It’s seriously disgraceful that people still cling to the Cosmological Argument. It’s like they think philosophy hasn’t progressed in the last few hundred years. The Cosmological Argument always comes down to infinite regress, self-contradiction, or fallacious Special Pleading, depending on how they try to weasel out of it. If your hypothetical agent is exempt from the laws of science and the rules of logic then that’s fine; but you may no longer pretend your position is logical or scientific.
Philosophy does not progress linearly. There are a lot of serious, non-theistic contemporary philosophers (John Searle, for instance) who argue that modern shift due to Descartes, Hume, Kant, etc. was far too extreme. Contemporary philosophy is also more amenable to, for instance, scholastic metaphysics than one would think. There are various strands of essentialist metaphysics (Saul Kripke, Hilary Putnam, Brian Ellis), dispositionalist causal theories (Stephen Mumford), causal pluralist theories (Nancy Cartwright), emergentist metaphysics, metaphysical/scientific realisms (Putnam again, E.J. Lowe, Searle again), etc., that no one would have thought possible in the middle of the modern philosophical “progression” that succeeded scholasticism.
It seems false to say that the cosmological argument “always” comes down to infinite regresses. Alex Pruss’s cosmological argument (in his book on the principle of sufficient reason) does not really invoke an infinite regress. That said, to make a claim about an infinite regress hardly constitutes a fallacy like “self-contradiction” or “special pleading.” That an essentially ordered causal series requires a first mover seems indisputable; what one might dispute more plausibly might be whether there are any essentially ordered causal series, as a Thomist would define them, or that the first mover could rightly be called “God.”
Regarding whether a first mover could be “exempt from the laws of science,” there are a few comments that should be made.
First, the argument purports to show that essentially ordered causal series require a first mover which, because it is purely actual, must be immaterial. It is not special pleading in the slightest to suggest that “the laws of science” would not apply to such an entity, since an opponent could not, I think, plausibly claim that the laws of science should apply to an immaterial being. (The suggestion seems dubiously coherent.) If they want to question whether an immaterial being is coherent, then that should be done at an earlier stage of the argument, and has nothing to do with whether or not the laws of science apply to it.
Second, the Thomist construes “the laws of science” as dispositions/powers of the material world. They are not abstract principles that act on the physical world to ensure it runs smoothly. Given the Thomists commitment, then, it is not special pleading at all to suggest that an immaterial entity is not constrained by the laws of physics. The laws of physics do not properly “constrain” anything, and those things to which they don’t apply (if there are any such things) are not properly “exempt” from them as a consequence of being “above” them or anything of the sort. The laws of science are abstractions from regularities rooted in powers intrinsic to the material world. (It should be noted that if the objector is a Humean and holds that the laws of science are brute regularities, then he is not himself committed to there being exceptionless.)
