Monday, January 9, 2017

Understanding Cosmological Arguments


At the Secular Outpost, Ryan M posted a reasonable effort at summarizing some of the common mistakes made by non-experts in philosophy of religion.  For this discussion, I'd like to focus on the first of those mistakes.
Mistake 1 - [Failing to understand basic cosmological arguments]

- Many non-experts, presumably all atheists, interpret cosmological arguments in general as having the following form:

1. Everything has a cause.
2. Therefore, the universe has a cause.
3. If the universe has a cause then God exists.
4. Therefore, God exists.
5. If God exists, then God does not have a cause.
6. Therefore, God does not have a cause.

The obvious issue is premise 1. Where can we find a cosmological type argument with a premise like that? Not in Aquinas, not in Duns Scotus, not in Leibniz, not in Aristotle, not in Koons, not in Pruss, not in Craig. Probably, no prominent defender of theism has used such a premise, and it's hard to tell if anyone has used it other than people misinterpreting arguments made by one of the listed philosophers.
The first thing I would note is that I'm not sure that this is actually the way many atheists interpret or understand the argument.  I don't see it quite that way.  And I certainly am aware that there are various formulations of it that don't all follow the same logical flow, but I want to focus on the second way of Aquinas specifically.  This is his argument for a first efficient cause of being, and I'll explain it the way I understand it.  Quodlibet states the argument this way:
1. We see in the world an ordered series of efficient causes.
2. No thing can be its own efficient cause,
    a. for this would require it to be prior to itself, which is impossible.
3. This series of efficient causes cannot be infinite,
    a. for all efficient causes follow an order: first, intermediate, ultimate.
    b. if there is no first cause, there can be neither ultimate nor intermediate
        i. for to remove the first cause is also to remove its effect
    c. in an infinite series there could be no first efficient cause, no ultimate effect, nor any intermediate cause; but this is plainly false.
4. Therefore, a first efficient cause is necessary; which we all call God.
I think  this version of the argument follows Aquinas fairly well.  Aquinas says that this is an a posteriori argument.  That means its premises are established by observation.  So with that in mind, it's reasonable to interpret the first two premises as rules of nature based on what is observed.  I'd say the rules he establishes are: 1 - that things have causes, and 2 - that the cause of anything is not the thing itself.  Aquinas does not explicitly say that everything has a cause, but since he's talking about what we observe, there's no reason on that basis to suppose that this rule shouldn't apply to everything, unless you want to smuggle in some additional unstated assumption (which theists surely do).  Actually the notion of an ordered series is elaborated in statement 3a, and it includes a first cause, but that first cause is NOT something that we observe.  So it is fair to say that Aquinas does indeed smuggle in an assumption here.

Statement 3 of the argument is supported by its own argument, given in 3a, 3b, and 3c.  Statement 3a declares that there must be a first cause before any intermediate or ultimate cause (because that is the order that causes follow, according to Aquinas).  Statements 3b and 3c merely reinforce the point that there has to be a first cause in any chain of causation, and that's the basis for concluding that there cannot be an infinite chain, which doesn't have a first cause.  This is the way I read Aquinas:  he's saying that there cannot be an infinite chain of causation, because there has to be a first cause.  And the insistence that there must be a first cause is nothing more than a bald assertion.

And this is problematic, because that assertion is the same as the conclusion of the argument, given in statement 4.  Try as hard as I might to see it differently, I cannot help but think that statement 3a begs the question, because the conclusion reached in statement 4 is assumed right there in statement 3a.  Begging the Question is a logical fallacy described as:
Any form of argument where the conclusion is assumed in one of the premises. ...  Begging the question is a form of circular reasoning. - Logically Fallacious
And there is an additional problem with this conclusion, because this thing that is regarded as first cause, whether it's God or something else, is assumed to be uncaused.  But that is not consistent with the implicit rule established in statement 1 - namely that things have causes.  But the first cause doesn't follow that rule.  It is the one and only exception.  And this might be seen as special pleading, another logical fallacy.
Applying standards, principles, and/or rules to other people or circumstances, while making oneself or certain circumstances exempt from the same critical criteria, without providing adequate justification.  Special pleading is often a result of strong emotional beliefs that interfere with reason. - Logically Fallacious
Now, I'll admit that there may be good reason for assuming that something exists without a cause, despite Aquinas' claim that this argument is based on observation (and I don't think he was saying that we observe such things).  So I'm willing to let Aquinas off the hook for this one.  But if this objection is valid (and I don't think it's so unreasonable, because the argument as presented fails to establish the necessity of a first cause, other than simply asserting that there must be a first cause), then there is justification for asking that oh-so-ignorant question: Who made God? 

Nevertheless, that brings us to another issue.  Why does the first cause have to be God?  And why does it have to be just one thing?  There is certainly no logical reason given in this argument that would preclude the possibility of one or more first causes that are not God.  This thing, whatever it might be, would be known as a brute fact.  But for some reason, Aquinas did not see fit to include this logical possibility in his second way.  And that's what we call a False Dilemma (or excluded middle).
When only two choices are presented yet more exist, or a spectrum of possible choices exists between two extremes.  False dilemmas are usually characterized by “either this or that” language, but can also be characterized by omissions of choices. - Logically Fallacious
So this is the way I understand Aquinas' cosmological argument.  I don't claim to be speaking for all atheists.  I know there are many people who think that Aquinas' logic is bullet-proof, but I have yet to see any explanation (beyond "Thats just ignorant") that satisfies my objections.  I would welcome any cogent explanations that clarify my possible misunderstandings about this argument.


5 comments:

  1. Aquinas' Second Way can be formulated like this:

    1. There are ordered series of causes.
    2. If there are ordered series of causes, then either an infinite causal regress exists or a first cause exists.
    3. Intermediate causes exist.
    4. If an infinite causal regress exists, then it is not the case that intermediate causes exist.
    5. Therefore, a first cause exists.

    I think you have interpreted Aquinas as saying that "If an infinite causal chain exists, then no first cause exists. But a first cause exists, therefore an infinite causal chain does not exist, so a first cause exists".

    More accurately, Aquinas makes an assumption about what would be the case if infinite causal regresses exist. i.e. Aquinas thinks there could be no intermediate causes or ultimate effects. Many Thomists are fond of saying taking the position that infinite causal regresses imply that the world could never make it to the present, so since the present exists then an infinite causal chain does not exist.

    The following links do a good job explaining issues with the argument:

    https://www3.nd.edu/~jspeaks/courses/2009-10/10100/LECTURES/3-second-way.pdf
    http://spot.colorado.edu/~heathwoo/Phil100/cosmo.html

    The special pleading charge doesn't work because Aquinas' first premise is not that "Everything has a cause". More accurately, Aquinas' first premise simply states "There exists at least one thing which has an efficient cause". As I have stated before, special pleading only occurs when you have a set of propositions which are inconsistent because one propositions violates a rule established in another proposition. i.e. one proposition states every x has a property P, and another proposition states that one x does not have P. In this sense, Aquinas isn't really special pleading.

    Your latter objections to the argument are good ones. First, you note that the argument fails to establish God as the first cause. Second, you note that the argument fails to establish that there is exactly one first cause rather than two or more. Now Aquinas does attempt in other words to fix these deficiencies since the Five Ways are actually supposed to be short summaries of his arguments, but I don't recall where Aquinas goes into such details. I think a good place to start would be Brian Davies' books covering Aquinas.

    ReplyDelete
    Replies
    1. I think you have interpreted Aquinas as saying that "If an infinite causal chain exists, then no first cause exists. But a first cause exists, therefore an infinite causal chain does not exist, so a first cause exists"
      - I was taking Aquinas at his word. He says "because in all efficient causes following in order, the first is the cause of the intermediate cause ..." He is literally saying that all ordered series include a first cause. I honestly don't understand how you can read Aquinas and not see that he is making the assumption of a first cause in his argument. He says it directly. It's right there in plain words.

      More accurately, Aquinas makes an assumption about what would be the case if infinite causal regresses exist. i.e. Aquinas thinks there could be no intermediate causes or ultimate effects.
      - That's right. And the reason there could be no intermediate or ultimate cause is because there must be first cause before there could be intermediate or ultimate effects. He says it right there in his argument. I take him at his word. He assumes a first cause. He doesn't argue that every effect must have a cause that precedes it, because that would not preclude an infinite causal chain. Instead, he PRESUMES a first cause to argue that there cannot be an infinite chain.

      Many Thomists are fond of saying taking the position that infinite causal regresses imply that the world could never make it to the present, so since the present exists then an infinite causal chain does not exist.
      - You may be right, but that is different from the argument Aquinas made. It is actually more like Craig's argument, about the impossibility of "traversing" an infinite series. Aquinas actually understands it better than Craig, because he realizes that the infinite series has no beginning (or no first cause), while Craig's unsophisticated (and mathematically incorrect) understanding of the infinite series says "you can't get from there to here", which DOES assume a beginning of the series, but argues that it is impossible to go the distance from that beginning to the ultimate end. On the other hand, what Aquinas says is "But if in efficient causes it is possible to go on to infinity, there will be no first efficient cause ..." Aquinas, unlike Craig, is mathematically correct, and these two arguments are not the same at all. But Aquinas DOES assume there must be a first cause.

      The special pleading charge doesn't work because Aquinas' first premise is not that "Everything has a cause". More accurately, Aquinas' first premise simply states "There exists at least one thing which has an efficient cause".
      - Let me try to explain this again. (And this is where I agree completely with Russell). Aquinas says he is basing his argument on observation. If that's true, then what we observe is that things have causes which are not the thing itself. We don't observe "first causes". So Aquinas' shtick about observing ordered series with first causes is just a lie. That's not what any human being has ever observed. If Aquinas wants to establish a rule of nature based on observation (which is what he seems to be doing), then he can't posit the existence of something that violates that rule without special pleading. Now, I agree that Aquinas doesn't explicitly state that rule. He's too clever for that. Instead, he carefully words his rule such that it doesn't directly claim that all things have causes. But the fact remains that everything observable to him does have a cause. so in a very real sense, he is engaging in special pleading, and I think Russell recognizes this, even if you don't.

      Delete
    2. Regarding your alternate formulation of Aquinas' argument, I would point to the statement
      4. If an infinite causal regress exists, then it is not the case that intermediate causes exist.
      This statement, on its face is self contradictory, because any cause before the ultimate effect is necessarily an intermediate cause. But Aquinas' only reason for making such a claim is that an intermediate cause cannot exist without a first cause. So in this formulation, the presumption of a first cause is smuggled in without being explicitly stated. It's still circular reasoning or question-begging (whatever you want to call it). I think there's no getting around it. Aquinas does not use valid logic.

      Delete
  2. A few notes:

    First, I think you ought to use multiple sources to determine exactly what Aquinas had in mind when making his argument. I posted some different sources, and Aquinas only used the Five Ways as summaries so you might also want to read some of Aquinas’ other works.

    Second, I’m not sure what to say about Aquinas and special pleading. I think you’re trying to say that Aquinas asserted that all sentences in his argument are known through observation, so Aquinas is special pleading since his argument contains a sentence that is not known by observation (e.g. that ordered series have a first cause). I’m not sure where, if anywhere, Aquinas claims that. In addition, I don’t know what resource you’re using from Russell. If I recall correctly, Aquinas actually says his argument is an a posteriori argument rather than a priori argument, and all this means is that the argument contains at least one premise which can only be known through observation. The first premise, that there are ordered series of causes, makes the argument an a posteriori argument.

    Third, you need to be careful when you claim an argument is invalid, or unsound. Suppose we have an argument with a set of premises {A, B, D} and a conclusion "C". To say the argument is valid is to say it is logically impossible that all of A, B and D be true while C is false. As a result, the following argument is valid:

    1. If (rocks exist and it is not the case that rocks exist), then 1 + 1 = 3.
    2. Rocks exist and it is not the case that rocks exist.
    3. Therefore, 1 + 1 = 3.
    In logic, the actual truth of premises is irrelevant to the validity of arguments. In the above argument, premise 2 is a contradiction. Since premise 2 is a contradiction, it cannot be true. Since premise 2 cannot be true, it trivially follows that it is logically impossible for all the premises to be true while the conclusion is false. As a result, the argument is valid. This would be an issue for your claim that Aquinas uses a contradictory premise since that would actually imply that the argument must be valid. Another issue that can creep up is claiming that if an argument is either circular or question begging then it is invalid. The following argument is question begging:

    1. All tigers are mammals.
    2. Therefore, all tigers are mammals.

    Despite being question begging, that argument is valid. An argument being question begging technically has nothing to do with the validity of the argument. If anything, an argument being question begging makes an argument less likely to be invalid since either the conclusion is contradictory, so the argument is valid due to the conclusion also being a premise, or the conclusion is true so there is at least one premise which implies the truth of the conclusion.

    The validity point might seem nitpicky, but it I figure that either you're using "validity" as logicians do, or you're using "validity" to be the same as "soundness" as logicians define it. In the former case, you would end up stating something false. In the latter case, you would end up sort of preaching to the crowd since I'm an atheist.

    ReplyDelete
    Replies
    1. I appreciate your comments.

      On the first point, yes, I agree that my knowledge of Aquinas could stand to be improved, but I still think his whole way of thinking is governed by his metaphysical and theistic presumptions.

      On the second point, you say "The first premise, that there are ordered series of causes, makes the argument an a posteriori argument." That's right. In the first premise, Aquinas says "We see in the world an ordered series of efficient causes", and it is on precisely that basis that I object to his argument. He's making assertions about causation based on what we observe, but on that basis, there is no justification for asserting that there must be a first cause.

      On the third point, I see what you are saying about validity. Good point.

      Delete