Re: [asa] CSI Forensics WAS Staggering drunk WAS Romans 1:20

We can describe one state as more "interesting" than another but that's subjective. A more objective way of describing the difference between (e.g.) a microstate with all the molecules in 1/2 the room & moving in the same direction & one with them distributed in a more uniform fashion is that the latter is more symmetrical than the first - it's invariant under a larger group of transformations. So the "interesting" state is one with "broken symmetry" - a concept which has been very fruitful in physics. Maybe it could be in biology too.

Shalom
George
http://web.raex.com/~gmurphy/
  ----- Original Message -----
  From: Iain Strachan
  To: George Murphy
  Cc: Dick Fischer ; ASA
  Sent: Friday, November 23, 2007 9:49 AM
  Subject: Re: [asa] CSI Forensics WAS Staggering drunk WAS Romans 1:20

  I think George's illustrations are very apt.

  A point that occurred to me recently was that all possibilities are equally likely for sure, but "interesting" possiblities are far less likely than "uninteresting" possiblities. All the distributions of gas that are evenly spread over the container are not interesting, but one where all the molecules are in half the room is very interesting indeed. It also boils down to the point I've made many times before about description length. If the molecules are restricted to one half of the room, then you need precisely one less bit per molecule to specify the positions, so the description length is reduced by O(10^24) bits.

  Likewise a Royal flush is much more interesting than the arbitrary hand that George gave!

  However, the key problem is that in ID, it seems to me, the proponents seem to think that "interesting/unlikely" is equated with the action of an Intelligent Designer, whereas it might be due to a physical process whose details we haven't yet fully fathomed out, which makes what was previously considered extremely unlikely, actually very likely.

  I also had a thought on whether (non)-randomness implied design - what started this thread. There are two examples:

  (1) Imagine chucking a bucket of water down a water-slide of the type you see at leisure centre swimming pools. Despite the randomness of the motion of the water molecules, overwhelmingly many of them go down the slide in a well-determined fashion. Now one might argue that this points to the fact that someone designed the water-slide to be a particular shape. Intelligent Design? No the analogy fails because:

  (2) Consider water flowing in a river. It also follows a twisty course in a deterministic fashion, but the processes of erosion and the laws of physics carved out the best path through the terrain. No-one designed the water slide - its shape arose naturally.

  Iain

  On Nov 23, 2007 1:59 PM, George Murphy <gmurphy@raex.com> wrote:

    I won't get into the debate on ID, Rom.1:20 &c here but want to contribute something that may help to elucidate the point about statistics that Iain is making below. When I 1st studied statistical mechanics from Sears I learned what I assumed then was - & still think should be - standard terminology but apparetly isn't, the distinction between a microstate and a macrostate. (Of course it's the concepts rather than the words attached to them that are important, but having distinctive words for different things can be helpful.) If we're talking about a container of gas, e.g., one specifies a microstate by giving the position and momentum of every molecule of the gas. A macrostate, OTOH, is specified by giving just the numbers of molecules with different positions and momenta (within limits specifying the precision of measurements).

    The basic assumption is that all microstates are equally probable. A given microstate with each molecules in my study in the north half of the room & travelling due north is as likely as a given microstate in which the molecules are evenly distributed throughout the room and in momentum space (i.e., moving in different directions with different speeds). But all macrostates are not equally likely. In the example, there are far more macrostates with the molecules evenly distributed than there are macrostates with all the molecules in one half of the room. The thermodynamic probability of a given macrostate is defined as just the number of microstates corresponding to that macrostate. (& entropy is then Boltsmann's constant times the log of that probability.)

    In poker, a hand with 10, J, Q, K, A of diamonds is exactly as likely as that with 2 & 6 of hearts, K of spades, and 6 & 9 of clubs. But a royal flush is lots less likely than one with a pair because there are lots more ways of getting the latter.

    Shalom
    George
    http://web.raex.com/~gmurphy/
      ----- Original Message -----
      From: Iain Strachan
      To: Dick Fischer
      Cc: ASA
      Sent: Friday, November 23, 2007 4:03 AM
      Subject: Re: [asa] CSI Forensics WAS Staggering drunk WAS Romans 1:20

      Why oh why do certain members of this list keep on making the same fallacious straw-man argument over and over again despite my many attempts to explain it? It seems to me that any mention of the word "probability" on this list results in yet another excuse to bash ID. I believe the correct response to ID is to point out that events that seem unlikely may well turn out to be highly likely once we've understood the processes involved.

      But this silly "low probability events happen all the time argument" is completely vacuous. Sure, if four players sit down at a bridge table and deal the cards, then the chance of them getting that set of four hands is one in 5.36x10^28 (The impressive figure given by D.F. Siemens in all its 67 digit glory in an earlier post turned out to be too big by a whopping 39 orders of magnitude, because he failed to take into account that the same bridge hands arise independent of the order in which the 13 cards in each hand are dealt). Nonetheless the 5.36e28 number is pretty impressive, but it doesn't surprise us because any set of hands has the same chance. But what if you shuffled the deck and dealt out the cards again and got the same set of four hands? Then you would be surprised, because it conforms to a specified target (it matched the previous hand).

      Likewise consider the case when two parents have a child by the usual means. Each sperm has 23 chromosomes, each one chosen at random from each chromosome pair. The same goes for each egg cell. Hence when I was born, I was one of 2^46 possible individuals that could have been conceived. Therefore God? as Dick would say. Of course that's a silly statement, but the point is that no one in the ID camp would make that claim. Dick's argument is just a straw man. But suppose my parents had a second child after me by the usual means and it turned out to be an identical twin genetically? At this point the 1 in 7x10^13 chance is significant because it conforms to a target (ie it's the same as me).

      Honestly, until you start to grasp this elementary point your arguments look every bit as foolish as some of the stuff coming out of RATE. Are you going to get it this time?

      Iain

      ............................................

  --
  -----------
  After the game, the King and the pawn go back in the same box.

  - Italian Proverb
  -----------

To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Fri Nov 23 11:14:51 2007