'Modally Challenged' comments:
I've run into this argument on several occasions and while the author(s) insist theists will accept the premises, it's more the validity I'd appreciate your take on.
1) If God is possible, then God is a necessary being.
2) If God is a necessary being, then unjustified evil is impossible.
3) Unjustified evil is possible.
Therefore, God is not possible.
In this post I explain the distinction between validity and soundness, explain why validity is a modal concept, and then use this fact to show that the modal distinction between the necessary and the contingent applies outside the sphere of human volition, contrary to what followers of Ayn Rand maintain. Finally, I demonstrate the validity of the above atheist argument.
A. Validity and Soundness
To say of a particular argument that it is valid is to say that its conclusion follows from its premises. So a person who gives a valid argument reasons correctly. But that is not to say that the premises are true or that the conclusion is true. One can reason correctly from false premises, just as one can reason incorrectly from true premises. The ideal, of course, is to reason correctly from true premises. Correct deductive reasoning from true premises will always bring one to a true conclusion. An argument that embodies such reasoning is said to be sound. A sound argument, then, is a valid deductive argument all of whose premises are true. It goes without saying that sound, like valid, is a terminus technicus. That being so, people are not entitled to read their own idiosyncratic meanings into them. Because 'valid' is a technical term, no one should be shocked that the following syllogism is valid: Every pig is a tiger; every tiger has three legs; ergo, every pig has three legs. After all, the conclusion follows from the premises: one who reasons this way reasons correctly, albeit from false premises. Not that anyone would give such a silly argument; but it makes the point that validity and truth are not to be confused. Validity and invalidity are properties of argument forms; truth and falsity are properties of propositions.
B. Validity as a Modal Concept
Some will be tempted to say that a valid deductive argument is one the logical form of which is such that no argument of that form has true premises and a false conclusion. But that leaves out something essential. There could be an argument A whose form F is such that no argument instantiating F has ever been given or ever will be given in which the premises are true and the conclusion false. But it doesn't follow that F is a valid argument form. So what we need to say is that a valid deductive argument is one whose form is such that no actual or possible argument of that form has true premises and a false conclusion.
This shows that validity is a modal concept. We cannot define it properly without the alethic modal concepts of possibility and necessity. Thus we say that in a valid argument the premises necessitate the conclusion, or that, given true premises it is impossible for the conclusion to be false. But the best way to put it is as follows. Let A be an argument with premises P1 and P2 and conclusion C. A is valid iff its corresponding conditional is necessarily true. Thus A is valid iff
1. Necessarily, if P1 & P2, then C
which is not to be confused with
2. If P1 & P2, then necessarily C.
To say that the premises logically (as opposed to causally) necessitate the conclusion is not to say that they render the conclusion noncontingent. A valid argument can have a contingent conclusion.
C. Validity/Invalidity and the Sphere of Human Volition
Let's apply the result of the preceding section to the question whether the modal distinction between the necessary and the contingent applies outside the sphere of human volition. The followers of Ayn Rand maintain that it does not, that it applies only within the sphere of the man-made, the sphere of that which can be affected by human will and choice. Whether an argument is valid or invalid, however, is independent of human volition. If an argument is valid, then its corresponding conditional is a necessary truth, as explained above. If an argument is invalid, then its corresponding conditional is not a necessary truth, which is to say that it is possible that its antecedent be true and its consequent false.
Since the validity/invalidity of arguments is one-to-one with the necessity/contingency of the arguments' corresponding conditionals, and since the validity/invalidity of argument forms is idependent of human volition, it follows that the modal distinction of the necessary and the contingent applies outside the sphere of human volition contrary to what the followers of Rand claim.
Note that it would be irrelevant to point out that people make arguments by uttering declarative sentences and arranging these sentences with words like 'since' and 'therefore.' No doubt people freely do things like utter and write down sentences. Sentences and arrays of sentences are man-made. But it doesn't follow that the validity/invalidity of the arguments expressed in these man-made sentences is itself man-made
D. The Validity of the Above Atheist Argument
The above atheist argument is valid. This can be demonstrated by deriving the conclusion from the premises using recognized rules of inference. Thus from (1) and (2) by Hypothetical Syllogism we deduce (4) If God is possible, then unjustified evil is impossible. From (4) and (3) by Modus Tollens we derive the conclusion that God is not possible.
So the argument is valid. But all that means is that the conclusion follows from the premises. Whether the premises are credible is a further question. Most theists will accept the first two premises, but will reject the third. Obviously, if one is a theist one believes that all evil has some sort of justification whether or not we understand precisely what that justification is. If you think that (3) is self-evident, then think again.
Leave a Reply