I see so many creationists argue against this law of nature, as if they actually know what it is. Just to clarify something for theists the 2nd law of thermodynamics is based on probability considerations. It's in some respects more a law of mathematics than physics: It's easy to throw a coffee cup into pieces (1), but the other way around, throwing the pieces back into a whole cup (2) is virtually impossible. But the latter isn't any harder than throwing a cup twice in exactly the same pieces scattered exactly in the same way (3). The difference between (1) and (2, 3) is that all end results count in (1) while only a single end result counts in (2, 3) so the first has a much higher probability. So, yes, evolution needs luck and thus many tries. That's why it took a billion years with the first very simple steps taking most of the time. The 2nd law is also often the center of (vague) philosophical discussions about the "arrow of time" often overlooking the probability aspects. Also, Speaking of probabilities and likelihoods... The odds of drawing any particular 5-card poker hand are smaller 1 in a hundred million. But every time I'm dealt a hand, I don't marvel at the extremely low likelihood of my getting dealt that hand; and I don't dismiss the reality of my holding that hand just because the odds of my getting it are so low. And to take this a step further, the odds of being dealt a specific hand is independent of the hand - any particular royal flush is just as likely as any specific arbitrary hand. Now, that royal flush looks highly ordered, but that doesn't make it less likely to get than any other particular hand. Another thing I want to discuss is the time-evolution of a sytem. Darwinian Evolution Theory requires a mechanism whereby certain specific outcomes evolve differently from other outcomes. In other words, all outcomes do not evolve identically, and this "selection" helps improve the odds of having complexity evolve from simpler things. Here's an example to illustrate this selection : Take a big box and fill it with two sizes of balls - 1 inch marbles and 5 inch balls. Chuck all these balls randomly into the box. Now, continuously tap the box (gently), so the balls can rearrange themselves. Slowly, you'll see the smaller balls settle to the bottom, with the big ones above them. Given enough time, all the little balls will form into a lower tier, and the big ones into an upper tier - a perfectly ordered state ! What caused this ordering is that some balls - the little ones - are better than others - the big ones - at finding their way through openings and getting to the bottom. If it were equally easy for both kinds of balls to head for the bottom, the probability of arriving at this ordered state would be incredibly small. But because of this mechanism of selectivity, this ordered state is virtually a certainty in the long run. So the very notion of an event's (nonzero) probability presupposes that the event can occur. If, for some reason, the event cannot occur, then, surprise, the probability of the event is zero. To suggest that very low probability events cannot occur without some form of intervention is disengenuous at best; dishonest at worst, and is just plain wrong to the core. ID? Who want's such a pessimistic -- life's too complex; just give up trying to understand -- spin centered doctrine? 1. How "unlikely" is the evolution into our present form? I have seen several estimates (and they ARE estimates in the roughest sense) of the chances of DNA's forming into its current form. Sure, the probability is very small. However, would anyone like to look at the probability of finding the top quark at the Tevatron several years ago? Let's see, they found... what, 8 events out of how many gazillion, gazillion events? And this is not counting those that are vetoed outright. Particle collider experiments are the MOST demanding scenario in terms of data acquisition, storage, and processing speed because of the mind-boggling quantity of data being gathered in just a fraction of a second. Someone should show these ID'ers the probability of finding a "positive" event out of all of these gazillion interactions and COMPARE that number with the numbers they're putting out for the DNA formation. Why this hasn't been done, I don't have a clue. The main point here is that just because the probability "order of magnitude" appears to be miniscule doesn't mean we have not seen such occurence already, even within our lifetime. When the probability for something to occur is small, but there are a gazillion candidates, the phase space for that to occur is still reasonable enough that it CAN (and has) happened. 2. Often, the calculation of such probability itself is highly dubious. It assume that one starts off already knowing the final phase space that one has to end up with. Let me give an example. Let's say I start off with 4 letters, A, B, C, and D. I have 10 slots to fill this letters with (repetition is allowed). I want to know what is the probability that, after a random selection, I end up with a sequence that such as BBDBDAEDDA One can do the straightforward calculation there. One can accurately argue that the phase space (or probability) of getting such a sequence is small. However, is this really what is going on with the evolution process? I would make a definite claim that it isn't. First of all, the formation of our DNA doesn't not happen all at once at the beginning of the evolutionary process. Natural selection dictates that based on the external ecological pressures, there will be traits that will be more favorable than others. Now unless I slept through reading the formation of our universe and earth, the earth a LONG TIME ago is not the same earth that's here today. There is just a different ecological pressures when it is mainly water/molten rocks/etc. Our current forms are just not "favorable" back then! As the environment changed, so does the external pressures, and different traits became more favorable. In other words, the DNA selection changes gradually. So maybe, using my example above, you have filled only the first 5 sequence of BBDBD_____. Now, the probability phase space to end up with BBDBDAEDDA is no longer as large as in the beginning. You have already established the first 5. It's like flipping a coin 4 times and asking for the probability that you end up with 4 heads. While the probability at the start is of course (1/2)^4, if you have already obtained 3 heads, then the probability of getting all 4 heads is just 1/2. For some reason, I haven't seen this argument put forth convincingly to people who are being seduced into believing in this probability game. 3. There is also the major assumption that ending up with BBDBDAEDDA sequence is the ONLY possible option. I mean, at the very beginning of the selection, how do we know that BBDBDAEDDA is the ONLY sequence that would produce anything worthwhile? Now, if we question that, then let's play this game.... Let's say I end up with ABACCAEDBB. There! I just got a sequence! Now, if we look at it AFTER THE FACT, someone can say "WHOA! The probability of getting that sequence is VERY low. How'd you do that?" I can just walk around and say "Oh, I'm very good at this" or "Well, I'm just a very lucky person". Yet, I didn't plan on getting that sequence. It just came up randomly. Obviously, someone who looks at it AFTER the fact, thinks I'm very lucky because it is VERY highly unlikely to get that sequence. In fact, ABACCAEDBB could be a new creature capable of producing anti-graviational effects via zero-point fluctuations! Remember, evolution never had a "final design" in mind. It is simply a trait being selected at that given moment due to all the external pressures and the available nitches in the ecology. So the sequence could easily be something else if our earth made a right instead of a left turn at Albuquerque.

Post edited by: vincent, at: 2007/01/23 22:16