12/21/08

The first premise of the KCA

I was doing a bit of link-chasing tonight when I ran across this article critical of William Lane Craig's KCA. Just a few thoughts...

Read more...




Although Dr. Craig's support for it is uneven, I find the arguments used by atheologians in this regard to be inadequate also. The energy used to argue "infinity" is energy wasted, when modern cosmology does not posit that the universe is infinite, and when the term itself is ontologically negatively defined. While infinity has great use in mathematics, it is a mathematical abstraction, nothing more: and we should not attempt to apply it any more than we should seek a perfect circle or the square root of -1.


This is a revealing sentence. I've always been struck by the claim of "ontological negativity", but I must admit that I don't quite understand it. For example, atheists will throw around this little zinger that we can't understand what "supernatural" is without the natural, or the immaterial without the material. Putting aside the "so what" answer, is that really true? Is infinity ontologically negative? And what would that mean? While I can't grasp what infinity is, I certainly believe an unlimited thing could exist with having some kind of negativity buried at its core. But I digress...

He does support it elsewhere by using two arguments: our observation of the caused entities around us, and causality as a principle of human thought. Dr. Craig is no doubt aware, however, that to infer a necessary causality on a whole -- the universe -- on the basis of observation of such attribute in the parts -- the existents around us - is a fallacy of composition. The attribute being transposed here, being caused, is relational and therefore cannot be transposed. Thus he cannot generalize from caused entities around us to the universe in this matter.


This is the heart of the matter. I'll ask you all to do a bit of thinking about what he just said about the fallacy of composition. Is it really true that it is improper to- in all cases and at all times- infer a characteristic from the parts to the whole? Bricks are hard, and I suggest that the hardness of a wall can be inferred from the composition of a collection of hard things. On the contrary, there are new properties that arise from collections of lower level properties. To flip it backwards, just because water molecules are wet does not mean that quarks are wet. What we need to consider, then, is the nature of causality and the universe as a whole and in parts.

I'd say that our observation of causality in everyday life is evidence that effects must have causes. There is no physical substance known as cause, and there is no reason to think that all this physical stuff around us necessitates causality. From those intuitions (which, I believe, are reasonable and ought to be believed by every thinking person) we come to the conclusion that causality governs the physical.

But back to our problem with the jump from parts to the whole. Of course, there are times when one ought not make that leap. But usually there is an obvious reason not to. For example, if I were to say molecules are tiny, and a wall is made of molecules, therefore this wall is tiny, you would cry foul. That is because there is something blatantly quantitative that carries over when you are adding things like sizes. But hardness does not add up this way (in the case of bricks and the wall) because it is qualitative. To the best of our knowledge, is there something qualitatively different about the universe-as-a-whole from the universe-in-parts? There are some differences, obviously. Some parts of the universe are very hot. Is the universe itself hot? Or, parts of the universe are dark, but that wouldn't suggest that the sum total of physical substances is dark (the question seems nonsensical).

All of this to say that the first premise of Craig's argument is something that rests on an "either you see it or you don't". If you don't see why causation is not just some physical maneuvering within the parts of the universe, then the best I can do is point out the same intuition in different lights until you get it. It seems like a lot of ad hoc escapism, but maybe there are good reasons for denying the universality of causation. I just don't know of any.

44 comments:

  1. "It seems like a lot of ad hoc escapism, but maybe there are good reasons for denying the universality of causation."

    IMHO I think that's rather missing the point of the critique...but anywho.

    Here's a different idea. I would like to see you play the devil's advocate to test the strength and the significance of the KCA. What are the possible weaknesses of the premises? Instead of trying to defend them, try your hardest to demolish them. What you add to the list of problems and weaknesses of the argument?

    Maybe you're not up for playing such a game, but I would be very interested to see what you would come up with if you tried.

    ReplyDelete
  2. The first premise is the cloudiest (for me). The nature of causation is mysterious and I don't know if we will ever fully understand it.

    There, I tried.

    ReplyDelete
  3. Dear sir - you are a fool, and if this post is at all typical of the reasons you have for declaring atheism to be dead, you have a great deal of learning yet to do. Expect to be featured on my blog soon.

    ReplyDelete
  4. Yeah, that's not a good thing. You can check it out for yourself here. I'd be interested in seeing if anyone connected to this blog even cares enough to answer me - my suspicion is that you'd all rather take whatever shots at atheism you can get whether or not they have a basis in facts or logic. But we'll see.

    ReplyDelete
  5. "...maybe there are good reasons for denying the universality of causation."

    Yes there are good reasons to doubt the universality of causation. They have been stated here a few times before but, obviously, not in a way that 'sticks'. Briefly, the idea of causality is deeply interconnected with the concepts of space and time (our intuition tends to focus on the time aspect but that is just us) and what Einstein showed was that space and time do not have existence except in relation to matter (keeping in mind that energy and matter are interchangeable via E=MC*C). So, if the mathematical singularity at the heart of the Big Bang is an accurate description then the entirety of our concepts of space, time, matter and causality cease to have any meaning that we can relate to.

    Now, it is certainly possible that the singularity is not an accurate description of the Big Bang and that when quantum mechanics is successfully brought into the picture we might be able to extend the ideas of space/time/matter/causality to beyond the concept of the singularity. That is the Big Band is not the beginning of the Universe but is just a phase that the Universe goes though.

    Now, these are very interesting and deep concepts but quite frankly they do not shed an iota of light on the question of the existences of Brahma.

    ReplyDelete
  6. Ooops, you may want to edit that entry. The author (me) of this article on the KCA is not Mariano (though I'd be happy to be confused for him), and he does not spam other sites for (solely, at least) attention. It is well within blog etiquette to let someone know you placed them in your blogroll. And if you've seen the scope of Mariano's project you'd know that it is simply unreasonable to want him to tailor a custom message for every source he put in said project.

    You seem to say that the pointing out an intuition is "whining". I don't think that is true. Much of what passes for philosophy is fleshing out an intuition- through thought experimentation and logical testing- to make it obvious to others. I think causation is one of those things.

    That being said, you cannot claim that I am committing the fallacy of composition because I am not saying that the whole (the universe) is caused because the parts are caused. I'm saying that causation as a metaphysical principle is evidenced by our experiences. Since I have no reason, metaphysical or otherwise, to deny causal principles to the universe itself, I feel warranted in saying that there are good reasons to believe the universe has a cause (if it has a beginning). Do you see the difference?

    ReplyDelete
  7. "The nature of causation is mysterious and I don't know if we will ever fully understand it.

    There, I tried."


    Aww come on....I don't think you tried very hard. Well ok, 'tis the eggnog season (or glögg, or glühwein), so you are forgiven :-)

    What I was hoping to see was a list of problems with the meanings of words, and of limits in our empirical knowledge, pointing to the inadequate empirical and logical support for the premises (this has been done before so I won't reiterate it).

    The KCA can not tell us anything new about nature; it is at best a tantalizing philosophical exercise. It is fun to try out our current knowledge on various test cases in, for instance, quantum mechanics - but we just don't know enough about quantum gravity to be able to draw any solid conclusions. And ultimately, even if the universe originated as a cause of something else, a teleological cause is far from the only possible explanation. As I mentioned recently, you should really read Paul Davies's book "The Goldilocks Enigma" if you haven't already.

    cheers

    ReplyDelete
  8. Good point about authorship, although the thought of further associating yourself with this piece really ought to give you pause.

    "...it is simply unreasonable to want him to tailor a custom message for every source he put in said project."

    Do you not understand how definitions work? Spam is precisely a message that the recipient doesn't want and hasn't asked for, and that is not tailored to the individual recipient. But this is a trivial point compared to the real topic.

    "Much of what passes for philosophy is fleshing out an intuition"

    But you have done no fleshing out here - again, you admit as much yourself. You just say that it's an intuition and try to get other people to accept it as such.

    "you cannot claim that I am committing the fallacy of composition because I am not saying that the whole (the universe) is caused because the parts are caused. I'm saying that causation as a metaphysical principle is evidenced by our experiences."

    This is the fallacy of composition: you're saying that we have experience with the parts of a thing (the universe) and that our experience with those parts applies to the thing itself. The word "metaphysical" here is just a shorthand description of this process - it doesn't help you at all.

    "Since I have no reason, metaphysical or otherwise, to deny causal principles to the universe itself, I feel warranted in saying that there are good reasons to believe the universe has a cause (if it has a beginning). Do you see the difference?"

    Did you read my post at all? Of course I see the difference: you're demanding actual evidence from your opponents while allowing yourself to form conclusions without it. While you may or may not actually have good reason to deny causality in the case of the universe, I have no good reason to accept causality in the case of the universe - am I now "warranted in saying that there are no good reasons to believe the universe has a cause (if it has a beginning)"? If we're given a 50/50 proposition, it's totally ad hoc to pick one direction over the other, and that's not even mentioning the fact that you had to rely on a second fallacy even to do that. How is this supposed to convince anybody?

    ReplyDelete
  9. Josh- you say:

    Much of what passes for philosophy is fleshing out an intuition- through thought experimentation and logical testing- to make it obvious to others. I think causation is one of those things.

    I'll agree with that. The trouble is, unless you are informed by data from the real world, you more often then not end up creating an elegant structure of words chasing their own tails, but not leading you any closer to a description of the Universe. Fun, but content-free.

    I'm saying that causation as a metaphysical principle is evidenced by our experiences.

    By "metaphysical", do you mean "magical"?

    Adonais said it: the KCA can not tell us anything new about nature. And as Larry points out at his blog, it has most of the same problems as the classical cosmological argument. It basically says: "I don't understand how the Universe began, so magic must have been involved, in the form of a Being exempt from causality".

    Happy Hannukah, everyone!

    ReplyDelete
  10. This is the fallacy of composition: you're saying that we have experience with the parts of a thing (the universe) and that our experience with those parts applies to the thing itself. The word "metaphysical" here is just a shorthand description of this process - it doesn't help you at all.

    I think you are confused. The fallacy of composition does not say that it is wrong to attribute properties from the parts to the whole. There are some instances when it is fallacious and some that are not. This first example is:

    All atoms are colorless, all cats are made of atoms, therefore cats are colorless.

    While the second is not:

    Bricks are made of atoms, this wall is made of bricks, therefore this wall is made of atoms.

    A cursory look through any introductory logic textbook will back me up on this.

    Did you read my post at all? Of course I see the difference: you're demanding actual evidence from your opponents while allowing yourself to form conclusions without it. While you may or may not actually have good reason to deny causality in the case of the universe, I have no good reason to accept causality in the case of the universe - am I now "warranted in saying that there are no good reasons to believe the universe has a cause (if it has a beginning)"? If we're given a 50/50 proposition, it's totally ad hoc to pick one direction over the other, and that's not even mentioning the fact that you had to rely on a second fallacy even to do that. How is this supposed to convince anybody?

    Actually, it is one of the first steps of philosophical argumentation to undercut defeaters (see here: http://rtmerrill.com/wcb/defeaters.php). If, say, I have good reasons to accept a principle (the principle of causality by means of intuition, for example) and I've shown that your defeaters (fallacy of composition) are undercut, then the first premise has become extremely viable.

    Besides all that, though, there is a second argument that you lump together with my argument from intuition. Consider the following proposition:

    Every effect has a cause.

    Whether I'm in the universe or not, this premise gains probabilistic support every time an effect follows a cause. It is metaphysical because it doesn't make reference to a particular set of circumstances (say, the universe we live in) but would be a principle that could not but obtain.

    ReplyDelete
  11. I'll agree with that. The trouble is, unless you are informed by data from the real world, you more often then not end up creating an elegant structure of words chasing their own tails, but not leading you any closer to a description of the Universe. Fun, but content-free.

    My defense of the first premise has been informed by raw date of the real world. Are you experiencing a vastly different universe than I am? :)

    By "metaphysical", do you mean "magical"?

    No; what would lead you to believe that's what I meant by the word metaphysical?

    Adonais said it: the KCA can not tell us anything new about nature. And as Larry points out at his blog, it has most of the same problems as the classical cosmological argument. It basically says: "I don't understand how the Universe began, so magic must have been involved, in the form of a Being exempt from causality".

    It depends. The KCA does tell you something new about the world if you don't already believe it was caused to exist. Besides, I'm arguing that we do understand how the universe works (at least in part) and because we understand it we can draw broad principles about events and how they relate- specifically, cause and effect.

    ReplyDelete
  12. Josh, you say:

    The KCA does tell you something new about the world if you don't already believe it was caused to exist.

    Only if you take it seriously, which I don't see any reason to. All it does is say "Everything we know has a cause, but it had to start somewhere without a cause, and since we don't understand how that could happen, it must have been magic. Therefore God exists."

    That's also why I asked you, whether by "metaphysical", you meant "magic". If not, what do you mean?

    ReplyDelete
  13. Only if you take it seriously, which I don't see any reason to.

    Why not? It is a valid argument that involves the beginning of the universe and causation. Are those un-serious topics? Disagreeing with the argument is not the same as not taking it seriously. The Argument from Evil is something I take very seriously, even though I think it is flawed.

    All it does is say "Everything we know has a cause, but it had to start somewhere without a cause, and since we don't understand how that could happen, it must have been magic.

    Not quite. I don't remember an appeal to ignorance or ambiguity anywhere in the argument, nor does the first premise say "Everything we know has a cause". It claims that all effects have causes, which, as I've argued, is a very intuitive principle (heck, it drives the entire edifice of science).

    That's also why I asked you, whether by "metaphysical", you meant "magic". If not, what do you mean?

    When I think of magic, I think of contrary to natural law. If that is the definition of magic you mean, then metaphysical is nearly the opposite. Metaphysical laws are ones that not only govern specific circumstances, but have to govern all circumstances. That's why I can't understand why you think the word magic is at all applicable to our conversation.

    ReplyDelete
  14. I was thinking about the first premise of the KCA at work yesterday, specifically about the criticism that, since the laws of physics break down before we get to the singularity, there's no way to know if the laws of causality still apply, and to believe that they do apply is mere speculation.

    This got me started thinking about the uniformity of nature, and I thought I found some parallels. It seems to me like someone could use the above criticism of the KCA and apply it to the uniformity of nature.

    "The uniformity of nature basically states that the future will resemble the past, and we know this from past experience. However, since the future hasn't happened yet, there's no way we can know that the uniformity of nature will still apply. Basically, it's mere speculation that the UoN will still hold."

    The parallels I find are, the principle of causality seems to be fairly well established and, as Josh said, drives the very discipline of science, which is concerned with finding causes of various phenomena. Even Hume said, "I never asserted so absurd a proposition as that anything might arise without a cause." However, there is supposedly a grey area, since before Planck time the laws of physics break down, so we don't know if causality still applies.

    This seems similar to the UoN, which is, like causality, confirmed in our experience. Events have causes, (with causality) and the future has resembled the past (with the UoN). But, as with causality, there is a grey area. With the UoN, it is the future. Since it hasn't happened yet, we don't know if it will hold. However, it seems rational to infer that the UoN will still hold, based on how *in the past*, the future has resembled the past.

    And with causality, it seems as least rational to believe that, if the principle of causality has been confirmed in our experience, that it seems at least rational to infer that it will still hold during that "grey area."

    Anyway, that's what was on my mind yesterday. I'm no philosopher, I just thought I saw some parallels between that criticism of the first premise of the KCA, and the uniformity of nature.

    Merry Christmas and/or Happy Holidays from blisteringly cold Edmonton!

    ReplyDelete
  15. By the way Josh, I have to thank you for recommending that book "The Last Superstition" by Edward Feser. I'm only starting the third chapter, but so far it's fantastic!

    ReplyDelete
  16. Thank you for your thought-provoking comments, Logan. And I'm glad you are enjoying Ed's book on atheism. Now if we could just convince you of libertarianism...

    ReplyDelete
  17. "It claims that all effects have causes, which, as I've argued, is a very intuitive principle "

    Yes but, and I tried to point this out before, what reason do you have to expect human intuition to be a reliable guide to truth in a domain that, if it even exists, is literally out of this world? This is just metaphysical extrapolation without any way to sanity-check the conclusion.

    @Logan:
    Those are two different questions (at least for now - we don't know if they have a common answer), regarding the nature of causality within the universe and the nature of causality before the universe existed. The notion of UoN can only apply to the first, and even then only on short time scales. The universe is vastly non-ergodic, so on longer time scales it will alter itself considerably (including, most likely, the laws of physics).

    For sure, the probabilistic nature of quantum mechanics casts doubt over our intuitive notions of causality, but even so those are questions of physics, not metaphysics.

    The metaphysics of all these questions are contingent on the various scenarios under consideration. For instance, in the bouncing universe scenario, the second category (metaphysics) might not need to concern us at all, because even as the universe passes through the "crunch" during a bounce, there is no singularity and the bouncing universe is "caused" by its predecessor self. In another version, the Hartle-Hawking boundary condition also gets rid of the singularity by making time space-like at the origin, but this leaves open the question of what (or indeed if anything) caused the universe to come into existence this way. At that point anyone's guess is as good as any other speculation - and the KCA is no different in this regard, it is a metaphysical speculation. Intuition or UoN can not validate its premises.

    ReplyDelete
  18. Folks, please go back and read your Intro to Logic text books. Pay attention to the distinction between validity and truth value.

    The first part of the KCA is valid but is essentially without content. It can be succinctly restated as "If the Universe has a cause then it has a cause." To which the only deep and philosophical response is: 'Yeah, so what?'

    Now it is the second part of the KCA, is also valid but without content: "If the Universe has a cause then we will name this cause 'X'. It is perfectly valid to assign a label to an idea.

    The problem with the KCA is the last step: 'X' is equivalent to the concept labeled 'God'. This is simply not valid.

    The bottom line is that all the KCA does is give a complicated way for a theist to say: I believe, my god created the Universe.

    ReplyDelete
  19. Josh, I have to ask you again: did you actually read my post? Because you're making the same mistakes now that I discussed there, which indicates to me either that you didn't read it at all or that (for whatever reason) you're incapable of learning. I need to know which one it is, so: did you read my post?

    There is no such thing, Josh, as a sometimes-fallacy. A fallacy is always a fallacy, and it always - always - means that any argument that uses it fails as logically invalid. That goes for both - yes, both - of the example arguments you presented. I defy you to cite a logic textbook - or any other textbook - that says otherwise.

    Now, about defeaters: of course I know that stuff, silly, I'm a logician. I know all about defeaters and epistemic probability and modal logic and all the other shell games you apologists like to play, and, as always, you don't know what you're talking about. Sure, if you'd provided some defeaters of my objection and if you'd provided some justification of your view, then you'd have some support for premise 1. Having failed to do either of those things, though, you have not provided and such justification.

    Finally, yes, let's talk about "every effect has a cause." I'm totally with you on that one - it seems to me, in fact, that "every effect has a cause" is true by definition. But is the universe an effect? I know you'd like to say that it is, because all the things we encounter are effects, but see, that's the fallacy of composition again. And then you're going to say, "Yeah, but I have this intuition," and I'm going to say, "Yeah, but intuitions are notoriously unreliable." So how does this help you?

    ReplyDelete
  20. ("...any such justification," of course)

    ReplyDelete
  21. Josh,
    Like my religious views, my political views are far from set in stone. By all means, let me know of any books you'd recommend that offer defenses of libertarianism. Or if you'd be willing to talk more about politics, give me your email address and I'll send you a message.

    ReplyDelete
  22. There is no such thing, Josh, as a sometimes-fallacy. A fallacy is always a fallacy, and it always - always - means that any argument that uses it fails as logically invalid. That goes for both - yes, both - of the example arguments you presented. I defy you to cite a logic textbook - or any other textbook - that says otherwise.

    This is just wrong. There are two types of fallacies- formal and informal. The fallacy of composition may be either informal or formal and is a "sometimes fallacy", contrary to what you have just claimed. Here are some sources that disagree with you:

    The fallacy of composition is the fallacy of inferring from the fact that every part of a whole has a given property that the whole also has that property. This pattern of argument is the reverse of that of the fallacy of division. It is not always fallacious, but we must be cautious in making inferences of this form.

    http://www.logicalfallacies.info/fallacyofcomposition.html

    It must be noted that reasoning from the properties of the parts to the properties of the whole is not always fallacious.

    http://www.nizkor.org/features/fallacies/composition.html

    From Schaum's Outlines series on logic (2nd Edition) on page 214 we see this:

    What about this argument?

    My liver is a part of my body.
    My liver contains at least a million cells.
    .: My body contains at least a million cells.

    Solution
    This argument has the form of compositional fallacies, but in spite of this it is valid. There is no counterexample and no fallacy.

    As with other formal fallacies, fallacies of composition cannot be spotted by considering the argument alone, since the form of the fallacy of composition has some valid instances.


    I have a hard time believing that you are arguing against this position.

    ReplyDelete
  23. I'm not sure why I said the fallacy of composition could be both formal or informal. To my knowledge it is only informal.

    ReplyDelete
  24. Then that book is full of it. The fallacy is a formal one, observe:

    1. Plb
    2. Ml
    3. :. Mb

    This is sentential logic, and basic sentential logic at that. Let Pxy be "x is a part of y," Mx be "x has at least a million cells," l be "my liver," and b be "my body." Which rule of logic do you use to get from 1 and 2 to 3? The answer, of course, is that there is no such rule: there's a premise missing. The fact that there's a premise missing is what makes this argument fallacious (every time), and formally so.

    This is basic stuff.

    Now, there are a few things more to be said. Yes, the argument about your liver has a true conclusion - as I've repeatedly said, fallacious arguments can have true conclusions. The reason the conclusion is true in this case is that the missing premise:

    2': for all x,y {(Mx & Pxy) -> My}
    (translated into English: if one thing has at least a million cells and is part of something else, that second thing has at least a million cells)

    is true, and moreover can be argued for trivially. But true though 2' may be, it's still missing from that argument, and that argument is invalid as a result, period. (If you like, I can even walk through the proof from 1, 2, and 2' - it'll take maybe 3 more lines.)

    Now, can the same thing be said of the missing premise in your defense of the first premise of the KCA? You want to say:

    1. Ct
    2. Ptu
    3. :. Cu

    In other words, the things we see (t) are caused (C), the things we see are part (P) of the universe, therefore the universe is caused. Once again, you're missing a premise - the argument is logically invalid. You'll need, of course, something similar to the above premise, something like:

    2'': for all x,y {(Cx & Pxy) -> Cy}
    (if one thing is caused and is a part of something else, that second thing is caused)

    But is this obvious? Hardly. Moreover, you haven't been arguing for it at all - you've just been deferring to your intuition. (In particular, consider that your God is often said to be omniscient: if so, the universe is a proper part of God, which falsifies 2''.)

    Finally, even if I admit - wrongly - that the fallacy of composition is sometimes not a fallacy, that wouldn't prove that it's not a fallacy in this case. That would bring us back once again to your intuition, which really is not good enough.

    ReplyDelete
  25. Whoa, hold the phone - did you just cite some random website and a test prep book at me? Do you have any idea what you're doing? Because test prep books are worse even than bad textbooks: at least bad textbooks try. Test prep books are designed to give you shortcuts even if those shortcuts have known flaws. The idea is that, over a large group of answers, you'll tend to guess right. But we're not doing a standardized test, we're doing philosophy, and the idea is that we don't want to guess.

    If you want a good, free resource from which to learn logic, I suggest starting with Gary Hardegree's stuff:
    http://www-unix.oit.umass.edu/~gmhwww/110/text.htm
    http://people.umass.edu/gmhwww/310/text.htm
    (and if you can get this far)
    http://people.umass.edu/gmhwww/511/text.htm

    ReplyDelete
  26. Sigh, another typo - replace "omniscient" in that earlier comment with "omnipresent."

    ReplyDelete
  27. Larry,

    I'm off for the next few days but I'll be mulling over what you've said in your last few posts. I hope you and yours (and everyone else) has a wonderful holiday season.

    ReplyDelete
  28. larryniven -

    I'm a little confused here. In your notation, shouldn't the KCA be something like this:

    1. Cb
    2. Pub
    3. :. Cu

    where the subject of the first premise (b) is "begin to exist", and it should be the predicate of the second premise, not the subject? This doesn't change the validity of your objections, I'm just a bit confused by your formulation. Also, I don't see the unsoundness of premise 1 so much as a fallacy of composition as a hasty generalization. But then I'm not a logician :-)

    ReplyDelete
  29. Eh, I guess b must be "things that begin to exist" to be a proper subject.

    ReplyDelete
  30. adonais, I'm afraid you have things a bit confused. The formal arguments I posted aren't supposed to represent the KCA itself, they're supposed to represent Craig's (and Josh's) suggested defense of the first premise of the KCA. If I were to do the KCA itself, it'd look more like:

    1. for all x {Bx -> (there is a y such that)[Cyx]}
    2. Bu
    3. :. there is a y such that {Cyu}

    Where Bx is "x began to exist," Cyx is "y was/is the cause of x," and u, again, is "the universe." This, actually, is a logically valid argument (again, I can walk through the proof if anybody wants), but it's not a sound argument (at least, not so as we can tell) if the first premise comes from a fallacy.

    As for the name of the fallacy, this is, fundamentally, where Josh is running into problems. A fallacy is just a name we give to a type of logical misstep in order to remember it better - a mnemonic, of sorts. But not every logical misstep has an associated fallacy, and not every fallacy belongs to only one logical misstep. So when Josh talks about the fallacy of composition, he understands it to mean "any argument that wrongly generalizes from the parts to the whole," whereas I identify it with the associated logical structure (because I find that this is a much more accurate way of reading people's arguments). You're right that it's a hasty generalization, because in some sense every fallacy of composition is, but ultimately the fallacy is the hole in the logic - the missing premise, such as 2' or 2'' - not the layman's definition that belongs to the name we give it.

    Merry Christmas, Josh - come back clear-headed, cause it'd be nice if we could wrap this up quickly.

    ReplyDelete
  31. "The formal arguments I posted aren't supposed to represent the KCA itself, they're supposed to represent Craig's (and Josh's) suggested defense of the first premise of the KCA."

    Aha, I see my mistake now. Again though, why I don't see the fallacy of the first premise as one of composition is that I don't agree with the unstated premise that all parts of the class actually share the predicated property of having a cause. I thought for that reason that generalization was a better description of the unsoundness. But maybe that's a distinction without difference in this case.

    cheers

    ReplyDelete
  32. Wait, no - I'm still confused :-) Something must be wonky with your characterization of Josh's defense of the first premise, since it ends with a conclusion ("Cu": the universe had a cause) that is identical to the conclusion ("there is a y such that {Cyu}": there is a cause such that it was the cause of the universe) of the KCA itself? (this is why I thought it was the KCA you were describing). Maybe this is what confused me: in your representation, it looks like the conclusion of the KCA is preempted in an argument for one of its premises? Something can't be right here.

    ReplyDelete
  33. Oh I see how it is - if there's something wonky, my characterization's at fault, is it? Actually, seriously, I don't think this is my translation doing that. The weirdness you're experiencing is what happens when one argues from a very controversial intuition without having a real argument to back up that intuition: you beg the question. You're absolutely right that this defense of premise 1 isn't a good one because it seeks to prove, in some sense, too much - but how would you translate it?

    ReplyDelete
  34. larryniven -

    Relax, have some eggnog, no need to get your hackles up. I'm grateful for your trying to bring some logical rigor to the table, and I appreciate the input. Don't suppose that I'm interested in defending Josh here; I'm just trying to obtain a clearer picture for myself.

    He did say initially that: "I am not saying that the whole (the universe) is caused because the parts are caused." Given that, I don't see how a characterization of his argument could end with a ":. Cu" statement. Though perhaps he contradicted himself later on this point.

    You ask, about his argument:

    "but how would you translate it?"

    Well first I would have to understand it, which I happily admit that I don't. It might be that it's not so much a formal argument (that you could write as a categorical syllogism) as merely a fuzzy sort of appeal to various institutions (intuition, experience, etc). That's how I have been treating it anyway. But let's wait for Josh to clarify his argument when he has digested the Christmas smorgasbord, or whatever they're having. Speaking of smorgasbords, there's one waiting for me right now..

    ReplyDelete
  35. Actually, all my other posts were me with my hackles up - that one was me being curious (plus, I'm not entirely sure that eggnog doesn't mildly disgust me). My moods are more clear in real life, apparently.

    Anyway, yes - I do remember Josh saying that. And it's a fair response on your part to say that you don't understand the argument (you're not, after all, the only one). But I'm basing my interpretation on things like:

    "My defense of the first premise has been informed by raw [data] of the real world"
    (If his defense consisted entirely of an intuition, this would not be the case. And, as I said earlier, empirical confirmation of an intuition in this case is not enough, because people have competing intuitions that are also "confirmed" by the same evidence.)

    Also:

    "I'm arguing that we do understand how the universe works (at least in part) and because we understand it [i.e., because we understand part of it] we can draw broad principles about events and how they relate- specifically, cause and effect."

    And (from the original post):

    "I'd say that our observation of causality in everyday life is evidence that effects must have causes."

    Now, if you ask me, your quote came in a moment of weakness for Josh where he realized that arguing from the parts to the whole would require him to defend a contentious premise that he doesn't really know how to defend and quite possibly never even seemed questionable to him (namely, that one can safely generalize from the parts of the universe to the universe itself in at least this one case). But maybe not - maybe, like Craig, he really does only want to argue from intuition (which, of course, I would only respond more negatively to). We'll have to wait until he gets back to see for sure.

    ReplyDelete
  36. Josh: the reason I don't consider the KCA worthy of serious consideration is because it is not based on evidence, but on lack of evidence and wordplay. The Argument from Design is a much more serious argument for the existence of God, imho, although Darwin et.al. have long since pretty much defenestrated it, along with the fact that it doesn't really "explain" anything, but merely adds another level of complexity which is unexplained. The Argument from Evil (theodicy) is not a good argument against the existence of all Gods (perhaps God is evil or at least arbitrary), but effective against some instantiations of the Christian God (who is generally presumed to be "good", whatever that means: ask the Amalekite babies what "good" is).

    You say:

    When I think of magic, I think of contrary to natural law. If that is the definition of magic you mean, then metaphysical is nearly the opposite. Metaphysical laws are ones that not only govern specific circumstances, but have to govern all circumstances. That's why I can't understand why you think the word magic is at all applicable to our conversation.

    Are there laws that govern all circumstances (in all possible or imaginable universes)? I don't know of any evidence for such laws. All that we know is provisional and subject to revision. The reason I asked if "metaphysical" meant "magical" to you is because you seem to think that "metaphysical" laws point to God's existence, and God is magical, in my understanding. I don't know of any "physical" (real, measurable) evidence for God's existence, do you? Perhaps I'm prejudiced, but I have an automatic distrust of conclusions that are built on constructions of words with no referents in the real, observable world.

    cheers from snowy Vienna, zilch

    ReplyDelete
  37. "(plus, I'm not entirely sure that eggnog doesn't mildly disgust me)."

    Actually I can't stand it myself, but I thought that this is what Americans drink instead of glögg, so I've been using it to substitute where I would otherwise have talked about glögg :-)

    ReplyDelete
  38. I don't mind eggnog myself, but here in Austria it's unknown- we drink a version of glögg called glühwein. Hic est.

    ReplyDelete
  39. Larry,

    I understand why the transfer of necessity is a formal fallacy. I understand why an undistributed middle is a formal fallacy. But it is unclear from your explanation (and none of the sources you provided supported your interpretation of the composition fallacy) why composition is a formally invalid rule. Indeed, it seems that the reason or grounds of wall being hard is just because it's constituent parts are hard.

    Researching your claim even further over my Christmas break has made me even more resolved that you are getting something wrong. In fact, the only things I could find on the fallacy of composition list it as a "material" fallacy, which is a problem with content not structure.

    ReplyDelete
  40. "it is unclear from your explanation (and none of the sources you provided supported your interpretation of the composition fallacy) why composition is a formally invalid rule. Indeed, it seems that the reason or grounds of wall being hard is just because it's constituent parts are hard."

    Okay, one more time: a wall is hard because (a) its constituent parts are hard and (b) given the way a wall is constructed (i.e., given the materials it's made out of and its physical properties), its constituent parts being hard implies that the wall itself is hard. (b) is the part that your argument is missing, which makes it invalid on its face. Moreover, not all (b)-type premises are true - specifically, the premise that Craig uses to get to step 1 of his KCA seems not to be true (or, at least, not to be uncontroversially true).

    Again, let's see what kind of (b)-type things we can come up with for Craig. We could say:

    (b') If a thing's constituent parts are caused, then that thing is caused.

    Except that doesn't work: we don't have enough evidence to say that, e.g., the laws of physics are caused, and the laws of physics are definitely a part of the universe. So maybe he wants to say:

    (b'') If a thing's constituent physical parts are caused, then that thing is caused.

    But as I said, now we run into problems with the idea of an uncaused omnipresent anything. Also, there's still the gap in the analogy used to make this claim, because when we experience this to be true, the composite thing is itself a physical object, but the universe is not just a physical object (remember, physical constants are an essential part of what gives this universe its identity; this is not so with the objects within it). You might be tempted to say something metaphysical, like:

    (b''') If a contingent thing's constituent (physical) parts are caused, then that thing is caused.

    Except that begs the question when applied to the universe. But ignore for the moment that it's not really clear which of these, if any, Craig actually wants: the point is that he doesn't tell us which he wants, which means we can't reasonably evaluate his argument on the basis of its premises (i.e., whether they're true or false or what). An argument that can't rightly be evaluated on the basis of its premises is, of course, a fallacious argument.

    ReplyDelete
  41. Okay, one more time: a wall is hard because (a) its constituent parts are hard and (b) given the way a wall is constructed (i.e., given the materials it's made out of and its physical properties), its constituent parts being hard implies that the wall itself is hard. (b) is the part that your argument is missing, which makes it invalid on its face. Moreover, not all (b)-type premises are true - specifically, the premise that Craig uses to get to step 1 of his KCA seems not to be true (or, at least, not to be uncontroversially true).

    Even though you are apparently backing off your original statement that instances of composition are always fallacious in logical reasoning (a formal fallacy) I think it is relatively obvious that your (b) is fallacious, or at least unnecessary. Consider the original example I cited with the wall and the molecules. The reason or grounds of a wall being made up of molecules is just because it is made up of molecules. Bricks are all made of molecules, and walls are made up of bricks. That walls are made of molecules is a fact just because they are composed of bricks which are in turn composed of molecules. I don't see how your (a) and (b) differ whatsoever.

    If you mean to say that there is something to say about the relation of constituent parts then I am in agreement. In fact, I said as much when I went on about the qualitative/quantitative factors in the OP.

    To apply this to Craig's argument, we can take causality to be a governor of events and a relation between them. So for every event that occurs there must be some sufficient cause. Depending on how one would like to define event (it doesn't matter, the universe fits in to all of them), we would see that the universe is not some qualitatively "other" that somehow possesses alternative characteristics than its parts. Indeed, the onus falls on the person who wishes that magical properties emerge sui generis than the defender of causality.

    ReplyDelete
  42. "...you are apparently backing off your original statement that instances of composition are always fallacious in logical reasoning..."

    This is not what I'm doing. Arguments from composition of the following type are always logically invalid: "A is made up of B-type objects, therefore A has some property or properties of B-type objects." This is the argument that Craig apparently uses, and it's a bad one.

    What I'm doing is saying that this kind of argument - like any fallacious argument - can be fixed (at least, logically) with the insertion of the right premises. In this case, being the right premise means explicitly linking as a general statement the fact of composition ("A is made up of B-type objects") to the fact (or not) of transitivity ("A has some properties of B-type objects"). Sometimes this kind of premise will be pretty harmless ("If an object x is made up of molecule-composed objects of type y, then x is also molecule-composed"), but in all the interesting cases, it won't be. Again, in Craig's case, he can't even form this premise with words - you're not doing any better, mind you - preferring instead to hint at it and beg off to people's intuitions. But if he wants to rely on his intuition, why not abandon the compositional argument altogether, as you do here:

    "To apply this to Craig's argument, we can take causality to be a governor of events and a relation between them [such that] for every event that occurs there must be some sufficient cause."

    Sure, we could just take the sufficient cause principle as axiomatic, but then why bother giving evidence for it? This is aside from events that (seem to) operate on different versions of causation altogether, too, but an axiom is an axiom, not a conclusion of an argument. If this really is axiomatic (dissenting evidence to the contrary), then neither Craig nor you should be arguing for it because that's not how axioms work.

    ReplyDelete
  43. Let me try this another way. Take the ad hominem fallacy - I think you'll agree that it is, indeed, a fallacy, and a logical one. But what if someone were to approach you on the street and say that a sign in a window displayed two distinct shades of blue, when you could only see one. You two could argue about it for a bit, but ultimately perception is perception, so the two of you would probably part ways still disagreeing. Let's say that this person, on leaving, tosses out the line, "Of course, if you're American, I should never have expected you to see it in the first place!" What would you make of that?

    It seems like an ad hominem fallacy, and a somewhat bizarre one at that, does it not? But as it turns out, many English speakers actually see fewer shades of blue than Russian speakers do, so it's entirely plausible that this person had good reason to say that bit about you being American.

    But so what? Remember, their argument - as stated, rearranging some verb tenses - is:

    1. You're American.
    2. Therefore, you shouldn't be expected to see this extra shade of blue.

    That's missing a premise, and no matter how true that premise may be and no matter how much the arguer may be aware of the premise and mean to refer to it in the argument, it ain't there. If you, as the person on the listening end, know this thing about English vs. Russian speakers, you could be charitable and insert the missing premise yourself, but in general, there's going to be no way for you to fill in the blank (and even if you fill it in with a good premise, you run the risk of filling it in with a premise other than the one the arguer intended). That is, in general, the argument - as presented - will have a gaping hole in it that it's beyond your ability to fix without substantively changing the argument. The argument, therefore, is still fallacious, despite it being fixable with really very little effort. You would be well within your rights to reject it out of hand.

    So what about Craig, then? It's the same thing: whatever you think his missing premise is, there is some missing premise or other. Sure, maybe it's something eminently rational, or maybe he wants to go the axiomatic route, or whatever - but it's not like he hasn't been given ample opportunity to correct this, first of all, and second of all, it's not our job to stare into the abyss of his argument until we feel the missing premise staring back at us. Craig's argument (or the shades-of-blue argument above) could be salvaged or completed by the addition of a premise, yes, but the act of adding that premise - which as you see is not always as simple or easy as it may seem - is tantamount to admitting that an extra premise was needed; that, in other words, the original argument was invalid.

    ReplyDelete