The Primary Dilemma of Irreducible Complexity

[MrSpooky]

I'm hoping to write a professional paper on this later.

 

I was sitting myself down for some delicious, delicious thai food this evening after a good day's work teaching in my Philosophy of Atheism class when an idea suddenly struck me. I mulled it over, sorted it out, and with an "OMGWTF" I scribbled it down in my notepad on my laptop. I'm glad I did.

 

Such catharses come to me occasionally, but sadly most are relegated to oblivion because I forget to write them down. Such a terrible, terrible shame.

 

I was pondering one problem that I've been trying to tease apart for the longest time, one of my favorite topics really. It was pseudoscience, namely, Intelligent Design.

 

A good chunk of ID is predicated on a methodology of presenting examples of "Irreducible Complexity." That is, "so-and-so is immensely complex in its components are such that the removal of one component would render it useless." That is, this IC system is complex in a manner that it can't be explained via natural law: it can't have arisen by evolution. When you boil it all down, IC as a whole predicates itself on the premise that so-and-so is INEXPLICABLE, and concludes that an intelligent agent must be at work (a deity of some sort).

 

There are many problems to this approach, God of Gaps and all that, and there's always George H. Smith's argument that "the supernatural is no explanation, it merely asserts the futility of an explanation." These are indeed broad and important points, but I'd like to present a more specific approach today in what I hope will be a good and final criticism of IC.

 

Remember that methods which seek to demonstrate "Irreducible Complexity" are based on the idea that a certain system is INEXPLICABLE. We shall refer to this as "the Principle of Inexplicability." It's not merely that so-and-so is extraordinarily difficult to explain, or that an explanation is not known at present. If this were so, one should be seeking an explanation in the first place rather than jumping to this conclusion of "Intelligent Design." Without the premise of conclusive inexplicability, ID merely becomes a "here there be monsters" on our scientific map of reality. It becomes at best a placeholder for the unknown.

 

Barring the nigh impossibility of proving a negative (as is known for those familiar with the Burden of Proof), a more intrinsic problem lies within the heart of IC.

 

For any good idea to be considered "scientific," it must in principle be falsifiable. Is this Principle of Inexplicability falsifiable? If it is falsifiable, then there exists the potential for an explanation. Ergo, the Principle of Inexplicability, by the very principles of science, must be a tentative claim. This renders the IC system not so inexplicable at all, and science must do as it has always done: experiment, discover, critically examine, and hopefully tease out an explanation soon. IC has no bearing on science if it is falsifiable.

 

What if the Principle of Inexplicability is NOT falsifiable? In principle, an IC system can then be proposed and the conclusion of ID is much closer to being a genuine conclusion. However, by sacrificing falsifiability for coherence, the ID scientist has also sacrificed his ability to refer to his hypothesis as "scientific."

 

This, I propose, is the PRIMARY dilemma in the IC camp. In order for IC to be scientific, it must be falsifiable. However,in order for IC to be tenable, it must be strong in its treatment of the Principle of Inexplicability and hence UNfalsifiable. Each of these possibilities the IDer turns to renders his argument moot. If he chooses to go the way of making IC scientific (and hence falsifiable) he sacrifices his ability to make his claim of inexplicability in the first place. If he chooses to go the way of making IC a tenable proposal for ID, he renders his argument unscientific.

 

Thus:

 

1. Before the proposal of IC systems can be made, one must accept the premise of the Principle of Inexplicability (that this system is not merely "currently unexplained" or "difficult to explain," but rather that such a system cannot be explained AT ALL).

 

2. If one accepts the Principle of Inexplicability, IC is unfalsifiable and hence unscientific.

 

3. If one makes the Principle of Inexplicability falsifiable, they sabatoge the Principle itself and hence IC is no longer tenable.

 

4. Therefore, given all the possibilities, an IC system cannot exist.

[sixoverme]

wow...you should definitely publish this

[Zach]

Great post, MrSpooky. Would you consider allowing me to post this on "Goose The Antithesis?"

[Celsus]

IC is falsifiable and at the heart it is based upon faulty methodology and approach. Basically, IC approaches life backwards. It take an existing thing (bacterial flagellum for instance) and say take X-protien out and it now longer functions. However, life doesn't work that way. Components which had other, simpler or different uses, gradually evolve over time to arrive at a new function and/or state.

 

An analogy would be to look at a car. Take away the tubeless tire from an automobile and it becomes non-functional for the purpose it is used for, namely driving on roads. an IC person would say....see this is evidence of IC. However reality shows that wheels evolved (all be it not biologically) from earlier wooden wheels with iron surfaces, to wooden wheels with hard rubber surfaces, to metal wheels with tubes and tires and finally to tubeless tires.

 

IC fallacy is that it approaches the issue in the wrong direction. It is bad science, which is why it is not accepted in the scientifc community. As Behe stated, it could be falsified and in his original book he challenged scientists to do so. They took up the challenge and not only showed his premise was flawed (working backwards), but his conclusions were also erroneous. In short, IC is red herring, a non starter out of the gate, another faith-based "God of the Gaps" abdication of reason.

 

Bruce

[MrSpooky]

By all means, Zach. I'm just hoping someone doesn't nab this idea and publish it first.

 

EDIT: Oh, just be sure to credit it to "Kevan Wang" please.

[Zach]

Thanks!

[MrSpooky]

Huzzah! Thanks and you're welcome.

 

Honestly, I feel pretty impressed with myself and yet I'm kinda wary of that bright-eyed "bwahaha I'm so awesome!" mood that novice philosophers tend to fall into.

 

I just hope it's a JUSTIFIED sort of "I'm so awesome" instead of the retarted kind like Troy Brooks's "Four-Step Perfect Proof for God."

[Penny1]

Anything that's possible, will be....

 

 

Christians wanted to believe that the Earth was flat and that everything in the entire Universe revolved around them, too. When Galileo proposed that the Earth revolves around the Sun, they sought to kill him. They couldn't hold back the truth, though.

 

Nice job.

[Mr. Neil]

Oooooh, I think I get it.

 

In other words, if inexplicability could be scientific (i.e., falsifiable), then it really becomes redundant, because it would have to be constantly subjected testing and experimentation, something that would happen anyway. However, if it wasn't falsifiable, then it's not science.

 

Absolutely brilliant!

[MrSpooky]

In other words, if inexplicability could be scientific (i.e., falsifiable)

 

Actually, if Inexplicability is falsifiable, it would no longer be inexplicable. It'd reduce to "currently unexplained," to be more precise.

 

Other than that, yeah. I gotta look up my Karl Popper to see how well-supported this is. ^___^

[♦ ficino ♦]

Hi, Spooky. I've been off here for a while - tons of stuff at work, plus just got back from UK. It was a thrill once again to stand on Darwin's grave in Westminster Abbey and soak up the vibes. Charles Lyell is buried not far from him.

 

On the dilemma of IC, I may be missing something, but isn't the inexplicability principle a metatheoretical premise, not a proposition that belongs to the purportedly scientific theory, ID (or IC) itself? To say that the IP is not falsifiable seems to me beside the point. Popper's falsifiability principle, for example, is not empirically falsifiable by appeal to empirical data as counterexamples, for it belongs to the philosophy of science, not to any particular scientific theory. Not all propositions must be empirically falsifiable to be meaningful. Metatheoretical claims stand or fall by their success in accounting for what we do when we use the theories in question. Of course you know this.

 

I don't know whether this remark bears on your position or not, but I thought I'd throw it into the mix.

 

(Man, I'm having a lot of trouble entering this reply over Netscape 7.0. Is there a site upgrade that 7.0 can't handle?)

[MrSpooky]

Oh I'm certain this is open to criticism, ficino. As I've told many of my friends, "I need to bone up on my Popper."

 

You can imagine them bursting out laughing.

[Zach]

Congrats, MrSpooky! Your argument was chosen for the most recent Carnival of the Godless!

[Amethyst]

Do you mind if I post this on my live journal? I will provide a link to the original post here, and give due credit, of course.

[Saviourmachine]

Agreement

I agree with Bruce and ficino.

 

IC and HMM

IC seems to be just as falsifiable as RC (reducible complexity). IC states that there are sets of simultaneous mutations needed to get some biological systems. Or in other words that incremental adding of mutations does not lead to every system known nowadays. Bruce is right that they should investigate sequences of additions and deletions too. Such a sequence can be presented by a Markov Model (HMM) for example.

When the machine is in a certain state several events can occur with each their own probablity. There is some chance that an addition occurs, or a deletion or an insertion, etcetera. The biggest chance is that it remains in that state (in this case). The transition probabilities are added to the edges between the states. Someone can calculate the probability that BOY can be generated from GIRL by only addition and deletion of letters while staying in the English dictionary. Moreover the most likely sequence (or shortest path between BOY and GIRL) can be found by the Viterbi algorithm. It is a hidden markov model because we can not be sure about the sequences in which additions and deletions took place, but we can calculate the probabilities for the different sequences.

 

Desired Result

RC has to proof that the current observed complexity (in terms of interrelatedness) generated by mutations that occur in some (constant or variable) rate falls in the available time window. IC has to proof that a much larger time window is needed. They can both use the same mathematical and empirical tools. And the result will have the unit "(Million) Years".

[MrSpooky]

Whoot! I haven't been following up much on the Atheist web circles, but this does sound neat that I'm being cited! =3

 

Hmm. I hope I don't come off as credit-grabby, but I should send the guy an email to correct him regarding the credit.

[MrSpooky]

By all means, Amethyst, post away.

[Saviourmachine]

That is, this IC system is complex in a manner that it can't be explained via natural law: it can't have arisen by evolution. When you boil it all down, IC as a whole predicates itself on the premise that so-and-so is INEXPLICABLE, and concludes that an intelligent agent must be at work (a deity of some sort).

In case nobody noticed, this is a strawman.

 

Give credit to your opponents. Take for example this text where Behe's problems are properly addressed. Is that so difficult?

[MrSpooky]

I'm still not sure what you mean, Savior. The Principle of Inexplicability (which I refer to as PI) as I've stated simply means that "one cannot explain so-and-so phenomenon via unintelligent, natural processes." This can be due to the fact that a system has a property such that it cannot be made by the known laws of nature (such as the blood-clotting biochemical cascade argument often made by Behe in which all the components supposedly need to exist at once for the system to work) or it can be an improbability issue such as Haldane's Dilemma.

 

Just about any argument regarding the inexplicability of a phenomenon would fall under the general realm of PI. I further subdivide PI claims into two sets... "weak" PI and "strong" PI. I don't use these terms in the argument posted above, but I'm sure you can tease them out.

[Saviourmachine]

I'm still not sure what you mean, Savior. The Principle of Inexplicability (which I refer to as PI) as I've stated simply means that "one cannot explain so-and-so phenomenon via unintelligent, natural processes." This can be due to the fact that a system has a property such that it cannot be made by the known laws of nature (such as the blood-clotting biochemical cascade argument often made by Behe in which all the components supposedly need to exist at once for the system to work) or it can be an improbability issue such as Haldane's Dilemma.

 

Just about any argument regarding the inexplicability of a phenomenon would fall under the general realm of PI. I further subdivide PI claims into two sets... "weak" PI and "strong" PI. I don't use these terms in the argument posted above, but I'm sure you can tease them out.

If you mean that the Principle of Inexplicability entails a probability of zero, like your texts shows, than your argument is flawed. Subsequently defining "weak" PI and "strong" PI does save you again, but you need to rewrite your whole article in that case. IMHO Behe stated that IC systems had a very small chance of arising spontaneously. That's why Dembski invented the term conceivability versus probability. Resembles your "weak/strong" debating technique by the way. If you mean with the Principle of Inexplicability that the probability - of creation of these IC systems by gradual evolution - is small, you are right. In that case Inexplicability can be renamed Hardly Probable. That renders many of your statements senseless.

[MrSpooky]

No, Saviour, I really don't think that affects my argument at all.

 

I fully take into account the fact that "infintesimally small probability" arguments are indeed a subset of PI claims. An "infintesimally small probability" argument exists IN VIRTUE of being inexplicable via natural law unguided by an intelligent entity.

 

I really think you're misunderstanding my point here. Improbability is simply a MANNER (and quite possibly the prime manner) in which a PI argument is instantiated.

 

EDIT: Also, please note that I DO NOT claim in my text that "inexplicability" implies "a probability of zero." Explicability does not equal probability, though a very small probability can in some cases entail inexplicability within a certain context.