Jump to content

Physics Guy

  • Content Count

  • Joined

  • Last visited

Community Reputation

180 Excellent

About Physics Guy

  • Rank
    Seasoned Member: Separates Light & Dark

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Of course I'm not interesting in making my case. Why on Earth should I be? No-one is paying me to research Mormonism. I'm a casual passer-by who saw something interesting in the Mormon shopwindow and ducked in to look closer. On closer looking I wasn't impressed. Sorry 'bout that, but it's true. Every few weeks I pass by again, and sometimes I look in again. I thought that's what you wanted: some interest. I'm the potential customer, not the merchant. You're blaming me, the customer, for not working harder to dig around in the back room of your shop in order to find the good stuff. What the heck kind of attitude is that? If you have better products back there, why not put them in the window?
  2. So far all I've seen are Mormon apologists asserting that the Book of Mormon is too complex for Smith to have composed it, and all I can do is shrug, because I just don't have that impression at all, to the point where I really just have no idea where the apologists are getting this idea that their Book would have been so hard for Smith to write. The only complexity I can see in it is the kind of complexity that is quite easy to make up, because it comes from rambling away without a clear plan. That's just my own subjective impression, but until "complexity" gets pinned down in specific terms, competing subjective impressions is all that there is, here. If someone were to be specific about exactly what objective features were supposed to have been hard for someone like Smith to have composed, then one could go and look at other texts by other writers and see how often those objective features have been achieved. One wouldn't even have to have the problem of comparing gifted/educated/professional writers with an unschooled farm hand, because these days there are web sites that get writing posted by all kinds of amateur authors. Amazon also lets people self-publish at $0.99 a book, so with a modest budget one could simply buy a large pool of finished unprofessional writing for comparison. It would be modern writing, but we're presumably not talking about comparisons of spelling and dialect. Crafting an intricate plot full of diverse characters should not come any more easily to the average person today than it did in the past. It might not be so hard to find out exactly where the Book of Mormon would lie on the spectrum of amateur writing, if it were considered as an amateur work of fiction. If it were totally off the usual charts for numerous objective features, that might be a strong argument that Smith didn't just compose it himself. It would be important to formulate the objective features in a way that even non-Mormons found reasonable, however. If one ended up effectively defining "complex" to mean "exactly like the Book of Mormon" then of course no other works would be so "complex". I have no interest whatever in undertaking this myself, but there seem to be quite a few Mormons working hard to convince non-Mormons about their faith. This might be a good project for some of them, because it would turn an old subjective argument into something more convincing—or, if the results turned out to be disappointing, they would at least let Mormon apologists know that they should drop an unconvincing argument and concentrate on better ones.
  3. One might well suspect that, given how outrageously small a chance is obtained in this paper. But it really isn't so. That's just how dangerously fragile their methodology is, multiplying so many probabilities together. That manifold multiplication is how they get their impressively low probability as a final result, but it inherently means that their result is incredibly sensitive to errors that are also multiplied many times. If ten of their discrepancies all get promoted from 50-fold effects to 50-million-fold effects, which could be perfectly reasonable, then that's a factor of 10^60 right there. And if treating 100 individual features as independent random guesses gives odds like 10^-100, but the 100 features are really only 10 independent suites of tightly correlated choices, then the real odds to be inferred from these features are only around 10^-10. So redoing their calculation correctly could realistically change the result by a factor like 10^90 or even much more. Considering JarMan's point that most of the paper's "hit" features are common features of ancient societies that Joseph Smith likely knew (he at least knew the Biblical ones), it might be that all those hundred-plus features only have a single collective probability around 10%, instead of 10^-100. That whole gigantic odds calculation could be nothing but smoke. Between the above two effects there is easily a factor of 10^150 up for grabs here. The final result is completely in doubt, because the same multiplication effect that gives them their stunningly small probabilities can come right back and bite them very hard. The carelessly applied Bayesian methodology that seems so amazingly powerful is in fact extremely fragile. Looking hard at the data can absolutely turn odds of 10^132 or whatever against into very high odds for the Book of Mormon not being authentic. The real problem with this paper is the failure to understand that extreme fragility. It would be one thing to say, Hey, look at this: if we take a wildly optimistic and naive shot at estimating Bayesian odds against the Book of Mormon, we can make the odds come out this insanely low! It's okay to say that. What you have to say in the next breath, though, if you're going to be remotely responsible, is that if you take a less optimistic shot then the odds come out extremely differently. Instead these guys seriously present a wildly optimistic shot with a wildly unreliable methodology as if it were rigorous. This is crackpot territory. It's really embarrassing for Interpreter. To put it in appropriately Bayesian terms: observing this single paper as a data point is enough to lower the odds that Interpreter is a legitimate academic journal by a large factor. A lot of careful hard work by other authors on other papers in Interpreter gets tarnished by association with this one piece of folly. "Oh, your paper is in Interpreter? The journal that published the 10^132 odds that the Book of Mormon is historical? Ha ha ha." Not what a serious Interpreter author wants to hear when they cite their publication—and not what Interpreter wants potential future authors to think, when they're deciding where to submit their next work. Whoever did peer review on this one really let the institution down.
  4. Having looked a bit more at the article I see another serious problem, besides the extremely grave problem that many obviously correlated features are treated as independent random guesses. It is that the maximum weight which can be assigned to any discrepancies between Nephite and Mayan societies is artificially and arbitrarily set at fifty. According to the authors' methodology there could literally never exist any discrepancy, however severe, which could make it more than fifty times more likely that the Book of Mormon was invented rather than authentic. This is absurd. Suppose I read a purported description of American society which makes it out that the office of US President is hereditary. Even considering only that one glaring error means that the odds that this purported description is bogus are way more than 50:1. It's not as though one political theorist in fifty thinks the presidency is hereditary. I'm not sure even one in a million thinks that. So artificially limiting the damage that can be done to a book's credibility to no more than a factor of fifty is completely unjustifiable. Serious discrepancies can boost the likelihood of being fake by way, way more than a factor of fifty. Insofar as the authors multiply many small probabilities together when they are obviously correlated, the authors wildly exaggerate the unlikelihood of Joseph Smith inventing a text with this level of correspondence to Mayan society. And insofar as they artificially limit the disconfirmatory effect of discrepancies, they wildly underestimate the severity of adverse evidence like Nephite steelmaking. And since both of these serious methodological flaws have exponentially strong effects on the paper's results, they really do undermine even the apparently overwhelmingly strong case that the authors have made.
  5. This points to a basic problem with the paper's basic methodology of multiplying a lot of probabilities together. You can only do that if the probabilities are all independent. If instead all the different features of Book of Mormon civilizations represent one common feature—"be like a typical ancient civilization"—then only one probability is really involved. You can estimate the probability that Joseph Smith would have an accurate picture of ancient civilization as high or low. It can't be too low, because he at least knew the Bible and the Bible portrays a lot of typical features of ancient civilizations. Even if you decide that probability is pretty low, however, you don't get to multiply a hundred probabilities together as if every little feature of Book of Mormon society would represent a separate and unrelated guess on Smith's part.
  6. Ideally, yeah. Maybe one could find other arguments in support of accuracy. Reporting grammatical errors, for example, would indicate that the reporter was not correcting the preacher's words. If the reporter simultaneously praised the preacher sincerely, then that would tend to rule out the possibility that the errors had been inserted by the reporter to mock the preacher. So it might be feasible to rule out Carmack's argument with this kind of data, by finding lots of reported EModE in old sermons, but not feasible to rule out the alternative hypothesis of overdone archaism, because if one found no EModE expressions in the reports of sermons then one couldn't tell whether the reporter had consciously or unconsciously corrected the preacher's bad King James grammar. So to support Carmack's argument might indeed need multiple corroborating accounts, which would be harder to find. Empirical evidence is often an unfair playing field that way, with decisive evidence one way being much harder to find than the other. As I said, it's going to be hard to close this loophole. It really is a big and obvious loophole, though. As soon as you think at all seriously about someone faking the Book of Mormon in the 19th century, the possibility that they might mess up the dialect comes up right away. As soon as you think at all seriously about exactly how a dialect might get messed up, you realize it would be a matter of systematic errors through applying mistaken grammatical rules. Closing the big loophole will be hard. It may well be impossible. If it is impossible, then that won't mean that there's proof that the Book of Mormon was composed in the 19th century. It will mean, though, that there is no proof on linguistic grounds that it wasn't.
  7. It would indeed be worth looking at sermons by uneducated preachers, if their speech has been preserved accurately. This was a good suggestion of yours, and Carmack seems to have recognized that it would be nice to do something along these lines; but he seems to think of it as an added luxury. He doesn't seem to realize that he really needs to close this loophole because until he does, he does not have a case against 19th century composition. It's not going to be easy to close this loophole. I'd be surprised if one could get a large enough sample from preachers to be confident of a pattern to which Smith would have had to conform. If the preachers were mostly just quoting the King James Bible, for instance, even in pastiche, then their accuracy in matching KJV grammar would not show what would have happened if they had tried to improvise an entirely new text of considerable length in a KJV style. On the other hand one might kill the whole Carmack argument against 19th century composition with one shot here, if lots of EModE usages turned up in rural preaching. So on the scientific principle of trying to falsify theories, rural sermons should be a high priority for Carmack to check. The possibility that Smith actually spoke archaic English in his local rural dialect is another big loophole in Carmack's argument. I haven't been talking about it because other people (mainly you, Clark) seemed to be covering it, but it's clearly another major issue. I can't blame Carmack for not (yet?) having covered these loopholes. They're going to be an awful lot of work to cover, and it's not clear that it will ever be possible to cover them convincingly because we may just never be able to find enough data. But you can't just ignore them and still claim to have demonstrated that Smith could not have composed the Book of Mormon because he didn't know EModE.
  8. I don't understand how you're using this concept of "support". The hypothesis that I don't think you've ruled out is that Smith tried hard to dictate just like the King James Bible, but through lack of expertise and lack of time to revise, he made systematic errors—he got King James grammar wrong. His divergences from King James usage would thus not be at all like those of educated writers producing texts for entertainment. Neither would they simply be random errors, however. He would be following systematic rules that were incorrect for his target dialect—but also happened to be more correct for an earlier period. I don't see how you can rule out this hypothesis by looking at any published pseudo-Biblical works by educated writers who were not trying to deceive. There is no reason why their pseudo-Biblical language should have anything to do with the systematic errors in imitating King James by an uneducated deceiver who didn't edit his text. I don't see how you can rule out this hypothesis by looking at the Bible or any other late modern texts, either. The hypothesis is that Smith made systematic mistakes. So his language would not simply be Biblical, nor would it resemble late modern texts that did not made mistakes in their own proper grammar. And I don't see how you can rule out this hypothesis just by accumulating multiple examples of Early Modern usage in the Book of Mormon. The hypothesis is that Smith's errors in imitating King James usage were systematic. Like anyone trying to speak in a dialect they don't properly master, he followed incorrect rules. If he got something wrong (for King James) once, he would get it wrong that way again and again—just as if he were following the right rule for Early Modern English. I don't see where you have even addressed this possibility. You seem to be simply ignoring it.
  9. My position is that you haven't shown how improbable the coincidence is. You've just shown (I assume, pending second opinions from experts) that there is a coincidence. My point is not that oral dictation makes it easier to produce archaic language. It's that oral dictation makes it easier to mess up archaic language, for example by producing Early Modern forms when you're aiming for King James Bible. I disagree with the bolded part. I don't pretend to be able to prove that the odds of hypercorrection are high, but I find it far from obvious that they are low. Can you cite any studies to support your low estimate?
  10. I'm not the one claiming to rule out an authorship theory on linguistic grounds. You are. I'm just pointing out loopholes in your argument. Just as a matter of logic, I don't have to do anything to prove that authors limited their archaism or that Smith overdid his fake archaism. You have to provide strong evidence that they almost certainly didn't and that he almost certainly could not have. Or else you have to back off from your claim that Smith could not have written the Book of Mormon. This isn't geometry. You don't need to prove things to mathematical certainty. But if you want to establish your conclusion beyond a reasonable doubt, you need to cover reasonable loopholes like mine. Those other authors clearly differed from Joseph Smith in ways that might reasonably affect their ability to produce controlled archaism. And how do we know that an author like Smith would not have produced grammar like the Book of Mormon by systematic error, hypercorrecting and overgeneralizing in a clumsy effort to imitate the King James Bible? Maybe the possibility seems remote to you, but as far as I know there has been no actual analysis of what fake archaism by uneducated con artists tends to be like. Raising a completely un-analyzed issue that is crucial to your conclusion is pointing out another reasonable loophole. I'm not saying anything at all against the hypothesis that someone from the 16th century wrote the Book of Mormon. I'm only attacking your apparent claim to have established your hypothesis as probable fact. If you can't cover the reasonable loopholes, you haven't established the conclusion beyond a reasonable doubt.
  11. I do strongly believe, myself, that there is no divine source for the Book of Mormon. I am not particularly committed to advocating that position. I do not feel that I have enough evidence or argument on my side that it should persuade convinced Mormons, and anyway I don't care what they believe. I'm not going to admit that the Book of Mormon has remarkable archaism exceeding that of the King James Bible or pseudo-Biblical texts, because I lack the expertise to make an informed judgement either way on this matter. I believe I have said before on this forum, and I happily say it again now: I do not challenge Carmack's claims that the Book of Mormon has remarkable archaism exceeding that of the King James Bible or of pseudo-Biblical texts by authors who were not at all like Joseph Smith. For the sake of argument I am perfectly willing to stipulate to these claims, pending more informed judgement by Carmack's trained peers. My point is that these claims do not contradict 19th century composition, because it has not been ruled out that fakery by an author like Joseph Smith or his associates could have produced similar remarkable archaism. This is the missing link in the linguistic argument against 19th century composition and without this link the chain does not hold. The fact that this simple logical criticism is not even being acknowledged, but instead is being deflected by accusing me of bias, is a worrying sign. If my simple criticism is misplaced, it should still be obvious that it's a natural issue to raise, and its simple rebuttal should just have been given.
  12. Several papers have been published in Interpreter to show that some features of the Book of Mormon language could have been composed in the 16th century, but that's not the same as showing that it couldn't have been composed in the early 19th century. The argument seems to be simply that the hypothesis of 16th century composition fits the Book's grammar so well that we don't have to ask how well any other hypothesis might fit it also. Empirical reasoning doesn't work that way, though. The history of science is full of dead theories which were all consistent with large bodies of evidence. They just happened to be inconsistent with some other evidence, and so they were replaced by other theories that could account for all the evidence instead of just some. Accumulating evidence consistent with your hypothesis is never enough to establish it. You have to present evidence against the alternatives to your hypothesis. So far I've only seen one paper, again in Interpreter, which even tries to show that 19th-century fake archaism could not have come out with grammar like the Book of Mormon. That paper examined pseudo-Biblical works written by educated authors with the intent of amusing book-buyers, rather than texts dictated by uneducated authors with the intent to deceive. So its conclusion that fake archaism could not have produced the Book of Mormon's grammar is like concluding that orange fruits can't exist, because all the fruits in a sample of apples were red. Is there anything more than that, not to show that 16th-century composition could have produced Book of Mormon grammar, but to show that 19th century fake archaism could not have done it?
  13. If something weird is going on for 25 years, I don't stop wondering why just because longer time scales also exist. For heaven's sake: this is a steady linear trend that persists across a total change of over a hundred standard deviations in the annual change! A linear function can have any slope, but the size of increase is certainly not irrelevant to whether something is linear. It sounds as though the Taylor series stuff didn't mean anything to you. If it really didn't, we're going to have a hard time discussing functions. Perhaps you could read the Wikipedia article on the subject? It's pretty clear, I think. At the midpoint of a logistic curve the curve looks linear ... because it is linear, there. A curve can't appear to be linear but be nonlinear deep in its soul. Curves don't have souls. They're just shapes. The way they look is the way they are. In this case the church membership has not grown linearly throughout its whole life, and it may well have stopped growing linearly in the past few years. Lots of things do show changes over time in the way that they grow. There's always a reason, however. Exponential growth levels off because a population reaches the limit of its food supply, or a predator species proliferates, or something like that. Scientists don't plot logistic curves and then simply sit back smiling as if the curve itself constitutes its own explanation. We ask why the growth is exponential in the early phase, why it straightens out into linearity, and then why it plateaus. We consider each of those questions important. If growth stays linear for an extended period, we wonder what's stabilizing the system in that behavior. (I can't believe I just wrote "a hundred standard deviations". That's crazy. But the standard deviation in yearly gain is something like 50-60 thousand, and the total gain is something like 9 million. Easily a hundred sigma. Does it really not register how strange this is?)
  14. Every smooth function looks linear on short scales, yes. All smooth curves become lines, when you zoom in. That's good to know, but even better is to understand why it is so. It's called the Taylor series. If you've never heard the term, you might Google it; if it rings a faint bell, you might review your freshman calculus. As you narrow the range over which you are looking, the higher order terms in the Taylor series all get smaller faster than the linear term. So if you narrow your view range enough, the linear term will be all that is left. The narrow field of view is the reason for trivial universal linearity. Whenever total change is small, in a smooth function, what small change there is will look linear. In this case, however: Church growth looks linear over at least 25 years. During this time the church increased in size by a factor of more than 250%. This is not just zooming in on a smooth function. It is not taking a narrow field of view. It is not a small total change. Church membership more than doubled by marching steadily up a straight line for a generation. If you don't want to recognize a pattern so clear and simple, why are you even trying to analyze data? What's the point of the exercise, for you? If this strikingly linear growth doesn't count for you as finding something, what kinds of patterns are you hoping to find?
  15. Linear growth is a simple fact about what happened during that thirty-year time frame. Perhaps you can explain what you mean by "a thing".
  • Create New...