80,000, and[/li][li] the distribution of names amoung those buried in ossuaries is the same as the distribution of names in inscriptions.[/ul][/li]
Given these assumptions, we can compute as follows.
In a population of 80,000 buried individuals, 2% are named James, of which 14% x 9% = 1.26% have a father named Joseph and a brother named Jesus. So there are 80,000 x 0.02 x 0.0126 = 20.16 such people.
What we’re looking for here is a confidence interval - a range within which we can be, say, 95% sure that the actual number of Joseph/Jesus/James combinations lies.
So we also need to know or make estimates for some other things:
-
Whether Hachlili’s numbers really do refer to percentages of inscriptions, or percentages of male names. If the former, then her estimate of the “Joseph’s as percent of all men” for example is not 14% at all, but “somewhere between 14% and 28% probably closer to 14%”. It makes a lot more sense to explicitly work with males only, since we KNOW that all characters referenced in this inscription are male
-
How many inscription data points she had to work with. If “14%” means “I looked at 100 names and saw 14 Josephs” the confidence range would be so wide as to be meaningless. If she had 1000 inscriptions to work with, her confidence as to the actual occurence of the name Joseph would be a lot better - something like (Waves Hands Frantically) a percentage point or so.
-
Total numbers of adult male deaths within the 2nd temple period. (In order to get that inscription on a tomb, the “Jesus” referenced has to be an adult,otherwise he would be neither famous nor in a position to bury someone, which implies that so is the “James”.)
4)What the expected number of brothers-per-adult male would be. If the poulation’s fairly static, then I’d expect that on average each couple would have maybe 2-3 children surviving to adulthood, which implies that the average number of siblings each person had would be around 1.5, and the average number of brothers would be around 0.75. This has to be factored into the calculation. Not everyone has a brother!
I might try to provide some actual numbers for a confidence interval later, when I’ve had time to remember more of my unfortunately-too-far-in-the-past stats lectures. But #2’s the real sticking point here. If you haven’t got enough data to go on, you can’t deduce anything.
It doesn’t matter, and it’s not like I haven’t addressed that issue, yet you keep ignoring it. The objections to the non-argument of the majority of historians (who are providing an opinion based on conjecture and not fact) are to be found in the thread linked, as well as many others. I respect their opinion, but you treat it like fact, or like it has any more weight than the opposite camp. Any number of historians saying Jesus existed and having little or no evidence to back up their claim, well that’s not an argument. It’s an unsupported position. It’s a hypothesis at best.
That is patently untrue. The standards of evidence I require are simply more stringent than post-humous secondary (or tertiary or worse) sources of (effectively) Christian propaganda, which you on the other hand have no trouble accepting. May I also remind you that among the sources you named as support for a historical Jesus were texts written hundreds of years after Christ and in one case that didn’t even mention his name! Nice evidence. We know practically nothing about the historical Jesus–we don’t even know if he existed. Again, the thread linked contains all these objections in greater detail.
No one is dismissing the gospels. However, the gospels do not provide evidence for the existence of a historical Jesus. They point in a direction, yes, but they are not concrete evidence even by relatively soft standards, and the use of soft standards on this difficult issue is something I have challenged before.
You’re parrotting my last post. A dispassionate scholarly investigation begins with the postulate that there may or may not have been a historical Jesus. That remains my position, and it certainly isn’t based on any agenda other than challenging inaccurate information issuing forth on this matter and confusing people. Get some real evidence, and I’ll be even more excited than the religious nuts (who don’t have doubts on the historicity of Jesus, whereas I do).
I accept the possibility of a historical Jesus, but I’m not saying that he existed or didn’t. Without sufficient evidence the historical Jesus remains a possibility, not a conclusion as you try to assert. It’s a quantum Jesus, in a sense–right now he both existed and didn’t exist. This issue is ruled (even among historians) by belief and opinion, often quite disconnected from fact (everyone is entitled to an opinion, but let’s not masquerade it as concluded fact).
And they disagree based on conjecture without sufficiently stringent evidence. Why, I wonder? Oh wait, this was already answered (and, as with many other points, never effectively challenged) in the thread linked.
Agenda indeed, JT. That ridiculous, ultra-biased, derisive article by Holding you keep plastering around as if it had any weight doesn’t exactly help your credibility as an unbiased or thorough scholar.
In this matter, as the ossuary matter is demonstrating, one has to be very careful and as unbiased as possible if one truly hopes to establish the truth.
—Lemaire’s pinning his ossuary’s date to Josephus’ mention of James, especially given that most historians agree that Josephus tended to exaggerate a bit on occasion.—
I still don’t quite get this: how can you “exaggerate” a date? A person? Maybe James wasn’t violently killed by people: instead that fig tree Jesus inexplicalby cursed fell over on him. But what’s the deal with doubting the date of James’ death? Lemaire already admits that his dating is based on the assumption that this is THE James: so what’s the problem?
Look, Apos, just drop it, okay? I’m sorry I ever made a slighting remark about Josephus. :rolleyes: I made it in the spirit of, “Oh, here’s this guy named Lemaire who’s fudging the facts in order to prove his theory, and to top it all off, he’s quoting Josephus as a cite, isn’t that just too perfect…”
Okay?
“Hear ye, hear ye, I now announce that Flavius Josephus is King of Historians.”
Happy now?
Aspidistra:
Hachlili’s information is contained in a series of articles, and as far as I can tell, none of them are posted online, and there’s no information that I can find on where she got her data. They’re probably written in Hebrew, too. But she sounds like an expert, and I have no reason to doubt her numbers.
http://archlgy.haifa.ac.il/staff/hachlili.htm
http://www.cfi.org.uk/yeshua.htm
I have some problems with the assumptions in this:
-
How can we make any kind of assumptions as to whether there are correlations between father’s, son’s, and brother’s names? We have no data, either way. For example, for all we know at this point, it may have been a longstanding tradition to name boys in alphabetical order, so James always came before Joseph and Joshua. yes, yes, I KNOW, I’m trying to make a POINT here…
-
Where did you get that “80,000 ossuaries” number? Cite? Why are you assuming that there are 80,000 ossuaries “out there”? Why are you assuming that all of them have inscriptions, and that all the inscriptions have names, and that we know what those names are, and that those names are men’s names?
-
Why are you assuming that the names on ossuaries would be the same names as on other inscriptions? Sometimes people choose to put a more important name on an official inscription, especially if they have several names anyway.
Also, your math makes no sense at all.
There is no logical reason for you to multiply 14% (the percentage of men named Joseph) times 9% (the percentage of men named Joshua) and then to say that the product of these two numbers–1.26%–equals the number of men who have a father named Joseph and a brother named Jesus.
It also makes no sense for you to multiply 80,000 (the total number of ossuaries) times 2% (the percentage of men named James) times 1.26% (the previous product that made no sense). Why did you do that? Just because it would give you that magic “20” number?
Explain your rationale behind your math operations here.
You cannot arrive at an answer merely by flinging numbers around, by multiplying numbers essentially at random until you come up with “20”. These percentages are not related to each other. There are no math operations that you can carry out with these numbers that will supply you with an answer to the question, “How many men named James had fathers named Joseph and brothers named Joshua?” That is an answer that can only be arrived at through the amassing of data–somebody would have to go and look at all the ossuaries and inscriptions and writings from the period and make a note of what everybody’s brothers and fathers names were, and THEN you’d have something to work with.
But there’s no information like that available here.
In a population of 80,000, if 2% of them are named James, then that means there are 1,600 men named James. But you have NO WAY of knowing how many of those 1,600 men have brothers named Joshua, or have fathers named Joseph. You have NO DATA on this, no information. The other percentages of men’s names apply to the population as a whole, NOT to the population of men named James.
In this same population of 80,000, if an additional 14% of them are named Joseph, then that means there are 11,200 other men named Joseph. But you have NO WAY of knowing how many of those 11,200 men have sons named either Joshua or James.
In this same population of 80,000, if an additional 9% of them are named Joshua, then that means there are 7,200 other men named Joshua. But you have NO WAY of knowing how many of those 7,200 men have brothers named James, or have fathers named Joseph.
DDG I don’t think it’s accurate to say we have NO WAY of knowing. It’s all probability distributions. In fact our “knowledge” that a 2% rate of James-naming in a population of 80,000 leads to 1,600 Jameses is * exactly the same * as our “knowledge” that the calculation of the naming rates would lead to 20 Joseph/James/Jesus combinations - BOTH of them simply represent probabilities based on the information we have. It’s just that in the second case you have to do more calculations to arrive at the point where you can make your estimate.
Now in the original article, when it says there “would have been about 20 instances” of the combination they’re not saying “yes there would be exactly 20 cases, not 19, not 21” - or if they are they need to be whacked hard with the Statistical Clue Stick. It’s that 20 is the most likely number and the way that they, and others in this thread, have calculated this number, is perfectly valid.
However, without a confidence interval it’s still relatively meaningless. What I would expect in a scholarly paper (and what I assume hasn’t made it all the way through to popular news articles) is a statement like.
“The number of such combinations would be 10 - 30 within a 95% confidence level” - this shows not only what the most probable number is, but how confident you are of your estimate.
None of which is to say that there COULDN’T have been 500 Joseph/James/Jesus combinations knocking around, only that if we found that this was so it would be Pretty Damn Wierd based on our current knowledge, and we’d probably want to investigate what extra factors lead to our data being so substantially screwed.
I’m sure that Hachlili, being a reputable scholar, has worked out the confidence intervals of her own data. It’s just that she hasn’t told us (nor do we have access to the raw info) which makes it more difficult to check her work.
Oh, and by the way, yeah I have a big problem with that “dated to 63” bit as well. In the article it’s portrayed like this is some extra new bit of data and … well, whaddayou know it confirms Josephus date. Whereas of course the causality is completely the other way round - it’s not CONFIRMATION of J’s date it’s TAKEN FROM J’s date
Oh, one other thing.
The calculation that Tyrrel used (just multiplying the three numbers together) does work if and only if the incidence of names is truly independant.
It’s probably not a bad assumption in this case, and it’s reasonable to think that the archaeologists would know whether or not the names were independent because they do have data on this: the inscriptions on a whole heap of bone-boxes saying “here lies X son of Y”. I assume that if, say, every bloke called James had a dad called Joseph SOMEONE would have remarked on this fact.
Unless we’re dealing with an Archaeologist With An Agenda in which case … well, anything’s possible.
–Okay?—
No. I have no problem with your characterization of Joe. He means nothing to me. I’m just confused as to why you’d note that he was prone to hyperbole when the bit of info in question is a date. I mean, it just struck me as being odd. What would an “exaggerated” date of death be like?
—There is no logical reason for you to multiply 14% (the percentage of men named Joseph) times 9% (the percentage of men named Joshua) and then to say that the product of these two numbers–1.26%–equals the number of men who have a father named Joseph and a brother named Jesus.—
Uh, there is a logical reason. It’s called conditional probability. And it works via the multiplication principle. Here’s a short primer.
If you want to get the chances of getting two heads in two coin flips, you multiply the probability of getting a heads for each flip: .5 x .5 The answer is .25 (That’s also the probability of getting two tails, and also the probability of either getting a tail and head, or a head and a tail).
The key to that, however, is that the two trials are independant: the outcome of one doesn’t affect the outcome of the other. The same assumption is being made with the names: that, given that you are named James, you are as likely to have a brother named Jesus as anyone else would be. Given that you are named James and have a brother named Jesus, you are still again as likely to have a father named Joseph as anyone else. Now, obviously, we don’t know if that’s really true. But it’s a pretty fair assumption: we know of no reason why it wouldn’t be true, it isn’t true of most cultures we’ve ever heard of, and it’d be a little odd if it was true.
With that assumption, one CAN simply multiply the probabilities to get an estimate. Certainly, it’s an estimate that’s only as good as the data it’s built on, which is itself just an estimate that’s pretty hard to generalize to a population. But it’s certainly a reasonable estimate given what data they have.
Sorry, Apos, but I’m a bit puzzled. Are you saying that independent distribution of names is an assumption which is likely to be true, or that it’s *not * likely to be true, but it the only assumption we can make with the data available?
If you are saying it is likely to be true, I’d have to disagree. In many societies distribution of names is frequently linked to regional origin or family connections, or both.
Okay, thank you, guys, for explaining the “math stuff”. I see what you’re getting at.
As far as I can glean about the Hachlili data, it’s just lists of names, with percentages of how common they were relative to the items she looked at. She didn’t do any “confidence intervals” or estimations, as far as I can tell. That’s why they’re only articles and not books, I think.
Well, just off the top of my head, there are societies that give the newborn baby the name of the just-recently-deceased sibling , so you could actually have had two brothers, both named “James”.
Colonial Period America was one of these societies. When a child died, the next baby to be born was sometimes given the name of the dead baby.
http://lonestar.texas.net/~mseifert/puritan14.html
That’s a good website, BTW, talking about other naming patterns and how frequent it was in Colonial America to name the baby after the parents or grandparents (as opposed to some completely new name).
So in families from that particular society, it would skew your probabilities, because a newborn baby would be slightly more likely to have been given a name that had already been given to someone, as opposed to a completely different name. And figuring out how likely it was that James would have a brother named Joshua would be influenced by whether the two boys had a father or grandfathers named Joshua. If Gramps was named Joshua, then the family was more likely to name the second boy Joshua. So in that society, at least, it would not be just as likely that the second boy would be named Joshua as any other name.
More on naming patterns in Britain and Early America.
http://www.rootsweb.com/~gamacon/naming_patterns.htm
We don’t have any information at this point about possible naming patterns among Jews in 1st Century Jerusalem, so where apparently Lemaire and Shanks would say, “Let’s go ahead and assume that a baby named James had just as much chance of having a brother named Joshua as he would of having a brother named anything else”, I would prefer to say, “We don’t have enough information to speculate.”
However, Lemaire and Shanks, being, respectively, a paleologist and an editor with heavy “Historic Jesus” agendas, prefer to make assumptions about naming patterns and to go ahead and speculate. That’s fine, as long as it’s clearly labeled as speculation–which it isn’t.
Well Cecil says Jesus was a real person, so- do we need anything else?;j
But seriously dudes- I beleive that the Roman Emperors lived- and that likely most of the Pharaohs did also- but I can do so without thinking they were Gods.
That box IS evidence- but not conclusive- but it doesn’t really matter. However, proving or disproving the box does not lend credence or doubt to whether or not Jesus was the “Son of God”. Even if the box is false- doesn’t mean Jesus didn’t exist, and even if the box is 100% legit- doesn’t mean that Jesus is the Messiah.
The first line was mean to be jocular- but it really isn’t. Most legit historians calmly accept Jeus as a real guy. Sober, non-biased sources such as Oxford do not even raise any doubts as to his “historical existance”.
If you chose to beleive in Jesus as the Son of God- all the lack historical evidence in the world won’t make a difference. If you chose to not beleive in Him- then even a copy of Pilates diary won’t convince you He is really the Messiah.