Oh dear. I have a great deal of admiration for Luke over at Common Sense Atheism. In particular, he has a very interesting philosophical approach to the whole question of atheism and in countering the arguments of the likes of William Lane Craig and Alvin Plantinga. I will be honest though – sometimes high-end philosophical chinstrokery leaves me a bit cold, and it can vanish down rabbit holes that are fairly irrelevant.
Luke's (uncharacteristic, it has to be said) banana skin here is in trying to take on Richard Dawkins' The God Delusion on William Lane Craig's terms. That should have been a red flag. A short synopsis of Dawkins' central argument in TGD:
Theists look at this extraordinarily complex and beautiful world and assume that since it is so complex, the best explanation must be a designer. Dawkins correctly (and despite some rather bizarre retorts to the contrary) points out that for this to be an explanation for complexity, the designer must be more complex, and therefore even more in need of explanation than what it allegedly created.
Luke contends that if you had to explain the origin of everything before it can be considered an explanation for anything else, you would never be done – you would enter an infinite regress of explanations. An example would be if you found a carefully crafted arrowhead - to explain that you would invoke humans. But then you would have to explain humans (such as via evolution). But then you would have to explain evolution and planetary formation and the Big Bang etc etc. This was Craig's way of missing the point, and I'm disappointed to see Luke swallowing it. The reductio ad absurdum really should have told him something was wrong.
So here we go. Dawkins was writing a (hugely popular) book for the mass market, not for high-end philosophers. Most people got what he was driving at. Some philosophers and theologians (notably Terry Eagleton and Karen Armstrong and Alvin Plantinga) got their knickers in a twist over Dawkins' failure to apply esoteric philosophical jargon in his arguments, as opposed to some straight talking. But that is of no import - these commentators have plumbed new depths of silliness that need no further comment from me.
But let's look at Luke's key objection to Dawkins here. Is there really an infinite regress of explanations? The answer of course is NO. Let's have a look at why that should be the case. In the case of the arrowhead, you invoke humans as your explanation. Fine. Now you have to explain humans. Exhibit B. Humans. All over the place. There ya go. The point here is what you are ULTIMATELY trying to explain is the arrowhead, and you've done that. The humans by this stage are a given.
This is manifestly not the case for a "designer", and Dawkins is quite correct to point out that if you disallow the complex coming from the simple (as some people still do, despite proper mathematical proof that they are wrong), then you are necessarily positing a more complex designer than the thing designed, which means you have even more heavy explanatory lifting to do to get yourself past that hurdle.
So why has Luke made this mistake? It is actually fairly simple - he has forgotten what we are trying to do with our explanation. Maybe it is a philosopher thing. Scientists, on the other hand, are well used to working with "black boxes" - we interconnect them, analyse their inputs and outputs, and try to infer their internal states. Where we can, we carry out experiments whose data help us open and unpack black boxes, to reveal...
...more black boxes inside. However, we can still explain the black boxes at one level if we know the system at the next level down, if we can verify and work with that.
In the case of the origin of the universe, that is a black box. However, theists propose that INSIDE that black box there is a BIGGER black box that you are not ALLOWED to open. That is a big claim, and no philosophical piddling is going to justify that - what is needed is EVIDENCE.
So was Dawkins right in his argument? I would suggest YES. I think Craig and Luke have got this wrong; they have lost the train of what the explanation is for in this context, and switched tracks mid-argument. Sorry boys - try again!