Tuesday, July 24, 2007

Worst. Apologetic. Ever.

Phritz and I have been having a... spirited... discussion about Pascal's Wager. Phritz's inanity aside, Pascal's Wager has the distinction of being the worst apologetic argument ever employed. It is literally invalid, unsound, and ridiculous at every point; as such, it is at least instructive as an example of what not to do in philosophy. (It's controversial whether or not Pascal himself intended the Wager as an apologetic argument; I'll leave the assignment of blame to better scholars than myself. It is manifestly the case, however, that many Christians and Muslims (and even agnostics) employ Pascal's Wager as an apologetic argument.)

To recap, Pascal's Wager sets up a probabilistic decision matrix on two dimensions: The existence or non-existence of God, and the presence or absence of belief:

God existsGod does not exist
Believe in GodInfinite rewardZero cost
Diselieve in GodInfinite penaltyZero reward
If there is a non-zero probability (Pascal himself puts the probability at 50%) that God exists, then the choice to believe in God dominates the choice not to believe.

Let's enumerate all the ways that this argument is bullshit through and through: Every assumption is unjustified and controversial.

First of all, the Wager demands an ontological commitment to objective probability; The Wager makes no sense at all expressed in terms of epistemic probability. It's entirely unclear, however, whether—despite its manifest utility—objective probability is at all physical.

Consider the case where I choose a poker hand of five random cards from a standard deck but don't look at them. In a linguistic sense, I can talk usefully about the probability of those five cards being a royal flush. But physically (unless we are willing to abandon all our notions of objective reality) those cards have exactly one definite content: The "objective" physical probability of the hand being a royal flush is either 1 or 0. We are merely expressing our epistemic ignorance—as we have not looked at the cards, we don't know their content—in terms of objective probability as a valuable shortcut. We must view any argument that relies exclusively on objective probability as inherently suspect.

The second general probabilistic difficulty is Pascal's assumption that the probability that God exists is not infinitesimal. But this assumption is completely unjustified, even if we assume arguendo that the probability is non-zero. If the probability were infinitesimal, then we can draw no conclusions about the resulting value, because simple multiplication of an infinity by an infinitesimal is undefined. Again, since Pascal predicates infinite reward, it seems legitimate consider also infinitesimal probability.

The two possibilities that Pascal enumerates are much too restrictive: We must consider not only the case that God exists, but several other conditionals:
  1. God exists and
  2. God rewards belief and
  3. God rewards the particular belief the believer adopts
Given the Christian context of the argument as well as the consistent Christian deprecation of "heresy", it seems entirely legitimate to discuss correct vs. incorrect belief. An honest decision matrix would embed all of these assumptions, each with its own probability:

God existsGod does not exist



RB: God rewards belief; RA: God rewards all belief and punishes only disbelief; RC: God rewards only correct belief and punishes incorrect belief as well as disbelief; BC: The believer believes, and believes correctly; ~ means "not".

We can turn then to the supposed "choice": Whether to believe or disbelieve. But belief itself is not a choice. I can choose to say things, to act in particular ways, but one cannot choose what to actually believe: Belief must be coerced, either by reality's evidence, one's intrinsic nature, torture or (if you go for that sort of thing) divine inspiration. I can't choose to believe in a God: It is simply a fact of my consciousness, over which I have no power of will, that I believe that no God exists. Furthermore, even if belief were a matter of choice, it seems difficult to understand how one could "choose" to believe correctly.

Not only are the rows and columns of the decision matrix unjustified in Pascal's formulation, the assignment of all of the costs and benefits are also unjustified. What does an "infinite" reward mean? An "infinite" punishment? These concepts are not directly comprehensible. To apprehend any infinity in a meaningful sense, one's mind must become infinite. But our minds are finite: There are only a finite (albeit quite large) number of states any finite mind is capable of: After some finite time, all possible states of reward or punishment would be exhausted.

And, furthermore, the assumption that God would reward or punish infinitely (even if that were coherent) is unjustified. We must add yet another category of columns—God rewards/punishes finitely vs. infinitely—each with its own individual probability.

Likewise, the assumption that belief or disbelief has zero cost is likewise unjustified. What sort of God would create an infinite reward for an action that had no cost whatsoever, a belief that had no ethical or moral dimension at all? Clearly belief in God, to be other than entirely vacuous, must have some cost, even if it were finite. And a non-zero cost renders the decision matrix invalid under an infinitesimal probability of God existing (or rewarding only correct belief).

In a sense, all conditionals are "true": If God exists and if God rewards belief and punishes disbelief and if God rewards any sort of belief and punishes only disbelief and if the probability that God exists is not infinitesimal and if God rewards belief infinitely or belief entails zero cost... if all of these things, then yes, it is a good bet to believe in God.

Of course, one could say with an equal amount of "truth" that if the Moon were made of green cheese and if we could convince a billion Chinese that cheese was tasty, and if we could get to the moon by flapping our arms then it would be a good bet to invest in my lunar cheese mining stock.

If you buy Pascal's wager, email me and I'll send you a prospectus.


  1. You flunked again, Bum. The Wager is NOT an argument, is it: it's an early form of a decision matrix. So there are givens, like many hypothetical situations. And given the religious context of Pascal's time (that's called history, genius), those givens are not so surprising. And regardless of what probabilities are assigned to the possibility of God existing, the matrix and consequences still hold (and you ain't one to assign probabilites to the possibility of anything).

  2. I can't choose to believe in a God: It is simply a fact of my consciousness, over which I have no power of will, that I believe that no God exists.

    That's a patent lie, Bum, about equal to like the Twinkie Defense. Deny free will, then. Fine: More of the Bum's BugWorld--don't make any more normative statements then, of any sort. Stalin and Hitler are just powerful primates, then. You sound like BF Skinner on crack at the homeless shelter.

  3. Warning: You're starting to bore me, Phritz.

  4. You bore me, clown. You are misreading the entire wager, for one; it's not a simple card-game like scenario regarding objective probability (since a mortal never finds out what the cards are). There's no objective way to assign numerical probabilties to the possibility of the Almighty. Besides, I did not offer it as an apologetic: you assume that it is. More of your usual deceit.

    Besides "bore" is another normative statement, implying intention. With no free will, no power of making decisions, simply reacting in good ol' behaviorist stimulus-- response style, the Barefoot Primate was conditioned to find some things "boring" (or good, or not good) But I think your conditioning was a bit off.

  5. If I bore you, don't read the blog.

    You have now officially bored me. Any repetitive comments will be deleted.

  6. This comment has been removed by a blog administrator.

  7. Phritz: That someone points out your mistakes bores you?

    You've already pointed out what you consider* the "mistake". Several times. Repeated assertion does change the truth or falsity of a criticism.

    Go get yourself a girlfriend, or at least a better hobby than posting inane blather on my blog: Sexually molesting farm animals might be up your alley.

    *In the loosest sense of "consider" that might apply to such simple nervous systems as sponges.

  8. Fuck you, fag. You're full of shit. A liar, a crass monolingual knave, a manipulator, an irrationalist, one step from Bukharin. I'm going to have your cheap ass shut down. Buh bye.

    Got dat? Buh bye. Fuck you bitch.

  9. I'm going to have your cheap ass shut down.

    I'm shaking in my boots.

  10. Um, wow?

    I'd point anyone from Blogger in the general direction of this thread to see who was being more abusive in the whole exchange.

    Phritz was at least interesting for a while before he devolved to personal attacks and name-calling. I think if he'd rather calmly explained why he felt you weren't addressing his critiques, instead of being rude, he could have contributed something. But this whole series of comments was unbe-fucking-lievable. What an ass.

  11. The best I can get for Phritz's critique is that "Pascal's Wager isn't an argument, it's a decision matrix." I don't think that's any better a critique than saying, "The argument from design isn't an argument, it's just a string of words."

    While it's unclear whether Pascal himself really intended the Wager to be an argument is a matter of scholarly controversy, or so I've been told. However, I've had any number of conversations—even with relatively "sophisticated" theists such as Kenneth—where Pascal's Wager was in fact used as an argument to at least believe in a God.

    I also specifically address Phritz's assertion that the conditional is "true" in some sense. So I really don't know where I've been deficient in responding to his criticism, unless he has any more actual points buried in his incompetent invective.

  12. Also, I suspect that if Perezoso/Phritz actually contacts Blogger with the complaint that "the bad blogger was mean to me," the only response is likely to be a brief chuckle as some intern puts the message into the crank file.

  13. It's your fault bum. You fed the troll.

    I often wonder what possesses people to argue with crazy folk. Gotta spend the time some way, I guess. Best of luck to you, Bum!

    Anonymous' post may be translated as: "I forfeit."

  14. BTW, contrast Anon's post with my own post a while back directed at Mr. Elliott and you will see the difference between infantile vitriol and self-deprecating satire. You may use it as a reference in future cases.


  15. Whatever the logical status of Pascal's wager, it's an IGNOBLE reason for believing in God.

  16. anticant: Yes. At the end of the day, that's probably the Wager's worst feature.

  17. Indeed. As has been pointed out ad nauseam, what kind of scheister is God if he welcomes to Heaven those who believe in Him because the belief is cheap and the reward infinite?

  18. Samuel Skinner
    Hey! We argue with nuts because we don't realize how crazy they are. Then when we understand we ask how anyone could believe such c@#p.

  19. I was scanning. Maybe I missed this. However, don't forget that the criteria for infinite reward and infinite punishment itself has infinite possibilities, with each possibility as good as the next. Only Christians go to hell, only theist go to hell, an ironic god, heaven and hell criteria without a god, people wearing pink shoe laces are punished, etc. It seems impossible to determine what is a "good bet" pending evidence or further argumentation. All hypothetical criteria have equal supporting evidence. No hypothetical does better under Occam's Razor as far as I can tell.

  20. To play (ahem) devil's advocate:

    I'm not understanding the premise that reward or punishment is infinite - infinite in duration, sure, but it seems you are saying infinite in variation. Not the same thing.

    Nor are the "Good bet" possibilities infinite in variation. The base assumption here is that there is some coherent, predictable belief system that adheres to some kind of ethical rules and is predictable - that is, as far as I can tell, the basis of every religion (even the non-theistic ones): conform to particular behaviors and beliefs, get this rather specific reward. Not an infinite variety of meaningless alternatives (pink shoelaces), not an infinite variety of rewards.

    Finally, I smell math abuse, a common danger of trying to apply mathematical operators to symbolic logic. If I manage to use the term infinite to two different things, that doesn't mean I can multiply them together - because math is abstract, and the things you're referring to aren't. That's why math, like evolution, is a theory - its algorithms. One plus one equals two, but one Cadillac plus one Subaru does not equal two Formula 1 cars (unless, perhaps, in heaven, which does tend to support your infinite variation reward scenario argument, I suppose).

    Note this isn't an endorsement of Pascal's Wager. I think it largely a moral argument, not a philosophical one (and certainly not one well expressed by symbolic logic), and as a moral argument a cowardly one. It also fails in theistic grounds, as pointed out.

    But I see in this argument what some call leaps of logic, a term I have other problems with but I've ranted enough for one Sabbath ;)

    1. compulsory posting/ also also8/18/15, 2:22 AM

      Quite sure the old anonymous comment was me now. Tried to resist, but
      1.) math being abstract doesn't negate its usefulness in understanding or framing a subject. For instance, in Ron's car example there doesn't fail to be two cars just because the cars are not abstract. Stating that there are two formula 1 cars would be a categorization error. Not math abuse.
      2.) I don't even mention multiplying infinities.
      3.) Theories in math don't mean the same thing as in science or in common usage.
      4.) It should bee obvious the pink shoelace example isn't used as a depiction of the basis of religions that you listed . It should be clear the examples used because those basis don't help to establish their verisimilitude of belief(s) or the after death utility so why not use something "meaningless" (to who? you?).
      5.) While behavior, actions, and beliefs (the state of belief) are not abstract the reward/punishment criterion (as well as afterlife, god(s), souls, and other supernatural aspects of religions) are entirely hypothetical (non-scientific use of hypothetical) , so talk of probabilities or number of hypothetical outcomes is limited by what you can frame. Does god like people that ingest 7*10^27 atoms per month? How about 8*10^28? How about 9*10^30. atoms how about (9+x)*10^(30+x) per month/year. It can be framed as infinitely easily in (maybe) an infinite number of ways precisely because the subject is entirely hypothetical even when using commonly accepted religions (including atheistic ones). What does a hypothetical being want anyway?. Religious afterlife propositions lacks any restraints that are Germain to "truth", what is real, or empiricism. The absurdity of the utility argument is better highlighted by there being an entirely contrary hypothetical to each hypothetical. I'm just harping on the first two conditionals.

    2. also. I missed the crux of Ron's point8/18/15, 1:00 PM

      I agree that infinitely good or bad reward in punishment isn't coherent but thats the way pascals argument tends to be phrased. Just replace "infinite" punishment/reward with same punishments/rewards. In my experience with religious people the reward/punishment tends to be "what you can imagine" and not a rather specific one.

  21. Weird. I was reading through this blog for the first time in years. I think I may have been who Ron was responding to, but I'm not sure. The point is, is that "meaningless" criteria are as well supported philosophically and empirically as anything that can be simply made up.Better supported actually because "predictability", "system", and "ethical rules" rules of actual religions are simply additional unsupported assumptions. It's obvious that religions don't propose "meaningless" but simpler and more plausible rewards and instead propose stories and codified behavior that add "meaning". So what? Doesn't show any verisimilitude of religions in comparison to "meaningless" alternatives. That there are potentially infinite sets of requirements for rewards is Germain to the wager- not potentially infinite variation of rewards mind you. That the anti-christian/muslim/raelean/atheist/theist nullifies the wager even if all the other issues with it are ignored is the point.

    1. "meaningless" is the only quote that is somewhat facetious. Religious "meaning" comes is given by the religious (and non-religious) and isn't in and of itself support for the factuality or truth of the claim. last edit. I'm having a hard time stopping. If you delete this would you just delete all of my comments?

  22. It's late. I mistyped. 'The point is, is that single criteria you come up with off the top of your head are as well supported as actual religious criteria.'

  23. also "verisimuilitude of actual religions in..."

  24. also "anti-whatever wager nullifies pascals wager" edit

  25. I haven't read this article (not blog) for the first time in years is what I should have said. I have read the more recent articles on this blog (great blog) in the past few years. I just didn't revisit this one. An edit for commenter function would be really nice. Is there one for people using a "hard" id? If so I think I'll use mine if/when I comment again.

    1. also...I'm very tired (not quite functioning properly)and just realized I'm being inconsiderate to the host. Apologies.

    2. Sadly, blogger offers no facility for editing comments, not even for me.


Please pick a handle or moniker for your comment. It's much easier to address someone by a name or pseudonym than simply "hey you". I have the option of requiring a "hard" identity, but I don't want to turn that on... yet.

With few exceptions, I will not respond or reply to anonymous comments, and I may delete them. I keep a copy of all comments; if you want the text of your comment to repost with something vaguely resembling an identity, email me.

No spam, pr0n, commercial advertising, insanity, lies, repetition or off-topic comments. Creationists, Global Warming deniers, anti-vaxers, Randians, and Libertarians are automatically presumed to be idiots; Christians and Muslims might get the benefit of the doubt, if I'm in a good mood.

See the Debate Flowchart for some basic rules.

Sourced factual corrections are always published and acknowledged.

I will respond or not respond to comments as the mood takes me. See my latest comment policy for details. I am not a pseudonomous-American: my real name is Larry.

Comments may be moderated from time to time. When I do moderate comments, anonymous comments are far more likely to be rejected.

I've already answered some typical comments.

I have jqMath enabled for the blog. If you have a dollar sign (\$) in your comment, put a \\ in front of it: \\\$, unless you want to include a formula in your comment.

Note: Only a member of this blog may post a comment.