Tuesday, March 01, 2011

Middlebrow Atheism. Part 1.

The video above was posted on PZ Myers’ blog under the heading “When will that fine tuning argument evaporate?” In this series of postings I will be working my way through the issues it raises. But before I do we should recall that the “hard” science of physics uses two main kinds of explanatory object to describe our world:

a) Functions (or laws) capable of generating patterns which putatively describe the configurations of our world. These functions by and large come under the Church-Turing thesis which conjectures that all such functions can be rendered computationally.
b) Disordered configurations that can be treated statistically.

Using my own short hand I refer to these two kinds of explanatory object as “Law and Disorder”. Given this pair of explanatory objects I’m mindful of three caveats:

a) Although physics has successfully used Law and Disorder to describe large swathes of reality there is no obvious reason why objects which do not come under the heading of either law or disorder should not exist.
b) Even if physics can successfully explain the whole Cosmos it is clearly impractical for all the patterns in our world to be reduced to physics. For example, history has no practical way of using “Law and Disorder” explanations in the thoroughgoing sense that physics attempts to use them. Thus history is likely to remain a very narrative intense discipline. In short history is practically, if not theoretically, irreducible too Law and Disorder.
c) Law and Disorder explanations, no matter how successful, are destined to leave an inevitable Grand Logical Hiatus, an irreducible kernel of brute fact. Attempts to explain a given law and disorder regime using law and disorder is liable to lead to a kind of “turtles all the way down” regress. It may well be that multiverse explanations are sign of this logical/mathematical barrier as attempts are made to explain the cosmic context by embedding it within higher contexts (a kind of “contexts all the way up” explanation)

Now to deal with the video item by item:

Video Item 01: Given that the overwhelming majority of the Universe in space and time is clearly sterile it is hardly appropriate to suggest that the Universe is finely tuned for life.

My Comment: The video comes up with a rough upper limit on the cosmic ratio of life to non-life in the order of 1 to 10 to the power of 100. The implicit conclusion is that given the presence of life in terms of space-time occupied is so trivial, then life has too small a representation to claim the universe is finely tuned for it. But this calculation gives a false impression: We should not be looking at physical space and time but combinatorial space, the space of possibilities unconstrained by an apparently arbitrary set of laws and constants. Amongst the class of all possibilities the overwhelming number of configurations is disordered. Inside that class the complex ordered configurations of life has a representation as negligible as 1 in 10 to the power of many trillions (and I’m probably understating it). Thus, even if there should be only a single planet in the whole visible Cosmos that contains a few replicating cells, this seems an unwarranted over representation of an otherwise very unrepresentative class of configuration. Therefore, from the perspective of the visible universe it seems that the cosmos (presumably, as a result of its particular L&R regime) is highly skewed toward life. It follows that the existence of life cannot be trivialized away as a non-problem – we clearly have a non-trivial question of why the visible Universe is overwhelmingly biased toward life. So in this sense, contrary to the claim of the video, the visible Universe is fine tuned.

Video Item 02: Since we don’t know what space of possibilities the universe has been taken from we can’t work out any probabilities in regard to its apparent fine tuning. Probabilities in this connection are meaningless.

My Comment: Yes and No. “Yes”, because we don’t know the engine that has generated our Cosmos, a Cosmos with its particular laws and constants which favour life. Thus we can’t assign a probability in the way that we can assign a probability to patterns generated by tossing a coin. We simply don’t yet have a well understood generating context for the Universe. In any case who knows whether the “universe generator” is a multiple trial system like a kind of cosmic tossing machine that creates universes?  Positing such rather makes the assumption that some kind of disorder mechanism akin to coin tossing is at work. In particular perhaps the mechanism that has generated our Cosmos lies between the extremes of disorder and the high order of physical laws, in which case the concept of probability may be meaningless.

And “No” because in spite of the fact that the probabilities of a configuration of fine tuned parameters cannot be meaningfully quantified, we are still left with the extreme rarity of the visible Universe in the space of all possibilities. Given that we don’t know why one possibility should be chosen over another, each possibility seems to be of equal weight. Given that disorder far outweighs order then it is clear that our universe is taken from a very a-typical class, thus leaving us with the non-trivial question of its selection. So contrary to the suggestion of the video we can still have an appreciation of the remarkable singularity, if not the improbability, of the visible universe

Video Item 03: The probability of a particular configuration generated by the tosses of a coin or the cards withdrawn from a shuffled pack has a vanishingly small probability and yet nobody claims it is miracle of creation given that these configurations arise routinely in gambling casinos.

My Comment: The reason why nobody bats an eyelid when a particular sequence of, say, coin tosses arises is to do with the class of the outcome. Disordered configurations have an overwhelming representation in the class of possibility. Hence, if a random process generates disordered configurations this is to be expected as such configurations are selected from an overwhelmingly large class and therefore the selection of an item from this class is overwhelmingly probable. However, if a process generates ordered configurations, especially the complex ordered configurations of life, this presents us with a non-trivial problem as these configurations are taken from a class that has a vanishingly small representation in the class of possibility. Thus, given that compared to the huge space of platonic possibility the visible Universe is in fact quite a small cosy place, we are left with a non-trivial question of why such a unique class is so favoured as to make an appearance in our tiny universe. Contrary to the suggestion of the video, then, the structures of the Universe are startling rather than banal

…to be continued…

No comments: