There is a section in the God Delusion (Dawkins, 2006) early on, where Dawkins’ addresses a certain confusion about how Russell had claimed in his early philosophical career that the proof of God’s existence by means of appeal to essence (namely, the Ontological argument) is actually valid. Dawkins then gives an anecdote where he gives a flippant variant of the ontological argument to prove that some trivial fact is necessarily true and he ends with the punchline: “They felt the need to resort to Modal Logic to prove that I was wrong.”
While the remarks that Dawkins makes on the chapter concerning A priorist arguments treats the enterprise as fruitless and a joke; he does make a half-serious point. Gaunillo, as Dawkins rightly attributes, gives the case of the ‘most perfect island’ that must necessarily exist, if we are to accept the inference that Anselm wants to use to prove God. One common response is to give add some caveats to the thing that we are trying to argue into existence. Adding caveats like ‘most perfect being’ cannot refer to contingent beings and are necessary by definition may attempt to exclude. Anthony Kenny, in his exegesis of Descartes’ ontological gives a hearing to this view with some comparison cases; firstly, Russell’s notion of the Gold mountain and secondly the more thorny issue of non-existent objects.
There is a case, if we accept the theorems of S5 logic where we might say that the ontological proof of Good is valid (which banks on the S5’s Rule of Necessitation). We can say that a maximally perfect being exists possibly. However, valid as it is; it does require some metaphysical steps to infer that it is the man with the beard. Dawkins addresses the so-called ‘Einstein’s God’ which is mischaracterised by theists, and he states from the offset that this is not the notion of God that he’s attacking. It is, however, this very ‘Spinozan’ notion of God (deus sive natura) that the ontological argument apparently proves, and not the Christian deity. Most Christian apologists who argue with the ontological argument in public debates always use the religious experience/testimony argument concerning the historicity of Jesus, because the S5 compatible proof allocates a non-religiously-affiliated God. What I find most interesting about this proof is the metaphysical world that it would entail, given S5 modal conditions (such as the issue of natural kinds, worldhood and perhaps the issue of universals).
In closing. Modality is no flimsy subject matter but one that has serious implications in term of systematic philosophy. I suppose my irk with Dawkins concerns a distinction between the protestant and catholic atheist which is so-joked about. My Thomistic tendencies would emphasise the role of reason and the a priori in terms of structuring reality. Aquinas, in his conciliatory effort to bring Aristotle to Christianity, believed that the Aristotelian method and reason itself must have a place in elaborating our view of the universe. Later theologians debated between the notions of analogia entis and analogia fides; whether our comprehension of the divine comes exlusively through scripture (and its empirical connotations) or our rational attempt to structure ultimate reality. The comment made my Dawkins seems insincere to the a priorism that supercedes its theological origins. I find it amusing that there would be such a distinction between a ‘protestant’ (empiricist) and ‘catholic’ (rationalist – or, reason + experience) atheist.