New member.  First post.

There was a point in the interview when Paulos said something that really struck me: The analogy of the deck of cards as it relates to biology.  Either I don’t understand what he’s getting at (which is possible) or his comment betrays a naiveté which is rather startling in a person with such good credentials.

As I understood his argument, he says that the extreme unlikelihood of the order we see in biology having been produced by chance mutations is irrelevant because biology has to look like something.  A specific biological process or configuration might seem prohibitively unlikely, but if we consider the very large pool of possible configurations and processes, we see that one of them is a certainty.

The analogy of the deck of cards illustrates this.  You are very unlikely to get any certain sequence of 52 cards by shuffling them, but you are assured of getting a sequence.  The biology we see is simply the hand of cards nature happens to have dealt.  It could have been a different hand, but it would have been a hand nonetheless.

I think I got that right?  If so, the argument is obviously and fatally flawed.

Surely Paulos is aware that simply getting an arrangement of biological parts is no more useful than getting just any old hand of cards.  In any card game, you aren’t after a hand.  You’re after a winning hand.  The vast majority of hands are weak or worthless (which is what makes the game possible).  Similarly, the vast majority of arrangements of biological matter are non-functional (as evidenced by any roadkill you might pass on your morning commute).

If you deal out cards, you’ll get a hand, sure, but it will most likely be a losing hand.  Paulos speaks as though every hand is a royal flush.

Does anyone else see this analogy as deeply flawed?

Well, there are a number of possible arguments he could be attacking here. You’re right that his argument doesn’t relate to the “747 from a junkyard” argument (that one is dealt with much better by Dawkins in Blind Watchmaker IIRC). Rather he’s attacking the argument that there’s such a vanishingly small probability that there would be humans on earth that it must have all been planned from the start.

That sort of inchoate argument is akin to saying that it’s so vanishingly improbable that I dealt out these 52 cards in precisely that sequence that it must have been planned from the start.

That’s only true if we regard all sequences as being equal.  If we define criteria for winning or losing sequences, the story changes very quickly.

Mr. Tweedy - 31 January 2008 03:46 PM

Does anyone else see this analogy as deeply flawed?

Yeah, well I agree it has that weakness but I also know the thread he’s trying to connect to. He returned to it later that people see patterns in the Bible text if they look hard enough—or people see a vision of the Virgin Mary in the shadows on the wall, etc.  I agree the evolution one doesn’t really work.

A lot of the stories in the news rely on reporting something unusual that happened that day—something unusual happens every day, so they report on it, and it’s a particularly unusual day without any unusual occurences.

On a mathematical example I remember a number of years back there was a story in the NYT about a woman who had won the lottery for the 2nd time. Incredible you say!  Just calculate the odds.  This is exactly like the deck of cards.  When you estimate how many lottery tickets had been sold over how long a time, it was just about time for the first person to have won twice.  Basically there had been enough lottery winners that it was no longer improbable that someone win it twice.

Mr. Tweedy - 31 January 2008 03:46 PM

Paulos speaks as though every hand is a royal flush.

Nope. Paulos speaks as if every hand is a hand. That’s all. In a card game we decide what is the winning hand, in life the decision is made by natural selection.

Mr. Tweedy - 31 January 2008 04:10 PM

That’s only true if we regard all sequences as being equal.  If we define criteria for winning or losing sequences, the story changes very quickly.

Why?. The point is that we cannot calculate probabilities after the fact because any final state would be very improbable, no matter if the sequence is a winning sequence or a losing one. And, of course, trying to calculate probabilities having just one sample is ridiculous.

George - 31 January 2008 06:29 PM

Mr. Tweedy - 31 January 2008 03:46 PM

Paulos speaks as though every hand is a royal flush.

Nope. Paulos speaks as if every hand is a hand. That’s all. In a card game we decide what is the winning hand, in life the decision is made by natural selection.

That’s not how natural selection works.  Natural selection is only able to distinguish between the best of a group of winners.  If a group of functional configurations is provided, then the most functional will be naturally selected, but amongst a group of non-functional ones there is nothing to select.  If you have a billion configurations that don’t work, none of them has any survival advantage over any of the others.  In life, you’ve got a card game with an infinite number of losing hands and a relative few that win.

For instance, it’s very easy to get an organism to mutate.  Mutation happens all the time, even under the most normal circumstances.  That’s dealing out hands.  But most of those hands are losers, as evidenced by people with Down’s Syndrome (etc).  Natural selection is good at picking out the best of any beneficial mutations, but it can’t make such mutations more likely.  It recognizes the winning hands, but it doesn’t produce them.

Cards are actually a good analogy.  Take the game of solitaire.  In this game, there are many winning and many losing hands.  Of the winning hands, some are better than others.  An exceptionally good hand will allow the player to win with just a few moves, while one that isn’t as good will allow a win only after a lot of tedious rearranging.  Among the winners, we can define which are best, which is what natural selection does.  But is any of the losing hands better than another?

Only winners are eligible for selection.

Mr. Tweedy - 01 February 2008 08:53 AM

That’s not how natural selection works.  Natural selection is only able to distinguish between the best of a group of winners.

I’m not exactly sure if I agree or not—it could be the wording here. At any rate, natural selection doesn’t distinguish between a group of winners. It distinguishes between a group of organisms, and the ones who reproduce more are the winners.

That is, unless you are using “higher fitness” to mean “winners”. That’s OK generally, except that there are cases in which the lower fitness organism ends up winning. It doesn’t happen often, but it does happen. E.g., the stronger animal may get struck by lightning and the weaker one reproduces.

It’s not easy to tell a priori which are the traits which provide higher fitness, BTW, and it cannot be done without also looking to the environment. That’s to say, it might well be that in some environments having Down’s Syndrome is a positive benefit, that is, one that provides higher fitness.

As I say, I am not precisely following your post so not sure if this makes a difference to what you’re arguing.

The original context of the cards analogy was as a refutation of the claim that the formation of biological features by random processes is prohibitively unlikely.

No matter how you peel out the deck, whatever results is going to be extremely unlikely.  That’s another thing connected to the creationist misuse of probability.  They say ‘Well, to go from A to B, from this biological entity to that one, you’d have to have this mutation which has a certain probability, that one which has a certain probability and so on,’ this whole sequence of probabilities.  The likelihood of all those mutations occurring (assuming independence) is the product of all those probabilities, which is extremely low.  So they say ‘Okay, that couldn’t have happened.’ But, again, something’s got to happen and the fact that any particular thing is unlikely doesn’t mean that the whole process is therefore dubious.

So a “winning hand” in this case is a working biological feature.  (It doesn’t matter specifically what the feature is, so long as it works and benefits the organism.)

Paulos essentially claims that any series of mutations will yield a useful feature ("something’s got to happen").  Every hand is a winner.  That is simply ignorant.

Hmm. I interpret the quoted paragraph differently. He’s not trying to refute the claim that the formation of biological features by random processes is unlikely. He’s trying to refute the claim that the formation of this particular biological feature is unlikely. E.g., it’s supposedly unlikely since in order for us to get this particular feature we had to go through a particular sequence of mutation events, every one of which is itself of low probability.

His point is that no matter what final state we get to, no matter what collection of biological features we have, each one will appear in retrospect to have been vanishingly unlikely. Or as SJ Gould would say, if you turned back the clock and re-ran the whole thing again, it’d be vanishingly unlikely to get precisely the biological features you see around you now.

As for the question of useful features ... again, what makes a feature useful is its particular environment, which includes its competition. A feature that might appear suboptimal in our environment (Down’s Syndrome, let’s say) might in fact be a “winning hand” in some other environment. Perhaps all the other humans are infertile, for example. I’m just saying this for clarification, since I’m not sure it bears on the argument.

Prof. Paulos’ mention of the “Yeah” religion reminds me of the religion I started a couple of months ago which I called “LOTU” for “Laws OF The Universe”. It’s a religion for atheists and critical thinkers. http://thelotu.com/

retrospy - 29 January 2008 10:27 AM

Kirk,

Sorry not to have replied sooner. Most all the parts of your reply were important, hence the length.

It appears your critiques of this book are that it doesn’t appeal to the upper echelon of logical reasoning.  I missed the point where this was Paulos intention.

To ask that the things offered as proofs are in fact proofs as written isn’t asking for much. Otherwise, I have to take Paulos’ arguments in the book on his say-so.

I can’t offer rigorous critique of any logical fallacies.

You are *much* too modest. You had the ability to understood the Pythagorean Theorem in eighth grade or earlier. If mathematical-style arguments *can* be understood by any inteligent person, then there is no-one better than Paulos alive to do it. Otherwise, we must accept mathematical proofs on the mere authority of mathematicians. I thought a slogan around here was ‘on the word of no-one’? Well, just because Paulos agrees with you or me or anyone doesn’t mean he’s given a single good argument; and just because you (and Steve Weinberg for that matter) *like* the conclusions of his arguments is a very weak reason to *accept* them - although to accept the word of authorities isn’t a *terrible* ground for believing, just a weak one.

I can say that your argument sounds an offal lot like a case for throwing out the baby with the bathwater.  Not to mention, you repeat that this book was hastily written, while it appears your critique of this book falls victim to the same logic.

An offal lot, eh? You accidental poet, you. True, the reviewer mentioned ‘hasty’ - i shouldn’t have. It doesn’t matter to a reader how long it took to write the book - good or bad is good or bad.

I don’t doubt that after you read this book you can find errors.  My argument is that sometimes errors are necessary to make a specific message palpable and set an individual on the right course for further inquiry.

No, that’s about as wrong-headed as it can get. Deliberate or lazy errors don’t reliably promote the truth no matter how small the ratio of error to truth. Perhaps you only meant “Don’t pick at small details; sometimes boring material needs punching up.” That *may* be so. All my professional life I’ve sold things - shirts, chemistry, George Berkeley - and i never had to lie about my products, nor has making mistakes encouraged the common . But if *you* insist that errors are *necessary* at times to enlighten people, it may indeed be so.

(1) a book that is based 5% on logical fallacies that makes a reader 20% less likely to fall victim to a logical fallacy.
(2) a book that is based 1% on logical fallacies that makes a reader 5% less likely to fall victim to a logical fallacy.

I would argue that they are both helpful books and a world that has both choices is better off than a world that has just one.

Alas, how do we know that either of these profiles match the book Irreligion? In fact it seems the ratio is much higher than 5%. Just on the arguments presented in the text reviews shown here, even the extract from the *sympathetic* choices have deep and confusing problems. The cracks in the book are pretty deep and neither about little things nor just for specialists to chew over. You understood the critiques i’d made, and those reviewers who pointed out problems, didn’t you?

Also along this point, what peaks some people’s interest to open a book may not work for others.  For instance, I have some ex-college roommates who have shown increasing interest towards atheist literature.

Scott, this is exactly why i’ve jumped into this thread. But I’m the teacher who’s being asked to recommend a book. It doesn’t really matter to me that i’m a theist, and my student’s not. I still want to recommend a book that *works* - if by ‘work’ is meant it purports to have arguments and proofs, well then I cannot recommend *Irreligion*. That’s a pity: the books I do recommend to young atheists are not so interesting to read, although they are more informative.

Bertrand Russell is a good choice, btw. Try also John Perry. He has a decent website too. He also has a little book from Hackett Publications called Dialogue on Good, Evil and the Existence of God. I happen to think that has the best exposition of a good atheistic argument about suffering. He also makes use of statistics in his argument, and correctly identifies the difference between a population and its members, a difficult concept for non-specialists to keep straight. Perry’s also a Sartre scholar, too: there’s a funny picture on his webpage of him on a flume ride as ‘John Perry descends into nothingness’.

Now then! As a theist, and as a putative expert (ho ho, har har) on reasoning, i happen to think it has all the defects of any of those non-theistic arguments, and its criticism of the theistic position is also defective. However!! I would indeed recommend it to a student.

So, there’s no good reason to recommend this book. First, there’s at least one other easy-to-read book out there (Perry’s) that does it better on at least one mathematical argument; second, it’s a little strange to further atheism and freethinking at the expense of argumentation, and at the expense of muddling a person’s head with things that are called proofs but aren’t. Lastly, the book might help make more atheists. But if the sampling of arguments from the reviews is any indication, then a fat lot of good *Irreligion* will do for a freethinker, unless he wants to fall back on an authority, and how free is that?

Kirk

dougsmith - 27 January 2008 09:42 AM

I believe that the only “new atheist” book that got a decent review from the NYTimes was Hitchens’s, for some odd reason. I think it was more “literary”.

Paulos has a letter to the editor responding to his book review
[Here - Feb 3 2008 NYT Book Review Letters Section]

I first read it in the paper but had a hard time finding the electronic version to post a link —couldn’t find it from NY Times WWW site.

Paulos responds specifically to the charge that he is “innocent of theology”, and I think gives a good analogy with an astronomer saying astrology is ridiculous—and no one really expects the astronomer to be able to draw up horoscopes and know the arcania…

Jackson - 03 February 2008 01:18 PM

I had to find it from his page—couldn’t find it from NY Times WWW site.

Well, it was published in the Book Review.

Jackson - 03 February 2008 01:18 PM

Paulos has a letter to the editor responding to his book review
[Here - Feb 3 2008 NYT Book Review Letters Section]
Paulos responds specifically to the charge that he is “innocent of theology”, and I think gives a good analogy with an astronomer saying astrology is ridiculous—and no one really expects the astronomer to be able to draw up horoscopes and know the arcania…

To continue your analogy: While it is true that it would be burdensome to expect that most astronomers know astrology, there is one astronomer who ought to know - the one who writes a book entitled ‘Why the Arguments For Astrological Influence Don’t Add Up.’ That astronomer, when he publically declares not just his ignorance of astrology but his pride in that ignorance, would be considered grossly incompetent to publish a book on astrology. Does this analogy still hold?

I shall pass over in silence the fact that comparing all theology to astrology is extreme - a move that ensures the book will lose most of its religious audience, who presumably need this book more than the atheists do. (After all, they’re the benighted ones.) It’s just another example that despite the book’s admonition to play nice with the religious folk, a New Atheist can hardly help but step on their feet and grind the heel in a bit more.

If young atheists need to anoint another secular saint, by all means; but freethinkers might want to think again.

There’s one crumb of comfort for the theist. This book will hardly *influence* anyone to be *effective* in arguing against theists and theologians with real bite.

Kirk