My post was concerning other philosophies and outspoken empiricists, but you bring up an interesting point. I don’t know what most scientists think about mathematics, but you’re more than likely right seeing as how a fair number of them are religious (at least within math we come to the same conclusions). I would think all truly hard empiricists see it as more of a tool, like logic and reason, than a form of true knowledge. Like logic and reason; Math is normally in concordance with empirical evidence, but that does not make it, like you said, empirical. I say normally because people can fail at them: Myself included.
A sign things may be changing for the better somewhat: John Horgan has put up two interviews with prominent physicists, cosmologist George Ellis (who seems to share my views on Lawrence Krauss’ latest book) and loop-quantum-gravitationalist Carlo Rovelli, claiming that ‘philosophical superficiality has harmed physics’.
(Of course, this opens up a more difficult question: if my opinions become mainstream all of a sudden, how do I fuel my persecution complex?)
Excellent interview with George Ellis, thanks for posting (I haven’t read the other one yet) – a fellow who seems to inhabit a fascinating intersection of cosmology, math, and philosophy. It’s good to see Krauss taken down a few pegs – the blowhard deserves it! As I said before, Krauss has some good observations about the extent to which physics has increasingly encroached on the former exclusive domains of philosophy and religion, but he’s wrong about implying that it has somehow obliterated them, or ever could. In particular, he extends the phenomenon of quantum fluctuations into a full-fledged argument not only for the spontaneous origin of the universe, but for atheism. Putting aside the anti-religious aspect, as Ellis points out, Krauss’ argument tells us nothing about the origin of the universe or how the laws of physics came to be. Krauss himself acknowledged that his book title (“A Universe from Nothing”) is intentionally provocative, and Ellis quite properly accuses him of using philosophy – badly – to try to put down philosophy!
Ellis makes a lot of other interesting observations that sound plausible: that we are approaching the limits of observational discovery both in high-energy physics and in cosmology, and that scientific discovery will be in the behaviors of complex systems rather than the discovery of new objects in the realm of the very small and the very large. Also his skepticism that either multiverse theory or string theory is scientifically meaningful because it will never be testable.
Hmmm… and then I read the rest of it…
Here, if I’m understanding him correctly, we seem to have a case of damn philosophers letting philosophical ruminations prevail over science again! Call me a “strong reductionist”, but complexity is the result of physical processes and can only have a physical existence. Any computer algorithm is nothing but a string of binary states, and it can do marvelous things only through the agency of intricate patterns etched in semiconducting silicon directing the flow of electric current through billions of switches – itself a physical object. Any idea in our minds is likewise just a pattern of neuron linkages.
He seems to be implying that something fundamentally new and different occurs at a sufficiently high level of complexity. There is an important sense in which this is true – a computer than can understand speech, talk back to you, and beat you at chess is a fundamentally different thing than an electronic calculator that adds and multiplies numbers – but the underlying physical processes are the same, and they are no less fully deterministic. You could theoretically build such a computer by taking apart a million calculators. You’ll have created complexity, but not changed the innate nature of its implementation.
wolfpup: my hackles rose at that phrase too. His point is trivially true – of the “So what?” variety – why, gosh, yes, human intelligence has let us create technological wonders. As you note, it is nothing more than a really complicated example of physics in action.
The only other possibility is that he is engaging in some kind of metaphysical spiritualism, positing a “creative soul” that only humans, as capable of philosophy, can possess, in which case he has left science way behind and has gone on to engage in a religious discussion.
Yeah, I’m not really on board with that part of Ellis’ opinion, either. I do think he made a good point regarding ethical and aesthetical judgments not being part of science, and in the sense that e.g. our ethical judgments may influence our actions, i.e. what physically occurs, there is a sort of ‘top-down causation’; but I believe it’s only a weak sort: if you fix the physical facts in some volume of spacetime, basically everything else goes along with it. But that may not be the appropriate level to talk about certain things: if I hit you, and you ask me why I hit you, an explanation involving the precise description of every atom bumping into atom, or every muscle cell contracting, or whatever micro-level you choose, may completely and fully suffice to describe the action, so to speak, but it’ll leave you wanting regarding an actual explanation. That explanation is framed much more usefully in terms of my intentions, for example.
But I don’t know whether Ellis was talking about anything of that sort; he’s a practising Christian, so he may well have had some spiritual component in mind, in which case, I’d disagree with his viewpoint. But that doesn’t invalidate his other points.
I’m glad to see us all in unilateral disagreement with Ellis on that point. As for the above, ethical and aesthetic judgments are subjective human perceptions that are by definition not within the realm of scientific interest, which is not a particularly insightful observation for Ellis to make. The only useful interpretation that I think one can put on his thesis is that as one increases a system’s level of complexity, one can achieve apparent intelligence, and then if one increases complexity even further one can achieve the aforementioned kind of purposeful ethical and aesthetic motivations that might be broadly defined as “spirituality”. Perhaps so, but that idea isn’t new, nor does it represent any kind of fundamental transition that transcends physical determinism: viz. Ray Kurzweil’s book The Age of Spiritual Machines – the title says it all.
The book, incidentally – which predates smartphones and tablets by many years – predicts the development of “wearable computers” (which in effect is already happening) and the eventual physical integration between computers and human minds and bodies. The other point, and the one behind the title, is that computers will evolve well beyond the point of “intelligence” and will eventually, in exceeding human intelligence, acquire and exceed those attributes that we associate with human spirituality. But they will still likely be using silicon to process bits, and if not, the technology will be functionally equivalent.
Well, it might be obvious to you, but many of the really strong reductionists and critics of the validity of any form of philosophical inquiry whatsoever indeed hold that human values are accessible to science—case in point: Sam Harris’ The Moral Landscape: How Science can Determine Human Values.
I think one can give a charitable reading to Ellis’ statements by considering the position of strong reductionism to be analogous to that of Leibniz’ calculemus, the attitude that for a sufficiently mature science, all differences of opinion can be resolved by a mere computation. That we know to be wrong: there exists a threshold of complexity beyond which such a first-principles reasoning is simply impossible, not only in practice (which is obvious), but even in principle. The reason is that any system that is computationally universal is essentially unpredictable: if you could, in any way, decide all non-trivial questions about its evolution, then you could also solve the halting problem, which you can’t. This follows from Rice’s theorem: you can’t, in general, decide what any given algorithm is going to do.
Of course, if you have sufficient computing power, you can just set up an explicit simulation of any given system, and speed it up; but even then, you won’t know when, or if, you’ll ever get answers to the questions you’re interested in.
In this sense, there is something new beyond a given threshold of complexity, something you won’t find using a Leibnizian approach.