[00:00:07] Speaker A: Welcome to ID the Future, a podcast about intelligent design and evolution.
[00:00:14] Speaker B: Hi there. I'm Andrew McDermott. Today I'm pleased to talk again with writer and speaker Kirk Durston. Durston holds a PhD in biophysics, a master's in philosophy, and undergrad degrees in physics and mechanical engineering. Author of several papers and peer reviewed science and philosophy journals, he has for decades studied the possibility that meaningful information, and genetic information specifically, is the fingerprint of intelligence. One of his current research projects involves working with a team of scientists to develop software that can help us better understand the submolecular structure of proteins and how they fold.
Kirk, glad to have you back with us.
[00:00:54] Speaker C: Well, thank you, Andrew. And I'm glad to be back here discussing these things.
[00:00:59] Speaker B: This is the third episode of a three part series exploring a series that you wrote recently for your
[email protected], and it was actually also cross
[email protected] and it looks at the differences between the major categories of experimental, inferential, and fantastical or science fiction. Today we'll look at this third category, fantasy in modern science.
Let's review a bit for those who haven't yet heard parts one and two of our discussion. Can you tell us what experimental and inferential science are and a few of the pitfalls associated with each?
[00:01:37] Speaker C: Sure. In my own experience with 21st century science or modern science, I could see that it really does need to be divided into three categories. And the first category that I would divide things into would be experimental science. And that is where you can actually do experiments, produce results, publish those results, and then other people can attempt to reproduce those results. And if they do produce them, reproduce them, then you have something that can be very trustworthy indeed.
The pitfall in that area would be what nature? An article in Nature called perverse incentives, where people may be tempted to, let's say, massage their statistical analysis or the results to get results so that they can qualify for funding next term or maybe for academic prestige. But there again, the accountability that experimental science requires is the ability to be reproduced. And so, in the end, even if people are massaging their results, it is still highly accountable in that anybody can come along and test those results using the same methods to see if they can reproduce them. The second major category was inferential science, and that occurs in situations where we make observations or we do an experiment, and then we make an inference to some conclusion that we can't actually test due to it occurring in the past, or it's just too huge of an experiment to actually be able to pull it off. But we make an inference from real data or real observations to a conclusion that we cannot actually test.
And if that leap, that deductive leap or abductive leap is fairly small, as it often is in, say, forensic science, then those conclusions are probably trustworthy. But if the larger the leap is, the less trustworthy those conclusions become. And this is the pitfall of inferential science, is that eventually you can infer almost anything if you're willing to take a big enough leap.
[00:03:42] Speaker B: Yeah. And when the lines start to blur between doing science and creative storytelling, fantasy science can result. What's happening when science becomes science fiction?
[00:03:55] Speaker C: Well, the big difference between inferential science and science fiction is that at least in inferential science, you're starting off with real results, and your leap is hopefully not that large. And there's at least the way to test partially some of your inferences.
But in fantasy science, the leap becomes so huge that you can't even test the process or the method that is supposed to produce that conclusion. So in inferential science, you may actually be able to test a possible method that would lead to that conclusion.
You don't know if that was the method because you're making an inference here. So that's inferential science. But in fantasy science, you can't even test the method, certainly not the result. So an example would be the multiverse. And a couple of scientists by the name of Joseph Silk and George Ellis published an article in Nature pointing out that really what we have occurring here is a threat to the integrity of physics itself, with some of these ideas of the multiverse and so forth, that we cannot even test. If we can't even test it, then we've left the realm of science, and we've entered the realm of creative storytelling or science fiction. Now, keep in mind, you can tell creative stories using mathematics. And so a very sophisticated model might use very sophisticated mathematics to arrive at a conclusion.
But every key aspect of that model still has to correspond to something in reality. And if we can't even test that, then what we've got is, in effect, a science fiction novel, but it's written in mathematics rather than a literary composition.
[00:05:41] Speaker B: Huh. That's a great point. We see a bunch of math, and we're tempted to think, well, they know what they're talking about, you know, but that's a really good point, that you can still dress up science fiction with mathematical terms and it's still fiction. Well, how has modern science suffered from scientism, which is the philosophical belief that science explains everything.
[00:06:05] Speaker C: Yeah, so what I see, and I think it's really atheism dressed up in a lab coat is one way of describing scientism. This idea that science explains everything. There's a couple of problems with that. First of all, you're basically committed to believing the conclusion, if that's the only materialistic conclusion on the table. For example, the idea that maybe, you know, there was an intelligent mind behind the origin and diversification of life. Well, if you're committed to scientism, no matter how good the evidence is, you have to rule that out because you have to come up with a natural explanation that does not invoke intelligence. And so consequently, if in fact, intelligence was required for the origin of life, scientism requires you to hold to a theory or a method of that in the end is completely false. But you're committed to it because scientism is the belief that there is a natural explanation for everything. So that's problem number one. You may wind up being committed to methods and theories that are completely false. And the probability of them being true is so it's essentially, for all purposes, a zero. And I'm talking about, let's say, for example, accumulating the protein families necessary for even the simplest life form. The probability of that which I can demonstrate is essentially zero. Very, very close. But you're committed. A person devoted to scientists is committed to that, no matter how widely improbable is. Second problem is that you ultimately, you see, natural processes are really causal chains that are unfolding, where this effect here was produced by that previous effect, which was produced by the previous one. And you work your way back to the origin of the universe. Or if you really want to believe in the multiverse, you could work your way right back to the mother of all, the start of that multiverse. And by the way, even in a multiverse, you have to have the first universe. You have to have a beginning. Now, the problem with that is that nature, just as a woman cannot reproduce herself or give birth to herself, so nature cannot possibly produce itself. So when we're looking at the origin of nature, where I'm defining it as everything composed of space, time, matter and energy and the laws of physics, you're saying, well, what caused that? If you're devoted to scientism, you find yourself in a checkmate position, because you have to say, well, there's a natural explanation for how nature came into existence. And in order for that to be true, you have to have nature somehow existing before it existed so that it's in a position to bring itself into existence, which is a violation of the principle of non contradiction. It's an absurd position to be in. And so, in the end, science does not, and in fact, it's logically impossible for science to explain everything, at least the origin of nature, in the first place, upon which science is built.
[00:09:04] Speaker B: So beware the over reliance on science alone to explain everything. Well, you mentioned the multiverse, which is a great example of science fiction. In fact, some people are actually ready to call that mathematical philosophy or fantasy instead of actual science, aren't they?
[00:09:22] Speaker C: Yeah, yeah. And those words have been, I have a few different articles and science journals saying, you know, we really need to call it what it is. It's not science, it's fantasy, or it's a threat to physics, or it's at best, philosophy of science, but it's not science.
[00:09:38] Speaker B: Well, turning back to your work on proteins for a moment, what have you learned about the structure of proteins and how it relates to the information that they bear?
[00:09:48] Speaker C: Well, the theory behind it is that if we back up to things that we build in general, any example of intelligent design, and let's just use, let's say, your car, a lot of people have cars, so they might be able to appreciate this. There are certain things where the people who are building that car, the tolerances are huge. There's not really anything carved in stone. There's not really tight things they need to adhere to. Let's say the shape of the, of the external body, for example. There's a lot of leeway there, but there's other things, let's say, where a shaft has to fit tightly on the inner race of a bearing or something, where the tolerances might be extremely tight. And so what you need to build that car is you don't need much information for things that are, that are. There's a lot of freedom in, but you need a lot of information for the things where the tolerances are, are very tight. So if you take that idea and apply it to the three dimensional structure of proteins, there may be areas within that protein sequence. It's ultimately the sequence of amino acids that folds into a three dimensional structure. And that in turn came from a sequence in the DNA, from a gene sequence. And so that information in the gene, you have to have extra information, say there's a particular gene, and you'll have a series of letters that encode the information to produce a particular protein that belongs to a protein family, and that particular sequence of letters, some of that protein structure. There's a lot of room for tolerance there, but others not so much. And so what you're going to look for in the information encoding proteins is areas within that sequence that seem to carry more information than other areas. And what that's going to tell you is which are the key or the crucial parts of that three dimensional structure. And so that's where my area of research was for my PhD project, and that's what I'm continuing on with this group of scientists we mentioned earlier, writing software that actually finds these key aspects of a protein by looking at the information density in a number of alignments of different sequences that will code for that same protein. To see where is this information concentrated here, because where it's concentrated is probably an important aspect of that protein. And then we can come up with sort of a the results will allow us to predict maybe even how that protein folds. And in my first paper, I showed that it is exactly that, at least for the first half of the protein, the last half that is. So it's useful for predicting how that protein folds, as well as which aspects of the three dimensional structure we can play around with and which aspects we better not mess with.
[00:12:40] Speaker B: So when you're actually looking at an image of a protein, you're seeing a visual representation of the information that is unique to that protein.
[00:12:49] Speaker C: You're seeing the end result of what that information encoded. It was digital information, and it produced that three dimensional structure. And the digital information has to be very specific for certain parts of that structure. And if we know which parts the digital information is highly specifying, then we can look at the three dimensional structure of a protein and we can say, okay, over here and there, that's very important. This section over here is there's probably a lot of room for freedom. So if you're a drug manufacturer, you know, you're manufacturing some sort of artificial protein that's helpful in treating some particular disease, and you want to patent it, you know, which parts you have freedom to manipulate and which parts you'd better get right.
[00:13:37] Speaker B: A very exciting part of the study of intelligent design, looking at proteins and how they fold and their structure. Well, if readers want to know more about that, about those particular projects you're working on, are you writing about that on your blog as well?
[00:13:55] Speaker C: That gets pretty technical. I have not really written anything yet. The best would be to refer to a paper, my last paper that I published on that, and it's available on my website. But the exciting thing is that we are working to produce, to reproduce to fine tune that software. And once we get that up and running, we're going to be able to do a lot of proteins and maybe even areas of the human genome run that through as well. But in the meantime, that paper there should explain the one I refer to on submolecular structure proteins. That will give the listener a bit of an idea, but it is a bit technical.
[00:14:37] Speaker B: Yeah, well, that's it for this episode, and that concludes our three part discussion. This is such a great reminder of the differences between experimental and inferential science and how pitfalls in both can lead to fantasy science or science fiction. It just goes to show that honest and careful examination of the evidence is just as important as the experimentation itself.
And I appreciate your discussion of the functional information and proteins as well, and that's been very helpful to review. As we know, information runs the show in biology, and seeing how that works at the molecular level is so interesting. Kirk, thanks again for your time.
[00:15:16] Speaker C: Oh, it's been my pleasure, and I enjoyed this.
[00:15:20] Speaker B: To read more of Kirk's posts, visit Kirkdurston and for more episodes of ID the future, including parts one and two of our discussion, visit idthefuture.com or subscribe and download through your favorite podcasts app. Until next time, Im Andrew McDermott for Id the future. Thanks for listening.
[00:15:42] Speaker A: This program was recorded by Discovery Institute's center for Science and Culture. Id the Future is copyright Discovery Institute. For more information, visit intelligentdesign.org and idthefuture.com.
