Luskin conflating design inferences

Posted 23 February 2008 by PvM

On Evolution News, Casey Luskin claims that the work by Wired Science to detect the names in the Venter genome sequence is an example of applying a design inference. Indeed, their application of a scientific design inference contrasts strongly with ID's failed attempts to extend the design inference to include areas of our ignorance. As is self evident, the example is nothing more than a pattern matching and has no relevance to the design inference approach as proposed by Intelligent Design Creationists which involves eliminating any and all known (and unknown) regularity and/or chance pathways.

It seems that the day when we can detect human intelligent design in biology has come much sooner than expected. But what if there are other sources of intelligent design in biology as well?

Luskin repeats here what Wilkins and Elsberry already argued in their paper "The advantages of theft over toil: the design inference and arguing from ignorance":

Intelligent design theorist William Dembski has proposed an "explanatory filter" for distinguishing between events due to chance, lawful regularity or design. We show that if Dembski's filter were adopted as a scientific heuristic, some classical developments in science would not be rational, and that Dembski's assertion that the filter reliably identifies rarefied design requires ignoring the state of background knowledge. If background information changes even slightly, the filter's conclusion will vary wildly. Dembski fails to overcome Hume's objections to arguments from design

Ian Musgrave presented a challenge to ID proponents to show how their design inference approach could be applied successfully to the Venter genome. Not surprisingly no successful design inference based on ID concepts was submitted. As the example by T. Ryan Gregory shows, finding 'design' in the genome is hardly trivial. Somewhat ironically, the word was found in the Plasmodium genome (remember Behe's arguments about malaria?). That settles it, in true Luskin fashion, we have seen how design can be found anywhere and everywhere. As Wilkins et al point out

So now there appears to be two kinds of design - the ordinary kind based on a knowledge of the behavior of designers, and a "rarefied" design, based on an inference from ignorance, both of the possible causes of regularities and of the nature of the designer.6

In other words, there is design which can be scientifically detected and is based on positive knowledge and there is rarefied design, which is the kind of design ID claims to be able to detect although it has failed to provide any successful applications of its concepts and ideas. The latter kind of design is based on an inference from ignorance. The example cited by Luskin shows how we need to understand motives and other relevant aspects of the designer before we can make a reliable design inference. Informed by the Venter that the genome contained "coded messages", the approach was simple and did not involve any of the proposed methods of detecting design proposed by ID Creationists. That by itself should be quite telling but Luskin's example shows that science does not reject design a-priori, undermining thus one of the many criticisms raised by ID creationists. It surprises me that ID creationists are still conflating these two very different methods of inferring design. But given the failure of ID to provide any examples of a successful application of their filter, it does not surprise me. After all, ID claims that it can fully incorporate science, the problem is that they have failed to show that they can do anything more than that. In other words, Intelligent Design remains a scientifically vacuous concept at best. Perhaps Luskin would like to attempt applying the ID design inference methodology to the Venter genome?

37 Comments

Stacy S. · 23 February 2008

Where did you find this please?

PvM · 23 February 2008

Evolution News

Stacy S. · 23 February 2008

Thank you :-)

CleveDan · 23 February 2008

ID has no published science to back it up.........no coherent theory but, now scientists signing their own name in their work is evidence for ID?? I suppose its no worse then anything else out of The DI

PvM · 23 February 2008

CleveDan: ID has no published science to back it up.........no coherent theory but, now scientists signing their own name in their work is evidence for ID?? I suppose its no worse then anything else out of The DI

So why is the DI pretending that there is any relevance to the work by Wired that applies to ID? Why causing such unnecessary confusion?

Glen Davidson · 23 February 2008

Yeah, gee, it didn't take long for someone to find actual design in the genome, nor did some purported dogma named "Darwinism" prevent anyone from doing so.

So do something other than whine for once, Casey, and find something similar in non-engineered organisms, if you're going to claim that they're designed. Otherwise I'm reminded of a familiar phrase, STFU.

Glen D

http://tinyurl.com/3yyvfg

harold · 23 February 2008

This reflects the same old argument against certain ID claims that I recall making in 1999, the first time I heard the claims.

Science recognizes "design" when knowledge of the "designer" is available.

The examples I used to like to use were beehives or bird's nests (just to emphasize that, of course, you don't need to be "intelligent" in a human sense to "design"). If a new type of beehive were discovered in the Amazon rain forest or some such thing, of course it would be recognized, but that's because we already know about bees and their designs. Even appeals to designs by "aliens" make the assumption that they would be so analogous to human design that our knowledge of humans would allow us to conjecture about the "designers". That certainly applies to SETI.

Although Luskin was crafty enough to recognize a good opportunity - take a few classes in a mainstream science program and at a law school, and screech that you believe in "intelligent design" and Jesus and vote Republican, and there's a six figure job at the DI for you, too, if you can handle the nausea - he appears to be genuinely more dull than many of his colleagues.

Glen Davidson · 23 February 2008

There's an enormous confusion that Luskin is perpetrating with his conflation of, say, EF "design detection" and normal sensible design detection that scientists engage in (such as "Wiredscience" did). The two methods are not even comparable, as one looks for rational arrangement, and the other simply looks for something which is unexplained, which to them means "supernatural." I wrote a little essay on Pharyngula about the differences, which I subsequently modified, and reproduce below:

[A] way of looking that the design-miracle connection that the IDists rely upon, while trying to hide it in some cases, is that IDists like Dembksi simply think of "design" as being something outside of nature. Whether it's humans or gods, creativity and design are doing what nature and its "lawful processes" are unable to do. That's why things "look designed" to them. They really don't go around pointing to similarities between our machines and life--except sometimes when prosetylizing--instead they're "pointing out" that natural processes can't make life. They say that things "look designed," because design is magic [because they think that minds are not natural], and magic is design, hence the miracles and creativity that God [is supposed to have] used to make life is "design" to them. That's also why "inferring design" is the default position whenever something can't be precisely explained. The point isn't that a "process of design" explains why a bird is as it is, the point indeed is that nothing explains it, which means to them that a miraculous process, like human consciousness and thought [as they conceive of consciousness and thought to be], did it. The fact that a bird isn't like our designs isn't important to them, because they don't think in terms of design being "natural" or "predictable," rather they see design as miraculous and unpredictable. The situation is that their vitalistic notions regarding the human mind are what make conversing with them useless. They were never talking about observed design when they said that "life looks designed," they were saying that "life looks magical and unexplainable except through the magic of godlike or god-given minds." They weren't trying to explain the unknown through the known, because to them, by definition, the human mind and its processes are unknowns. They were trying to explain the unknown (to them) by the unknown (to them), explaining the mystery of life through the mystery of life. The whole project is hideously circular, then. The mind is supernatural and unpredictable, and only something supernatural and unpredictable could give rise to the mind. Hence God, who is also supernatural and unpredictable mind, gave rise to mind. The only flaw in that circularity is that life is predictable in many of its aspects via evolutionary theory. That is why evolutionary theory must go. Were they going to try to show that evolutionary theory doesn't work in science, either as an organizer or as a useful idea? Of course not, their entire reason for ...[despising] evolution is that it explains life, and they need life to be necessarily dependent for its origins upon God. Therefore they have to "prove" that life can't be explained by MET, and we're quite familiar with their usual bumbling, misapprehension, and ideology-driven prevarication in their attempts to do so. Evolution destroys their miraculous mind which has to come from another miraculous mind, which means that evolution must be destroyed. To their minds, saying that "life looks designed" makes perfect sense in their desire to overthrow evolution, because they're operating on the assumption that "it doesn't look like it appeared naturally" is exactly the same thing as "look designed" to them. Dembski's EF depends on that notion--he doesn't care about showing that life looks or is designed by comparing design with design, he's implicitly counting on the equation "not natural=designed." All that the EF is even supposed to do is to show that life is "not natural." That would mean nothing at all with respect to design if Dembski didn't presuppose that "not natural=designed." So not even the IDists think that life "looks designed" in the ordinary sense, they really mean that life "doesn't look natural," which to their minds means that it was produced by the supernatural, IOW, a mind.

Glen D
http://tinyurl.com/3yyvfg

Gary Hurd · 23 February 2008

Good job Pim. In a new blow against ID creationist codswollop, Cosmological Fine Tuning, well beloved by creationists like Guillermo Gonzalez, might have a simple explanation. In an article today from Nature,

Quote: Scott Funkhouser of the Military College of South Carolina (called The Citadel) in Charleston has shown how this number ( 10^122 ) — which is bigger than the number of particles in the Universe — keeps popping up when several of the physical constants and parameters of the Universe are combined 1. This ‘coincidence’, he says, is surely significant, hinting at some common principle at work behind the scenes.

The news item concludes,

Quote: Funkhouser notes that 10^122 is about equal to the cube of 10^40. Is there some reason why the two sets of large numbers should be linked in that way? It would follow, Funkhouser says, if there happens to be a certain mathematical relationship linking the mass of a nucleon (a proton or neutron) with the speed of light, the gravitational constant, Planck’s constant and the cosmological constant. That sounds pretty exotic, but in fact such a link was proposed in 1967 by the Russian physicist Yakov Zel’dovich based on an entirely separate argument. And he’s not the only one to suggest this. “The interesting thing is that the relationship has been proposed before by several different authors, each with a different explanation,” says Funkhouser. And in a separate paper 3, he’s come up with yet another justification for it. “I have shown that a simple model for the origin of our Universe involving ten spatial dimensions leads naturally to this relation,” he says. It follows if seven of the dimensions shrank while the other three “puffed out” to form the reality we observe today.

Joe Mc Faul · 23 February 2008

Casey Luskin asks:

"It seems that the day when we can detect human intelligent design in biology has come much sooner than expected. But what if there are other sources of intelligent design in biology as well?"

There are. Ever see a beehive? A bird's nest? A spider web?

Bob O'H · 24 February 2008

As Wilkins et al point out

et al. means "and others", i.e. that there is more than one other author. WHAT HAVE YOU DONE TO OUR WES?

John S. Wilkins · 24 February 2008

Good to see that the ordinary/rarefied distinction is being used. I was kinda proud of that - for once I made a distinction that was simple and to the point... I think Wesley helped me with that.

Torbjörn Larsson, OM · 24 February 2008

In a new blow against ID creationist codswollop, Cosmological Fine Tuning, well beloved by creationists like Guillermo Gonzalez, might have a simple explanation.

Hmm. I'm of the opinion that a naturally motivated anthropic principle, as opposed to the religious anthropic argument (i.e. bait-and-switch a posteriori likelihoods for a priori probabilities) is also a simple explanation for any number of finetunings. Perhaps even simpler, and also a decided blow against creationism. But the proposed explanation will appear more reasonable for most and, well, it is certainly a new blow. Pile it on!

“The interesting thing is that the relationship has been proposed before by several different authors, each with a different explanation,” says Funkhouser.

That is interesting, as it often means there is something there that nature tries to tell us. (What? So now only biologists are allowed to be anthropomorphizing natural processes? Well, I never.) I look forward to see what becomes of this.

WHAT HAVE YOU DONE TO OUR WES?

What is this sudden attack on "Al"? He is one of the most, if not the most, productive authors I've seen.

Bobby · 24 February 2008

It seems that the day when we can detect human intelligent design in biology has come much sooner than expected.

Funny, I thought they had been detecting design in biology for over a decade now.

So why is the DI pretending that there is any relevance to the work by Wired that applies to ID? Why causing such unnecessary confusion?

Sowing confusion is the DI's mission.

abb3w · 24 February 2008

Let me point out to something which evidently has not occurred to the bulk of those sane and sensible opponents to ID out there. The Cdesign Proponentsists are as much strangers to the inherent character of Design as they are to Intelligence. DESIGN ITSELF IS AN EVOLUTIONARY PROCESS of iterated change and competitive selection. This is why evolution looks like "intelligent" design.

I point out George Basalla's book "The Evolution of Technology" (suggested to my attention by a historian of my acquaintance) for those interested in more information on this particular viewpoint.

IDiot · 25 February 2008

If the secret codes were designed by humans, WHO DESIGNED THE DESIGNERS?

Torbjörn Larsson, OM · 25 February 2008

Let me point out to something which evidently has not occurred to the bulk of those sane and sensible opponents to ID out there. [...] DESIGN ITSELF IS AN EVOLUTIONARY PROCESS of iterated change and competitive selection.

I can't vouch for others, but IIRC I used it early on when I started to discuss science on blogs, and later rejected it. Because it is IMHO both a piss-poor analogy and one of those that easily delude the reader on the characteristics of most of the process: 1) Evolution has no functional goal. Human design have functional targets, even if they may be changed during the design process. This prompts the occasional reader to wrongly conclude that evolution is a targeted process. 2) Evolutionary fitness by reproductive success is equivalent to a search for "good enough" traits, and functional such. Human design targets "best of breed" functionality, and accepts intermediary non-functional or sub-functional test "generations". (To pitch in another couple of failed analogies.) This prompts the occasional reader to wrongly conclude that species are "optimally fit", that "mutants are unfit", that evolution is a global instead of local search, and that there isn't a fixation of traits. 3) Evolution is hereditary. Human design has no genetic equivalent. This prompts the occasional reader to wrongly conclude that extinct species are equivalent to "failed design tests" and that there is a "ladder" towards "final targets". Does the proffered reference engage these problems? Because if it doesn't, I can't see why it should go onto my reading list.

David B. · 25 February 2008

So the DI are presumably adamant that none of the names detected in Venter's designed genome occur anywhere in the natural genome of, say, a human being?

Because if they did then either:
(a) the supposed 'design inference' can't reliably tell the difference between a created and evolved gene sequence, or,
(b) Craig Ventner isn't 'playing God', he really is God!

I should think the odds of finding 'VENTNER' expressed somewhere in the human genome are actually quite high, but clearly that can't be so. No, quite obviously the DI are right and the presence of any human readable words (in English naturally), or recognisable names or places are clear indications of the artificial origin of our genomes.

But why stop there, if the work of Michael Drosnin has taught us anything it's that our 'unnamed creator' is infinitely subtle in his handiwork. I should think a careful search of our DNA coding with an eye to 'skip-codes' should turn up lots of really interesting messages.

Nigel D · 25 February 2008

There's one other flaw in Luskin's claim...

How on Earth could anyone detect design in a DNA sequence by pattern-matching, when the whole lot is supposedly "designed" in the first place? Is human design really so different from other kinds that it sticks out so obviously that it can be detected by simple pattern-matching?

Asn-Ile-Gly-Glu-Leu :-)

Bobby · 25 February 2008

I should think the odds of finding ‘VENTNER’ expressed somewhere in the human genome are actually quite high

The odds are 100%. All you have to do is cook up the right encoding scheme. Likewise you can find the entire bible, Moby Dick, Darwin's deathbed confessions, and confessions that ID is a scam.

Torbjörn Larsson, OM · 25 February 2008

a piss-poor analogy

It occurs to me that the marketing of designs is a much closer analogy to the process of evolution. (Where the absence of genetics is less confusing to boot.) What do you think, could we "sell" creationists on their various gods as women's shoes salesmen listening to the given name "Al"? [Which may stand for "Al lah lah", for all I know.]

Jason Failes · 25 February 2008

"But what if there are other sources of intelligent design in biology as well?"

Then we'll study it.
That's how science works: First you need a phenomenon to study.

That's what makes us so snippy at you IDers (besides the obvious forensic links to old-school creationism): You go on and on about academic freedom, and repression, and poor-me-why-don't-they-take-my-ideas-seriously, in a complete absence of phenomenon. Show us the sites of ancient gene-splicing, and mainstream scientists will be chomping at the bit to do some "intelligent design" research. Until then, you have nothing, at it is dishonest, to say the least, to pretend that you have anything at all, much less something scientific.

Rrr · 25 February 2008

Torbjörn Larsson, OM:
Let me point out to something which evidently has not occurred to the bulk of those sane and sensible opponents to ID out there. [...] DESIGN ITSELF IS AN EVOLUTIONARY PROCESS of iterated change and competitive selection.
I can't vouch for others, but IIRC I used it early on when I started to discuss science on blogs, and later rejected it. Because it is IMHO both a piss-poor analogy and one of those that easily delude the reader on the characteristics of most of the process: 1) Evolution has no functional goal. Human design have functional targets, even if they may be changed during the design process. This prompts the occasional reader to wrongly conclude that evolution is a targeted process. 2) Evolutionary fitness by reproductive success is equivalent to a search for "good enough" traits, and functional such. Human design targets "best of breed" functionality, and accepts intermediary non-functional or sub-functional test "generations". (To pitch in another couple of failed analogies.) This prompts the occasional reader to wrongly conclude that species are "optimally fit", that "mutants are unfit", that evolution is a global instead of local search, and that there isn't a fixation of traits. 3) Evolution is hereditary. Human design has no genetic equivalent. This prompts the occasional reader to wrongly conclude that extinct species are equivalent to "failed design tests" and that there is a "ladder" towards "final targets".

This takes me back to an old song by the recently departed prolific Swedish comic singer-songwriter Povel Ramel where his favourite male lead singer claimed to be perfect as "Erik Anders Svensson Typing, the ultimate form of humanity". Long may they both live in memory.

It occurs to me that the marketing of designs is a much closer analogy to the process of evolution. (Where the absence of genetics is less confusing to boot.) What do you think, could we “sell” creationists on their various gods as women’s shoes salesmen listening to the given name “Al”? [Which may stand for “Al lah lah”, for all I know.]

Well(ington) aren't you mixing your footwear here? Maybe Sand-Al, hah? :-) Brings to mind the absent-minded professor who came to class wearing one black and one brown shoe and, when the discrepancy was pointed out to him, exclaimed, "I know. And the strange thing is, I have another pair exactly like these in my closet at home."

abb3w · 25 February 2008

Torbjörn Larsson, OM: 1) Evolution has no functional goal. Human design have functional targets, even if they may be changed during the design process. This prompts the occasional reader to wrongly conclude that evolution is a targeted process.

First: yes, the requirement of a functional target is an additional criterion of design. That someone cannot distinguish the criterion of a subcategory from a criterion of a supercategory... well. These are the same people who don't seem to understand that all cats being quadrupeds does not make all quadrupeds cats. (No, I don't want to know about your three legged cat...) This, however, does not change the fact that Design is a specific form of the more general category of Evolutionary Processes. More to the point, the "target" often changes during the design process, both at the level of the individual Technologist (such as a hammer maker) and especially at the level of the Art as his (such as hammer making in general). The first example I can think of for the former is the development of the microwave oven while working on Radar research; for the latter, the development of the X-Box as a gaming-only console from the general purpose "PC" computer. (The Bell-to-Modern telephone arguably falls into both categories.) Your second point I believe oversimplifies the nature of Evolution; feel free to elaborate on it. However, I would assert in passing that if anyone believes Microsoft Word is "optimally fit", that ought to qualify as grounds for throwing them in a loony bin. =) "Good enough to make money" is the real criterion for most engineering design.

Torbjörn Larsson, OM: 3) Evolution is hereditary. Human design has no genetic equivalent. [...] Does the proffered reference engage these problems? Because if it doesn't, I can't see why it should go onto my reading list.

The last point is specifically addressed in multiple sections in the book, yes. The "genetic equivalent" is the understanding by the individual technologist of history- and state-of-the-art; that is, the "know-how" on making something. The cover (visible at Amazon and Cambridge Press) helps illustrate this with a sequence of hammers.

Torbjörn Larsson, OM:It occurs to me that the marketing of designs is a much closer analogy to the process of evolution.

Real world engineering design is always constrained by market forces. The competition of one designer against others is an inherent part of real world design (Bell and Gray, for example). If there is a "niche" for your widget, more of them will be made (either by the originator or another designer who reverse-engineers it); contrariwise, if you can't get people to buy your gizmo because they see no need for gizmos, it doesn't matter how good it is as a gizmo per se. Even when the designer is also the sole intended consumer, the law of supply and demand has its effects, in the trade offs of time, effort, and available resources. (I believe Basalla's book also covers some of this.)

Torbjörn Larsson, OM · 25 February 2008

Maybe Sand-Al, hah? :-)

ROTFL! Much better!!! Possibly of Ramel class.

Torbjörn Larsson, OM · 25 February 2008

the criterion of a subcategory from a criterion of a supercategory

Eh? Which are your categories, and why do they put Design under evolutionary processes if Design has an added criterion?

Your second point I believe oversimplifies the nature of Evolution; feel free to elaborate on it.

Indeed it does. Under rapid evolutionary changes (possibly the criterion is that there is no time for fixation - I'm not a biologist) one can observe quasispecies (viruses; putatively abiogenesis). "The analogue of fitness for a quasispecies is the tendency of nearby relatives within the cloud to be well-connected, meaning that more of the mutant descendants will be viable and give rise to further descendants within the cloud." But I think that removes the search from primarily interaction with the environment to primarily internal interaction. And well-connectedness is a distributed trait, not an individual one. Sure, there are some designs that are like later - web pages comes to mind. But it isn't a general property of designs.

However, I would assert in passing that if anyone believes Microsoft Word is “optimally fit”, that ought to qualify as grounds for throwing them in a loony bin. =)

Well, I claimed "targets", and allowed for sub-functionality. But your point is taken, such an idealized model wouldn't describe what really happens in Design and what characterizes the search. I still think my point stands, until a realistic model of Design comes to mind for the causal reader instead of the idealized I described.

The “genetic equivalent” is the understanding by the individual technologist of history- and state-of-the-art;

Actually, that may make sense. Yes, a populations alleles may be considered a data base of working solutions. And state-of-the-art solutions "mutates" and "fixates". I may have to yield that point.

Real world engineering design is always constrained by market forces.

Yes, but it is a separate process. (Cost is a part of design, but such a factor is isolated.)

Henry J · 25 February 2008

Rrr: Brings to mind the absent-minded professor who came to class wearing one black and one brown shoe and, when the discrepancy was pointed out to him, exclaimed, “I know. And the strange thing is, I have another pair exactly like these in my closet at home.”

Exactly like, or symmetrical to? ;)

Torbjörn Larsson: Human design have functional targets, even if they may be changed during the design process.

Don't remind me about designs changing - at work sometimes the specs change so often that it's like aiming at a moving target to code to them. Henry

abb3w · 26 February 2008

Torbjörn Larsson, OM: Eh? Which are your categories, and why do they put Design under evolutionary processes if Design has an added criterion?

The category is Evolutionary processes, which work by iteration of variation, reproduction, and competitive selection. Design shares all of these criteria; therefore all design is an evolutionary process; design also requires intelligent intention, meaning not all evolution qualifies as a design process. Biological taxonomy works the same way: all chordates have a spinal cord, but to be a mammal they also must meet the additional criterion of having hair/fur and milk. Still not getting the implications of your second point, unless you're arguing against design being obviously evolutionary to the layman, in which case...

Torbjörn Larsson, OM: I still think my point stands, until a realistic model of Design comes to mind for the causal reader instead of the idealized I described.

Possibly as evidence against people finding it intuitive, but not against its inherent validity. However, I'd argue that this makes it even more important to try to hammer home. Our society does not (yet) rely on the effect of evolution in biology on a day-to-day or year-to-year basis, but it most certainly does rely on design.

Torbjörn Larsson, OM:
Real world engineering design is always constrained by market forces.
Yes, but it is a separate process. (Cost is a part of design, but such a factor is isolated.)

Err... you're not an engineer, are you? Cost is not so much isolated as omnipresent.

Ichthyic · 26 February 2008

Our society does not (yet) rely on the effect of evolution in biology on a day-to-day or year-to-year basis

wrong.

we do, you just haven't bothered to figure it out yet.

it's not hard, really.

Pim would be happy to walk you through it, I'm sure.

stevaroni · 26 February 2008

Evolution has no functional goal. Human design have functional targets... This prompts the occasional reader to wrongly conclude that evolution is a targeted process.

While evolution may not have targets per say, mother nature is generous with her rewards for good designs, and even more generous with her punishments for bad ones. Specifically, the good designs live long enough to breed, while the bad ones, well... don't. It's important to keep this in mind when the creobots start babbling about the odds of "random mutations" adding up to human genomes. Yes, given a truly random search, the odds are indeed daunting, and even after 4 billion years we'd still be slime molds. But when amplified through the feedback loop of natural selection, where less-than-optimal designs are constantly culled - often by things with teeth - the odds improve phenomenally. Nature doesn't really need intelligent designers - she runs the biggest product testing lab of all time.

Torbjörn Larsson, OM · 26 February 2008

The category is Evolutionary processes, which work by iteration of variation, reproduction, and competitive selection.

Ah, I see. You are abstracting away evolution as a physical process and design as a social or technical process. Yes, in a restricted sense they could be considered such categories. But explicitly promoting that view is a further conflation that a casual reader would have to battle with to avoid confusion.

Design shares all of these criteria; therefore all design is an evolutionary process;

Not really. Look, a minimal definition of evolution could be fulfilled of a lot of processes; it's not the intention to distinguish between say biological evolution and evolution simulation in software, and in fact it can not do so. One minimal definition of evolution is "Evolution is a process that results in heritable changes in a population spread over many generations." This is fulfilled by both biological populations as well as software evolutionary agents populations. In your case you could say that a population is the contents of your library of state-of-the-art functional designs for some related subset, say vacuum cleaners of various constructions. So we need to supplement the definition to distinguish between the processes, as it matters - design and evolution is sufficiently different to not being subject to the same theories. And it is the theories that helps us do exactly that. Thus, the process of biological evolution. And these differences is also why analogies between these processes may be confusing instead of helping.

unless you’re arguing against design being obviously evolutionary to the layman,

I am pointing out that design is often described or thought of as the idealized process, and that the layman may then draw the wrong conclusions I listed.

However, I’d argue that this makes it even more important to try to hammer home.

I'd argue against, as long as you can't make design theory and evolution theory exactly identical. I'm still standing for my initial assessment, to conflate these different processes by way of an analogy is making a piss-poor analogy and one of those that easily delude the reader on the characteristics of most of the evolutionary process.

you’re not an engineer, are you? Cost is not so much isolated as omnipresent.

As a matter of fact I'm trained as one and has worked as one, my PhD notwithstanding. Also, I'm not an economist, but all of that is besides the actual argument. What I'm saying is that producers don't sell the innards of a product by item. Thus, one makes cost a factor in production but it is isolated from the product market. (Different markets.) It is fairly easy to see that we have a market process and a design process, and that they are different (albeit correlated) phenomena. That is all I'm claiming right here.

Steven Laskoske · 26 February 2008

Torbjörn Larsson, OM: It occurs to me that the marketing of designs is a much closer analogy to the process of evolution. (Where the absence of genetics is less confusing to boot.)

Actually, it appears that the marketing of ID is no different than the marketing of "Scientific Creationism". Creationism has certainly been under enough selective pressure. Unfortunately, the only mutations have resulted only in cosmetic changes. (Hmmm... perhaps that can be considered protective coloring. Nah. We saw in Dover that the coloration didn't provide any real protection.)

Steven Laskoske · 26 February 2008

Don't remind me about designs changing - at work sometimes the specs change so often that it's like aiming at a moving target to code to them. Henry

As the phrase in computer programming goes: "Writing to spec is like walking on water. It works better if it is frozen."

abb3w · 27 February 2008

Torbjörn Larsson, OM:
The category is Evolutionary processes, which work by iteration of variation, reproduction, and competitive selection.
Ah, I see. You are abstracting away evolution as a physical process and design as a social or technical process.

I'm possibly biased by the people I work with. Also, it seems to me that viewing design as anything but a "social or technical process" is the abstraction, removing it from the environment in which it must operate.

Torbjörn Larsson, OM:
Design shares all of these criteria; therefore all design is an evolutionary process;
Not really.

Counterexample?

Torbjörn Larsson, OM: Look, a minimal definition of evolution could be fulfilled of a lot of processes; it's not the intention to distinguish between say biological evolution and evolution simulation in software, and in fact it can not do so. One minimal definition of evolution is "Evolution is a process that results in heritable changes in a population spread over many generations." This is fulfilled by both biological populations as well as software evolutionary agents populations. In your case you could say that a population is the contents of your library of state-of-the-art functional designs for some related subset, say vacuum cleaners of various constructions.

"State of the art" is a grossly misrepresentative term; just as archaea remain alive, hand-made bows are still in use by some human cultures. There are living fossils in both technology and biology; check your local server room. =) The main difference between biology and technology is the nature of speciation, because sexual reproduction eventually limits diffusion of information. And, yes, I would hold that there are a very large number of processes that are evolutionary in the more general sense of the word.

Torbjörn Larsson, OM: And these differences is also why analogies between these processes may be confusing instead of helping.

Strictly speaking, I'm not trying to make an analogy; I'm trying to point out a taxonomic isomorphism. To wit, both the process of evolution in biology and of design in technology are both examples of the broader category of evolution.

Torbjörn Larsson, OM:
unless you’re arguing against design being obviously evolutionary to the layman,
I am pointing out that design is often described or thought of as the idealized process, and that the layman may then draw the wrong conclusions I listed.

Fair enough. However, that their natural inference is erroneous does not mean the classification is. My bias is towards better education about design, due to the multiple forms of benefit.

Torbjörn Larsson, OM:
However, I’d argue that this makes it even more important to try to hammer home.
I'd argue against, as long as you can't make design theory and evolution theory exactly identical.

I believe we differ due to differing objectives. You appear to desire the public understand evolution for it's own sake, where I merely consider it part of a larger goal of increased human understanding. Thus, I'm advocating a better understanding of design for it's own sake foremost, and consider the undermining of the "Intelligent Design" argument a (viciously satisfying) side benefit.

Torbjörn Larsson, OM:It is fairly easy to see that we have a market process and a design process, and that they are different (albeit correlated) phenomena. That is all I'm claiming right here.

While not agreeing, I don't think it's central.

Torbjörn Larsson, OM · 27 February 2008

I'm sorry if I don't advance my argument much this time. I have only time for a response on points:

it seems to me that viewing design as anything but a “social or technical process” is the abstraction, removing it from the environment in which it must operate.

Are we know measuring who has the biggest abstraction? :-P Really, I'm just starting where the processes touch ground. Now let me see the argument why design phenomena and theory is part of biological evolution...

Counterexample?

You can’t make design theory and evolution theory exactly identical.

“State of the art” is a grossly misrepresentative term

Different alleles in different populations.

The main difference between biology and technology is the nature of speciation, because sexual reproduction eventually limits diffusion of information.

Which theory predicts this?

However, that their natural inference is erroneous does not mean the classification is.

That is not my main point, as I was arguing that it is a piss-poor and confusing analogy. But in the bottom lies the different phenomena and theories related to the different processes.

I don’t think it’s central.

It is, if my claim is that the market process is a closer analogy to evolution than the design process. I believe such an analogy has been noted before.

abb3w · 28 February 2008

Torbjörn Larsson, OM: Really, I'm just starting where the processes touch ground. Now let me see the argument why design phenomena and theory is part of biological evolution...

It might be interesting to attempt making that case; I'm not, because I'm grossly unqualified for such a novel line of research (not to mention unconvinced of it). What I'm saying is biological evolution and technological design are both examples of a more general category of evolutionary processes, and that this is the reason evolution looks like design.

Torbjörn Larsson, OM:
Counterexample?
You can’t make design theory and evolution theory exactly identical.

On the one hand, that's not a specific counterexample; and on the other, you can't show a dog is a kind of cat, but you can show the structural similarities mean they are both mammals. (I find this parallel in your argument here to the creationist stance amusing.)

Torbjörn Larsson, OM:
“State of the art” is a grossly misrepresentative term
Different alleles in different populations.

Oh, I agree; however, the phrase is like calling humanity "the apex of evolution". The impression it conveys gives the casual reader vast opportunities for mistaken impressions -- or even not-so-casual listeners. Thus my phrasing as "grossly misrepresentative". The smart money is still on the beetles. =)

Torbjörn Larsson, OM:
The main difference between biology and technology is the nature of speciation, because sexual reproduction eventually limits diffusion of information.
Which theory predicts this?

In biology? Well, the diffusion of traits through an interbreeding population is a byproduct of standard genetics models, which are readily approximated by partial differential equations. Google Scholar turns up a couple hundred thousand hits, so this is hardly controversial. As for speciation, I'm using Mayr's definition as it occurring (approximately from memory) when a sub-population acquires characteristics which promote or guarantee reproductive isolation. Ergo, no (or progressively reducing) sexual interbreeding, and diffusion of traits stops, as there is no routine means for mutations in group A to pass to the offspring (or offspring^2, in the case of sterile hybrids) of group B. EG, cats won't ordinarily receive mutations in dogs, barring independent redevelopment, unusual viruses, or human genetic engineering. (Asexual reproduction, of course, would seem to limit diffusion even more quickly.) Both of these are AFAIK as uncontroversial as anything is in biology these days. In technology, however, it's relatively common for solutions in one area of technology to get retooled for use in other areas that long ago diverged. The relation of the reciprocating engine to the automatic submachine gun springs to mind, but that one may be anecdotal only. If you really want examples here, I could track down someone who does History of Technology for a living. I probably could have phrased that more clearly the first time.

Torbjörn Larsson, OM: That is not my main point, as I was arguing that it is a piss-poor and confusing analogy. But in the bottom lies the different phenomena and theories related to the different processes.

And I repeat: I’m not trying to make an analogy; I’m trying to point out a taxonomic isomorphism. I'll admit you might be correct that the general public may be detrimentally confused on first encounter, but I'm not convinced that it would be useless to try and undermine "Intelligent Design" by showing its proponents' ignorance of the subtleties of Design as well as of Biology.

It is, if my claim is that the market process is a closer analogy to evolution than the design process. I believe such an analogy has been noted before.

That the forces in economics are also evolutionary in character, I agree; there are others. However, I don't see how one can use economics to attack "Intelligent Design".

Torbjörn Larsson, OM · 28 February 2008

What I’m saying is biological evolution and technological design are both examples of a more general category of evolutionary processes, and that this is the reason evolution looks like design.

I think we are in general agreement here. Both processes can be strait jacketed to fulfill a minimal definition of evolution (a property that I think you call a "taxonomic isomorphism" for some reason - I think it should be taxonomic equivalence), never designed (!) to throw out non-biological processes. However, I don't think design looks like evolution (to use your original order of inclusion) as they display different phenomena and needs different theories. I don't think I can add anything to this discussion that I haven't already said.

that’s not a specific counterexample;

Huh? I pull down my book on axiomatic design theory (Suh, N.P., “The Principles of Design”), and I don't find a theory that a biological system follows. For example, Suh's definition of total design information have a "design range" that a designer tries to satisfy. As evolution doesn't do that, it doesn't contain "design information". I could go on, but again I would repeat myself.

however, the phrase is like calling humanity “the apex of evolution”.

Well, that would be my complaint, wouldn't it? I placed it in context, and you as a reader (albeit not a casual one) find it gives mistaken impressions.

the diffusion of traits through an interbreeding population

So you define "traits" as information. But information is collected from the environment by all populations, all the time, and expressed in the genome as the frequency of current alleles. Diffusion of traits is diffusion of what is (to be) tested, not what is learned. Learning is established at fixation in regards to the over all process, once fixated remains fixated with respect to environmental pressures, so it is a characteristic of a whole population. I don't see that you present a theory of design similar to population genetics however. Again coming back to the radically different phenomenas to describe.