Monday, June 02, 2008

A proposal

About a year ago, If I remember correctly, I got into a conversation with someone regarding the Multiple Universe theory. I asked him why he considered it scientific. Evidence, he said. Like what?, I asked. The math, he said.

The math. I disengaged at that point. I was dealing with a zealot and didnt want to get into his spiraling down into fractal wrongness. But now that I think about it, I may have been harsh in my assessment. He was just regurgitating the views of prominent theoretical physicists. In this new phase of science that began with Einstein, evidence is not as important as mathematical beauty. Paul Dirac once famously said that ''it is more important to have beauty in one's equations than to have them fit experiment.'' Therefore, if you postulate a multiple universe in an equation youre creating, and have that equation turn out to be beautiful, then you are assured that multiple universe is possible -- never mind that you postulated its existence in the first place -- and therefore multiple universes is a legitimate scientific theory and not science fiction. Beauty is truth; truth beauty. That's all you need to know.

The same is true of Darwin's theory. It is such a simple, elegant theory, and as such has proven seductive that natural and social scientists have been mesmerized into applying it in all aspects of life. Darwin's theory isnt survival of the fittest. That's a tautology. Of course the fittest survive. (Sure, sometimes the luckiest survive, but one couldnt build a scientific theory out of 'survival of the luckiest'.) Darwin's theory says that random mutations occur all the time in living organisms and then nature selects, through a blind, unintelligent process, without goals or purpose, who gets to pass on their genes. See how this process is not applicable to politics, culture, ideas, society in general? But the theory is so friggin beautiful one can't resist it. In biology teacher blackshama's words in a comment in a post over at Philippine Commentary, it has great explanatory power.

And the evidence? This is what Richard Dawkins has to say about the evidence:
MOYERS: Is evolution a theory, not a fact?

DAWKINS: Evolution has been observed. It's just that it hasn't been observed while it's happening.

MOYERS: What do you mean it's been observed.

DAWKINS: The consequences of... It is rather like a detective coming on a murder after the scene. And you… the detective hasn't actually seen the murder take place, of course. But what you do see is a massive clue. Now, any detective…

MOYERS: Circumstantial evidence.

DAWKINS: Circumstantial evidence, but masses of circumstantial evidence. Huge quantities of circumstantial evidence. It might as well be spelled out in words of English. Evolution is true. I mean it's as circumstantial as that, but it's as true as that.
Dr. Dawkins has this penchant for picking the most inappropriate analogy for his theory. He did this in the third chapter of The Blind Watchmaker when he tried to demonstrate natural selection by writing a program wherein the phrase METHINKS IT IS LIKE A WEASEL emerges from a random string. He was able to do it in 40 or so generations, which is pretty impressive, except that the target phrase is already programmed into the system. The system knew what it was looking for. In trying to illustrate Darwin's theory, he inadvertently illustrated some form of intelligent design theory.

He did it again with this interview with Bill Moyers. The crime scene analogy is a favorite among advocates of intelligent design, and here he is, using the same thing, and in a wrong way too. Darwin's theory does not allow a murderer. It assumes that no intelligent agency is responsible for the crime. Darwin's theory only allows accident as a cause of death in the crime scene. In fact, Darwin's theory will say no crime was committed at all.

A year ago, I said that it was impossible to model Darwin's theory using computers. That's because you always come back to the programmer, that is, you always come back to an intelligent agent who programmed the whole thing, who determine who gets to survive and all that. METHINKS IT IS LIKE A WEASEL all over again. Here is a critique of the attempts so far from the Panspermia site. Dawkins's little foray into computers is included.

However, with the powerful supercomputers we have nowadays, we can at least try to mimic Darwinian evolution using computers. Here's how it can be done:

1) Take large set of characters and arrange them in random fashion. For example: 2882482u9hedi1whei1uei1uer1urei1urncuio3`ur`2ru03r93949124012488*(&*^(*&)&(^&)(*YUOHHFTR%&&RFft78&*t88t87T78tT7t78tGyIkbh H+_Ouiuu009)8u0098u()890898292-3-83-234-14-2`3-48(***_(*)_+U(I)*_*_(++_*_**_*I_(NBOUHGTUKUJUGPUU@##$DRTFYT(*?>><L<L>O}{{
...

You need a very, very, very long string. This represents nature before life emerged.

2) Next, you will need an algorithm to simulate natural selection. This algorithm will have to be randomly generated and must not have any human intervention in it. The selection process will have to be 'meaningless' to the humans and can vary in complexity. For example, you can run find-and-replace sequences wherein the strings to be searched for are random strings and will be replaced by random strings. For example, in one instance, the program looks for the string 4995HUJijpp^55$ IF they are followed by the string 29820*(&)&BY*O, if in turn this occurs after the 679th character of a 1 MB worth of string is a G, etc... and replace it by the string iooo-012489*(*(_)=+, or it could be a simple algorithm like replace all A with G.

3) Next, you will need to reboot the system at random intervals and the machine will try to run the program created by steps 1) and 2) above. This represents the input of energy that we postulate happened to the primordial soup. If nothing happens, it will repeat processes 1), 2), and 3). The system does this until a self-replicating string emerges, or more accurately, until steps 1) and 2) comes up with a recognizable program to step 3) and results in a string that replicates. This represents the common ancestor.

4) The system then introduces random errors based on some random algorithm in the replicating string such that at random intervals the copies of the replicant will not be exactly the same as the "common ancestor." The 'errors' will just be one bit at a time.

5) An algorithm similar to step 2) is then applied to the whole thing again. We will not need step 3) anymore as it is assumed that the replicators can take energy from the existing environment.

With any luck, using ultra-powerful supercomputers, we'll end up with the same biodiversity from a common ancestor that Darwin's theory predicts using blind processes.

18 comments:

cvj said...

I haven't read about Dawkin's attempts but isn't natural selection responsible for the emergence of drug resistant bacteria? Or is your point more about irreducible complexity which we earlier discussed?

Also, in the area of Game Theory, haven't there been programs even back in the 70's (like that of Robert Axelrod) which modeled the evolution of cooperative behavior or are you talking about something else entirely like simulating actual evolution in a computer?

In that case, from what i read, you cannot model the universe (and everything that comes with it) using a classical computer but you can do so with a quantum computer. That technology is still in its infancy. Also, we would need an understanding of quantum gravity which is also as yet unsolved.

As to MWI and Copenhagen which we discussed in DJB's blog, the reason why both are considered scientific is because neither interpretations can be ruled out as implications of the underlying equations, and everyday experience can no longer be used as a guide in this area.

Jego said...

but isn't natural selection responsible for the emergence of drug resistant bacteria?

Yes. This is beyond dispute. What natural selection (or in this case, artificial selection) did was eliminate unfit bacteria, not create a new species of bacteria. Natural selection has not been shown to create new things, just destroy them. And of course there's this other thing that it promotes uniformity and not diversity.

The trouble with attempts to model Darwin, and I suspect this is also true of the Game Theory thing, is that they are too 'smart'. By necessity, the model should be dumb, goalless, or purposeless. No judgement is allowed as to who or what should emerge. So the kind of computer, whether classical or quantum, is not important.

As for the term 'scientific', it has been shown that it is simply a matter of consensus (or politics, if you will). The hard-and-fast rules such as testability, replicability, etc., have since been thrown out the window. Philosophy rules in that roost. As what we have seen in theoretical physics, beauty is now a criteria, whatever that is, more than experiment. As laymen, that is difficult for us to take, but that's the way it is in their world. Our role as free people then is to be skeptical of everything that comes out of their findings and not accept them just because they say so.

cvj said...

I see that your basic question is that how can there be a program without a programmer? Isn't that another version of the 'God of the gaps' argument? Do you accept this as valid? And if yes, can such acceptance be considered scientific (on your own terms)?

As for modeling Darwin, i think that has to wait until we develop a way to model the Universe (which is the point of my first comment). In the meantime, there are other ways to show that species are related, i.e. by looking at similarities in DNA.

In ever human activity, you cannot take away the social aspect. Scientific consensus (aka orthodoxy) is good in that it protects us from cranks. (Remember that 'electron is positive' guy?)

When did scientist announce that they are throwing out testability, replicability etc? When physicists talk about beauty, i think they do so in a mathematical sense which takes us back to your friend's comment about 'the math' which i take to mean that the math cannot rule out 'Multiple Universes' as an interpretation.

Jego said...

can there be a program without a programmer?

The answer is obvious. No.

Isn't that another version of the 'God of the gaps' argument? Do you accept this as valid? And if yes, can such acceptance be considered scientific (on your own terms)?

God of the gaps is too presumptuous. The 'murderer' didnt leave fingerprints. We just know that a murder occurred. Whether or not God did it doesnt change the fact that a 'murder' occurred. Dawkins et al is using a variant of god of the gaps which is Darwin of the gaps.

In the meantime, there are other ways to show that species are related, i.e. by looking at similarities in DNA.

Sure, but Im still not convinced that DNA is fool-proof. That 98% similarity with chimpanzees is too arbitrary. What criteria did they use to establish that? It seems like a number out of the blue. Remember, this number has been in existence even before chimp and human genomes were mapped. Going back to the program analogy, programs written using C++ would look related even though they can be orders of magnitude different from each other. Im not saying we're not related to chimps. Chances are we are. But DNA evidence is too, I dont know what to call it... facile?

Scientific consensus (aka orthodoxy) is good in that it protects us from cranks.

Yes of course. But it's a double-edged sword. Who gets to say who the cranks are? Scientific breakthroughs come from 'cranks'. Ive read some physicist say that if Einstein tried to publish today, he wouldnt make it past the first editor. Science then was more open. Science now is too big. And when something gets too big and powerful, that's when we should crank up the skepticism.

When did scientist announce that they are throwing out testability, replicability etc?

They didnt. They relegated it to the 'nice to have' category.

When physicists talk about beauty, i think they do so in a mathematical sense ...which i take to mean that the math cannot rule out 'Multiple Universes' as an interpretation.

They cannot rule it out because they cannot rule it out. It's that circular. They proposed its existence in the first place. The math followed after the fact. String theory is the same way. String theory, like multiple universes, like the existence of God, is one of those things that belong to the 'not even wrong' category, and yet...

cvj said...

Couldn't the 'obviousness' be a consequence of language used or, alternatively, because you are arguing by analogy? Isn't obviousness an unreliable guide since nature has confounded human intuition before.

I think the more accurate way to put it is that they cannot rule it out because of the math, i.e. the fact followed the math.

I'm aware that Lee Smolin and Peter Woit and a hosts of physicist have criticized String Theory. Will have to read that one of these days. I'm glad though that we're in agreement on the existence of God belonging to the not even wrong category.

BTW, you might be interested in checking out Seth Lloyd's "Programming the Universe" which has a discussion along the lines of your blog entry.

Jego said...

Isn't obviousness an unreliable guide since nature has confounded human intuition before.

Intuition has always been a reliable guide. It can be fooled of course, but on the whole, it has proven reliable. In fact most of our theories start with intuition. As Einstein said:

"The supreme task of the physicist is to arrive at those universal elementary laws from which the cosmos can be built up by pure deduction. There is no logical path to these laws; only intuition, resting on sympathetic understanding of experience, can reach them.

They start with intuition and subject these intuitions to rigorous testing. Einstein's intuition was guided by his sense of beauty, and the rest is history. Darwin, we could also argue, was guided by his sense of beauty; his theory is truly elegant. And the thing about him is his sense of science. He is not dogmatic. A fair portion of his book Origin... was dedicated to critiquing his own theory, telling readers how to falsify it. Scientists now, it could be argued, have done just that. It's just that it's hard for Darwin supporters to let go of the orthodoxy, as Thomas Kuhn has observed.

I think the more accurate way to put it is that they cannot rule it out because of the math, i.e. the fact followed the math.

I couldnt see how this could have happened. More likely, the idea came first, then the math to support the idea which included postulating the idea. The math couldnt conceive of multiple universes without the mathematician putting in parameters that supported it; it's a creation after all, like a work of classical art. It's not like some physicist playing around with some equations and saying, "I was fiddling around with some equations and out came this thing of great beauty and I couldnt make sense of it at first, until I thought, 'Hey! This would make sense if there were multiple universes.'"

But not to worry. If youre a Multi-Universe advocate, the mainstreaming of this idea is coming soon via the popular media. In fact, I could see its appeal. It's just not science.

Thanks for the Seth Lloyd tip. I'll go look for it.

cvj said...

In the case of MWI and Copenhagen, i think the fact followed the math on the question of what happens to Schrodinger's Cat (who, at one point, is both dead or alive) in the thought experiment of the same name. In the Multiple Worlds Interpretation (MWI), there is a version of reality where the cat met the opposite fate as it did in this world.

In the case of Multiple Universes, unless i remember (or understood) wrongly, it came about as an explanation as to why the Cosmological Constant is just at the right value to allow us to exist. Its value seems to have been arbitrarily set, neither to small to cause the Universe to collapse, or too big to cause it to expand too fast for matter to clump together. So some surmise that there must be other universes where the Cosmological Constant took on a different value such that life could not exist. (In fact, there may be infinite such universes.) We just happen to be in the Universe where the value is suitable to our existence.

I really do not have any opinion on MWI or Multiple Universes either way but Bencard's existence is otherwise hard to explain :-)

Jego said...

Yes, Ive read about that, too. The Multi-verse could be traced back to the Cosmological Constant. However, if beauty were the criteria, the CC doesnt pass. Einstein was deeply offended that he had to introduce it to his theory. He felt it ruined its beauty so he had to abandon it. But somehow physicists cant get rid of it since Einstein's equations needed it. They chose the preservation of Einstein over the preservation of beauty. Physicists such as Magueijo, Moffat, and others are looking for a new theory of gravity to replace Einstein's, and would ground these equations with actual observations (IOW, no more dark stuff) but good luck with that. Einstein is some sort of infallible deity in that community and to criticize him is like career suicide. (Magueijo in his book knew that he was probably committing career suicide.)

As for Bencard, he is beyond the scope of science. I remember when he was arguing against the role of the jury and I kept thinking, Please God, dont let anybody from the american bar see what he's writing. They could take his license away. :-)

cvj said...

My understanding is that many physicists are looking for a quantum theory of gravity to replace (or at least extend) Einstein's General Relativity.

Is Magueijo the one who said that the speed of light might have been different in the past? I haven't read his work, but it occured to me that he may be on to something while reading Lisa Randall's Warped Passages. On the chapter where she discusses Branes, Randall wrote that there is a relationship between the speed of light and the distance between two Branes. So if we (aka the known Universe) are on one Brane (labeled 'A'), and then there is another Brane (labeled 'B'), if the distance between the two was different sometime back, then this might any past variations in the speed of light.

Jego said...

Lee Smolin I think is on the quantum gravity thing. John Moffatt is on the Modified Gravity theory (MOG) which AFAIK isnt quantum.

Yes Magueijo is the variable speed of light guy which he arrived at with Andreas Albrecht, but Moffatt was the first to propose it.

Anonymous said...

Jeg,
multiple universes.
I have wonder if thise I see in the cartoons and comicsthat if enter a blackhole,you will enter a different universe.

also, in cartoons and comics like in the justice league, and challenge of the suoerfriends way before that the concept of prallel universe, where they went to each other's universe one or more times.

and of course superman's phantom zone.

sorry puro cartoons have not read a book about it yet.

Anonymous said...

"but isn't natural selection responsible for the emergence of drug resistant bacteria?"


In the series house,gregory house blames doctors.

always prescribing antibiotics at the slightest hint of fever.

TV na naman,sorry been watching a lot lately instead of reading.

Jego said...

That's exactly where Multiple Universes should belong, KG. In science fiction. Is it possible? Of course. (Invisible pink unicorns are therefore possible within the context of an infinity of universes.) But is it science? Noop.

And I love watching House. It shows the scientific method at work. It's educated guesses followed by a lot of testing.

Jego said...

In the context of an infinity of universes, a monkey banging on a typewriter, did manage to produce the Complete Works of Shakespeare. :-D

Anonymous said...

Ok jeg,til next time.

cvj said...

Jeg, in Seth Lloyd's book which i referenced above, he distinguishes between Monkeys typing into typewriters and Monkeys typing into computers [which is then interpreted by a program].

Jego said...

I see. I'll have to go look for Seth Lloyd's book. There are some intrepid souls in the internet whose mission is to provide these things for free for the masses. Some call them pirates, but I call them rebels. :-D (To assuage Prof Lloyd, I buy hard copies of internet books if they are good.)

But nevertheless, in an infinity of universes, a monkey at a typewriter could have indeed banged out the complete works of shakespeare in at least one of them. And in the Many Worlds theory, there may even be several monkeys who did just that.

(Any word on MLQ3's comments section? It seems to have gone kaput.)

cvj said...

Yeah, maybe Manolo's comments section is taking a much needed break.

Seriously, with all his problems with the availability of his website, i wonder if he should consider just having it hosted somewhere else.