From - Fri Dec 09 03:12:17 2005 X-Mozilla-Status: 0001 X-Mozilla-Status2: 01000000 FCC: mailbox://dharris234%40mindspring.com@auth.earthlink.com/Sent X-Identity-Key: id1 X-Account-Key: account1 Date: Fri, 09 Dec 2005 03:12:17 +0000 From: David Harris X-Mozilla-Draft-Info: internal/draft; vcard=0; receipt=0; uuencode=0 User-Agent: Mozilla Thunderbird 1.0.7 (Windows/20050923) X-Accept-Language: en-us, en MIME-Version: 1.0 To: ExI chat list Subject: Re: [extropy-chat] against ID References: <200512080330.jB83Ute22037@tick.javien.com> <6.2.1.2.0.20051208113709.01c8f760@pop-server.satx.rr.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit For the past year I have been the main computer developer at DarwinDay.org, which is encouraging the celebration of Science and Humanity on Darwin's birthday, February 12. We invite everyone to visit DarwinDay.org and participate in some fashion. In the course of this work I look at all the emails we receive at "Unsubscribe", resulting from invitation emails that we send to science-related professionals. Some of these unsubscribe messages have attempts at arguments against evolution and in favor of creationism. This has motivated me to think hard about what is wrong with ideas like "Intelligent Design". I've bought two of Karl Popper's books on the theory of science, and recalled a course in philosophy of science that was taught by my graduate advisor at Stanford, Edwin Parker. My understanding is consolidating around the idea of "testable (falsifiable) predictions" as the mark of what is science. This argument is straight from Popper, hopefully expressed in a catchier way. And there are other aspects worth expressing. The idea is that science is limited to theories that can make predictions that are unexpected, risky, and clearly stated, and those predictions must be falsifiable/testable by other people. If no one else can perform the test, the prediction is outside science. This restricts science to things that MAKE A DIFFERENCE in the world that everyone shares, the observable natural world. If some theory (including Intelligent Design) makes a claim, ask "What testable predictions do you make based on this theory?" If a theory doesn't make testable predictions, we can rightly call the theory "non-science", or "boring" or "useless". If a theory does make a testable prediction, we can perform the test and try to falsify the theory. The progress of science is made through testing a lot of theories and throwing out the false ones, leaving only the ones that have survived a lot of attempts at falsification. Those that are tested in powerful ways repeatedly and just never seem to get falsified gain respect and gradually come to be considered scientifically "true". But it only takes one convincing falsification to throw them out. For instance, in my graduate days in molecular biology, we believed (as part of what was provocatively called the "Fundamental Dogma of Molecular Biology") that "one gene makes one protein". Recently everyone was able to count the genes in the results of the Human Genome Project listings, and there were about 30,000 genes there. But by other means it is believed that there are about 100,000 different proteins in a human. So, the prediction is falsified and the Fundamental Dogma is FALSE in that aspect! And the reaction of scientists is generally "Great! That 1 to 1 idea was false and now we know not to believe it. So how DO those genes lead to multiple proteins?" And people began to think up new theories like alternative splicing of gene products, and RNA with catalytic power, etc. One of my "unsubscribers" asked about what was before the Big Bang, and I asked if there was any prediction about that time which could be tested. Apparently no one has made a testable prediction, so the issue of BEFORE the Big Bang is NOT science, however intrigued we may be about it. We must keep our work focussed on theories that make a difference. As I read the various arguments about evolution and other scientific matters, I am struck at how few people who are scientists or supporters of science are able to make this argument clearly. To be science, a theory must make risky, testable, predictions. Even the famous essay about "Nonoverlapping Magisteria" (at http://www.stephenjaygould.org/library/gould_noma.html) by the late Stephen J. Gould does not make the point clearly. As I reflected on my years at Stanford, I recalled my study of medical decision analysis, in which our decisions are represented as branches in a tree, with choices at the notches where branches diverge. Some notches are choices we can make, with percentages for the likelyhood of various outcomes. This is the kind of knowledge science can give us: what is likely to happen if we make a certain choice. We predict that if ducks are raised with pigs and people, as in parts of China, the duck, pig, and human genes will be transferred to the other nearby species, and we can find the genes by sequencing individuals and using computers to search for identical genetic sequences. More specifically, we predict that bird influenza genes are likely to mutate and spread to humans, and create highly infectious flu strains that spread worldwide in a pandemic. But the decision analysis has another aspect: the outcomes have VALUES attached to them, usually expressed in money prices. This works well for decisions like whether I buy a lottery ticket. If I don't buy it, the expected value of my choice is zero. If I do buy it for a dollar, the chance of winning is like one in 20 million, and there is one chance in 40 million that I will win that jackpot. So there is a one half chance on average. Multiply the one dollar times the one half chance and the "expected value" of my buying the lottery ticket for a dollar is that I will get back 50 cents. But in many decisions, the VALUE associated with different outcomes is a matter of dispute. If a stem cell researcher kills a 100 cell blastocyst to make cell lines, what value ought we to place on it? To me, since there is no nervous system in the blastocyst, the value for killing a blastocyst is approximately zero. But for people who believe that the blastocyst is a full human being with a "soul", the value assigned may be highly negative. To make a decision in this systematic way, one needs to know what will happen, and what values to place on the outcomes. Together those two pieces of information allow one to calculate expected values and to choose the choice with the higher expected value. gts wrote: > On Thu, 08 Dec 2005 13:04:01 -0500, Damien Broderick > wrote: > > >>> The second definition seems on the surface to be quite >>> reasonable, perhaps even an improvement on the first, but it >>> lacks the requirement that science be about *natural >>> explanations*. In Kansas, *any* explanation for natural >>> phenomena now qualifies as science, including for example >>> astrology as an explanation for human personality. >> >> >> But what is your objection to examining astrology as an explanation >> for human personality? > > > None, technically. My use of language here is evolving. Astrology > should be rejected as science because it doesn't explain or predict > anything. > > Similarly, Intelligent Design doesn't explain or predict anything. It > suffers for two reasons: 1) there is no solid evidence that > so-called "irreducibly complex" structures really exist in biology, > and 2) even if they do exist, the so-called theory of ID does not > explain how those structures come into existence. > > I found this quote of Behe (the chief protagonist of ID), from a > lecture in which a physicist pressed him for an explanation of how > structures alleged to be irreducibly complex (for example bacterial > flagella) come into existence. > > ===== On November 11, 2002, Larry Arnhart reported on a lecture by > Behe at Hillsdale: At Hillsdale, after his public lecture, I > challenged Behe in a small-group discussion to give us a positive > statement of exactly how the "Intelligent Designer" creates bacterial > flagella. As usual, he was evasive. But I didn't let him get away. > And finally, he answered: "In a puff of smoke!" A physicist in our > group asked, "Do you mean that the Intelligent Designer suspends the > laws of physics through working a miracle?" And Behe answered: > "Yes." > > > > http://wiki.cotch.net/index.php/Puff_of_Smoke ===== > > So it happens in a puff of smoke! The least Behe could do is explain > how the Intelligent Designer's smoke does this magic. How does the > magical smoke enter into the physical world, and how does it > rearrange organic materials? What is the mechanism? If he could > answer that question then he might have a falsifiable theory. > > I agree with you that "natural" is a troublesome word. I have used > "naturalism" to mean something close to "positivism," but that is I > think the wrong way to approach the problem. > > -gts