Quote:
|
Originally Posted by Sauron
|
I'm posting this here, as a follow-up on the Feedback Forum.
In case anyone is still following this conversation, the book which McDowell drew upon - and which started the cottage industry of "prophecy probability" - is Peter Stoner's work,
Science Speaks. McDowell's Ch11 of ETDAV is based in large part upon this work, and McDowell quotes Stoner frequently throughout that chapter. Well, that book is now in
online version. Stoner's grandson, Donald Stoner, has taken it upon himself to publish the work in HTML format, along with some photos from the book, book covers that show the progress of the volume through each printing, and the occasional update or scientific edit to his grandfather's original treatise.
This is going to make discussion of this work so much easier, since all participants can now have access to the work directly.
The book starts out by offering a defense of the Genesis creation account. Stoner then takes us on a tour of scientific theories that he says used to be considered correct, but have been overturned. Stoner makes several mistakes in doing so (i.e., calling the Nebular Hypothesis incorrect), but in general he seems to be an old-earth type of creationist. He plays fast and loose with the text and plasticizes it badly in order to match the genesis account up with the state of scientific knowledge; for example, he tries to resolve the order of creation conflict where the sun/moon are created after the earth and vegetation. He tries to convince us that when genesis says that God "made" the sun and moon, what it really meant was "God parted the clouds so that the moon and sun were visible".
By similar sleight of hand, he comes up with this statement:
Thus we find that the thirteen things named in Genesis are in the same order that geology finds them. We must therefore ask the question, from what source did Moses obtain this order?
But in order to get these thirteen things to all line up correctly, he engages in some very flexible interpretations and re-ordering of events. He then goes on to play word games over how long a "day" is in genesis; all stock in trade for anyone who's dealt with old-earth creationism.
Anyhow, enough of that. I'm going to cross-post this in the Evolution and Creation forum, in order to get more professional commentaries on Stoner's comments on evolution, the human fossil record, etc.
Now on to the probability and prophecy section. I have to say that I was underwhelmed with what I found here. After the big PR production that this work has received by McDowell, Hal Lindsey, Coral Ridge Ministries, I was expecting far more red meat here than what I actually found. And after reading
One Who Replies' comments here, and appreciating the level of technical terms, I also had high expectations for Stoner's signature work on this topic:
Quote:
Understanding the single and corporate laws of chance requires some hard work. Borel proposed 10(to the negative 50) as a universal probability bound below which chance could be (to a mathematical certainty) precluded. In order to reliably use probability models in context with prospective historical events, Borel demonstrated a practical need to analyze the end product with attention to 1. The Rules of Compound Probability 2. The mathematical parameters of multiple equilibria and 3. The mathematical boundaries of non-predictability (Borel demonstrated the necessary methods of avoiding non-ergocity and factoring out insignificant externalities to avoid corrupting the data).
Dr. Peter Stoner was a particularly capable mathematician and student of the work of Borel(and other mathematicians and early researchers in applied statistical analysis, which later came to be called 'operations research' and then game theory). Dr. Stoner was the Chairman of the Mathematics Dept at his college.
[...]
Dr. Stoners' work passed peer review then and continues to do so today.
Dr. Josh McDowell, a magna cum laude theology graduate, studied Dr. stoners' work and also studied the work of Borel and others,and though not academically sophisticated in advanced statistical analysis, used Borel's methods and those of Dr. Stoner to make a similar analysis of Biblical events and historical probabilities. McDowell's work was peer reviewed by the Committee of the American Scientific Affiliation as well as by Dr. Peter Stoner, Dr. Harold Morowitz of Yale University,Dr. Gerald Abramowitz, Dr. Fred Meldau, and quite frankly, has been undergoing peer review ever since. As to DR. McDowell's methodology it pases muster with peer review.
|
With a send-up like that, my expectations were set for a rather lengthy read and some time spent checking formulas.
But imagine my disappointment when Stoner presents no advanced statistical formulas or analysis whatsoever. None. Zero. Zip. Nada. In fact, he point-blank tells the reader that he will use nothing but straightforward elementary probability to discuss the prophecies:
Quote:
This chapter is being written so that the person who is in doubt about the inspiration of the Bible may be given strong evidence upon which to anchor his faith. Only a very few of the great number of prophecies which could be considered are being used, and only a very brief account is given of the fulfillment of these prophecies; just sufficient to make evaluation possible. Other books are suggested for more complete treatments. In some cases references are given to histories and other books, so as to facilitate the checking of the facts from sources outside of the Bible.
I am making use of the well-known principle of probability. If the chance of one thing happening is one in M, and the chance of another independent thing happening is one in N, then the chance that they shall both happen is one in M times N. The proof of this theory will be found in any college algebra text. I shall only illustrate the truth of the principle. Suppose one man in every ten is bald, and one man in 100 has lost a finger, then one man in every 1,000 (the product of 10 and 100) is both bald and has lost a finger. To show the truth of this conclusion, suppose you take 1,000 men at random and sort out all that are bald. Since one in ten is bald you will have just 100 bald men in the 1,000. The other 900 are not bald, therefore cannot both be bald and have lost a finger. Now go through these 100 bald men and look for men who have lost a finger. Since only one man in 100 has lost a finger, you will find only one in this group. This man then is both bald and has lost a finger, and he is the only man who fills the two conditions. In other words, we could say that the chance of a boy becoming bald is one in ten, and the chance of a boy losing a finger is one in 100. Therefore, the chance that a particular boy shall become bald and lose a finger is one in ten times 100, or one in 1,000. Thus the theory is seen to be true.
|
Basic math, involving none of the higher order analysis that OWR mentioned in his post. In addition, Stoner seems to understand that this formula above (in bold font) only applies to calculating the probability of
independent events. All well and good. But then Stoner gets the formula for dependent events wrong:
Quote:
If the events are not entirely independent, a somewhat different theory of probability must be used.
If the chance of one thing happening is one in M, and after it happens the chance that a second and related event will happen is one in N, then the chance that both will happen is one in M times N.
|
Notice the formula for dependent events presented here (also in bold) is the same formula as Stoner presented for *independent* events above (bold). Almost word-for-word the same formula! Yet the same formula cannot apply to both dependent and independent events; but here we see Stoner doing that. Now to be fair, it is remotely possible that this was a transcription error by the grandson; but given the number of revisions that Don Stoner has made (visible by his comments) it would be very strange that this key item was overlooked.
***Note that the correct formula for dependent probability can be found here.
Stoner continues with an example, and a declaration:
Quote:
Let me illustrate. What is the chance that a young man will eventually lose both his first and second fingers? We cannot find the answer by asking the questions: One man in how many men has lost his second finger? and then multiplying the two results. The loss of the first and second finger is often related. The same accident that takes the second finger is very apt to take the first finger also. We must ask the questions: One man in how many men loses one of his first two fingers? And one man in how many men after losing one of his first two fingers also loses the other? Then we may multiply these results.
This principle will be carefully adhered to throughout this treatment of probability in the following pages.
|
But this is not sufficient. If Event 1 occurs, then some probability figure must be given for dependent Event 2 occuring. Without adjusting that second figure for dependency, the dependent nature of Event 2 isn't being reflected in the final probability figure. And contrary to Stoner's declaration, the probability estimates given in this section do not reflect a knowledge of dependency between linked events. Stoner continues to mathematically treat dependent events as though they were independent, stand-alone events.
In my earlier post on this topic I pointed out that Stoner failed to present any justification for his original probability figures; i.e., why was 1/250 chosen instead of 1/300? Stoner admits this, and tells the viewer
not to expect any such justification:
Quote:
|
The probability of the fulfillment of the items of various prophecies is difficult to arrive at. The abilities used in this treatment are an attempt to give the chance of the prophetic items being fulfilled if they had been written only with human knowledge. The author has used the mean of the estimates given him by some seven hundred college students over a period of more than ten years as a starting point, but he has carefully weighed these and changed all that seemed questionable. In nearly all changes the estimates were made more consecutive. no attempt will be made, however, to defend these estimates. The reader should feel perfectly free to make the estimates which seem reasonable and conservative to him. He should then compute with his own estimates the probability of the fulfillment of the prophecies. The author believes, however, that the reader may wish to increase as many values as he would decrease and thus make very little change in the final results. At any rate his computed results will be conclusive.
|
It should be noted that these OT prophecies (and their probabilities) were not set by this class on Christian Evidences by Intervarsity, offered at Pasadena City College. That only applies to the messianic prophecies, mentioned later.
Stoner then goes on to discuss several of the OT prophecies that appear in McDowell's book; indeed, it looks like ETDAV borrowed heavily from this section, including all of Stoner's mistakes about history and archaeology.
- mistaking the mainland suburbs of Tyre for the island city;
- ignoring the fact that the Tyre prophecy specifies Nebuchadnezzar, not Alexander, as the agent of fulfillment;
- calling the prophecies about the people of Moab and Ammon fulfilled by recent events, even though those nations went extinct and haven't existed as a people for centuries;
- grossly underestimating several conquests, including Edom;
- mistaking Petra for Bozrah;
- not understanding the agriculture and topography of Samaria;
- incorrect statements about the status of Ashkelon;
- misconstruing the term "bald" when applied to Gaza;
The last section on prophecy talks about the messianic prophecies. It should be noted that only this section was considered during the aforementioned class at Pasadena City college:
Quote:
The Inter-Varsity Christian Fellowship at Pasadena City College sponsored a class in Christian evidences. One section of the work of this class was to consider the evidence produced by the fulfilled prophecies referring to the first advent of Christ. The students were asked to be very conservative in their probability estimates. They discussed each prophecy at length, bringing out various conditions which might affect the probability of any man fulfilling it. After discussion, the students agreed unanimously on a definite estimate as being both reasonable and conservative. At the end of the evaluations the students expressed their feelings thus: If any one were able to enter into the discussions and help in placing the estimates, as they had done, that person would certainly agree that the estimates were conservative. The estimates used in this chapter are a combination of the estimates given by this class on Christian evidences combined with estimates given me later by some twelve different classes of college students, representing more than 600 students. I have carefully weighed the estimates and have changed some to make them more conservative. If the reader does not agree with the estimates given, he may make his own estimates and then carry them through to their logical conclusions.
|
I'll leave the analysis of those prophecies to someone else; my interest is in ancient near east and archaeology. The book closes with several attempts to impress the reader with how big some of these probability numbers are, as well as the personal challenge to accept christianity.
Don Stoner, the grandson, seems to acknowledge that his grandfather's work hasn't stood up well on the assumptions it has made. In his own section, he writes:
Quote:
|
Critics of my grandfather's book have suggested many possible errors. Among these, that the estimates may be too high, or that the events are statistically connected - so that fulfilling one prophecy will virtually guarantee fulfillment of another. Although my grandfather repeatedly invited critics to supply their own estimates and see what happens, I have found that most critics are content to skip the exercise; so, I have included this chapter as an example of what sort of numbers a person playing the devil's advocate might obtain. Every attempt has been made to keep these estimates conservative beyond any reasonable challenge:
|
The kinds of errors that Don Stoner mentioned above (in red) invalidate any such attempt at proving prophecy with probability, yet in a misguided attempt to honor his grandfather's memory he plows ahead anyhow. He also repeats the mistake that our friend OWR made: insisting that critics provide their own numbers. But as I demonstrated, that is not necessary. Stoner has made the claim; it is up to him (or his fans) to defend it. Skeptics are not required to provide better numbers, merely to point out procedural flaws in how Stoner's math was done.
08-28-2005, 09:53 AM
For prophecy and Josh McDowell fans.....
