What is Science?

TCS: Tech Central Station asks: Is Intelligent Design a Bad Scientific Theory or a Non-Scientific Theory?

So what is the mark of genuine science? To attack this question, Popper examined several theories he thought were inherently unscientific but had a vague allure of science about them. His favorites were Marx’s theory of history and Freud’s theory of human behavior. Both attempted to describe the world without appeal to super-natural phenomena, but yet seem fundamentally different from, say, the theory of relativity or the gene theory.

What Popper noticed was that, in both cases, there was no way to prove to proponents of the theory that they were wrong. Suppose Jim’s parents moved around a lot when Jim was a child. If Jim also moves around a lot as an adult, the Freudian explains that this was predictable given the patterns of behavior Jim grew up with. If Jim never moves, the Freudian explains — with equal confidence — that this was predictable as a reaction to Jim’s unpleasant experiences of a rootless childhood. Either way the Freudian has a ready-made answer and cannot be refuted. Likewise, however much history seemed to diverge from Marx’s model, Marxists would always introduce new modifications and roundabout excuses for their theory, never allowing it to be proven false.

Popper concluded that the mark of true science was falsifiability: a theory is genuinely scientific only if it’s possible in principle to refute it. This may sound paradoxical, since science is about seeking truth, not falsehood. But Popper showed that it was precisely the willingness to be proven false, the critical mindset of being open to the possibility that you’re wrong, that makes for progress toward truth.

I agree with this 100%. In fact, this is actually enshrined in American jurisprudence as the Daubert standard. This is the standard that courts apply to expert testimony, and the ultimate question on it is falsifiability.

If we examine ID in this light, it becomes pretty clear that the theory isn’t scientific. It is impossible to refute ID, because if an animal shows one characteristic, IDers can explain that the intelligent designer made it this way, and if the animal shows the opposite characteristic, IDers can explain with equal confidence that the designer made it that way. For that matter, it is fully consistent with ID that the supreme intelligence designed the world to evolve according to Darwin’s laws of natural selection. Given this, there is no conceivable experiment that can prove ID false.

And this is where I disagree. ID (which isn’t yet a theory, simply a collection of hypothesis) proposes that design can be detected mathematically (the signal in the noise) and should therefore be falsifiable. In fact, the techniques they are proposing have already been shown to work in fields like cryptography.

On the other hand, Evolution falls squarely on the Freudian example above. What does Evolution do when new evidence is found? It simply shoehorns it in with whatever looks like the best fit. It makes the presumption of common decent and then shoves the jigsaw puzzle together in whatever way seems to fit closest (regardless of whether it is a proper fit or not.)

Evolution has failed to account for the apparent error-checking functions that weed out mutation in genetic material. Evolution has failed to show a way to perform a regression analysis on genetic material and show common decent. Evolution has failed to show any significant speciation.

It is sometimes complained that IDers resemble the Marxist historians who always found a way to modify and reframe their theory so it evades any possible falsification, never offering an experimental procedure by which ID could in principle be falsified. To my mind, this complaint is warranted indeed. But the primary problem is not with the intellectual honesty of IDers, but with the nature of their theory. The theory simply cannot be fashioned to make any potentially falsified predictions, and therefore cannot earn entry into the game of science.

This is flat wrong. The entire premise of the ID thought is that you cannot design something without leaving evidence of that design. The evidence is the very unlikelihood of it having occurred by random chance and selection.


  1. R says:

    “the signal in the noise”

    Why do you insist on using this patently inaccurate term? Did I not already post about how you’re using it incorrectly?

  2. Phelps says:

    You posted about it. You are still wrong, too.

  3. R says:

    No, you’re wrong. Infinity.

  4. Jim Darkness says:

    I think everybody’s wrong. Including me.

    But I might be wrong about that.

  5. Cosmic Siren says:

    Strange, according to the upper level mathematics I took as a engineering major many years ago and the statistical analysis I did as a Quality Assurance Specialist before I was married – Phelps is using it correctly.

    I also remember my father visually showing me the concept using an occiloscope when I was like 6 years old.

    Originally, it means to distinguished the carrier signal – or valid information – from the normal background radiations and static. In data analysis, it means to distinguish the predictable pattern among the outliers.

    I used to apply the Youden analysis to lab technician test data to determine correlation. It is a method for determining whether the results from Technician A matched close enough with Technician C to be use validly in other research.

    With every testing procedure, there is always some data that will land outside of the correlation circle due to random chance. I had this happen to one of my own test results. It just happened that I had received a sample that was still in spec, but was different enough from everyone else’s that my data fell out of the inclusion circle. However, the rest of my data was almost dead center of the correlation circles. We retrieved my samples and the other technicians tested them and got the same results I did.

    Now, the “signal” part was that I had as a technician an established pattern of doing my testing accurately. The “noise” was due to normal variations in our samples – it was my outlier data point.

    You get this type of noise all the time when testing for quality. The goal is to determine what is a pattern of compliance and what is a pattern of noncompliance.

  6. Citizen Quasar says:

    Whatever type of model that one wants to use, scientific theories are based on Occam’s Razor and Rand’s Razor.

    The proposed teaching of Intelligent Design is merely a witlless attempt by Christians to perpetuate belief in the creation story as put forth in the book of Genisis.

    The proof is thus: It is entirely possible for a hyperdimensional being to have put the entire Universe into motion. Without multiplying assumptions beyond necessity, evolution becomes the vehicle for affecting the design. The only reason to add another assumption, that ID stands in contrast to evolution, is to leave room for the Old Testament version of things.

    This also brings to mind my observations of Christians in particular and mystics in general. They are dishonest. They are lying when they say that this flaw in their logic is not for the explicit sinister purpose of deception.

    I guess that’s OK according to them because the Ten Commandments do not say “Thou shalt not lie.” They say “Thou shalt not bear false witness against thy neighbor.”

    I fucking hate Elizabethan English. It’s written that way because the bitch had a lisp. For the past 45 years I have heard every religious kook that came along speaking Elizabethan to try to sound accurate and correct. Do I have to live until the twenty-second century to see some mental progress from humanity?

  7. Citizen Quasar says:

    Also, considering the sorry state of affairs in the world, if we are the result of an intelligent design, it is a pretty sorry goddamned design. Any such “almighty god” that did make such design, mankind with all its social imperfections, disease, death, starvation, and suffering in general is a sadistic son-of-a-bitch of a Devil and not worthy of any respect, praise, or consideration, from me or anyone else.

    Oh that’s right! It’s all part of God’s great hidden purpose and I, as a mere human, am just not wise or gifted enough to see it. It is only if God chooses to reveal his GREAT PURPOSE to me that I might see. Just like he REVEALS intelligent design to his chosen sheep who then inflict it on the rest of us as their condemned inferiors.

  8. Phelps says:

    A Yugo is a poor design. It is still a design.

    There is no reason that a hyperdimensional being needs to be involved, anymore than Evolution rules one out. Since ID only opines as to life on Earth, there is no reason that we couldn’t have been seeded by an older civilization from another planet that we have no observations of. In fact, that is one of the options often cited.

  9. Citizen Quasar says:

    My bad. I view evolution as part of the Law of Time and Chaos. Perhaps my frame of reference is too broad.

  10. Citizen Quasar says:

    On second thought, seeding of the Earth by a species merely from another planet evokes a reductio ad absurdum.

    That is, where did that species come from? Were they seeded also or did they come about by evolutionary implimentation of a design by a hyper-dimensional being?

    It just keeps going until one gets beyond elementary space-time. Then the IDers, by applying Rand’s Razor and Occam’s Razor to their fantasies are left with pure unadulterated Christian creation via their book of Genesis.

    In order for this not to be so, they have to through science out the window, which is what they are really, sneakily trying to do anyway.

    They are not of this earth(reality). They hate this earth(reality) because it makes them have to use their rational faculties instead of being whim-worshipping cry babies.

    The proposed argument from Intelligent design just goes to prove that the mystics cannot design an intelligent argument to support their unfounded, 3rd century beliefs.

  11. Phelps says:

    On second thought, seeding of the Earth by a species merely from another planet evokes a reductio ad absurdum.

    That is, where did that species come from? Were they seeded also or did they come about by evolutionary implimentation of a design by a hyper-dimensional being?

    That’s a philisophical argument, not a scientific one. If the evidence points to turtles all the way down, then it is what it is. There could be an entirely different mechanism on another planet, but we can’t know right now because we have absolutely no evidence of what the situation is there.

    Science is about seeing what the evidence is, not about making people happy, whether they are theists or naturalists or athiests.

  12. R says:

    “Strange, according to the upper level mathematics I took as a engineering major many years ago and the statistical analysis I did as a Quality Assurance Specialist before I was married – Phelps is using it correctly.”

    Clearly you were instructed incorrectly; either that or you never understood the lesson as Phelps’ attempted application of signal and noise theory to observed biological phenomena is incorrect.

    Background noise is due to 1) ambient electromagnetic activity not related to the carrier signal or 2) the carrier signal itself reflecting off of the environment which modulates it in some way. However, none of these alter the original signal.

    The biological tendencies of species to 1) stay around some common ‘configuration’ (i.e., a resistance to chance) and 2) carry one’s evolution over some genetic ‘rise’ and ‘fall’ into another ‘well’ of stability, all affect the original genetic line.

    What Phelps is attempting to describe as analogous to electromagnetic noise is the outlying mutations that go nowhere, and thus can be discarded using a purely statistical manner of looking at things.

    You’re not the only one who went to college. Yay for your upper level mathematics courses. I took them, too. I also took courses in signal analysis and microelectronics.

  13. guy in the UNLV jacket says:

    Who cares? Nobody knows where it all came from and nobody ever will. They need to drop it.

  14. R says:

    So I guess the question is: “Do you know what signature you’re looking for, and would you know it if you saw it?”

  15. Phelps says:

    What Phelps is attempting to describe as analogous to electromagnetic noise is the outlying mutations that go nowhere, and thus can be discarded using a purely statistical manner of looking at things.

    No. Digital images are a good example of the idea. A JPEG is not a random arrangement, but it has the signature of randomness. If the JPEG is altered in an image editing program, it will leave a signature of having been manipulated by that particular program. Photoshop, GIMP, Paintshop — they all leave a signature that can be detected. This is a signal (Edward Tufte defined information as “differences that make a difference”) in the noise of the original image. ID hypothesizes that a like signature can be demonstrated in the known genomes. (And that is certainly a falsifiable hypothesis.)

    The signature of random mutation selected by environmental pressures will be distinct from that of intelligently altered genes (regardless of whether ID or Evolution or something else is correct.)

  16. Phelps says:

    That’s the hypothesis.

  17. R says:

    I disagree. A hypothesis would predict an outcome. Your signature is the predicted outcome.

    But since you don’t even know what you’re looking for, there is no way you can call that a scientific hypothesis.

  18. Mexigogue says:

    No, you’re wrong – – INFINITY!!

  19. Korgmeister says:

    Well, based on your argument, I can accept that ID as you put it is certainly worthy of being considered as a hypothesis.

    But I think that’s still too weak to be teaching in schools. Speculating on hypotehesis goes on in College. We already have a hard enough time teaching things we’re certain about in high school without also introducing nascent hypothesis that are in fashion with a few special interest groups.

    That’s as bad as teaching Marxism to kids.