iDesign at UCI: Defending ID 4: Falsifiability of Intelligent Design

Art said...

Design Paradigm has a post related to this one.

3/08/2006 5:25 PM

X best explains Y when X best predicts Y. That's why falsifiability is vital.

Genetics, inheritance, sex, and mutation make definite predictions, including common descent and observed transitional fossils. And the more we know about biochemistry, the better our predictions.

Neo-Darwinian evolution has within it explanatory mechanisms that are individually falsifiable. We do not expect to be able to guess all of these mechanisms correctly the first time. However, we must expect to be able to use scientific inference to eventually identify these mechanisms.

In contrast to NDE, ID explains nothing because it predicts nothing. All you can say about ID is that, in principle, a designer could have been involved in evolution, and that a sufficiently powerful (and long-lived) designer could have created any arrangement of organic matter we might observe.

However, ID is NOT explanatory. There's a difference between saying that that "magic X caused Y through unknown mechanisms," and "cause X predicts effect Y according to consistent rules." ID is an example of the former. Biological ID never proposes any falsifiable mechanisms. As long as this is the case, the detection of biological ID will be fundamentally impossible.

Look at it this way. What is the difference between your saying that "life was designed" and my saying that the world is the way it is exclusively due to natural laws without my specifying what any of those laws actually are? There's no difference.

If I tell you only that "the Sun generates energy due to natural laws," would you accept that statement as an explanation? No. You would demand that I provide you with detailed laws of particle physics that predict stars, and that I be able to test those laws in the lab.

The same goes for cosmological ID. Saying that there's some invisible mechanism that fine-tunes the universe for what we see isn't explanatory when you cannot say what it is about your model that predicts what we see. This goes for both CID and multiverse models. Any multiverse model that fails to predict something about our universe also fails to be explanatory. Likewise, any cosmological ID model that fails to predict something observable is equally non-explanatory.

As for your naturalistic ID scenarios (alien design, Stonehenge, pocket watches, etc.), these work because they rely on rule-based mechansisms. It is the predictive mechanisms within naturalistic ID that make these models scientific. It's when you deviate from predictive mechanisms that you get into trouble.

3/08/2006 10:17 PM

Art said...

Neo-Darwinian evolution has within it explanatory mechanisms that are individually falsifiable.

Is this an implicit suggestion that NDE as a whole is not falsifiable, but that its individual components are falsifiable? A similar argument is made by some evolutionists: ID is not falsifiable, even though some components of ID, like irreducible complexity, are falsifiable.

However, ID is NOT explanatory.

ID is explanatory -- the mechanism for achieving the spectacular complexity in life is design. Design provides a reasonable explanation for complex and elegant phenomena that are out of the reach of stochastic processes.

Here are some predictions of ID:
1) Junk DNA will have utility.
2) Organs thought to be useless in various animals will be found to have an important function.
3) The area of biomimetics (which copies designs found in nature) will become extremely successful.
4) More instances of "isomorphic instantiation" will be found (e.g. the "LED in butterfly" example).
5) The correlation between habitability and discovery will be strengthened ("Privileged Planet Hypothesis").

3/09/2006 12:05 AM

Doctor Logic said...

A similar argument is made by some evolutionists: ID is not falsifiable, even though some components of ID, like irreducible complexity, are falsifiable.

ID has no falsifiable components, as we will see...

ID is explanatory -- the mechanism for achieving the spectacular complexity in life is design. Design provides a reasonable explanation for complex and elegant phenomena that are out of the reach of stochastic processes.

Replace the words "ID" and "design" with the word magic, and the meaning doesn't change. ID only "feels" explanatory.

Take your 5 so-called predictions and tell me why they are predicted. What are the fundamental rules of ID?

Why is all junk DNA useful (as science already claims that it is)? Why are vestigial organs not so (cough, appendix)? Why do snakes have useless limbs?

Evolution explains these things because large scale biological structures need only to have once had a utility, but need not have continued utility.

Tell me, what are the laws of ID that require your five claims to be true? Show me the argument behing the Junk DNA claim.

Not one of these five are predictions of generic ID. They key point of data is that ID does not predict common descent. Human designers do not use common descent because common descent is unnecessarily limiting and inefficent. If you want to claim ID predicts common descent, you had better say what falsifiable laws of design you are proposing.

BTW, IC is a fact. There are systems that don't work when you take out a single component. The problem is that IC has nothing to do with falsifying evolutionary mechanisms because IC structure can evolve. Meanwhile, CSI is a metric of how designed something looks. CSI is not a test of whether or not something was designed. That's the hypothesis.

3/09/2006 6:40 AM

RLC said...

"...when in fact it is really making a much stronger claim:..."

Actually, it is a weaker claim than the conventional creationist's "God did it" (GDI). GDI is empirically incorrect, but up front in its reference to magical processes as a causal mechanism. ID, however, sheepishly avoids investigation, or even mention, of a mechanism and consequently says next to nothing.

Although ID poses as science, a brief comparison with legitimate investigation of intelligence (e.g. archeology) demonstrates that the observation and hypothesis (in arch.) deal with motivations and methods of the designer, and are subsequently tested on that basis. ID, on the other hand, takes dubious observations, proposes a non-specific hypothesis, and stops there. No testing. Which means no science. It looks slightly more scientific than GDI at first, but after you lift the lab coat...

"If this is your guiding principle, then there is no concievable scenario in which design might be a legitimate explanation for any feature of biological life..."

No, this is silly. Design is already a "legitimate explanation" for many features of biological life. Humans have artificially designed organisms for millenia. Why is this a "legitimate" explanation and ID is not? Because human design is a singular, and extraordinarily well evidenced, example of design. Therefore reference to this as causal agency is not controversial. But reference to non-human (and especially non-natural) intelligence or design involves a staggering multiplicity of unwarranted assumptions. When the evidence for this putative causal agency is as solid as that for human design only then will ID enter the realm of legitimate explanation.

"...and that in this case the impossibility of ruling out chance/necessity with cartesian certainty is obviously irrelevant."

You miss the point. And it was one you made in your previous sentences. Yes, it is logically possible that we might someday discover evidence of intelligent alien intervention in the genome. But that evidence will have to be empirical, and testable. And for that to be the case it will have to involve uncovering something of the motives and methods of the aliens in question. In other words it will have to significantly reduce the multiplicity of assumptions to a point where we feel justified in saying something stronger than "we don't know."

Only when we know all there is to know of the universe will the "impossibility of ruling out chance/necessity with cartesian certainty" be irrelevant. Only then will "we don't know" be a less rational conclusion than unevidenced design.

"But I just don't see how the claim that ID is not falsifiable can stand."

Anyone who claims that ID is not falsifiable is speaking in the broad causal context. There are specific components of ID "theory" that can, and have been, refuted (e.g. irreducible complexity).

The point is that there is no evidentially demonstrated connection between ID's observations and its causal agency, and if there is no empirical connection there can be no falsification of the activity of that agency.

3/09/2006 9:34 AM

Wedge said...

Tell me, what are the laws of ID that require your five claims to be true? Show me the argument behing the Junk DNA claim.

Not one of these five are predictions of generic ID. They key point of data is that ID does not predict common descent. Human designers do not use common descent because common descent is unnecessarily limiting and inefficent. If you want to claim ID predicts common descent, you had better say what falsifiable laws of design you are proposing.

doctor_logic,
Regarding common descent: If you look at the technological evolution of (say) cars, you'll actually find something quite similar to what we see in the fossil record. You are correct that ID doesn't specifically predict common descent, but common descent is perfectly consistent with ID. A successful theory need not be so specific that it is consistent with only one prediction of every event. There is a lot of wiggle room in evolution as well - for example, there might be junk DNA because it accumulated over the centuries, or there might not be because the cell eliminated the dead weight. Both are consistent explanations.

Regarding Art's predictions:
1) and 2) (junk DNA and vestigial organs) follow from the fact that the Designer can be expected to follow Occam's razor and create structures that exhibit the marks of intelligence, not the marks of chance. This ties in to how ID is different that "God did it". "God did it" is consistent with a designer who mimics chance, but ID is not consistent with that possibility.

I actually take a different approach than Art does here. Chance and necessesity are indeed powerful forces, and they have been operating on their own for quite a while now. My prediction is that there is a principled way to distinguish the initial (designed) genetic information in an organism from the acretions of chance that have occured since it was created. In other words, it should be possible to "look through" mutations and see a skeleton of the original design.

3 & 4 (biomimetrics and isomorphic instantiation) follow from the fact the the Designer was presumably more intelligent than we are, and this his solutions can be expected to be at least as good or better than ours.

1-4 follow, I think, from the principles of good design and are therefore "generic". 5, the priveledged planet hypothesis departs slightly from generic ID, by assuming that the designer is also interested in our development and scientific progress.

In addition, there's no necessary reason for ID to remain generic. Most ID theorists have private suspicions about the character of the designer, and there is plenty of room for arguing whether Denton's deism is more explanatory (i.e., predictive) than Dembski's Christian theism.

3/10/2006 12:26 PM

Doctor Logic said...

wedge,

A successful theory need not be so specific that it is consistent with only one prediction of every event.

NDE and ID are both meta-theories. Theories are the things that make predictions, not predictions about predictions.

Naturalism and supernaturalism are not scientific theories. They make predictions that predictive theories will be found or not found respectively.

It is these predictive theories about mechanisms that constitute explanations.

Common descent is vital to the mechanisms proposed to underly evolution, specifically, to natural selection. Natural selection as a mechanism for evolution requires birth, death and inheritance. Natural selection predicts common descent, and natural selection is an explanation of the apparent evolution of species. It does not (and need not) explain everything at once. As long as it is predictive, it is partially explanatory.

As we know, the evidence for common descent is overwhelming.

Clearly, common descent was a highly successful test of evolutionary theories because it is a test of the mechanisms of those theories. However, as you point out, ID neither requires nor favors common descent. This is because no testable mechanisms of ID have been proposed.

So common descent is not evidence for ID. Indeed, no observable fact is evidence for ID until you propose a testable, falsifiable mechanism underlying ID.

Could a designer have used common descent or natural selection? Sure. But ID doesn't predict these things. ID is only scientific when it proposes testable mechanisms.

You still have not explained how you arrived at your predictions. To begin with, I don't understand how Occam's Razor is relevant to the designer's task. We humans use Occam's Razor to select simpler theories over more complex ones when each theory has comparable predictive power. That's because computing costs energy, and we humans have only finite energy and computing power. Are you claiming that the limits of the designer's computing power have a bearing on this prediction?

If we skip the OC bit, you say "the Designer can be expected to ... create structures that exhibit the marks of intelligence, not the marks of chance." Why is a lack of Junk DNA a mark of design vs. chance? It seems to me that the question of Junk DNA is irrelevant to ID.

ID can produce junk:

1) Front-Loaded ID with or without Junk
2) Evolutionary mechanisms produce Junk
---
3) Prediction: Junk

Evolution can also produce Junk-free genomes:

1) No ID
2) Gene expression is a function of "Junk"
---
3) Junk is functional, i.e., not Junk
---
4) Prediction: No Junk

So, are you claiming either:

1) ID was not front-loaded and had no Junk
2) Evolutionary mechanisms are irrelevant
---
3) Prediction: No Junk

or:

1) ID was front-loaded and had no Junk
2) Evolutionary mechanisms don't produce Junk
---
3) Prediction: No Junk

Don't forget to rule out these options:

1) ID was not front-loaded, but had Junk
2) Evolution was irrelevant
---
3) Prediction: Junk

and many more!

Suppose you claim that ID was front-loaded. What predictions can you make from this claim?

Would you like to know what I might expect to see if life forms were actualy designed?

a) Little or no common descent.
b) A multitude of life based on different chemical or electronic substrates.
c) Life that exists without the need from replication from scratch (i.e., a focus on long-term maintenance instead of manufacture)
d) Life manfuactured from immutable patterns instead of continuously mutating.
e) A purpose to all life beyond survival.
f) Designs that are isolated and modular.
g) Diagnostic points that were used for testing and calibration during design and manufacture.
h) Mechanisms for quality control with respect to fidelity to a common design (i.e., ensuring that we are all perfect copies).
i) Design and manufacture across timespans comparable with the operational life of animal species like humans (I don't take a billion years to design a car that will last 10 years).
j) Large populations of life forms that are unsuited to their environment, and which have no inbuilt capacity for replication.

Sorry for the long post.

3/10/2006 10:54 PM

Art said...

Septeus7,

Thanks for the post, but please remember not to let the comment degenerate into ad-hominems.

3/22/2006 4:20 PM

Doctor Logic said...

LOL!

77, you should seek psychiatric help.

EAM sounds remarkably similar to Lamarckism. TD Lysenko would be proud.

PEH is based on the absurd premise that evolution has stopped.

Biotic Message Theory says that life does whatever life is observed to do (including looking evolved), and that this should be taken as proof of God because there's no alternative explanation. Sorry, but meta-explanations are not explanations.

The rest appears to consist of uncivilized waffling about CSI (which has already been conceded as just another way of saying things "looks designed"), and blanket denials of scientific facts.

Why is a lack of Junk DNA a mark of design vs. chance?

Because it is another instance of specific complexity which is more likely explain by intelligent than by know mechanisms.

Here we go again. Just replace "specified complexity" with "looking designed".

If there's front-loading and the physics produces junk, then you'll get junk anyway. If the physical mechanism doesn't produce junk, you'll not have junk either way.

Let's examine your responses:

a) Non-sequitor.

The mechanisms of natural selection require common descent. You can have multiple independent trees, but you must have trees with branches. ID does NOT require trees at all. Sure, you can twist claim that there are models of ID that have trees, but they're not a requirement. You're just tuning ID to match what you observe. You're still not being predictive.

b) False, See the Biotic Message and Front Loading Theories. Darwinism has actually predicted b.

See (a). Not only does ID not need trees, it doesn't need common substrates. Can a theory of magical origins account for a common substrate and common descent? Sure. No predictions yet, though.

c) Replication isn't from scratch, its from a prior organism.

Like automobiles?

d) Impossible, von Neumann machines require some variation.

Who said anything about self-replication?

e) Question Begining.

I assume you meant to write "question begging." If life was constructed by an intelligence, what's it for? Apparently, it's just for survival. Oh, I forgot - God got lonely.

f) Thats no fun. And besides isn't the citric acid cycle modular? There are countless examples of modular physiology. Write to John Davidson, he'll happy to tell you about them.

No, life isn't remotely modular. It's spaghetti code that's massively non-linear.

g) I'm not entirely sure there aren't such point. It would be a good area to do research that Darwinist would never consider.

Neither would a bricklayer, but physicians search for such things continuously. Find them they do not.

h) That is physically impossible given the nature of physically reproducing systems and the time involved. Further more that would be an increbilible stupid way to build systems to survive geological time periods. Furthermore, quality control mechanism regarding the genetic code and other features of life are numerous.

Not physically impossible for God or the Q continuum. Not even impossible for humans. And do you suppose QC mechanisms have selection advantage?

i) Assumes the Designer cares, something Dr. Illogical can't know.

No. Assumes the designer has a reason for doing things.

j) Hello, You might as well say a Intelligent Designer never living life.

I wouldn't say that because I have no idea what it means.

Bluster is no substitute for knowledge and rationality.

3/23/2006 8:58 PM

Defending ID 4: Falsifiability of Intelligent Design

9 Comments: