The Late War/Bayesian probability Article at Interpreter

[_bschaalje]

The apologists move into full damage control mode by (a) contracting an outside specialist to weigh in on the Late War (b) vying for the appearance of fair-mindedness by not dismissing the Johnson's findings outright.

I am not an outside specialist, but it's fun to be thought of that way. I've been responding to critics since 1991 when I wrote about David Holmes' paper on Book of Mormon vocabulary.

All that being said, the Schaalje piece should be taken seriously

Well thanks for that. I'm not all that serious though. Here's something else I wrote about the Late War:

The irony of the whole Late History incident (‘hidden in plain sight’) is that you have all missed the obvious conclusion, eh. The skip has just curled his stone into the centre of your house and left you laying zero. Duane and Chris Johnson are Canadians (and very likely Vancouver Canucks fans. too). The War of 1812, the war highlighted in the Johnsons’ ‘discovery,’ is the ONLY WAR IN HISTORY in which Canada has won a battle over its neighbour to the south (you can google the famous Battle of Beaver Dams, which the Canadians won after being warned that the ‘Americans are coming’ by their own FEMALE Paul Revere, Laura Secord). The galling part for Canadians like the Johnsons is that most Americans assume they won the war, the battle, everything. The thought makes a loyal Canadian’s maple syrup boil as she sits on her chesterfield watching CFL on CBC. Hence the Johnsons want to promote the “Late War” as the first sinister strike by an ignored but proud race against Americans on MDB. This is the opening volley of a new COLD war, so to speak. Yes, the Blue Jays won the World Series twice, but not even that can soothe centuries of resentment by MacIntosh-toffey loving militant Canadians like Duane and Chris Johnson, and probably Greg Smith too. Don’t be had, eh. See this for what it is.

I will try to answer some of your points during the day, especially about the

Steaming pile of horse turd. Incredibly lame that the Mormon Interpreter decided to "publish" this.

[_Equality]

My dad used to say figures don't lie but liars can figure. Also that there are lies, damned lies, and statistics.

If this guy's Bayesian analysis is scientifically sound, surely other statisticians who are not faithful Mormons could corroborate it? Duplicate the results? Come to the same conclusion? Or maybe they wouldn't. It'd be nice to see the probability that objective statisticians might arrive at when crunching the numbers. If there is any value to the exercise, one would expect that multiple analyses would converge on a consensus number. And if not, then that would also give us valuable information.

[_Always Changing]

Well thanks for that. I'm not all that serious though. Here's something else I wrote about the Late War:

So, in other words, it is way early to debunk the Johnson Bros'. work. However, the Interpreter asked you to do something, so you did, even though there is not yet enough data available to even be able to debunk it? There isn't even enough data available to support it-- it is all preliminary.

Chris Johnson wrote:

we would still caution readers to wait until our findings are verified by the peer review process, before drawing any conclusions. We encourage critical examination of our claims, and personal examination of the texts revealed by the research.

Nice work.

[_Fence Sitter]

Nice work.

LH Check your PM's please.

[_robuchan]

Hey everybody, I just determined there is a 60% chance it will rain today. I assigned values that sounded right to me and pumped it through my equation. I don't know the first thing about weather patterns, but trust me, I know stats!

[_Res Ipsa]

What about a Bayesian analysis of the probability that the Book of Mormon is an ancient text?

Why is it that I am not at all impressed with the notion of a Bayesian analysis?

Is it that a number of people have used it to support propositions that ultimately contradict each other in the larger picture?

I don't know, but I continue to find myself in shrug mode.

Given that the prior probability that the Book of Mormon is an ancient text is vanishingly small, the impact of this additional evidence is small because it confirms a substantial amount of pre-existing evidence. So, there is a kind of a "well, duh" aspect to the new study in that sense.

The main impact, as others have commented, is on apologetic arguments based on the structure of the Book of Mormon's language.

[_Always Changing]

Given that the prior probability that the Book of Mormon is an ancient text is vanishingly small, the impact of this additional evidence is small because it confirms a substantial amount of pre-existing evidence. So, there is a kind of a "well, duh" aspect to the new study in that sense.

This is exactly why we can afford to be rational and objective in examining the data.

My whole project has become an intellectual exercise, answering how many different ways I can prove that Mormonism is a man-made religion, and not a very good one, at that.

The Johnson Bros' data does conclusively show that the Book of Mormon is not unique in its use of KJV language. In fact, it is somewhat inferior to other productions of its type from between 1770 and 1830. But we already had access to that information from Shalev's work.

[_bschaalje]

I am sorry if my blog on Interpreter upset anyone here. That was not my purpose. My purpose was also not to make up numbers that would lead to a predetermined conclusion.

The purpose of my blog article was simply to point out that Bayes’ rule can help in sorting out the confusing claims and arguments and observations associated with the The Late War and the Book of Mormon. There is another (but still equivalent) form of Bayes rule that might make this point more clearly and less controversially than the form in my blog:

odds(H|E)/odds(H) = prob(E|H)/prob(E|H*)

We don’t need to guess individual values for any of the quantities here to show how Bayes’ rule can help us in thinking about The Late War and Book of Mormon authorship and influence questions.

This equation says that the ratio of (posterior odds of H to prior odds of H) is equal to the ratio of the [probability of a true positive (sensitivity) to the probability of a false positive (specificity)]. If a true positive is much more likely than a false positive, the posterior odds increases relative to the prior odds. If a true positive is about as likely as a false positive, the posterior odds stays about the same as the prior odds.

This is important in the Late War situation because there is not much information about how sensitive and specific the Johnsons’ procedures are. In my opinion, sensitivity is decreased and specificity is increased by at least two features of the Johnsons’ study:

1. the massive search model tends to produce false positives.
2. the dependence of weights on a randomly selected corpus (from books of many genres between 1500 and 1830) tends to affect sensitivity and specificity in unpredictable ways; I can conceive of ways in which sensitivity is decreased and specificity is increased.

We don’t have much to go on at this point when evaluating the meaning of the Johnson’s work. If Bayes’ rule can focus our attention on aspects of the procedure that need to be assessed and improved, we are the better for it.

[_Maksutov]

I am sorry if my blog on Interpreter upset anyone here. That was not my purpose. My purpose was also not to make up numbers that would lead to a predetermined conclusion.

The purpose of my blog article was simply to point out that Bayes’ rule can help in sorting out the confusing claims and arguments and observations associated with the The Late War and the Book of Mormon. There is another (but still equivalent) form of Bayes rule that might make this point more clearly and less controversially than the form in my blog:

odds(H|E)/odds(H) = prob(E|H)/prob(E|H*)

We don’t need to guess individual values for any of the quantities here to show how Bayes’ rule can help us in thinking about The Late War and Book of Mormon authorship and influence questions.

This equation says that the ratio of (posterior odds of H to prior odds of H) is equal to the ratio of the [probability of a true positive (sensitivity) to the probability of a false positive (specificity)]. If a true positive is much more likely than a false positive, the posterior odds increases relative to the prior odds. If a true positive is about as likely as a false positive, the posterior odds stays about the same as the prior odds.

This is important in the Late War situation because there is not much information about how sensitive and specific the Johnsons’ procedures are. In my opinion, sensitivity is decreased and specificity is increased by at least two features of the Johnsons’ study:

1. the massive search model tends to produce false positives.
2. the dependence of weights on a randomly selected corpus (from books of many genres between 1500 and 1830) tends to affect sensitivity and specificity in unpredictable ways; I can conceive of ways in which sensitivity is decreased and specificity is increased.

We don’t have much to go on at this point when evaluating the meaning of the Johnson’s work. If Bayes’ rule can focus our attention on aspects of the procedure that need to be assessed and improved, we are the better for it.

Thank you, professor, for your pertinent and professional response. It deserves a thoughtful analysis.

[_robuchan]

I am sorry if my blog on Interpreter upset anyone here. That was not my purpose. My purpose was also not to make up numbers that would lead to a predetermined conclusion.

The purpose of my blog article was simply to point out that Bayes’ rule can help in sorting out the confusing claims and arguments and observations associated with the The Late War and the Book of Mormon. There is another (but still equivalent) form of Bayes rule that might make this point more clearly and less controversially than the form in my blog:

odds(H|E)/odds(H) = prob(E|H)/prob(E|H*)

We don’t need to guess individual values for any of the quantities here to show how Bayes’ rule can help us in thinking about The Late War and Book of Mormon authorship and influence questions.

This equation says that the ratio of (posterior odds of H to prior odds of H) is equal to the ratio of the [probability of a true positive (sensitivity) to the probability of a false positive (specificity)]. If a true positive is much more likely than a false positive, the posterior odds increases relative to the prior odds. If a true positive is about as likely as a false positive, the posterior odds stays about the same as the prior odds.

This is important in the Late War situation because there is not much information about how sensitive and specific the Johnsons’ procedures are. In my opinion, sensitivity is decreased and specificity is increased by at least two features of the Johnsons’ study:

1. the massive search model tends to produce false positives.
2. the dependence of weights on a randomly selected corpus (from books of many genres between 1500 and 1830) tends to affect sensitivity and specificity in unpredictable ways; I can conceive of ways in which sensitivity is decreased and specificity is increased.

We don’t have much to go on at this point when evaluating the meaning of the Johnson’s work. If Bayes’ rule can focus our attention on aspects of the procedure that need to be assessed and improved, we are the better for it.

This is fine. I think you WAY overstepped your bounds when you pulled out completely random values of .5, .5, .8 as assumptions and then described them as "generous to the critics", "skew things towards the critic's position", and "worst-case (from our point of view)", implying they are valid and not simply to illustrate your very generic point about false positives.