# Does Bayesian Fuzziness Add to the Analysis?

In comment 30 to this post Elizabeth Liddle writes:

I can think of lots of ways of testing specific design hypotheses, but they all involve a hypothesis involving a postulated designer. And IDists insist that this is irrelevant – that “Design detection” should only involve the observed pattern, not any hypothesis about the designer. This is ludicrous, frankly.

Let’s explore one of Lizzie’s prior forays into design detection, and we’ll leave it up to the onlookers to decide which side is “ludicrous.”

In a prior post I posed the following question to Dr. Liddle:

If you were to receive a radio signal from outer space that specified the prime numbers between 1 and 100 would you conclude (provisionally pending the discovery a better theory, of course) that the best theory to account for the data is ‘the signal was designed and sent by an intelligent agent?’

Dr. Liddle responded:

Yes. And I’ve explained why.

She expanded on her explanation:

Barry, I did NOT make the inference ‘based upon nothing but the existence of CSI’!

My inference had nothing to do with CSI.

It was a Bayesian inference based on two priors:

My priors concerning the probability that other parts of the universe host intelligent life forms capable of sending radio signals (high)

My priors concerning the probability that a non-intelligent process might generate such a signal (low).

Dr. Liddle’s problem can be summarized as follows:

1.  Denying the design inference based on the prime number sequence is not an option.  The inference is so glaringly obvious that to deny it would be absurd.  Even arch-atheist Carl Sagan admitted this signal was obviously designed (when he used it as the basis of his book “Contact”).  Therefore, were Dr. Liddle to deny the obvious design inference she would instantly lose all credibility.

3.  Her solution:   “I know.  I’ll admit the design inference but cover up my admission with Bayesian fuzziness, and that will obscure the fact that I used the methods of the ID proponents while I continue to denounce those very methods.”

Notice how Dr. Liddle’s Bayesian “priors” add absolutely nothing to the design detection methods advocated by ID proponents.  Here is a graph of the explanatory filter:

Let’s run the prime number sequence through the explanatory filter to see how.

1.  We observe an event (i.e. a radio signal specifying the prime numbers between 1 and 100).

2.  Is it highly contingent?  Yes.  We can exclude mechanical necessity.

3.  Is it highly complex and specified?  Yes.  We can exclude chance.

4.  The best explanation for the data:  Design.

Now let’s see if Dr. Liddle’s Bayesian analysis adds anything to what we already have.

Prior 1:  Estimate of the probability that other parts of the universe host intelligent life forms capable of sending radio signals:  High

It is obvious that one’s prior estimate of the probability of the existence of intelligent life forms in other parts of the universe is utterly irrelevant to the design inference.  How do I know?  By supposing the exact opposite of course.  Let’s assume that a person believes there is practically zero chance that other parts of the universe have intelligent life (as we have seen on this site, there is very good reason to believe this).  If that person were to receive this signal he would have to revise his conclusion, because the signal is obviously designed.

We see, therefore, that whether one’s Bayesian prior regarding the probability of the existence of intelligence life forms is 0% or 100% makes absolutely no difference to the design inference.  From this we conclude that Dr. Liddle’s first prior adds nothing to the analysis.

Prior 2:  Estimate of the probability that a non-intelligent process might generate such a signal:  Low

This, of course, is the explanatory filter by another name.  How do we know that the probability that a non-intelligent process might generate such a signal is low?  Because it is highly continent, complex and specified.

It is important to see two things:

1.  When Dr. Liddle correctly inferred design from the prime number sequence she had one and only one data point:  A radio signal specifying the primes between 1 and 100.

2.  Dr. Liddle knew nothing about the provenance of the radio signal.  In other words she made a design inference based on nothing but the pattern itself while knowing absolutely nothing about the designer.  When she made her design inference she did not first make a hypothesis based on the “postulated designer,” for the simple reason that there was not a scintilla of data upon which to base that hypothesis other than the pattern itself.

Conclusion:  Though she tried to cover it up with Bayesian fuzziness, Dr. Liddle did the very thing she now says is “ludicrous.”

## 120 Replies to “Does Bayesian Fuzziness Add to the Analysis?”

1. 1

First of all, you might like to look at my later response to Timaeus in the same thread, here.

Second, I wonder if you could link to my original response to your question.

Third, I utterly reject this characterisation of my “problem”:

Dr. Liddle’s problem can be summarized as follows:

1. Denying the design inference based on the prime number sequence is not an option. The inference is so glaringly obvious that to deny it would be absurd. Even arch-atheist Carl Sagan admitted this signal was obviously designed (when he used it as the basis of his book “Contact”). Therefore, were Dr. Liddle to deny the obvious design inference she would instantly lose all credibility.

3. Her solution: “I know. I’ll admit the design inference but cover up my admission with Bayesian fuzziness, and that will obscure the fact that I used the methods of the ID proponents while I continue to denounce those very methods.”

I am utterly uninterested in “continuing to deny the methods of ID proponents” and thus utterly unmotivated to cover up any putative “admission” whatsoever. This assumption regarding my integrity (or rather lack of it) is extremely offensive.

Or would be if I took it seriously, which I do not. So that’s all fine. Harrumph.

So I will attempt to respond to your OP:

You conclude design based on the following reasoning steps:

1. We observe an event (i.e. a radio signal specifying the prime numbers between 1 and 100).

2. Is it highly contingent? Yes. We can exclude mechanical necessity.

3. Is it highly complex and specified? Yes. We can exclude chance.

4. The best explanation for the data: Design.

I use a different reasoning method, and come to the same conclusion. Here is mine, in full (as I don’t currently have access to my previous response I will do it again from scratch, and show my working):

What I want to know is the probability of Intelligence, given the Signal. I will write that as:

P(I|S).

using Bayes rule, I get:

P(I|S) = P(I)*P(S|I)/P(S)

And I can expand the denominator to:

= P(I)*P(S|I)+P(¬I)*P(S|¬I)

So let me take a stab at some priors:

Apparently I said then (and I would say again now):

My priors concerning the probability that other parts of the universe host intelligent life forms capable of sending radio signals (high)

My priors concerning the probability that a non-intelligent process might generate such a signal (low).

So let me set my prior on the probability of intelligent life somewhere, quite high, say .01.

And let me set my prior on the probability intelligent life being able to send radio signals to earth as lower, let’s say .00001

And let me set my prior on the probability of non-intelligent mechanisms sending such a signal even lower, let’s say, .00000001

So P(I)=.01

and P(S|I)=.00001

Substituting above, we get, for the numerator:
.01 * .00001

Now, for the denominator, we need

P(¬I), which is 1-.01 = .99

and

P(S|¬I), which I have set low, at .00000001

Which gives us, for the denominator

.01 * .00001 + (1-.02)*.00000001

Which, unless Excel is lying to me, gives me a posterior probability of an intelligent source, given the signal, i.e. of P(I|S) of

0.91

Which is clearly high enough to take to the bank.

Note that I can tinker with these priors as I want, but as long as I put the probability of intelligent life, and the probability of intelligent life being capable of sending a radio signal as substantially more than the probability of a non-intelligent mechanism emitting a signal, then I can safely conclude design.

But at no point have I used CSI. I set my probability of the signal given a non intelligent source at way lower than 500 bits – only 27. So even with a much more lenient prior for non-intelligent origin than Dembski’s I can still conclude “intelligent source” with considerable confidence.

This is not “fuzziness” – it is a way of extracting an estimate of the confidence that one can place in a conclusion, given one’s prior beliefs about what is likely to be the case. And because my prior belief in the likely existence of intelligent designers capable of sending a signal is substantially greater than my prior belief that any non-intelligent mechanism is likely to generate such a signal, I end up with a variant on the “inference to best explanation” method, with a tad more quantitative rigor.

And the beauty of the Bayesian approach is that it allows us to revise our estimate as more information comes available. Should we discover some crystal lattice structure that leaves voids in a prime number pattern, and can act as a radio signal filter of some kind (I’m speculating wildly here, deliberately), then we’d have to revise our priors substantially, and might, like Jocelyn Bell, have to abandon the LGM for something less exotic but possibly no less interesting.

I entirely agree that my method does not “add” to yours. But nor does it require yours. It does not, as I said, require me to compute the CSI of the signal. Which is just as well, because I do not think that CSI is computable!

And nor, in fact, do I think your first step works either – it isn’t really separable from the second, which is why Dembski in the end rolled them both up into one, by requiring that for the P(T|H) parameter in the CSI calculation, the H should be “the chance hypothesis that takes into account Darwinian and other material mechanisms”.

In other words, if you do that,you don’t need Step 1, because it has already been done in your CSI computation for Step 2.

Except that my case is that it isn’t possible to do, for anything other than a very tightly constrained null (for instance, the null that a coin is fair).

Look Barry, as you will see from my later response to Timaeus: I honestly don’t have a dog in this hunt. Or not much of a dog anyway, maybe a small theological ferret.

I’d be truly delighted to see a well-founded approach to design detection in the domain of living things. I even suggested some possible avenues that I would consider potentially fruitful, or at least, not beset by the extremely grave problems I see in CSI and its derivatives.

Especially as I rather belatedly discovered today,that Dembski’s 500 bit university probility bound, far from being “conservative” (although more conservative than Seth Lloyd’s) was actually based on an estimate of the number of particles in the observable universe which is extremely unlikely to be coterminous with the entire universe! So, in effect, we do not have an upper bound on probabilistic resources of the universe anyway.

But as I’ve said, many times, and as this post exemplifies: I would be perfectly happy to conclude “Design” on a much more lenient cutoff criterion than any UPB, given a better founded rationale.

Obviously you will differ.

But I rest my case (is that the right expression?)

2. 2
Mark Frank says:

Let’s assume that a person believes there is practically zero chance that other parts of the universe have intelligent life (as we have seen on this site, there is very good reason to believe this). If that person were to receive this signal he would have to revise his conclusion, because the signal is obviously designed.

I am sorry but I wouldn’t jump to the conclusion it was designed. If I was virtually certain there was no possibility of intelligent life I would examine the possibility of another cause rather seriously. Clearly we have something rather extraordinary happening but then we have already accepted that intelligent life would be extraordinary. One line of thought might be a natural mechanism that generates a stream of increasing numbers but is filtered by another mechanism that eliminates any number that is a multiple of previous numbers in the sequence. Sure I have no idea how such a mechanism could work but then if I don’t accept the possibility of intelligent life I have no idea how it could be designed either.

One of the big advantages of Bayesian thinking is that makes you estimate the probability in a structured way rather than just saying “it’s obvious”. The world is full of examples of people jumping to the wrong “obvious” conclusion because they have not adopted Bayesian thinking (including many famous legal cases which you will no doubt be aware of).

3. 3

Yes, indeed, Mark. And that is of course where we both disagree with Barry.

We simply do not accept that we can make a design inference in a vacuum. In fact, I could (and probably should!) have responded to Barry with a much shorter post:

We cannot evaluate the probability of an event without considering the generative processes that might have produced it.

It’s the error at the heart of Dembski’s method. And it’s hiding there, in plain site, in the old EF: You can only reject “mechanical necessity” if we have exhausted not only known mechanisms but also unknown mechanisms. Which is clearly impossible. So the filter is doomed from the start.

4. 4
Barry Arrington says:

Elizabeth, let’s go back to your original statement:

I can think of lots of ways of testing specific design hypotheses, but they all involve a hypothesis involving a postulated designer.

Where in comment 1 do you make any hypothesis about the designer other than that he/she/it is “intelligent life”? The answer, of course, is that you do not. Your own comment 1 confirms the point of my post, which is that a design inference can be based on nothing but the observation of the pattern itself without reference to any information about the designer other than that he/she/it is intelligent. I will address other issues in your response in later comments, but I wanted to first thank you for confirming the point of the OP.

5. 5
scordova says:

Bayesian ranking works if you have alternative mechanisms and your list of mechanisms is complete. It will fail if the mechanism in play is not on your list.

We may not be able to prove an intelligence exists in the formal sense to create the aliens that gave the SETI signal, but we can use CSI to demonstrate the SETI signal resembles a design. CSI is provisionally a measure of resemblance of design based on available distribution functions.

FWIW, pulsars were once considered possibly of alien origin because of their clock like precision….

Suppose someone did think pulsars were evidence of ID? So what? The damage done would hardly be worth notice in the scheme of things.

But in the realm of uncertainty, the real question is which idea (ID or Darwinism) is the better wager? Darwinism is a terrible wager. Not even in terms of eternal life, but if we devalue human life, when in fact it is special, what is the cost of Darwinism if Darwinism is wrong?

We might never be able to settle with ID is true or not, but like a businessman wagering on the best investment in light of unending uncertainties, what is the best wager on truth. I’m not betting on Darwin, it’s a negative expectation bet on many levels. That wager reasonable inference that I’ve not seen any rebuttal for.

6. 6

Bayesian ranking works if you have alternative mechanisms and your list of mechanisms is complete. It will fail if the mechanism in play is not on your list.

Can you explain this, Sal? I don’t understand what you are saying.

FWIW, pulsars were once considered possibly of alien origin because of their clock like precision….

Suppose someone did think pulsars were evidence of ID? So what? The damage done would hardly be worth notice in the scheme of things.

Yes indeed they did. It was the Cambridge team with Jocelyn Bell, who, as it happened, had been at my boarding school and I was friends with her younger sister. It was all very exciting – hence my reference to the “LGM” hypothesis – the Little Green Men.

7. 7
scordova says:

Can you explain this, Sal? I don’t understand what you are saying.

Sorry, my wording was awful. We can rank possibilities in Bayesian reasoning according to various mechanisms, but what if the real mechanism is missing? What if the mechanism can’t be described in mechanical terms (like intelligence).

8. 8

Barry:

Where in comment 1 do you make any hypothesis about the designer other than that he/she/it is “intelligent life”? The answer, of course, is that you do not.

Indeed.

Your own comment 1 confirms the point of my post, which is that a design inference can be based on nothing but the observation of the pattern itself without reference to any information about the designer other than that he/she/it is intelligent.

My method would work just as well if my prior for a living designer was infinitessimal, but my prior for a non-living designer was, say, .01.

In fact, rather than re-do the calculation, just substitute :

So let me set my prior on the probability of an intelligent life designer somewhere, quite high, say .01.

And let me set my prior on the probability an intelligent life being able designer choosing to send radio signals to earth as lower, let’s say .00001

It works just as well. We still do not need to know anything about the designer, and we can still conclude design, given those priors. They are not mine, but for someone, perhaps yourself, with good reasons for having them (direct experience of the Divine, for instance) they could be perfectly reasonable.

I will address other issues in your response in later comments, but I wanted to first thank you for confirming the point of the OP.

But I haven’t. Your OP seemed to be saying that my Bayesian approach “added nothing” to the EF. I took the example of a SETI scenario, and plugged in priors about an Extra Terrestrial, because I thought that was what we were talking about.

But as I have just shown, the Bayesian approach works just as well for a supernatural designer – as I point out, it’s a rather more algebraic equivalent of the Inference to Best Explanation, which quite a few IDers are quite keen on.

My point is not that the Bayesian approach adds anything to the EF, but that it is a much better alternative. And certainly does not preclude making a Design inference. It just that with the Bayesian approach, you have to make your priors explicit. With the CSI approach they are hidden (in that P(T|H) I keep banging on about).

I think you have assumed (well, you have explicitly said this) that because I am critical of ID, I am nitpicking at the methodology because I don’t like the inference.

This is simply not true.

9. 9

Sorry, my wording was awful. We can rank possibilities in Bayesian reasoning according to various mechanisms, but what if the real mechanism is missing? What if the mechanism can’t be described in mechanical terms (like intelligence).

Well, we can certainly rank the posteriors, given our priors and likelihoods. I don’t see that I’ve done anything wrong. I don’t think I have to describe intelligence in mechanical terms to give it a substantial prior.

(This is slightly weird: I am trying to persuade an IDer that it is just fine to put a prior on an unknown mechanism, and that unknown mechanism is a Designer…. Ah well :))

10. 10
Barry Arrington says:

Liddle: “But I haven’t [confirmed the point of the OP].”

Of course you have. The point of the OP is that one can make a design inference based on nothing but the existence of the pattern. A corollary to the point of the OP is that your statement that one needs to make a hypothesis about the designer before one can make a design inference is false.

In fact, as you now concede, one need know absolutely nothing about the designer (other than that he/she/it is intelligent, i.e., capable of design) to make that inference. You confirmed that point in your comment 1 and you reconfirmed it in your comment 8. Thank you.

11. 11
Barry Arrington says:

Mark Frank: “One of the big advantages of Bayesian thinking is that makes you estimate the probability in a structured way rather than just saying ‘it’s obvious.’”

Liddle: “Yes, indeed, Mark.”

Here’s the problem with Liddle’s analysis. Prior to receiving the signal we have zero data upon which to reasonably assign any probability whatsoever to the existence of intelligent life other than that on earth. The most famous attempt, of course, is based on Drake’s equation. But as Crichton has pointed out:

The problem, of course, is that none of the terms [in the equation] can be known, and most cannot even be estimated. The only way to work the equation is to fill in with guesses . . . As a result, the Drake equation can have any value from ‘billions and billions’ to zero. An expression that can mean anything, means nothing. Speaking precisely, the Drake equation is literally meaningless . . .

Liddle’s estimate of the probability of intelligent life in the universe is meaningless, and her estimate of the probability that intelligent life in the universe can send a radio signal is also meaningless.

Liddle and Frank say that the advantage of Bayesian thinking is that it makes you estimate the probability in a structured way.

Uh, explain to me again how pulling probabilities out of nothing but thin air* and putting them in an equation is an advantage.

Of course it is not an advantage. Again, it is obvious what Liddle is doing (and Frank is applauding). Liddle has pulled meaningless probabilities out of thin air and plugged them into a Bayesian equation. The result looks rigorous until you realize that when you boil it all down it is all but meaningless and if by chance it does have some meaning, that meaning is “my gut tells me there is intelligence out there and they sent the signal.” Yeah, that’s rigorous alright.

*I had to resist with all my might the temptation to write “your a– –” instead of “nothing but thin air,” but that would have been vulgar. 🙂

12. 12
RDFish says:

Hi Sal,

What if the mechanism can’t be described in mechanical terms (like intelligence)

The behavior of humans and other animals that we call “intelligent” cannot presently be explained in mechanical terms, because we don’t understand how that behavior is generated. To say intelligence can’t be described in mechanical terms in principle is an unsupported assumption, however.

Moreover, if we cannot explain what “intelligence” is in mechanical (or non-mechanical!) terms, then it is not helpful to offer it as an explanation for anything. It is not “intelligence” that designs cars and computers and Mt. Rushmore, it is “human beings”. It is not “intelligence” that designs eyeballs and blood clotting cascades and flagella, it is… nobody has any idea.

The “design hypothesis” is not a meaningful explanation of anything at all. It is like trying to explain planetary motion with the “movement hypothesis” without saying what is doing the moving – it tells us precisely nothing.

Cheers,
RDFish

13. 13
CentralScrutinizer says:

E Liddle: Barry, I did NOT make the inference ‘based upon nothing but the existence of CSI’! My inference had nothing to do with CSI.It was a Bayesian inference based on two priors: My priors concerning the probability that other parts of the universe host intelligent life forms capable of sending radio signals (high) My priors concerning the probability that a non-intelligent process might generate such a signal (low).

So, Dr Liddle, I’m wondering, when it comes to the particular DNA replication system that we find in nature, and the processes going on within cells, what do you consider the probably of those being the result of natural forces, and what probably of them being the result of a human-like intelligence?

14. 14
Mark Frank says:

#11

Barry – if the prior probability is really unknown then it forces you to confront that fact and therefore the incredible uncertainty in your conclusion. That uncertainty is real and it is better to recognise it. An alternative that brushes over the uncertainty and thus allows you to announce that something is certain is not superior.

15. 15

But I haven’t based it on “nothing but the existence of a pattern”, Barry!

I’m really mystified as to why you should be telling me I’m saying things that I plainly have not said!

I’m saying that in order to make a design inference using Baysian reasoning, we need to consider our priors for the probability of a number of things, including the probability of a Designer. Ideally that probability should be based on some kind of data, if you want to have confidence in your posterior. You also need a prior about the probability of the signal without a designer. And you should probably also have a probability for the probability that if there is a Designer, he/she/it would send a radio signal, but we can set that as 1 if you like.

That is a long way from “nothing but the existence of a pattern”! But it DOES include the probability of an immaterial designer if that is what you are interested in.

So reports of my concessions are greatly exaggerated!

A corollary to the point of the OP is that your statement that one needs to make a hypothesis about the designer before one can make a design inference is false.

I am not aware of having said such a thing, although I’d like to see the quote and check the context. Certainly if you wanted to use null hypothesis testing, you would need a hypothesis about the designer, and I’ve certainly said that. I’ve also said, as you quote in the OP, that “I can think of lots of ways of testing specific design hypotheses, but they all involve a hypothesis involving a postulated designer.”

However, null hypothesis testing is not the only way of making an inference, and I have suggested the Bayesian reasoning given above. As the priors for an immaterial designer are poorly supported (though not absent), I don’t think one could have a great deal of confidence in the inference, but an inference is nonetheless perfectly possible, and indeed reasonable. But a Bayesian inference is not a test of a hypothesis – it is a way, as Sal says, of ranking hypotheses – specifically, of ranking the probability of a hypothesis being true, given the data, but the ranking is extremely dependent on the input priors.

On the other hand, if we did have a specific design hypothesis, for example the “front-loading” hypothesis, that would be very testable, or could be. It could make specific predictions not predicted by an evolutionary model and so could go head-to-head with evolution.

As I keep saying, Barry, I have no problem in principle with attempting to detect design, and I have even no problem in principle, in trying to figure out how to detect a Designer who transcends physical laws.

The point I was trying to make in the exchange you quote is that I do not think that the EF, or indeed CSI, work. I think they represent a completely invalid application of Fisherian null hypothesis testing, and I think ID could do a lot better.

16. 16
CentralScrutinizer says:

RDFish: Moreover, if we cannot explain what “intelligence” is in mechanical (or non-mechanical!) terms, then it is not helpful to offer it as an explanation for anything. It is not “intelligence” that designs cars and computers and Mt. Rushmore, it is “human beings”. It is not “intelligence” that designs eyeballs and blood clotting cascades and flagella, it is… nobody has any idea.

So if we were to continously detect the first 100 primes from 2 to 541 from a distant source, you would not at least provisionally put the tag of “intelligently generated” on the phenomenon? How would you tag it?

17. 17
scordova says:

To say intelligence can’t be described in mechanical terms in principle is an unsupported assumption, however.

Agreed, thank you for the correction.

Sal

18. 18
Barry Arrington says:

Liddle quotes Arrington: “A corollary to the point of the OP is that your statement that one needs to make a hypothesis about the designer before one can make a design inference is false.”

Liddle then writes:

I am not aware of having said such a thing, although I’d like to see the quote and check the context. Certainly if you wanted to use null hypothesis testing, you would need a hypothesis about the designer, and I’ve certainly said that. I’ve also said, as you quote in the OP, that “I can think of lots of ways of testing specific design hypotheses, but they all involve a hypothesis involving a postulated designer.”

Dr. Liddle, for the life of me it appears as though you are saying that the phrase “a hypothesis about the designer” is not equivalent to the phrase “a hypothesis involving a postulated designer.” If that is the case, please explain what you meant by the latter and how it differs from the former. Keep in mind that my overall point is that one can make a design inference while knowing absolutely nothing about the designer other than that he/she/it is intelligent. If you concede that statement, then we don’t disagree.

19. 19
Barry Arrington says:

Mark Frank:

Barry – if the prior probability is really unknown then it forces you to confront that fact and therefore the incredible uncertainty in your conclusion. That uncertainty is real and it is better to recognise it. An alternative that brushes over the uncertainty and thus allows you to announce that something is certain is not superior.

Mark, it seems like you are saying that if we cannot reliably estimate the probability of the existence of other intelligent life prior to receiving the Signal then there is “incredible uncertainty” in the conclusion that the Signal is the result of intelligent design.

If that is what you are saying, then your statement is preposterous and patently absurd. It is the exact opposite of the truth. Far from being incredibly uncertain, the design inference based on the signal is all but certain. As I said, even a man who was perhaps the most famous materialist atheist of the latter half of the twentieth century (Carl Sagan) would agree that the design inference is all but certain.

Can it be that you really believe the gibberish that you wrote? I have a hard time accepting that.

20. 20
scordova says:

The “design hypothesis” is not a meaningful explanation of anything at all. It is like trying to explain planetary motion with the “movement hypothesis” without saying what is doing the moving – it tells us precisely nothing.

Cheers,
RDFish

A slight but respectful disagreement. Planets move according to an elipse. It would be wrong to say the mechanism is an elipse, the mechanism is gravity that creates the elliptical motion.

Designs look to be designed. It is formally wrong to say design is the mechanism of design. It is correct however to say it looks analogous to something a human could design if the human hand the resources and intelligence. Whether intelligence is the root cause, there is nothing wrong to say it looks like the product of human-like intelligence or something greater. I don’t argue vigorously whether it is a scientific claim, but the impression of design in certain cases is clearly there.

In the case of pulsars, the impression of design was false. No great loss, no disaster. If ID is wrong, no great disaster to the human enterprise either, imho.

21. 21
Barry Arrington says:

Liddle:

I’m saying that in order to make a design inference using Baysian reasoning, we need to consider our priors for the probability of a number of things, including the probability of a Designer. Ideally that probability should be based on some kind of data, if you want to have confidence in your posterior.

And as I explained in comment 11, your priors are literally meaningless. You’ve pulled them out of thin air.

Keep in mind that I am not criticizing Bayesian analysis as such. Indeed, I use it all the time at the poker table.

I am saying that Bayesian analysis in this instance does nothing but cover up the fact that you know nothing about the probability of the existence of intelligent life prior to getting the Signal, and that is a huge coverup. In other words, when the Bayesian priors are garbage the Bayesian posteriors will also be garbage, a classic example of the GIGO phenomenon at work.

22. 22
CentralScrutinizer says:

RDFish: The “design hypothesis” is not a meaningful explanation of anything at all. It is like trying to explain planetary motion with the “movement hypothesis” without saying what is doing the moving – it tells us precisely nothing.

Wrong. Because designers are intelligent, that is, intelligence manifests various properties. What would you call something that has the following properties? …

1. “Understanding” of natural forces.

2. “Foresight” of future events based on the arrangements and relationships of various natural forces.

3. “Intent” to alter the course of future events.

I call it “intelligence.”

Apparently SETI agrees, because their entire objective is to find something “out there” that exhibits the marks of something that only an intelligent source could generate. The term “intelligence” is not vacuous, and neither is a design hypothesis.

Let’s try to explain the first 100 primes from a distant radio source. It would either be due to chance and necessity of natural forces, or Something With the Properties Cited Above, i.e., “intelligence.” Both are plausible explanations given what we know about entities that have those properties, namely, humans.

A design hypothesis in this case would be one that says, “an intelligent entity or entities caused the transmission of these primes. It is not plausible that cause and necessity of natural forces did it.” That’s essentially the hypothesis SETI is operating under.

Sidebar: does this mean the intelligence has to be from a “human?” I don’t think so. What cogent case can you make that any such intelligence would necessarily be human?

23. 23

Barry:

Here’s the problem with Liddle’s analysis. Prior to receiving the signal we have zero data upon which to reasonably assign any probability whatsoever to the existence of intelligent life other than that on earth. The most famous attempt, of course, is based on Drake’s equation. But as Crichton has pointed out:

The problem, of course, is that none of the terms [in the equation] can be known, and most cannot even be estimated. The only way to work the equation is to fill in with guesses . . . As a result, the Drake equation can have any value from ‘billions and billions’ to zero. An expression that can mean anything, means nothing. Speaking precisely, the Drake equation is literally meaningless . . .

Liddle’s estimate of the probability of intelligent life in the universe is meaningless, and her estimate of the probability that intelligent life in the universe can send a radio signal is also meaningless.

I think this is a pretty fair point. I think trying to weigh up probabilities in the absence of any data on which to construct a frequency distribution is pretty pointless. What emerges from a Bayesian posterior is only as good as the priors you plug in. If they are well-founded, you can have confidence in your posterior. If they aren’t, well, it’s GIGO. Nonetheless, for the example you gave me, my inference would be “design” because on the little data I have, the priors for intelligent signal-sending life are higher (after all, I know of at least one intelligent signal-sending life form) than for a prime-number-generating non-intelligent process (frequency distribution of data = []). So simply by using the fact that I have some reason to think an intelligent signaller is possible and no reason to think a non-intelligent signaller is possible, Design will come out ahead.

So using the very little data I have, I would conclude Design, using that reasoning.

But then you asked me to consider it again, for an non-living designer. Well, I simply have no basis in which to make such a prior estimate at all, but as an exercise, I showed you how it could be done, possibly for someone such as yourself who does consider that a non-living designer is a substantial non-zero possibility, presumably based on your personal experience.

Liddle and Frank say that the advantage of Bayesian thinking is that it makes you estimate the probability in a structured way.

Uh, explain to me again how pulling probabilities out of nothing but thin air* and putting them in an equation is an advantage.

Not quite thin air, Barry, as I explain above. But pretty thin air, I agree, in this case, not surprisingly as it is a completely fictional scenario.

Of course it is not an advantage. Again, it is obvious what Liddle is doing (and Frank is applauding). Liddle has pulled meaningless probabilities out of thin air and plugged them into a Bayesian equation. The result looks rigorous until you realize that when you boil it all down it is all but meaningless and if by chance it does have some meaning, that meaning is “my gut tells me there is intelligence out there and they sent the signal.” Yeah, that’s rigorous alright.

I didn’t claim it was rigorous. It’s as flakey as heck. But if a gun was held to my head, and I had to choose, at least it would give me a basis on which to pick, using the tiny amount of data at my disposal. A Bayesian inference is only as good as the support for your priors.

*I had to resist with all my might the temptation to write “your a– –” instead of “nothing but thin air,” but that would have been vulgar.

No problem, Barry. I have heard (and used) worse 🙂

But let me try to clarify the point that both Mark and I are making:

To compute CSI, you need to pull an number out of your ahem.

Specifically you need to plug in P(T|H) where H is “the relevant chance hypothesis taking into account Darwinian and other mechanisms”.

There is no way to compute this. You just have to pull it out of your donkey based on your priors about the probability of “Darwinian and other mechanisms”. It is, indeed, the exact equivalent of P(S|¬I) in my equation above. But unlike my equation, it does NOT require you to make any estimate of P(I). It is therefore precisely as flakey in one parameter, and simply omits the other. And then it adds another parameter, which as far as I can see is based on a mistaken assumption regarding the upper bound on the size of the universe.

But the worst thing about it is that that P(S|¬I), though present, is never, ever, in my experience of seeing CSI arguments, calculated for anything other than “random independent draw”. In other words, the first step of the EF, never actually makes it into the second step. So while my figure might have been a wild stab, at least it was based on some kind of data, and balanced by an estimate of P(I) and P(S|I) given more, CSI simply buries all this donkey extraction stuff, and then triumphantly claims Design, p<10^-150!

My more modest p=.91 for the probability of Design, given the data, seems rather more seemly.

My Little Pony extraction, perhaps 🙂

24. 24

And as I explained in comment 11, your priors are literally meaningless. You’ve pulled them out of thin air.
Barry

Keep in mind that I am not criticizing Bayesian analysis as such. Indeed, I use it all the time at the poker table.

I am saying that Bayesian analysis in this instance does nothing but cover up the fact that you know nothing about the probability of the existence of intelligent life prior to getting the Signal, and that is a huge coverup. In other words, when the Bayesian priors are garbage the Bayesian posteriors will also be garbage, a classic example of the GIGO phenomenon at work.

On the contrary, as I explain above, it does the reverse of “cover up” the paucity of the data that supports the priors. It makes it absolutely explicit. But as my post there has only just loaded (for some reason it got stuck in cyberspace for a short while) I’ll say no more until you’ve had a chance to read it.

25. 25
Barry Arrington says:

Liddle “But I haven’t based it [i.e., the design inference”] on ‘nothing but the existence of a pattern’, Barry!”

Of course you have Elizabeth, your denials to the contrary notwithstanding.

As I have shown, your Bayesian priors are meaningless. Therefore, your Bayesian analysis is meaningless. Yet you and I both know that the design inference is indisputable. [Setting aside ludicrous appeals to crystal lattice structures that leave voids in a prime number pattern].

And what is that indisputable design inference based on if it is not based on a wholly unsupportable Bayesian analysis? Why, it is based on nothing more than our knowledge that nature is incapable of generating the Signal while intelligent agents produce Signals of this sort routinely. In other words, the inference is based on nothing but the pattern itself and that fact that it contains complex specified information that nature cannot produce.

26. 26
LarTanner says:

the fact that you know nothing about the probability of the existence of intelligent life

Other than the other fact that apparently intelligent life lives on Earth.

27. 27

Barry:

Dr. Liddle, for the life of me it appears as though you are saying that the phrase “a hypothesis about the designer” is not equivalent to the phrase “a hypothesis involving a postulated designer.” If that is the case, please explain what you meant by the latter and how it differs from the former. Keep in mind that my overall point is that one can make a design inference while knowing absolutely nothing about the designer other than that he/she/it is intelligent. If you concede that statement, then we don’t disagree.

Let me try to say what I mean, in the hope that it may make sense of what I have previously said 🙂

Dembski’s EF, and CSI concepts are based on Fisherian null hypothesis testing. I think it is misapplied. I think if you want to use null hypothesis testing, or indeed any frequentist form of hypothesis testing, you need to have a specific hypothesis about the designer, or design method, that will enable you to properly calculate an appropriate null.

However.

I do think that it is perfectly reasonable to make an informed (even minimally informed) inference about which of two (or more even) general hypotheses regarding a phenomenon (Was it designed? Was it not designed?) are more likely, given the data, using a Bayesian approach, which, as I say, is extremely similar to “inference to best explanation”. But, as you (and I, simultaneously as it happens!) state: GIGO. Your inference is only as good as your priors. If the data supporting your priors is very weak, then you probably shouldn’t stake too much on the table when you make your decision. However, if it’s very strong, then you might.

But you are not testing your hypotheses, in the Popperian sense. You are simply trying to make the best decision between two alternatives given the best information you have. And the elegance of Bayes, as you will appreciate, is that it allows you to pull out an often counter-intuitive posterior given quite intuitive priors.

So we can probably agree (let me slightly rephrase this from your formulation) that it is perfectly reasonable to make an inference about which of two hypotheses is the more likely to account for your data, without fleshing out the hypotheses much at all (although I would argue that you do need to make a stab at something about your priors).

But it is not a test of your hypothesis, in the Popperian sense.

Thus I see no reason to think the Design hypothesis for life is unreasonable, although it’s not what comes out of the equation given my own priors.

What I do think is faulty is the EF/CSI method, which gives you a positive for Design with a p value to die for, but which, I suggest, is invalid.

28. 28
Barry Arrington says:

Liddle: “So simply by using the fact that I have some reason to think an intelligent signaler is possible and no reason to think a non-intelligent signaler is possible, Design will come out ahead.”

You write that you have no reason to think a non-intelligent signaler is possible, but surely you understand this is wrong. Non-intelligent sources send out radio signals all the time (quasars, for example). It is not the fact that a signal has been sent that is important. Both intelligent and non-intelligent sources send signals. The issue is whether there is anything about this particular signal that would lead to a design inference. And obviously there is.

Tell me Elizabeth, what is there about this particular signal that separates it from a signal generated by, say, a quasar.

Hint: The answer has the following three words that you can arrange in the correct order: information, specified, complex.

29. 29

While you catch up on my posts, Barry, because I think we have come slightly adrift, let me present a nice puzzle to you all reprobate gamblers (Sal, Barry, whoever else :))

You are a prisoner, about to be shot. But you are offered a possible out.

You are presented with a large bag, apparently filled with coloured balls.

You are told that the balls in the bag are from a factory that makes only red and blue balls.

You are invited to withdraw one ball from the bag.

It is red.

You are invited to withdraw a second, but first you must guess its colour.

If you are correct, you go free.

If you are incorrect, you are shot.

Which do you choose?
And why?

30. 30
Barry Arrington says:

Liddle and Arrington agree on this:

Your inference is only as good as your priors. If the data supporting your priors is very weak, then you probably shouldn’t stake too much on the table when you make your decision. However, if it’s very strong, then you might.

Liddle and Arrington agree that her priors here are:

Not quite thin air, Barry, as I explain above. But pretty thin air, I agree, in this case,

And that leaves the trap that Mark Frank fell into and I pointed out in 19. Let me explain it step-by-step:

1. Liddle can have no real confidence in her Bayesian priors. They are, as she concedes, pulled from pretty thin air.

2. Liddle also concedes the common sense conclusion that a Bayesian inference is only as good as the priors on which it is based.

3. Therefore, the Bayesian inference in this case is very weak.

4. This highlights the futility of applying a Bayesian approach in this case. The design inference based on a Bayesian analysis is very weak. Yet we know with all but certainty that the design inference is correct.

5. Therefore, an analysis that gives weak (non-existent really) support to a certain conclusion is useless.

31. 31
kairosfocus says:

RDF: Being human is neither necessary nor sufficient to be intelligent in the relevant sense of design. beavers are (albeit limitedly so) evidently intelligent in some of their dam building behaviour. Humans who do not know a fair amount of relevant things and have relevant skills cannot design computers or software, and so forth. It seems that “intelligence” is one of those things that points where you would not go. KF

32. 32

Barry:

You write that you have no reason to think a non-intelligent signaler is possible, but surely you understand this is wrong. Non-intelligent sources send out radio signals all the time (quasars, for example). It is not the fact that a signal has been sent that is important. Both intelligent and non-intelligent sources send signals. The issue is whether there is anything about this particular signal that would lead to a design inference. And obviously there is.

Tell me Elizabeth, what is there about this particular signal that separates it from a signal generated by, say, a quasar.

Hint: The answer has the following three words that you can arrange in the correct order: information, specified, complex.

Ah. I see the problem.

I would consider the fact that it was a rather special signal the major reason for stacking my priors for an intelligent agent. No issue with that at all. I think this is going to be resolved quite easily.

The issue here is, simply, the definition of “complex specified information”.

If you are defining CSI as something like:

A signal high in Shannon entropy (which Dembski, somewhat confusingly, calls “complexity”) i.e. lots of bits, but also spelling out something meaningful and thus “specified” and “information”

– no problem.

Yes I have certainly used CSI, thus defined, in reaching my Bayesian conclusions.

The problem is: this is not how Dembski defines it, and I was assuming we were using Dembski’s definition, seeing as we seemed to be using his EF!

Dembski has defined CSI in a number of ways, and one of them includes what I would call the “alpha” value in Fisherian null hypothesis testing – he sets it as p<10^-150, and then includes that value in his definition of CSI, which I always found rather odd. But ignoring that part, and just concentrating on his definition of a “complex, specified” pattern, his criteria are that it should have high “complexity” which he defines, somewhat confusingly, as high Shannon entropy (lots of bits) but, in addition, high compressibility (“ease of description” as I recall), more usually referred to as low “Kolmogorov” complexity, ironically.

Using Dembski’s definition of a complex specified pattern, the pulsar sequences have CSI: They are binary (on off on off), and they are long (they go on interminably) and therefore any sequence of signals has vast Shannon entropy (as many bits as the sequence is long). However, they are also extremely compressible (repeat “on off” N times).

So even just using that part doesn’t work.

But if we just define the kind of sequence we have (fictionally) observed as “a lengthy signal binary signal with high Shannon entropy (and thus large channel capacity) consisting data that can be readily interpreted as a meaningful message) – fine.

It’s just not how Dembski defines CSI.

(Although even if we define CSI as above, it’s still not the only data I used, and I would have dearly liked more).

33. 33
Barry Arrington says:

Liddle:

Using Dembski’s definition of a complex specified pattern, the pulsar sequences have CSI: They are binary (on off on off), and they are long (they go on interminably) and therefore any sequence of signals has vast Shannon entropy (as many bits as the sequence is long). However, they are also extremely compressible (repeat “on off” N times).

How in the world could you say that a simple “on-off” pattern is in any conceivable sense of the word “complex”? That you would suggest that anyone would say that an on-off pattern is complex is nothing short of astounding. Elizabeth, words have meaning. It is not licit to do linguistic violence to those meanings.

34. 34
Barry Arrington says:

Liddle:

I would consider the fact that it was a rather special signal the major reason for stacking my priors for an intelligent agent.

Game, set, match to Barry

35. 35
Mark Frank says:

Barry #19

Mark, it seems like you are saying that if we cannot reliably estimate the probability of the existence of other intelligent life prior to receiving the Signal then there is “incredible uncertainty” in the conclusion that the Signal is the result of intelligent design.

Pretty much yes.

If that is what you are saying, then your statement is preposterous and patently absurd. It is the exact opposite of the truth. Far from being incredibly uncertain, the design inference based on the signal is all but certain. As I said, even a man who was perhaps the most famous materialist atheist of the latter half of the twentieth century (Carl Sagan) would agree that the design inference is all but certain.

I don’t know what Carl Sagan wrote but the only reasoning you have given to support your case is my claim is absurd. As this is a situation unlike anything we have ever had to address it would be naive to argue from “this is absurd”.

Meanwhile I give you some maths:

Bayes formula:

P(Design|Data)= P(Data|Design)*P(Design)/(P(Data|Design)*P(Design)+P(Data|Notdesign)(1-P(Design))

I take it you accept that?

The formula relies heavily on P(Design). Therefore any uncertainty in P(Design) is reflected in uncertainty about P(Design|Data)

If there is something wrong with this reasoning you might indicate which step.

36. 36

Barry

Liddle:

I would consider the fact that it was a rather special signal the major reason for stacking my priors for an intelligent agent.

Game, set, match to Barry

Well, as long as we use Barry’s CSI definition, sure.

But in that case I declare a Barry-Lizzie draw, and Dembski gets the wooden spoon.

How in the world could you say that a simple “on-off” pattern is in any conceivable sense of the word “complex”? That you would suggest that anyone would say that an on-off pattern is complex is nothing short of astounding. Elizabeth, words have meaning. It is not licit to do linguistic violence to those meanings.

Tell that to Dembski, Barry. I was using his definitions, because I thought that was what we were talking about. It was what I was talking about.

In his 2005 paper, Specification: the Pattern that Signifies Intelligence, he actually gives an example of something very close to your fictional signal: a series of ones and zeros that, when interpreted as binary digits, turn out to form the Champernowne sequence:

1|00|01|10|11|000|001|010|011|100|101|110|111|0000|0001|0010|0011|0100|0101|0110|0111|1000|1001|1010|1011|1100|1101|1110|1111|00.

He says that this sequence has the signature of Design because it has a combination of two properties:

Pattern-simplicity, which he describes as “easy description of pattern”

and

Event-complexity, which he defines “difficulty of reproducing the corresponding event by chance”, i.e. its Shannon entropy, namely 2^100 bits, and which would be possessed by any 100 digit string consisting of equiprobable 1s and 0s.

So Dembski’s “event complexity” would be identical for this sequence and a sequence of alternating ones and zeros, and “pattern simplicity” doesn’t come much more simple than a repeating 2 bit pattern.

So, Barry, please don’t blame the messenger. I was using Dembski’s own terms. If you find that it is not licit to do such linguistic violence to those meanings, then perhaps you should get in touch with the guy you inherited this blog from 🙂

37. 37
Axel says:

‘It is not “intelligence” that designs cars and computers and Mt. Rushmore, it is “human beings”. It is not “intelligence” that designs eyeballs and blood clotting cascades and flagella, it is… nobody has any idea.’ – RDFish

How about a comatose ‘human being’? Do you really think, a comatose ‘human being’ is a better candidate as the likely designer.

There is no design without intelligence. Materialists don’t have the tools to be able to dispense with such sophistries, in favour of common sense.

It doesn’t matter how eminent and celebrated a philosopher, departure from a Christian understanding of the fundamental nature of man – in this case the soul (memory, will and understanding) – can only end in tears before bedtime, with endless sophistries leading up blind alleys. False assumptions are most deleteriously seminal in the thought-processes of materialists.

38. 38
keiths says:

Barry Arrington:

Game, set, match to Barry

Barry,

Your declarations of victory crack me up.

39. 39

Axel:

There is no design without intelligence.

Well, said. And there is also no design without Work.

40. 40
Joe says:

Elizabeth proves that she doesn’t understand how science operates:

It’s the error at the heart of Dembski’s method. And it’s hiding there, in plain site, in the old EF: You can only reject “mechanical necessity” if we have exhausted not only known mechanisms but also unknown mechanisms. Which is clearly impossible. So the filter is doomed from the start.

Geez if that were so then no one could ever reach any scientific inference. Science does NOT require absolute proof and that is what Lizzie is asking for.

Science works with what we know- it cannot and does not wait for what tomorrow may or may not uncover.

And the power is in the hands of teh anti-design people- all they have to do is actually step up and demonstrate that blind and undirected processes can do it and the design inference falls.

41. 41
Axel says:

We see in nature, what we see in artifice, i.e. intelligence and purpose – although to a virtually supernaturally exalted degree?

So, reason demands, surely, that we extrapolate from our own relatively clumsy designs that intelligence and purpose are invariably characteristic properties of a sentient being, who must have employed such artifice in creating the natural order.

How could random chance mimic a brilliantly subtle mind, such as that of say, David Berlinsky – never mind, to the point of being such a creative thaumaturge as would be required to design and create the universe?

It is not just a question of random chance’s ability (or otherwise) to create, is it? It’s the question of how such consummately, such sublimely, beautiful artistry, and the power to construct the universe (not to mention, keep it going), could converge at such an impossible height of perfection. By random chance? Likely? Surely not.

42. 42
RDFish says:

Hi CentralScrutinizer,

So if we were to continously detect the first 100 primes from 2 to 541 from a distant source, you would not at least provisionally put the tag of “intelligently generated” on the phenomenon? How would you tag it?

Like the SETI folks say, we’d tag it “life as we know it” – the only things we know of that might produce such a signal are complex life forms like us (which is precisely the sort of thing ID attempts to explain!).

43. 43
RDFish says:

Hi Sal,

I still have not successfully made my point regarding the equivocations surrounding “intelligence”. Let me try again:

RDF: The “design hypothesis” is not a meaningful explanation of anything at all. It is like trying to explain planetary motion with the “movement hypothesis” without saying what is doing the moving – it tells us precisely nothing.
SAL: A slight but respectful disagreement. Planets move according to an elipse. It would be wrong to say the mechanism is an elipse, the mechanism is gravity that creates the elliptical motion.

Here is what I mean: Imagine being a pre-Newtonian who does not know that the same force that causes apples to fall from trees is what causes planets to move around the Sun. Then you try to explain observed planetary motions by saying “Ah – I believe it is movement that accounts for it!” and proceed to explain that only the “movement hypothesis” can explain what we see in the night sky. Obviously this would be a meaningless hypothesis – it tells us precisely nothing about what is going on, and simply says that something (what don’t know what) explains the movements we observe. My point is that saying the “design hypothesis” accounts for the designs we observe in biology, etc. is just as vacuous – it tells us precisely nothing. It simply says that something (we don’t know what) explains the complex form and function we see in biology.

Designs look to be designed.

I don’t know what this means. It’s like saying “Blue things look to be blue”.

It is formally wrong to say design is the mechanism of design.

I don’t understand this either.

It is correct however to say it looks analogous to something a human could design if the human hand the resources and intelligence.

I do understand what “human design” means – it means a human being devised a plan for something.

Whether intelligence is the root cause,

This doesn’t actually make any sense at all. “Intelligence” is not the “root cause” of anything at all. This is like saying “athleticism” is the root cause of running.

Cheers,
RDFish

44. 44
RDFish says:

Hi Axel,

RDF:
‘It is not “intelligence” that designs cars and computers and Mt. Rushmore, it is “human beings”. It is not “intelligence” that designs eyeballs and blood clotting cascades and flagella, it is… nobody has any idea.’
AXEL: How about a comatose ‘human being’? Do you really think, a comatose ‘human being’ is a better candidate as the likely designer.

Sorry, I don’t understand. Humans are only capable of designing and building things when their brains are in good working order. However, many people (including me) often have the experience of solving technical (design) problems during sleep, because conscious thought is not necessary in order to accomplish many complex tasks.

There is no design without intelligence.

This is like saying there is no running without athleticism. It’s just true by definition.

…in this case the soul (memory, will and understanding)….

Do you not believe that memories are stored in the brain? How would you explain our neuroscientific findings regarding memory formation, retrieval, loss, etc. if memories are not dependent upon neurological processes?

So, reason demands, surely, that we extrapolate from our own relatively clumsy designs that intelligence and purpose are invariably characteristic properties of a sentient being, who must have employed such artifice in creating the natural order.

Sentient beings are, as far as we know from our uniform and repeated experience, chock-full of exactly the sort of complex form and function ID attempts to explain. So it doesn’t seem to be possible for a sentient being to produce the first complex form and function – it’s a bootstrapping problem, you know? There is no CSI without thinking, no thinking without CSI. Quite a mystery!

Cheers,
RDFish

45. 45
keiths says:

Lizzie:

It’s the error at the heart of Dembski’s method. And it’s hiding there, in plain site, in the old EF: You can only reject “mechanical necessity” if we have exhausted not only known mechanisms but also unknown mechanisms. Which is clearly impossible. So the filter is doomed from the start.

In fairness to Dembski, he does attempt to address that issue. From his article:

Granted, this would eliminate all the chance hypotheses in {Hi}i?I, but would it eliminate all chance hypotheses überhaupt? Probabilistic arguments are inherently fallible in the sense that our assumptions about relevant probability distributions might always be in error. Thus, it is always a possibility that {Hi}i?I omits some crucial chance hypothesis that might be operating in the world and account for the event E in question. But are we to take this possibility seriously in the absence of good evidence for the operation of such a chance hypothesis in the production of E? Indeed, the mere possibility that we might have missed some chance hypothesis is hardly reason to think that such a hypothesis was operating. Nor is it reason to be skeptical of a design inference based on specified complexity. Appealing to the unknown to undercut what we do know is never sound epistemological practice. Sure, we may be wrong. But unknown chance hypotheses (and the unknown material mechanisms that supposedly induce them) have no epistemic force in showing that we are wrong. Inquiry can throw things into question only by taking other things to be fixed. The unknown is not a fixed point. It cannot play such a role.

I would argue that in the case of SETI, we are reasoning as follows:

1. We are unaware of any unintelligent mechanism that could produce the pattern in question.

2. We are aware that intelligence can produce the pattern in question.

3. When we receive such a pattern, we are weighing two (partly subjective) probabilities against each other:

3a. The probability that there exists an unknown unintelligent mechanism capable of producing the pattern, versus

3b. The probability that there exists an intelligent being (or beings), besides us, capable of producing the pattern and within sending distance of Earth.

If you regard 3b as more probable than 3a, you’ll infer design. If you regard 3a as more probable than 3b, you’ll infer unintelligent causes. If you’re uncertain about the relative probabilities, or if it’s “too close to call”, you’ll remain agnostic.

46. 46
RDFish says:

Hi CentralScrutinizer,

Because designers are intelligent, that is, intelligence manifests various properties.

Saying “designers are intelligent” is like saying “runners are athletic”. It is simply true by definition.

What would you call something that has the following properties? …
1. “Understanding” of natural forces.
2. “Foresight” of future events based on the arrangements and relationships of various natural forces.
3. “Intent” to alter the course of future events.

I call it “intelligence.”

What would you call something that has the following properties:
1) Running
2) Jumping
3) Throwing
I call it “athleticism”

Now, imagine we look through a telescope at another planet and see traces of movements that look like running, jumping and throwing. How would you explain these traces? What would you learn by saying “the cause of these traces is athleticism“?

Cheers,
RDFish

47. 47
RDFish says:

Hi CentralScrutinizer,

My point above is that it is a mistake to reify intelligence. Intelligence and athleticism are analogous concepts, but we do not tend to reify primarily physical abilities in the way we reify primarily mental abilities. Trying explain how a cheetah runs so fast by saying it is because of “athleticism” is ridiculous. It is equally ridiculous to try and explain how a human being designs a computer by saying it is because of “intelligence”.

Cheers,
RDFish

48. 48

keiths:

Yes indeed. I addressed this in my response to Winston Ewart, who also raised it, and I look forward to Winston’s response.

Here is what I wrote at TSZ (the relevant portion)

**************************************************

Me at TSZ:

But how does Dembski defend this approach? He writes

At this point, critics of specified complexity raise two objections. First, they contend that because we can never know all the chance hypotheses responsible for a given outcome, to infer design because specified complexity eliminates a limited set of chance hypotheses constitutes an argument from ignorance.

Yes, indeed, this critic does. But Dembski counters:

In eliminating chance and inferring design, specified complexity is not party to an argument from ignorance. Rather, it is underwriting an eliminative induction. Eliminative inductions argue for the truth of a proposition by actively refuting its competitors (and not, as in arguments from ignorance, by noting that the proposition has yet to be refuted). Provided that the proposition along with its competitors form a mutually exclusive and exhaustive class, eliminating all the competitors entails that the proposition is true.

OK, but…

But eliminative inductions can be convincing without knocking down every conceivable alternative,a point John Earman has argued effectively. Earman has shown that eliminative inductions are not just widely employed in the sciences but also indispensable to science.

Hold it right there. When Earman makes his plea for eliminative induction, he says:

Even if we can never get down to a single hypothesis, progress occurs if we succeed in eliminating finite or infinite chunks of the possibility space. This presupposes of course that we have some kind of measure, or at least topology, on the space of possibilities.

Earman gives as an example a kind of “hypothesis filter” whereby hypotheses are rejected at each of a series of stages, none of which non-specific “Design” would even pass, as each requires candidate theories to make specific predictions. Not only that, but Earman’s approach is in part a Bayesian one, an approach Dembski specifically rejects for design detection. Just because Fisherian hypothesis testing is essentially eliminative (serial rejection of null hypotheses) does not mean that you can use it for eliminative induction when the competing hypotheses do not form an exhaustive class, and Dembski offers no way of doing so.

In other words, not only does Dembski offer no way of computing the probability distribution under P(T|H) unless H is extremely limited, thereby precluding any Design inference anyway, he also offers no way of computing the topology of the space of non-Design Hypotheses, and thus no way of systematically eliminating them other than one-by-one, never knowing what proportion of viable hypotheses have been eliminated at any stage. In other words, his is, indeed, an argument from ignorance. Earman’s essay simply does not help him.

Suffice it to say, by refusing the eliminative inductions by which specified complexity eliminates chance, one artificially props up chance explanations and (let the irony not be missed) eliminates design explanations whose designing intelligences don’t match up conveniently with a materialistic worldview.

The irony misser here, of course, is Dembski. Nobody qua scientist has “eliminated” a “design explanation”. The problem for Dembski is not that those with a “materialistic worldview” have eliminated Design, but that the only eliminative inductionist approach he cites (Earman’s) would eliminate his Design Hypothesis out of the gate. That’s not because there aren’t perfectly good ways of inferring Design (there are), but because by refusing to make any specific Design-based predictions, Dembski’s hypothesis remains (let the irony not be missed) unfalsifable.

*******************************************************

But in specific response to the passage you cite: I am not necessarily asking that the null embrace every possible unknown non-design hypothesis, but Dembski specifically requires that when computing P(T|H), H should be “the relevant chance hypothesis that takes into account Darwinian and other material mechanisms“. Darwinian mechanisms are not “unknown material mechanisms”.

And I have never seen any attempt to do this. I simply do not know how you would begin. That’s why no-one would ever use Fisherian null hypothesis testing in the way Dembski proposes.

Better to be explicit about your prior for P(T|¬Design) as in the Bayesian formulation and go with your hunch. The CSI calculation merely disguises what Bayes makes explicit. And indeed, boils down, given so little data on which to make an informed guess, to your reasoning steps above.

Your last sentence is as true of CSI as it is of any overtly Bayesian method.

49. 49
CentralScrutinizer says:

RDFish: Like the SETI folks say, we’d tag it “life as we know it” – the only things we know of that might produce such a signal are complex life forms like us (which is precisely the sort of thing ID attempts to explain!).

As I understand them, ID “attempts to explain” the proximate source of life on earth.

(BTW, SETI’s “life as we know it” falls under their Astro-Biology section, not their SETI Observing Projects section. see http://www.seti.org/faq#csc24 )

For all we know, extra-terrestrial intelligent life may be “implemented” by very different means. Who’s to say to the contrary? SETI can say nothing about this in advance and neither can you. All that matters are signs of intelligence.

What SETI observation projects are interested in is “coded information.” See http://www.seti.org/faq#obs9

I think the first 100 primes would qualify, regardless of the nature of the sender, that we’re dealing with an “intelligent source”, don’t you?

50. 50
Axel says:

‘This doesn’t actually make any sense at all. “Intelligence” is not the “root cause” of anything at all. This is like saying “athleticism” is the root cause of running.’

No. That is a false analogy, RDF. They are derived from different parts of speech. It is not given to us mortals ‘to athleticism’, but it is given to us ‘to perceive’.

‘intelligent’:

Interesting derivation… the present participle of the verb, ‘intellegere’ – ‘to perceive’ (literally, to ‘choose between’).

Athletes don’t athlete, but intelligence may perceive, discriminate, judge, select, etc. The confusion arises from their both being definable as, showing a particular functional aptitude, whereas, ‘intelligent’ may also answer to the definition of ‘perceptive’, ‘perceiving’, discriminating, etc.

One example give in the Free Dictionary is:

3. Showing sound judgment and rationality: an intelligent decision; an intelligent solution to the problem.

[Latin intelligns, intelligent-, present participle of intellegere, intelligere, to perceive : inter-, inter- + legere, to choose; see leg- in Indo-European roots.]
– Free Dictionary

Note the infinitive, ‘to choose’, akin to ‘to discriminate, to select, to judge, etc. In computer science, they apparently talk about ‘data-storage, intelligent terminals’.

‘athleticism’, however, is a simple, abstract noun, describing a general physical aptitude or compendium of physical aptitudes, while ‘intelligence’ is far less limited in the scope of its functioning than the human body and its parts, yet is far more specific in its operation and purposiveness; while its root meaning is closely associated with that of an active verb, ‘to perceive’.

The one describes possession of marked, physical aptitudes, gererally; the other, a marked aptitude closely associated with a very particular verb – with an end product, of which it is the root cause!

51. 51
Axel says:

No, RDF. ‘Trying explain how a cheetah runs so fast by saying it is because of “athleticism” is ridiculous.

But to attribute it to the cheetah’s natural athleticism makes perfect sense.

52. 52
wd400 says:

Barry, If we can get back to the OP, I think we’ll find where you are going wrong here.

It is obvious that one’s prior estimate of the probability of the existence of intelligent life forms in other parts of the universe is utterly irrelevant to the design inference. How do I know? By supposing the exact opposite of course. Let’s assume that a person believes there is practically zero chance that other parts of the universe have intelligent life (as we have seen on this site, there is very good reason to believe this). If that person were to receive this signal he would have to revise his conclusion, because the signal is obviously designed

Nah. The heart of Bayesianism that you update your beliefs based on evidence. If you data that very strongly supports an hypothesis (that is the likelihood, P(data|h) is high, and P(data|!h) is low) then you will rightly update your prior and have strong belief in h. But your prior still matters, if you have very good reason to belive there is no life outside of earth, then same data will give you a different posterior probability.

The prior might be of little practical importance when data is absolutely overwhelming supportive of an hypothesis. That might be the case for 100 primes (though you cold possibly imagine other generating processes (they could even, like cicada lifecycles, evolve!).

But it’s certainly not the case for biology. The Bayesian formulation also makes the glaring importance (biology|no-design) stand out. That’s what was missing in the original formulation of CSI, and something many authors here don’t seem to grasp.

53. 53

CentralScrutinizer:

So, Dr Liddle, I’m wondering, when it comes to the particular DNA replication system that we find in nature, and the processes going on within cells, what do you consider the probably of those being the result of natural forces, and what probably of them being the result of a human-like intelligence?

I don’t know CS. I don’t have enough information to be able to make decently informed priors. My prior for natural forces is higher because I know that biochemical forces exist, whereas I have no evidence of any other force that can move molecules around and make them inter-react.

As Barry says, it’s pretty well GIGO at this point, except that with my back to the wall, I’d go for natural forces.

Nobody want to have a go at my puzzle at 29?

It’s quite relevant 🙂

54. 54
CentralScrutinizer says:

RDFish: What would you call something that has the following properties: 1) Running, 2) Jumping, 3) Throwing. I call it “athleticism”

That’s perfectly legit. If we define athleticism as the properties that an entity can have of being able to run, jump and throw, then there’s no reason why “athleticism” can’t be used as a shorthand to explain a certain effect.

Q: Why did Joe win Olympic Gold in the Marathon?

A: Because of his superior athleticism

IOW, because of his superior skills in running, jumping and throwing. It’s a valid explanation. So you didn’t make your point.

55. 55
CentralScrutinizer says:

Dr Liddle: My prior for natural forces is higher because I know that biochemical forces exist

What evidence exists that chance and necessity of biochemical forces led to the self-replicator we have?

whereas I have no evidence of any other force that can move molecules around and make them inter-react.

Humans can do that, with understanding, foresight, and intent. Moreover, we have a pretty decent understanding of Earth’s self-replicator, and there is no obvious reason why we shouldn’t be able to master the whole shebang some day.

…that with my back to the wall, I’d go for natural forces.

Why?

56. 56

CS: Humans can move things. They can use tweezers, and Gene guns and test-tubes.

They can, in other words, exert force on matter, so that it travels through a distance. Molecular forces can do this two. But I have no evidence that anything can move anything without exerting a force on it.

So, absent evidence for some additional force in the universe we have not yet discovered, I’d back the ones we know about.

Of course the those forces themselves could have been designed to do the job. But in that case all we will be able to detect from withinthe universe are tne natural forces that the Designer created.

57. 57
keiths says:

Lizzie,

But in specific response to the passage you cite: I am not necessarily asking that the null embrace every possible unknown non-design hypothesis, but Dembski specifically requires that when computing P(T|H), H should be “the relevant chance hypothesis that takes into account Darwinian and other material mechanisms“. Darwinian mechanisms are not “unknown material mechanisms”.

I certainly agree that Darwinian mechanisms aren’t unknown. However, you seemed to be making a stronger claim earlier:

You can only reject “mechanical necessity” if we have exhausted not only known mechanisms but also unknown mechanisms. Which is clearly impossible. So the filter is doomed from the start.

My point is that you can salvage the filter if you can make a decent argument for why this probability is low:

3a. The probability that there exists an unknown unintelligent mechanism capable of producing the pattern…

It isn’t ironclad; you can get false positives if your estimate is wrong. But at least the filter isn’t doomed.

And of course I agree that the bigger problem with P(T|H) is that nobody can compute it for the known mechanisms (including Darwinian evolution), much less the unknown ones!

58. 58
wd400 says:

re: Liz a at 29.

I’d call red. Prior to the 1st draw we know nothing about the relative frequency of red or blue. The first draw gives us some information, no matter how meagre about what the frequency of red balls is likely to be so we stick and hope.

(working on the back on an envelope, presuming uniform distribution of the frequency of red balls across trials, each with 100 balls per bag, you survive about 65% of the time with this strategy )

59. 59
CentralScrutinizer says:

Nobody want to have a go at my puzzle at 29?

I’d hit the guard over the head with the bag and grab his gun.

60. 60
CentralScrutinizer says:

Dr Liddle: [Humans] can, in other words, exert force on matter, so that it travels through a distance. Molecular forces can do this two. But I have no evidence that anything can move anything without exerting a force on it.

Humans do more than merely exert force on matter, they can do so with foresight. What biochemical force does that?

61. 61
CentralScrutinizer says:

Dr Liddle,

Let me ask you, can physical forces create a computer? Or does it require a human, with their understanding, foresight, and intent?

62. 62
CentralScrutinizer says:

BTW, I’m going to the grocery store now. Anyone need anything?

63. 63

I know they do more than merely exert force on matter, CS.

But they can’t build anything at all unless they exert force on matter.

Without the force-on-matter part, they can’t do anything at all.

64. 64

wd400 @58.

Me too, wd400.

Anyone go with blue?

Anyone else going for red?

It’s the reasoning I’m interested in.

65. 65
CentralScrutinizer says:

Dr Liddle: Without the force-on-matter part, they can’t do anything at all.

Agreed. And I would assume whatever/whoever is responsible for creating Earth’s bio-replicator would have that ability.

66. 66
Axel says:

‘However, many people (including me) often have the experience of solving technical (design) problems during sleep, because conscious thought is not necessary in order to accomplish many complex tasks.’

Oh no, the mind operates far more abstrusely than you give it credit for. Are you not conscious of your dreams? I am of mine; though the memory of them usually falls away quickly.

But even if I weren’t, even if I were not conscious of them, that is when the mind operates optimally, that is, at a much deeper and smoother level than during our waking hours, as you just pointed out.

This, I believe, has implications for our understanding of the nature of our intelligence. I think perhaps more people than ever are familiar today with the idea that, as Aldous Huxley suggested, the brain is a reducing valve for survival in time. Otherwise, we’d be so close to the Beatific Vision, we’d probably be ‘transported’, probably and starve. There are reasons to believe that unlettered people score highly in this kind of truly heavy-weight IQ test. Christ chose the poor to be rich in faith.

Consequently, as I mentioned in another post, I actually think intelligence is one of the great levelers, although to appearances, it is the very antithesis. The key function of the soul, the will, being the operative one, so that we base our world-view and values – the synthesis or lack thereof being too abstruse to be approached by other than the heart – on the assumptions we choose.

Christ’s precepts and his ultimate judgment of us, as related in the Gospels, are all based the disposition of our hearts not our heads – or rather, the former, before the latter. Belief and knowledge are a continuum, as are their secular counterparts. We know what we wish to know; which is what I believe philosophers call ‘voluntarism’. You may be more familiar with it. I only did Philosophy I, though I think I passed it.

In the end, we are all wishful-thinkers. It just happens that God has made the world to conform with the hopes and wishes he instills in us, to some extent innately, and to some extent via baptism. Wishful thinking has got a worse name than it deserves. Why should the truth be cold, hard, undesirable, not to be wished for, not to be hoped for?

Beauty is very, very very closely associated with love. I mean primarily, but not exclusively, moral beauty, but not personal physical attractiveness, which, of course, is not a sure indicator of character. Although virtue, character, sometimes does shine out of a face.

67. 67

keiths:

Yes indeed, but not if you are using Fisherian testing.

I absolutely agree, but my beef was with Dembski’s Fisher model. For a Fisher model, you really do have to specify your null carefully, and when you reject it, be clear that that null is the null you rejected, not any other null.

Or at any rate, that’s how I see it. In Fisherian hypothesis testing you only reject the null. What you conclude is more likely, given that rejection, is something else.

68. 68

CS

Agreed. And I would assume whatever/whoever is responsible for creating Earth’s bio-replicator would have that ability.

Right. But I have considerable evidence for the ability of molecules to assemble and disassemble via known fundamental forces, but none for any additional force.

Hence my elevation of the natural prior.

But do note that I do not, thereby, reject the hypothesis that those fundamental forces were themselves designed. But in that case they are still natural forces.

69. 69
CentralScrutinizer says:

Dr Liddle: I have considerable evidence for the ability of molecules to assemble and disassemble via known fundamental forces, but none for any additional force.

But you don’t have considerable evidence for them to assemble into Earth’s self-replicator, except from prior instances of the self-replicator.

Hence my elevation of the natural prior.

I don’t see why. Especially since the self-replicator uses coded information. What else has the known ability to generate that except intelligent entities?

70. 70
CentralScrutinizer says:

… which is one of the things SETI is looking for:

http://www.seti.org/faq#obs9

71. 71
CentralScrutinizer says:

From SETI website,

“Other tell-tale characteristics include a signal that is completely polarized or the existence of coded information on the signal.”

http://www.seti.org/faq#obs9

72. 72

Obviously your priors are different CS. I told you I couldn’t give you well-founded priors, just what I’d go for if backed against the wall. You’d go for the other thing. Fine.

The problem is that neither has enough information to be anywhere near decisive either way which is my whole point.

Try the ball game.

73. 73
Axel says:

‘Do you not believe that memories are stored in the brain? How would you explain our neuroscientific findings regarding memory formation, retrieval, loss, etc. if memories are not dependent upon neurological processes?’

What scientific findings, RD? There is an article online by Pim van Lommel positing that the mind is like a receiver, rather than generated by the brain. I can’t find the link now but will try again tomorrow. It was extrapolated from an out-of-body experience, monitored in a very sophisticated fashion by medical people involved. Must go.

We seem to know very littele else about the mind

74. 74
CentralScrutinizer says:

Dr Liddle: I told you I couldn’t give you well-founded priors, just what I’d go for if backed against the wall. You’d go for the other thing. Fine. The problem is that neither has enough information to be anywhere near decisive either way which is my whole point.

So what you seem to be telling me is that the fact that the self-replicator has coded information, and that the only known source for such are entities with intelligence (understanding of nature, foresight, intent) does not matter to you.

OK, fair enough.

75. 75
CentralScrutinizer says:

Dr Liddle: Try the ball game.

You did not specify color ratio of balls in the bag, or their distribution characteristics. So IMO, there is not enough information in your puzzle to make any more than a random-ass guess.

76. 76

No, I know I didn’t. That’s the point of the puzzle – you have very little information.

You have been told that the balls could be red or blue. You have no idea of the proportions. You don’t even know if the person is telling the truth or not.

And the first ball you draw turns out to be red.

That’s it. Your life depends on it. What do you decide, and why?

77. 77

CS

So what you seem to be telling me is that the fact that the self-replicator has coded information, and that the only known source for such are entities with intelligence (understanding of nature, foresight, intent) does not matter to you.

OK, fair enough.

Oh, it matters. But the lack of a known moving force weighs a little more heavily with me. I know that intricate and beautiful objects can result from natural forces without a mind. I do not know that intricate and beautiful objects can result from a mind with no natural forces.

78. 78
CentralScrutinizer says:

Dr Liddle: I know that intricate and beautiful objects can result from natural forces without a mind. I do not know that intricate and beautiful objects can result from a mind with no natural forces.

Well, when you find natural forces sans intelligence generating coded information, you let me know, OK? 🙂

79. 79
CentralScrutinizer says:

But the lack of a known moving force weighs a little more heavily with me.

BTW, there is a lack of a known moving force when it comes to the existence of the “natural forces” of space-time. So there’s at least one precedent for the idea that such things are possible. (Not that I argue that the proximate cause of Earth’s OOL necessarily be outside of space-time, because I certainly do not. Although, I think there are good arguments why the ultimate source may be.)

80. 80
MrMosis says:

Dr. Liddle @ 77:

I do not know that intricate and beautiful objects can result from a mind with no natural forces.

What is a mind with no natural forces, exactly?

81. 81
RDFish says:

Hi CentralScrutinizer,

For all we know, extra-terrestrial intelligent life may be “implemented” by very different means. Who’s to say to the contrary? SETI can say nothing about this in advance and neither can you. All that matters are signs of intelligence.

Say we receive a signal that doesn’t appear to be caused by known physical mechanisms outside of complex living organisms. We might then assume that complex living organisms of some sort sent the signals, and that would be our best explanation. But to say “intelligence” sent the signal doesn’t actually mean anything beyond “the signal was sent by something that was capable of sending the signal”, which obviously tells us nothing at all.

RDFish: What would you call something that has the following properties: 1) Running, 2) Jumping, 3) Throwing. I call it “athleticism”
CS: That’s perfectly legit. If we define athleticism as the properties that an entity can have of being able to run, jump and throw, then there’s no reason why “athleticism” can’t be used as a shorthand to explain a certain effect.
Q: Why did Joe win Olympic Gold in the Marathon?
A: Because of his superior athleticism

I suppose this explanation does tell us something by eliminating other possibilities. For example, other explanations could be “because he bribed the judges” or “for political reasons” or “because he used steroids”.

However, if you ask “How does Joe manage to run so well?”, then the explanation of “Because of his superior athleticism” becomes vacuous – it is no explanation at all, since you are merely labeling the observation.

If somebody can run well, but not throw or jump at all, is that person still athletic? Or somebody who can’t run, throw, or jump, but plays championship golf? Or ping-pong? Or lifts weights? Or is an expert marksman? Or swims? Which of these are athletic?

There is no right answer, of course. Something “athletic” may have one or more of these abilities, or none of them at all (maybe it can fly, or dig, or rematerialize on a different planet…)

Not only that, but it may achieve these things by radically different means. Maybe it uses bones and muscles, or hydraulics, or pneumatics, or solenoids, or…

And this is why saying something is “athletic” tells us precisely nothing, and that is why “athleticism” cannot serve as an explanation of anything.

Likewise, “intelligence”.

Cheers,
RDFish

82. 82

Mr Mosis:

What is a mind with no natural forces, exactly?

I have no idea, which is why my priors are low for such a mind having moved molecules together to make life.

83. 83

SC:

Well, when you find natural forces sans intelligence generating coded information, you let me know, OK?

Well, natural forces generate coded information every time a cell in your body divides.

I guess you mean wrote the coding system?

84. 84

CS

BTW, there is a lack of a known moving force when it comes to the existence of the “natural forces” of space-time. So there’s at least one precedent for the idea that such things are possible. (Not that I argue that the proximate cause of Earth’s OOL necessarily be outside of space-time, because I certainly do not. Although, I think there are good arguments why the ultimate source may be.)

See my last paragraph in my post to you at 68.

85. 85
Joe says:

Elizabeth:

Well, natural forces generate coded information every time a cell in your body divides.

How are you defining “natural”, Lizzie? Metghinks you are equivocating, again.

86. 86
Joe says:

So more blah, blah, blah from the anti-IDists and still nothing that demonstrates taht they even understand science…

87. 87
Joe says:

RD:

But to say “intelligence” sent the signal doesn’t actually mean anything beyond “the signal was sent by something that was capable of sending the signal”, which obviously tells us nothing at all.

Look RD, we understand that you are scientifically illiterate, but really? Saying some intelligence did something changes the investigation.

Do you think that archaeologists or geologists are better equipped to understand Stonehenge?

88. 88
groovamos says:

M.F. : One line of thought might be a natural mechanism that generates a stream of increasing numbers but is filtered by another mechanism that eliminates any number that is a multiple of previous numbers in the sequence.

This ain’t even the half of it. Nobody has mentioned here that SETI is searching for narrowband emissions in the lower end of the EM spectrum. No physical process is known to generate such and this in itself ain’t even 10% of it. Suppose such an emission is found. Very quickly, before it stops, the centroid of the spectrum of such would be determined. Then the signal would be translated to “baseband” by four-quadrant multiplication by a sinusoid of the centroid frequency, OR nonlinear processing after algebraic addition of such sinusoid. Then the alias spectral band(s) of the translated signal must be removed by a suitably selected lowpass filter. Then the resulting signal would be digitized at a suitable sample rate to avoid spectral alias distortion. These last 2 steps would be done after careful analysis of the baseband signal bandwidth. Then of course this digitized signal would be stored.

Then for the hard part UNLESS the signal were intended for an unknown observer. Assuming the spectrum centroid is accurately determined as a carrier (a big question could be offered here) then there would have to be a determination of the modulation employed by the presumed intelligence. If the signal were to be intended for an unknown receiver, like us, it could be a simple modulation like frequency shift keying. Here I will make a statement: there is NO POSSIBILITY of a natural, non-intelligent process that will generate a signal like this, a varying signal with built-in character synchronization, and employing two frequency points to code on the EM spectrum. And this is even before we get talking prime number encoding OR more complicated modulation schemes. Put it this way: ZERO, Zilch, Nada, Null probability. All this hedging of bets on this thread with overly intellectualized reasoning otherwise amounts to nothing.

While at this, I should challenge: look at all the intelligence required to go through the analysis steps I mention above. Then the determination of an intelligent emitter at the end of the search, knowing for sure that an information system has been discovered. Information makes no sense (haha, ironic that last word) outside the context of mind.

89. 89
Mark Frank says:

there is NO POSSIBILITY of a natural, non-intelligent process that will generate a signal like this, a varying signal with built-in character synchronization, and employing two frequency points to code on the EM spectrum

History is full of statements by eminent men saying that such and such is impossible who turn out to be wrong.

90. 90
groovamos says:

OK if there is a non-zero probability of such, then what is it? If I state that there is a probability of zero that stable liquid water droplets are orbiting the earth at 1000 Km altitude, are you going to come back with something about eminent men and their sayings? This is sound argumentation?

91. 91
CentralScrutinizer says:

RDFish: Say we receive a signal that doesn’t appear to be caused by known physical mechanisms outside of complex living organisms. We might then assume that complex living organisms of some sort sent the signals, and that would be our best explanation.

SETI sky scanning is looking for evidence of intelligence, namely, evidence of beings that possess understanding of nature, foresight, and intent.

But to say “intelligence” sent the signal doesn’t actually mean anything beyond “the signal was sent by something that was capable of sending the signal”, which obviously tells us nothing at all.

No, it means that we found something that appears to have been sent by intelligence, i.e, sent something that possesses understanding of nature, foresight, and intent.

Moreover, SETI is not trying to explain intelligence. It is trying to find evidence of it based on what is thought to be plausible manifestations of it, such as “completely polarized emissions” and “coded information.”

I suppose this explanation does tell us something by eliminating other possibilities. For example, other explanations could be “because he bribed the judges” or “for political reasons” or “because he used steroids”.

Thank you for yoru partial concession. However, it also can tell us that he possessed superior athleticism than his competitors. This is not a vacuous explanation for the simple fact that I defined athleticism, namely, as possessing the ability to run, jump and throw. Are you saying the ability to run, jump and throw are vacuous ideas? Or that one athlete cannot possess those attributes in greater and lesser degrees?

However, if you ask “How does Joe manage to run so well?”, then the explanation of “Because of his superior athleticism” becomes vacuous – it is no explanation at all, since you are merely labeling the observation.

This is correct, but irrelevant. As stated above, SETI doesn’t attempt to explain intelligence- it is looking for signs of it. Likewise, ID proponents are not attempting to explain intelligence. They are looking for signs of it. Ponder the difference.

If somebody can run well, but not throw or jump at all, is that person still athletic?

In that case they could be said to exhibit some of the attributes of athleticism, but not all, as I have defined athleticism. Likewise if a retarded person can understand some things, but has no concept of “the future”, i.e, no foresight, such a person could be said to possess some of the attributes of intelligence, but not all, according to my definition of intelligence.

Or somebody who can’t run, throw, or jump, but plays championship golf? Or ping-pong? Or lifts weights? Or is an expert marksman? Or swims? Which of these are athletic?

If depends on how you define athleticism. Under my definition, golf isn’t specifically mentioned. Are you saying that nobody has defined intelligence? If that were true, then ID would be a vacuous term. I think “intelligence” is pretty well defined, and I have offered a definition of my own to you several times. What part of it are you having trouble with?

And this is why saying something is “athletic” tells us precisely nothing, and that is why “athleticism” cannot serve as an explanation of anything.

It only tells you nothing if, A) there is no definition of “athletic, or B) you are trying to explain what it is from the definition. SETI isn’t trying to explain intelligence and neither is ID. They are looking for signs of it. Both groups have a pretty decent idea how to define the term.

92. 92
CentralScrutinizer says:

Dr Liddle: Well, natural forces generate coded information every time a cell in your body divides. I guess you mean wrote the coding system?

See #69: “But you don’t have considerable evidence for [natural forces] to assemble into Earth’s self-replicator, except from prior instances of the self-replicator.”

93. 93
Phinehas says:

Blue.

If the factory makes red and blue balls, they would typically be packaged as, all red, all blue, or some mix. If I pull out a red ball, I know I’m not dealing with a bag of all blue balls, so I can remove that possibility from the list. If this is an all red bag and I choose red, then I am guaranteed to live.

If the bag is some mix of red and blue balls, there are three possibilities. 1. There are more red balls than blue balls. 2. There are more blue balls than red balls. 3. There are the same number of blue and red balls.

The fact that I’ve pulled out a red ball means that (1) is the most likely scenario, in which case my odds of getting another red are no less than 50%. I don’t know enough probability math to figure out exactly how much more likely (1) is given the fact that I’ve drawn a red ball, but I don’t really need to know, since, intuitively, it is above 50% and I can still compare it to the other possibilities.

As (2) approaches the worst case scenario for selecting red (in other words, I got the one red ball and all blue are left), the odds of that actually being the distribution I am facing, having drawn a red ball randomly, start at less than 50% and go down to approach zero. So, any potential distribution in which there is increased benefit for selecting a blue ball is balanced against the fact it is less likely to be the actual scenario I am facing. I’m not sure how perfectly these probabilities balance out, but intuitively, the combination would not rise above 50%.

If I am facing (3), and have picked a red ball, then I’ve got slightly better odds of picking a blue ball with my next choice. But what are the odds that I am facing an exactly even distribution?

Well, here’s where the human element might come it. An exactly even distribution sounds like a logical choice for someone setting up the scenario described, so I can definitely see a human tendency to pick it as a specification. On the other hand, logic might not be the highest priority of someone who is basing my life or death on probabilities, so clearly I cannot choose the wine in front of me. I’m going to assume someone who’d set up this kind of scenario is a sadist (probably from Australia, which, as everyone knows, is entirely peopled with criminals) who is most likely interested in having some fun at my expense, so there’s a pretty good chance that they’ve built up an immunity to Iocane and this whole thing is rigged.

So, I’m picking blue, since it is my favorite color.

94. 94
CentralScrutinizer says:

groovamos: This ain’t even the half of it. Nobody has mentioned here that SETI is searching for narrowband emissions in the lower end of the EM spectrum.

True, although I did point to a paragraph on their website at #70 and #71 that sort of mentions it. 🙂

http://www.seti.org/faq#obs9

95. 95
RDFish says:

Hi CentralScrutinizer,

No, it means that we found something that appears to have been sent by intelligence, i.e, sent something that possesses understanding of nature, foresight, and intent.

How could we possibly establish that the transmission of, say, prime numbers required that the sender had these attributes? We can’t of course – we would have absolutely no idea what any attributes of the sender might be. Only to the extent that we established a likelihood that the sender is a complex organism similar to ourselves could we tentatively infer that they experienced conscious thought like we do.

However, if you ask “How does Joe manage to run so well?”, then the explanation of “Because of his superior athleticism” becomes vacuous – it is no explanation at all, since you are merely labeling the observation.

CS: This is correct, but irrelevant.

No, this is both correct and critically relevant. Once again:
1) Observation: Joe runs the marathon quickly. Vacuous explanation: Joe is athletic.
2) Observation: Joe designs a computer. Vacuous explanation: Joe is intelligent.
3) Observation: Biology exhibits CSI. Vacuous explanation: The cause was intelligent.

Likewise, ID proponents are not attempting to explain intelligence. They are looking for signs of it. Ponder the difference.

You need to ponder the problem. You can’t find “intelligence”, because it isn’t a thing. It is a very loosely defined property of living things, like “athleticism” (or “living”!).

In some sense, every living thing is intelligent, and every intelligent thing is living. You imagine that perhaps something could have the same abilities that living things exhibit without the benefit of being a complex organism; well, you can imagine all sorts of things, but as far as anyone knows, this is impossible.

RDF: Or somebody who can’t run, throw, or jump, but plays championship golf? Or ping-pong? Or lifts weights? Or is an expert marksman? Or swims? Which of these are athletic?
CS: If depends on how you define athleticism.

Thank you for this concession. You should now begin to appreciate the problem.

Under my definition, golf isn’t specifically mentioned. Are you saying that nobody has defined intelligence? If that were true, then ID would be a vacuous term.

The problem is not that there are no definitions of “intelligence”; rather, the problem is that there are arbitrarily many different definitions of course. What makes ID a vacuous term is that ID fails to provide one particular definition. And if ID is to be empirically supportable, the definition must be scientific – that is, it needs to describe something that we can test for.

I think “intelligence” is pretty well defined, and I have offered a definition of my own to you several times. What part of it are you having trouble with?

You say that intelligent things “possess understanding of nature, foresight, and intent.” Ok, so I have something here in my laboratory (I call it a “booja”). Can you please tell me how to determine if this booja is intelligent?

Conversely, I have something else here in my lab that is called a jooba, and I’m certain the jooba is intelligent. Can you please tell me one single thing that I can observe about this jooba?

RDF: And this is why saying something is “athletic” tells us precisely nothing, and that is why “athleticism” cannot serve as an explanation of anything.
It only tells you nothing if, A) there is no definition of “athletic, or B) you are trying to explain what it is from the definition.

Of course there are plenty of definitions of “athletic” – just look in some dictionaries. That doesn’t help.

Cheers,
RDFish

96. 96
Joe says:

I, for one, am very glad that RDFish is not a paid investigator. Nothing would ever get accomplished with him around.

97. 97
Mark Frank says:

Groovamos

OK if there is a non-zero probability of such, then what is it?

We know it is extremely low and in most contexts we would rightly dismiss it. But if you are confronted with an extraordinary results you need to look for extraordinary explanations. Now, as discussed many times, human-like aliens are not only a better explanation of the prime number sequence then some filtering process but for most of us they are more likely to exist. However, we were considering the hypothetical assumption that we dismissed the possibility of aliens.

If I state that there is a probability of zero that stable liquid water droplets are orbiting the earth at 1000 Km altitude, are you going to come back with something about eminent men and their sayings? This is sound argumentation?

I don’t know what kind of phenomenon would be best explained by stable liquid water droplets orbiting the earth at 1000 Km altitude – but if there were some extraordinary outcome with no other conceivable explanation then it would be sound to consider if there were some way such drops were possible. Nature is full of bizarre events. On the other hand in the absence of such a phenomenon it is sound argument to dismiss the water drops as a possibility.

98. 98

Thanks, Phinehas?

Anyone else want to have a go? There is, obviously, no correct answer, it’s the reasoning I’m interested in 🙂

99. 99

Ignore the question mark, Phinehas! Finger slipped on new keyboard!

100. 100
Joe says:

Dembski’s EF, and CSI concepts are based on Fisherian null hypothesis testing.

No, the EF works fine without Fisher.

As for your puzzle, I wouldn’t choose anything. I would tell them to shoot me and that is all the entertainment they will get.

101. 101
Joe says:

Elizabeth:

I know that intricate and beautiful objects can result from natural forces without a mind.

But from where did those “natural forces” come from?

102. 102
Phinehas says:

C’mon. No props for the Princess Bride references?

*sigh*

I suppose I’ll just have to go start a land war in Asia.

103. 103
Phinehas says:

Maybe a more obvious reference?

Blue…no, yelloooooooooooooooooow.

104. 104

Sorry, now! Never seen it 😮

105. 105

The smileys on this board are weird. That was supposed to the :embarrassed: smiley

106. 106
Phinehas says:

Liz! @99:

No worries! I’m having the same issue with my exclamation mark!

–Phin!

107. 107
Phinehas says:

Being an evomat is almost understandable. Not having seen the Princess Bride is a different matter altogether!

Well, maybe I can rectify one if not the other. Here is a taste.

108. 108
groovamos says:

M. F. Nature is full of bizarre events.

I can state absolutely that there is zero possibility that there can be a mindless, natural source for a Lexus ES330. It would be bizarre that you might want to maintain the possibility of a nonzero probability for such an occurrence. I use the same logic going back to my first post. You have no way of convincing anyone that there is a nonzero probability for any of these occurrences and frankly to maintain so is bizarre in itself.

109. 109
Mung says:

Elizabeth Liddle:

That’s it. Your life depends on it. What do you decide, and why?

My life depends upon God and God alone, and red balls or blue balls are not going to change that.

110. 110
111. 111
Mung says:

“Of course design is detectable!”

– Elizabeth Liddle

112. 112
CentralScrutinizer says:

RDFish: How could we possibly establish that the transmission of, say, prime numbers required that the sender had these attributes? We can’t of course – we would have absolutely no idea what any attributes of the sender might be.

We do know that humans possess certain properties that allow us to create such coded signals. And that is what SETI is looking for. That is what ID is looking for.

RD: However, if you ask “How does Joe manage to run so well?”, then the explanation of “Because of his superior athleticism” becomes vacuous – it is no explanation at all, since you are merely labeling the observation.

CS: This is correct, but irrelevant.

RD: No, this is both correct and critically relevant. Once again: 1) Observation: Joe runs the marathon quickly. Vacuous explanation: Joe is athletic.

You are shifting ground. “Running quickly” is indeed part of my definition of athleticism and therefore in your example would be circular and therefore a vacuous explanation. My example was, Q: why did Joe win the competition? A: superior athleticism. Not vacuous given my definition of athleticism.

2) Observation: Joe designs a computer. Vacuous explanation: Joe is intelligent.

Not vacuous. It would be vacuous if “designing computers” was part of the definition of “intelligence.” It’s not in my definition. Observation: Joe designs a computer. Non-vacuous explanation: because he possesses understanding of nature, foresight and intent.

3) Observation: Biology exhibits CSI. Vacuous explanation: The cause was intelligent.

Sidebar: At this stage, I’m not convinced that “CSI” is a knock-down argument in favor of intelligence causation. Interesting, and perhaps promising, but not knock-down.

I think codes are a very strong indicator of intelligence as I define it. So I would reword your example to say: Observation: coded information and coding/decoding in the DNA self-replicator that have no evidentiary support for plausible pre-systemic chemical affinities between the encoded data medium and the decoding mechanism; Explanation: best inference to cause- intelligent agency, i.e, some entity with an understanding of nature, foresight, and intent.

Not vacuous.

(Whether or not you agree with my take on intelligence being the best inference for the aforementioned bio-mechanism is beside the point. The point is, “intelligent cause” is not a vacuous explanation. It may be wrong, but it’s not vacuous.)

Me: Likewise, ID proponents are not attempting to explain intelligence. They are looking for signs of it. Ponder the difference.

RDFish: You need to ponder the problem. You can’t find “intelligence”, because it isn’t a thing. It is a very loosely defined property of living things, like “athleticism” (or “living”!).

Intelligence as we know it and as I define it can exhibit certain effects, which is what SETI is looking for. And what ID is looking for. Not vacuous.

In some sense, every living thing is intelligent, and every intelligent thing is living. You imagine that perhaps something could have the same abilities that living things exhibit without the benefit of being a complex organism; well, you can imagine all sorts of things, but as far as anyone knows, this is impossible.

SETI is looking for the effects of human-like intelligence (something that understands nature, has foresight, and intent), and so is ID. Not vacuous.

RDF: Or somebody who can’t run, throw, or jump, but plays championship golf? Or ping-pong? Or lifts weights? Or is an expert marksman? Or swims? Which of these are athletic?

CS: If depends on how you define athleticism.

Thank you for this concession. You should now begin to appreciate the problem.

I didn’t concede anything. Limiting the scope of a putative source is not a concession. I can limit the scope of athleticism to simply “running, jumping and throwing” and use that term in a non-vacuous explanatory way, as I have already demonstrated: Q: why did Jim win the marathon (an effect), A: superior athleticism (a non-vacuous causal explanation.)

Say I’m a talent scout looking for people who are more likely to win, I can use the term this way: Q: what are you looking for? A: people who win marathons. Why? Because, as I understand it, winning marathons is a effect of people with superior atheticism.

SETI is looking for signs of intelligence. Because, as we know it, intelligence has the properties that can create recognizable effects such as coded information. ID is the same. Non-vacuous.

The problem is not that there are no definitions of “intelligence”; rather, the problem is that there are arbitrarily many different definitions of course. What makes ID a vacuous term is that ID fails to provide one particular definition. And if ID is to be empirically supportable, the definition must be scientific – that is, it needs to describe something that we can test for.

Maybe somebody should nail it down in a robust and rigorous way. But then I ask: what is SETI looking for? Where is their robust and rigorous definition? I think you’re picking at nits. SETI is looking for signs of human-like intelligence, something that can create coded information, like us, because its understanding of nature, foresight, and intent. ID is looking for the same.

You say that intelligent things “possess understanding of nature, foresight, and intent.” Ok, so I have something here in my laboratory (I call it a “booja”). Can you please tell me how to determine if this booja is intelligent? Conversely, I have something else here in my lab that is called a jooba, and I’m certain the jooba is intelligent. Can you please tell me one single thing that I can observe about this jooba?

I would have to examine their effects. What are they? If they emitted a radio signal that encoded the first hundred primes and nothing else I would conclude that either that it was intelligent, or that its creator was intelligent. If they rearranged matter in ways that are extremely unlikely given the chemical affinities, then I might be tempted to think the same.

113. 113
CentralScrutinizer says:

RDFish,

… And if they rearranged matter in ways that are extremely unlikely given the chemical affinities, and produced a system of coded information, storage, transmission, replication, then I would hard pressed to not think the same.

114. 114
RDFish says:

Hi CentralScrutinizer,

RDF: How could we possibly establish that the transmission of, say, prime numbers required that the sender had these attributes? We can’t of course – we would have absolutely no idea what any attributes of the sender might be.
CS: We do know that humans possess certain properties that allow us to create such coded signals. And that is what SETI is looking for. That is what ID is looking for.

Human beings are complex living things, with sense organs, nervous systems, muscles, and other complex physical systems that enable us to perceive, store and retrieve memories, generate plans, and design and build other complex systems. SETI does indeed assume that anything out there that might send us a signal is just this sort of thing – “life as we know it”.

But I think you’d agree that is what ID is “looking for”: If ID posits that extra-terrestrial complex life forms are the reason for complex biological systems on Earth, then we might as well assume we are simply descendents of those complex life forms, and not the products of its bioengineering!

RDF: 2) Observation: Joe designs a computer. Vacuous explanation: Joe is intelligent.
CS: Not vacuous. It would be vacuous if “designing computers” was part of the definition of “intelligence.” It’s not in my definition. Observation: Joe designs a computer. Non-vacuous explanation: because he possesses understanding of nature, foresight and intent.

The explanation is vacuous without some specific definition. You have provided a specific definition (although it is not one that has been established as the technical definition associated with ID theory).

So the problem with your definition is not that it is vacuous; rather, it is that we have no evidence that this definition applies to the cause of living things.

Think about it: How do you know that Joe the computer-maker possesses a conscious understanding of nature, and conscious foresight and intent? Answer: Because we know these things about human beings of course, and because Joe is a human being. If we knew nothing of what caused the computer, you could not make this inference.

RDF: 3) Observation: Biology exhibits CSI. Vacuous explanation: The cause was intelligent.
CS: Sidebar: At this stage, I’m not convinced that “CSI” is a knock-down argument in favor of intelligence causation. Interesting, and perhaps promising, but not knock-down.

I use “CSI” as shorthand for the “complex form and function” that convinces us that living systems need some explanation. I’m not interested in debating just what features of biology represent “CSI” or “irreducible complexity” or whatever – I’m quite convinced that the complex machinery we observe requires an explanation that we have not yet come up with.

I think codes are a very strong indicator of intelligence as I define it.

That is a completely unsupported opinion. You have precisely one data point: human beings generate codes, and human beings have conscious intent. You cannot therefore conclude that anything that generates codes must likewise have conscious intent. The only thing we know of that has conscious intent requires a well-functioning brain in order to experience that – and brains are ridiculously complex organs, with more “CSI” that anything else we know of.

So if you are going to try and reason that like effects have like causes, you end up with the hypothesis that whatever created DNA mechanisms in biological systems was something with a very complex brain. But again, ID doesn’t want to conclude that some complex life form created life on Earth. Rather, ID wants to reject that part of the inference. ID argues that even though the cause of life wasn’t itself a complex organism, it was still similar to human beings in that it experienced conscious intent, etc. But of course there is no evidence that anything like that has ever existed, and no idea how it could.

Whether or not you agree with my take on intelligence being the best inference for the aforementioned bio-mechanism is beside the point. The point is, “intelligent cause” is not a vacuous explanation. It may be wrong, but it’s not vacuous.

I agree that your definition is not vacuous: Your definition of intelligence is “having understanding of nature, foresight and intent”. Even though we do not understand how “foresight” or “understanding” works in human beings, I understand each of these attributes subjectively, because I consciously experience what we call understanding, planning, and intentions. You could also try to verify these attributes by asking me questions, or test that I can generate plans by giving me novel problems to solve. You cannot, however, ascertain if the cause of living things likewise experiences these things.

SETI is looking for signs of intelligence. Because, as we know it, intelligence has the properties that can create recognizable effects such as coded information. ID is the same. Non-vacuous.

As we know it, “intelligence” is a property of complex living things. SETI is searching for this. If that is what ID is “searching for”, then ID has another set of problems, as I’ve explained above.

RDF: The problem is not that there are no definitions of “intelligence”; rather, the problem is that there are arbitrarily many different definitions of course. What makes ID a vacuous term is that ID fails to provide one particular definition. And if ID is to be empirically supportable, the definition must be scientific – that is, it needs to describe something that we can test for.
CS: Maybe somebody should nail it down in a robust and rigorous way.

Only if ID is supposed to be taken seriously. You have provide one definition that is non-vacuous, but still cannot be operationalized in the context of ID. What you’ll find is that there is no definition that will meet ID’s objectives of being non-vacuous and objectively verifiable in the context of ID.

But then I ask: what is SETI looking for?

Signals that human beings (or similar life forms) might send out into space. If we find some, we’ll conclude that there is life elsewhere in the galaxy, and try to find out where they are living.

Where is their robust and rigorous definition?

SETI is not a theory!!! It is a search. They describe what they are searching for, and what makes them think they’ll find it (i.e. the odds of finding a highly evolved life form on some other planet). If they ever find something interesting, we will have to come up with theories to explain what might be responsible (perhaps by looking at the planet that the signals are originating from).

RDF: You say that intelligent things “possess understanding of nature, foresight, and intent.” Ok, so I have something here in my laboratory (I call it a “booja”). Can you please tell me how to determine if this booja is intelligent? Conversely, I have something else here in my lab that is called a jooba, and I’m certain the jooba is intelligent. Can you please tell me one single thing that I can observe about this jooba?
CS: I would have to examine their effects. What are they? If they emitted a radio signal that encoded the first hundred primes and nothing else I would conclude that either that it was intelligent, or that its creator was intelligent. If they rearranged matter in ways that are extremely unlikely given the chemical affinities, then I might be tempted to think the same.

Ok, so even though I have told you that a jooba is intelligent, you cannot tell me one single thing that I can observe about this jooba. That means that calling something “intelligent” has no observable consequences. In other words, nothing follows from determining that something is intelligent. In other words, “intelligent” is a scientifically vacuous description.

Cheers,
RDFish

115. 115
CentralScrutinizer says:

RDFish: Human beings are complex living things, with sense organs, nervous systems, muscles, and other complex physical systems that enable us to perceive, store and retrieve memories, generate plans, and design and build other complex systems.

Right.

SETI does indeed assume that anything out there that might send us a signal is just this sort of thing – “life as we know it”.

They put that under their biological section, not their space scanning section, but I don’t have a problem with it up to a point. “As we know it” implies just the sort of properties I have been saying with my definition of “intelligent.” How they are implemented is another matter. I would expect that some entity with an understanding of nature, foresight, and intent, such as humans have, would have physical processes analogous to ours. But I wouldn’t expect them to have neurons and synapses. However, I would not be surprised if they did. But it’s beside the point. SETI is looking for the effects of intelligence, not the explanation of it. Same for ID.

But I think you’d agree that is what ID is “looking for”: If ID posits that extra-terrestrial complex life forms are the reason for complex biological systems on Earth, then we might as well assume we are simply descendents of those complex life forms, and not the products of its bioengineering!

If, in some sense, we turned out to be the “descendents” of some ETI it would not offend me in the least. It’s possible. But irrelevant to my point, which is, “intelligence” is not a vacuous term. Saying we’re looking for the effects of intelligence is not a vacuous statement.

The explanation is vacuous without some specific definition. You have provided a specific definition (although it is not one that has been established as the technical definition associated with ID theory).

Neither has SETI established a technical definition, although they have specified some of it’s effects, e.g, “coded information”, “narrowband signals”, etc. But it’s pretty obvious what everyone has in mind. I think you know that. Same for ID.

Think about it: How do you know that Joe the computer-maker possesses a conscious understanding of nature, and conscious foresight and intent? Answer: Because we know these things about human beings of course, and because Joe is a human being. If we knew nothing of what caused the computer, you could not make this inference.

I agree. We know about humans. We know they can make computers. And we know what proeprties they have that allows them to do so. We know we can make coded information and narrow band signals, and we know why. That’s why SETI is looking for these effects, the effects of human-like intelligence. ID looks for those effects in earth’s biology.
I’m not sure what exactly you disagree with, except your complaint that “intelligence” hasn’t been rigorously defined in some way you are satisfied with yet. However, it’s not stopping the SETI crowd, and I don’t see it as a serious objection, because I think “intelligence” is reasonably defined and understand by those looking for it.

CS: I think codes are a very strong indicator of intelligence as I define it.
RD: That is a completely unsupported opinion. You have precisely one data point: human beings generate codes, and human beings have conscious intent. You cannot therefore conclude that anything that generates codes must likewise have conscious intent. The only thing we know of that has conscious intent requires a well-functioning brain in order to experience that – and brains are ridiculously complex organs, with more “CSI” that anything else we know of.

Bizarre. You rebuff me and then support what I’m saying in the same breath. Zero data points would be completely unsupported. You acknowledge that humans can do it. So that’s one data point in my favor. And that “data point” is one helluva “data point.”

Humans are the only known source of novel coded information systems. (Earth’s DNA replicators can replicate and pass on coded information, but they do not create novel systems. And its source is unknown.)
We know the properties in humans that allows us to create coded information- why they can, and why they do do it. And it takes a very heavy bit of physical “machinery” (human brain) wired up a certain proposterously complex way to acheive it, and that it’s the only known source. That plus the fact that no plausible mechansim or cause/necessity is even remotely in view, I stand by my opinion that at present coded information is a strong indicator. This could change, but so far I have no reason not to think so. These reasons are is why SETI thinks the signs of intelligence are worth searching for. We can agree to disagree on this. But it’s far from an unsupported opinion.

So if you are going to try and reason that like effects have like causes, you end up with the hypothesis that whatever created DNA mechanisms in biological systems was something with a very complex brain.

You could be right. And if that turned out to be true, I would not be offended.

But again, ID doesn’t want to conclude that some complex life form created life on Earth. Rather, ID wants to reject that part of the inference.

I’ve never read that from any of the prominent ID writers, but you could be right. I, personally, would not necessarily “reject that part of the inference.”

ID argues that even though the cause of life wasn’t itself a complex organism..

You could be right But I’m not familiar with such statements from the ID luminaries. Anyway, “ID” is like “Christianity”, there is apparently lots of ideas about what that means. But I think the thread that runs thru it all is that something with human-like (or better) understand of nature, foresight and intent, brought about the first DNA replicator system. Beyond that, I think it can get ugly.
I’m not trying to convince you to accept any particular form or position of ID except to say that the idea of “intelligence” is not a vacuous source. SETI is looking for effects of it, and so is ID. Whether or not any particular persons likes or dislikes the idea of a designer with a complex brain is beyond the scope of my interest. And if that proximate source of Earth’s DNA replicator turns out to be a ET with a very big brain, that would not hurt my feelings whatsoever.

I agree that your definition is not vacuous: Your definition of intelligence is “having understanding of nature, foresight and intent”. Even though we do not understand how “foresight” or “understanding” works in human beings, I understand each of these attributes subjectively, because I consciously experience what we call understanding, planning, and intentions. You could also try to verify these attributes by asking me questions, or test that I can generate plans by giving me novel problems to solve.

I agree.

You cannot, however, ascertain if the cause of living things likewise experiences these things.

Not yet. But looking for the signs is all we have to go on at this point.

As we know it, “intelligence” is a property of complex living things. SETI is searching for this. If that is what ID is “searching for”, then ID has another set of problems, as I’ve explained above.

I don’t see those as problems, but I see how others might see them as problems.

Only if ID is supposed to be taken seriously.

It seems to me that a lot of people take ID seriously, including myself.
As for the scientific community at large, well, they have historically be on the slow end of things when it comes to paradigm shifts. I’m not worried about it. But we’ll see.

You have provide one definition that is non-vacuous, but still cannot be operationalized in the context of ID. What you’ll find is that there is no definition that will meet ID’s objectives of being non-vacuous and objectively verifiable in the context of ID.

Sure I did. Coded information systems are only know to originate from entities that have understanding of nature, foresight and intent. You can add to that, “that stem from their complex brains.” OK, so, “coded information systems are only know to originate from entities that have understanding of nature, foresight and intent because of their complex brains.” These are “intelligent beings.”

I have no problem with that.

That may not be appreciated by some folk who think God designed DNA directly without other intermediate intelligences in the chain, but it’s not a problem for ID, per se., and not for me at all, personally.
As for being objectively verifiable, for some things I may have to be satisfied with the inference to the best explanation. Time will tell. But ID is not vacuous.
Anyway, I think I’ve stated what I need to. Anything beyond this will be beyond the scope of my interest.

Signals that human beings (or similar life forms) might send out into space. If we find some, we’ll conclude that there is life elsewhere in the galaxy, and try to find out where they are living.

You bet. But even if we could not make contact, it would be reasonable to assume an intelligent source, likely one with complex brains, but not conclusively until we could examine them.

SETI is not a theory!!! It is a search.

It’s a search in an attempt find evidence supporting an implied conjecture: namely, that it is very likely that it takes a human-like intelligence to produce “coded information”, “narrowband signals”, and the like. Otherwise search for these things would be a waste of time.

FWIW, I don’t think ID is a theory, either. Like SETI, it’s a conjecture looking for evidence. The conjecture underlying both enterprises is, “it takes human-like intelligence to produce these effects.”

They describe what they are searching for, and what makes them think they’ll find it (i.e. the odds of finding a highly evolved life form on some other planet).

“Highly evolved life” is just another term for “human-like intelligence”, given what their searching for. And they say just that on their web site. Here’s how they put it:

“What is the premise of SETI?

“It is an effort to detect evidence of technological civilizations that may exist elsewhere in the universe, particularly in our galaxy. …we have some ability to discover evidence of cosmic habitation, and in the specific case of our SETI experiments to find beings that are at a technological level at least as advanced as our own…Other tell-tale characteristics include a signal that is completely polarized or the existence of coded information on the signal.”

In other words, human-like intelligence exists, and they can produce technological societies like humans do, that can produce the effects that human-like technological societies can produce, and we hope to find such effects.

Now, on earth, we’ve found coded information from a non-human source… in our own DNA replicator. It’s right there for all to see. Why is coded information a reason to suspect intelligence “out there” and but not “in here?” Particularly in light of lack of natural chemical affinities for the components. I think Meyers makes a strong case for this. Hardly, “unsupported” as you claim.

If they ever find something interesting, we will have to come up with theories to explain what might be responsible (perhaps by looking at the planet that the signals are originating from).

Even if we find signs of intelligence, we may never be able to determine the physical structure of the entities that generated the effects. We could assume it was complex brains. But it wouldn’t matter. All that matters to get started is to make a conjecture, and put telescopes out to the sky looking for evidence. ID is “putting the telescope” inward, so to speak. And I think the DNA replicator is the smoking gun – a clear sign of intelligence – for reasons previously given: coded information in an extremely unlikely chemical combination. Whatever it is, it isn’t a vacuous conjecture.

Ok, so even though I have told you that a jooba is intelligent, you cannot tell me one single thing that I can observe about this jooba. That means that calling something “intelligent” has no observable consequences. In other words, nothing follows from determining that something is intelligent. In other words, “intelligent” is a scientifically vacuous description.

Perhaps I misread what your intent here. I think I now know what you wanted. If you tell me (given my definition) that jooba is intelligent, I would look at its effects and see if it produced coded information or narrowband radio signals. If it did, I would conclude that it had intelligence or was made by an intelligence.

Not vacuous

Cheers!

116. 116
RDFish says:

Hi CS,

They put that under their biological section, not their space scanning section, but I don’t have a problem with it up to a point. “As we know it” implies just the sort of properties I have been saying with my definition of “intelligent.”

It also implies just the sort of properties I have been pointing out that are requisite for intelligent behavior: the complex form and function of nervous systems, sense organs, muscles, and so on – exactly the sorts of things that ID purports to explain in the first place.

If, in some sense, we turned out to be the “descendents” of some ETI it would not offend me in the least. It’s possible.

If that were the case, then ID would simply be false: Life on Earth would not be the result of intelligent design by these ET life forms, but rather it would be the product of their biological reproduction. There have been lots of folks who have suggested this in the past (Francis Crick, the Raelians, and so on) but the idea never caught on because there is no evidence that it is true.

But irrelevant to my point, which is, “intelligence” is not a vacuous term. Saying we’re looking for the effects of intelligence is not a vacuous statement.

Again, it is vacuous unless you provide a specific definition for the term, and it is scientifically meaningless unless something objectively observable is specified in the definition.

RDF: The explanation is vacuous without some specific definition. You have provided a specific definition (although it is not one that has been established as the technical definition associated with ID theory).
CS: Neither has SETI established a technical definition, although they have specified some of it’s effects, e.g, “coded information”, “narrowband signals”, etc.

SETI is not obliged to provide a definition of “intelligence”, because it is not attempting to explain anything by reference to the term. Biologists needn’t provide a rigorous definition of “life” (also difficult), because there is no theory that invokes “life” as an explanation for anything. SETI and biology simply use the terms “intelligence” and “life”, respectively, as descriptions of what they are looking for or studying, respectively.

But it’s pretty obvious what everyone has in mind. I think you know that. Same for ID.

Yes, what everyone has in mind is their own anthropomorphic projection of human minds. My point is that it is fallacious to imagine that just because human minds have certain properties (including consciousness) and can build complex machines, then anything which builds complex machines is likely to have those same properties. Again, if you concede that it is likely that the cause of life on Earth were complex organisms with complex sense organs, brains, and so on, then yes, it would be more likely that they experienced conscious thought like humans do. If that is what ID is arguing for, then it needs to say so, and we wouldn’t have these debates – ID would just be another little cult that believes we came from aliens and doesn’t answer where they came from. But ID refuses to limit itself to that hypothesis, because most ID folks believe that an immaterial, transcendent god is the intelligent being responsible. But that hypothesis is even worse, since there is nothing of the sort in our uniform and repeated experience.

I agree. We know about humans. We know they can make computers. And we know what proeprties they have that allows them to do so.

Brains, eyes, hands…

I’m not sure what exactly you disagree with, except your complaint that “intelligence” hasn’t been rigorously defined in some way you are satisfied with yet. However, it’s not stopping the SETI crowd, and I don’t see it as a serious objection, because I think “intelligence” is reasonably defined and understand by those looking for it.

Let us say that SETI finds a signal similar to what humans would transmit. We would infer that a civilization of life forms similar to humans was responsible, and that inference would provide a great deal of information about what we think regarding the cause of the signal: We could look for the temperate planet these beings lived on, look for the liquid water they needed to survive, the buildings they lived in, their electrical power generation facilities, the transportation systems they’ve constructed, agriculture and waste disposal facilities, communication infrastructures, spaceships, and so on.

Now let us say that ID decides that terrestrial biological systems contain evidence that something has “designed” them. What does that inference tell us? Precisely nothing, because the term is vacuous without further qualification.

Bizarre. You rebuff me and then support what I’m saying in the same breath. Zero data points would be completely unsupported. You acknowledge that humans can do it. So that’s one data point in my favor. And that “data point” is one helluva “data point.”

??? My point was that one data point cannot support the inference you are trying to make:

1) Everything that produces CSI has conscious intent
2) Biological systems contain CSI
3) Therefore whatever produced biological systems had conscious intent

You cannot assume that (1) is true simply on the basis that humans are conscious. Not only is your sample size too small (N=1!), but everything we know about how human beings manage to design things confirms that we require well-functioning complex brains in order to do it, so the very first biological CSI could not possibly have been produced by human-like thought processes.

RDF: So if you are going to try and reason that like effects have like causes, you end up with the hypothesis that whatever created DNA mechanisms in biological systems was something with a very complex brain.
CS: You could be right. And if that turned out to be true, I would not be offended.

I’m glad you wouldn’t be offended (by the way I think you’re a very reasonable thinker), but that isn’t my point. My point is that ID conflates two very different hypotheses:

1) Life on Earth came from complex life elsewhere (ID-ET theory)
2) Life on Earth came from something radically different from living things but could still think like human beings (ID-god theory)

The problems with #1 are that (a) it doesn’t answer where ET life originated, (b) it is simpler to assume we are their biological descendents rather than the products of their bioengineering, and (c) there is no evidence that ETs exist.

The problems with #2 are that (a) we have no idea how anything could think like a human but without the benefit of a complex living brain/body, and (b) there is no evidence that any such thing exists.

So, neither of these two hypotheses give us any reason to suppose they are correct.

But I think the thread that runs thru it all is that something with human-like (or better) understand of nature, foresight and intent, brought about the first DNA replicator system. Beyond that, I think it can get ugly.

I’m not sure it’s “ugly” (I personally have no trouble with religious claims, as long as they don’t aim to co-opt scientific status). But just saying that the cause of DNA replication systems had the sort of general conscious understanding, foresight and intent that human beings experience goes far, far beyond any evidence or supporting argument. One would expect that something with these attributes could explain their designs in grammatical language, design other sorts of things, play board games, understand romance and comedy, enjoy music, and generally have conscious beliefs and desires the way we do. But none of these expectations can possibly be tested against the evidence, and we have no reason at all to think any of them are true.

I’m not trying to convince you to accept any particular form or position of ID except to say that the idea of “intelligence” is not a vacuous source.

Since nothing specific follows from labelling something “intelligent”, it is vacuous as an explanation.

RDF: You cannot, however, ascertain if the cause of living things likewise experiences these things.
CS: Not yet. But looking for the signs is all we have to go on at this point.

In that case, you’re saying that ID’s hypothesis is not merely “design”, but rather “something that had mental abilities and experiences similar to human beings”. Moreover, you are conceding that as yet, ID has no evidence whatsoever for supporting its hypothesis.

OK, so, “coded information systems are only know to originate from entities that have understanding of nature, foresight and intent because of their complex brains.” These are “intelligent beings.”

I have no problem with that.

Well, ok, but if ID was actually positing that life on Earth came from life elsewhere, we would not be on this forum debating the matter. The reason for the popularity of ID is because it comports with people’s religious beliefs about a transcendent god, and if ID was actually about alien life forms, the majority of people who debate here wouldn’t be interested.

It’s a search in an attempt find evidence supporting an implied conjecture: namely, that it is very likely that it takes a human-like intelligence to produce “coded information”, “narrowband signals”, and the like. Otherwise search for these things would be a waste of time.

No, the conjecture is that if complex living things evolved on Earth, they likely evolved on other Earth-like planets. And if complex living things here figured out how to transmit signals through space, so might other life forms.

Again: If you’re talking about complex life forms being responsible for DNA or SETI transmissions, then I understand what you are talking about. If you’re talking about some unspecified type of “designer” being responsible for these things, then you have said nothing at all. If you say the designer had conscious awareness and intent, then you are saying something concrete, but without support and untestable.

Now, on earth, we’ve found coded information from a non-human source… in our own DNA replicator. It’s right there for all to see. Why is coded information a reason to suspect intelligence “out there” and but not “in here?” Particularly in light of lack of natural chemical affinities for the components. I think Meyers makes a strong case for this. Hardly, “unsupported” as you claim.

Either the source is a life form or it isn’t. If ID posits a life form, it’s just another “we came from aliens” hypothesis, like Crick’s panspermia, that has been languishing with little attention for lack of any reason to believe it’s true. And if ID posits something that isn’t a life form but still has conscious intent, well, neither Meyer nor anyone else has provided any reason to think that any such thing can exist.

People have talked about us coming from ETs, or from God, for a very long time, and ID brings nothing new to the debate. Continuing to prove that evolutionary theory doesn’t account for biological complexity just tells us something I for one already know: We can’t explain the origin of biological information. It doesn’t tell us one single thing about what the answer might be.

Perhaps I misread what your intent here. I think I now know what you wanted. If you tell me (given my definition) that jooba is intelligent, I would look at its effects and see if it produced coded information or narrowband radio signals. If it did, I would conclude that it had intelligence or was made by an intelligence.

I have a jooba here in my laboratory. I am telling you that this jooba is “intelligent”. Now, given that information, can you tell me one single thing that I can observe about this thing? If you ask me if it produces coded information or narrowband radio signals, I will tell you no, it does not (at least as far as I can tell). Can you tell me something – anything – about this jooba based on the fact that you know it is “intelligent”? You cannot, of course, because the term “intelligence” is scientifically vacuous, even given your own definition. Nothing observable follows from saying something is “intelligent”.

Cheers and thanks for the thoughtful discussion,
RDFish

117. 117
Joe says:

If the word “intelligence” is scientifically vacuous then archaeology and forensic science are both scientifically vacuous.

But then again RDFish’s posts are scientifcally vacuous, so perhaps that is the whole problem…

118. 118
RDFish says:

Hi Joe,

If the word “intelligence” is scientifically vacuous then archaeology and forensic science are both scientifically vacuous.

For the hundredth (thousandth?) time: Archaeology and forensic science analyze the results of human actions. That has nothing to do with trying to explain how biological systems originated.

Cheers,
RDFish

119. 119
CentralScrutinizer says:

RDFish,

I would have to agree that there is a difference between SETI and ID: that SETI is looking for signs of intelligence (effects of entities that have human-like properties), and ID is trying to explain the origin of life on Earth.

OK, let’s say we create an organization, and call it, “SIDA”, the Search for Intelligently Designed Artifacts, whose mission is finding the effects of human-like intelligence no matter where it comes from, whether outer space or earth. SIDA is not interested in explaining the origin of the creators of any of the artifacts it finds. SIDA has adopted the same criterion for detection of intelligently design artifacts, such as narrowband signals, and coded information.

If SETI finds a signal with coded information, given no plausible “natural” explanation, and given that coded information systems are known products of intelligence, they can reasonably put that in the category of the “this is likely, given what we know, that this was designed by a human-like intelligence.” This can spur on some effort to communicate with the origin of the signal. Lacking any information about the source of the signal, we cannot be sure any signal was produce by a human-like intelligence.

Now, I put my microscope to Earth’s DNA replicator and find is has a coded information system. Given no plausible “natural” explanation, even strong reasons to think that one cannot exist (lack of chemical affinities), and given that coded information systems are known products of intelligence, I can reasonably put that in the category of the “this is likely, given what we know, that this was designed by a human-like intelligence.” This can spur on some effort to communicate with the origin of the system. Although, SIDA suffers from a disadvantage because we have no idea where to look for the source. Lacking any information about the source of the system, we cannot be sure the system was produce by a human-like intelligence.

SETI and SIDA are essentially equivalent in nature, just different in scope.

In addition to narrowband signals and coded information systems, if something like F/CSI analysis can eventually become a knock-out approach to discovering artifacts that are the products of human-like intelligence, then it would be a powerful tool in SETI’s and SIDA’s toolbox.

CS: If you tell me (given my definition) that jooba is intelligent, I would look at its effects and see if it produced coded information or narrowband radio signals. If it did, I would conclude that it had intelligence or was made by an intelligence.

RD: I have a jooba here in my laboratory. I am telling you that this jooba is “intelligent”. Now, given that information, can you tell me one single thing that I can observe about this thing? If you ask me if it produces coded information or narrowband radio signals, I will tell you no, it does not (at least as far as I can tell). Can you tell me something – anything – about this jooba based on the fact that you know it is “intelligent”? You cannot, of course, because the term “intelligence” is scientifically vacuous, even given your own definition. Nothing observable follows from saying something is “intelligent”.

I would look for effects that it had an understanding of nature, foresight, and intent, since that’s my definition of intelligence. Narrowband signals and coded information systems are two candidates. If it did, I would put it in my, “”it is likely, given what we know, that this was designed by a human-like intelligence.” How is this vacuous?

RD to Joe: For the hundredth (thousandth?) time: Archaeology and forensic science analyze the results of human actions. That has nothing to do with trying to explain how biological systems originated.

But archaeology and forensic science don’t merely analyze the results of human actions. They look at putative effects and compare those to a set of criteria, namely, what is known about human actions. SETI and SIDA are analagous. In all cases, they is looking at a putative artifact and comparing it against a set of criteria.

120. 120
RDFish says:

Hi CentralScrutinizer,

OK, let’s say we create an organization, and call it, “SIDA”, the Search for Intelligently Designed Artifacts, whose mission is finding the effects of human-like intelligence no matter where it comes from, whether outer space or earth. SIDA is not interested in explaining the origin of the creators of any of the artifacts it finds. SIDA has adopted the same criterion for detection of intelligently design artifacts, such as narrowband signals, and coded information.

We agree that we’re not trying to figure out what caused the cause of these things we observe, but rather simply what caused the observed things (DNA, flagella, narrowband signals, etc) themselves.

The problem with “SIDA” is that it assumes that anything which produces things that are similar to what humans produce is going to be “human-like”. The point I am trying to make is that we have no way of knowing in what ways the cause is “human-like” and in what ways it is not. For that reason, we learn precisely nothing by labelling the cause “human-like” or “intelligent”.

If instead we determined that there was a high probability that the cause of what we observe was a complex form of life as we know it, then we at least gain some real knowledge of what the cause was. So, if ID was about finding extra-terrestrial life forms that may have designed life on Earth, then ID would not be a vacuous endeavor (it would simply fail for lack of evidence). As it is, however, the hypothesis of ID is empy of scientific meaning, because we would have no idea what observable attributes ought to be associated with the hypothetical cause of what we observe.

If SETI finds a signal with coded information, given no plausible “natural” explanation, and given that coded information systems are known products of intelligence,…

This statement is itself highly misleading. “Intelligence” does not produce anything, of course, any more than “beauty” or “strength” or “athleticism” or “determination” produces things. It is just idiomatic use of language where we reify these attributes and say these attributes cause things. In fact, it is human beings that cause things, not their abstract characteristics.

Here is my point in a nutshell:
Saying that a complex life form produced something (e.g. a narrowband signal or DNA) has meaning. Saying that a human being produced it adds a tremendous amount of specific meaning. Saying that it was produced by something with “understanding of nature, foresight, and intent” is meaningful, but can’t be ascertained simply by looking at the result – it must be ascertained by interacting with the thing (organism, entity, process, system, whatever) that is supposed to have these attributes. Saying only that it was produced “by intelligence” is meaningless.

RDF: Nothing observable follows from saying something is “intelligent”.
CS: I would look for effects that it had an understanding of nature, foresight, and intent, since that’s my definition of intelligence. Narrowband signals and coded information systems are two candidates. If it did, I would put it in my, “”it is likely, given what we know, that this was designed by a human-like intelligence.” How is this vacuous?

I’ve already said that the jooba does not produce narrowband signals or coded information. Still, I assure you that the jooba is intelligent. If the term “intelligence” was scientifically meaningingful, it would refer to something that can be somehow characterized in terms of observables. So I ask you again, tell me one thing that I ought to be able to observe about this jooba given that it is intelligent.

But there is nothing, of course, that you can tell me about the jooba! Not one single thing. Can it ride a bike? Play guitar? Solve a crossword puzzle? Hunt a bear? Read a book? Dam up a river? Grow a crop of corn? Write a novel? Come up with a philosophical argument against abortion? Play a video game? Write a song? Flirt with a woman? Find the acorns it hid last fall?

You have no way of knowing which of these things – if any – may be true of the jooba. And for the very same reason, when ID says that the cause of biological complexity is “intelligent”, it is a completely vacuous statement.

But archaeology and forensic science don’t merely analyze the results of human actions. They look at putative effects and compare those to a set of criteria, namely, what is known about human actions. SETI and SIDA are analagous. In all cases, they is looking at a putative artifact and comparing it against a set of criteria.

When archaeologists declare they’ve discovered an “artifact”, we immediately know a great deal about the cause of the artifact, simply because the cause was a human being and we know a lot about human beings. When a forensic scientist declares some event was caused by a “suspect”, we likewise know a lot about what sort of thing we’re looking for: A human being. If SETI found a narrowband signal and some theorist declared it was likely sent by an “a civilization of extra-terrestrial life forms”, we would know something about what was being suggested (they likely lived on temperate with liquid water, built electrical power generation facilities, etc etc). In contrast, if SIDA found some putative artifact and some theorist declared it was produced by “something intelligent”, it would tell us nothing whatsoever beyong what we already know – it somehow produced the artifact in question!

Likewise when ID says the cause of biological complexity is “intelligent”, it is saying nothing at all about what we could observe to be true about the cause.

Cheers,
RDFish