The Concept of Design

To move towards a resolution of some of the questions posed in the comment threads to recent posts it is necessary to back up and try to clarify some of the fundamental terms in the debate. One of them is 'design.'

Our starting point must be ordinary language. As David Stove points out, "it is a fact about the meaning of a common English word, that you cannot say that something was designed, without implying that it was intended; any more than you can say that a person was divorced, without implying that he or she was previously married." (Darwinian Fairytales, p. 190, emphasis added.) In other words, it is an analytic proposition that a designed object is one that was intended in the same way that it is an analytic proposition that a divorced person is one who was previously married. These are two conceptual truths, and anyone who uses designed object and divorced person in a way counter to these truths either does not understand these concepts or else has some serious explaining to do.

I should think that Richard Dawkins has some serious explaining to do. Consider the subtitle of The Blind Watchmaker. It reads: Why the evidence of evolution reveals a universe without design.

Now I think I understand that. What Dawkins will do in his book is argue how the modern theory of evolution shows that the natural universe as a whole and in its parts is in no way the embodiment of the intentions and purposes of any intelligent being. Thus a bat, a piece of "living machinery," is such that "the 'designer' is unconscious natural selection." (p. 37) The scare quotes show that Dawkins is not using 'designer' literally. What he is saying, putting the point in plain English, is that there is no designer. For if there were a designer, then he would be contradicting the subtitle of his book, which implies that no part of nature is designed. So far, so good.

Unfortunately, on the same page Dawkins says the following about Paley:

His hypothesis was that living watches were literally designed and built by a master watchmaker. Our modern hypothesis is that the job was done in gradual evolutionary stages by natural selection.

But now we have a contradiction. We were told a moment ago that there is no designer. But now we are being told that there is a designer. For if the design job is done by natural selection, then natural selection is the designer.

Now which is it? Is there a designer or isn't there one? Dawkins cannot have it both ways at once.  If there is no designer, then natural selection cannot be the designer.  What this contradiction shows is that Dawkins is using 'design' and cognates in an unintelligible way.

Some will say I am quibbling over words. But I am not. The issue is not about words but about the concepts those words are used to express. I am simply thinking clearly about the concepts that Dawkins et al. are deploying, concepts like design.

If you tell me that design in nature is merely apparent, and that in reality nothing is designed and everything can be explained mechanistically or non-teleologically, then I understand that whether or not I agree with it. But if you tell me that there is design in nature but that the designer is natural selection, then I say that is nonsense, i.e. is unintelligible.

One cannot have it both ways at once. One cannot make use of irreducibly teleological language while in the next breath implying that there is no teleology in nature. The problem is well expressed by Stove:

. . . ever since 1859, Darwinians have always owed their readers a translation manual that would 'cash' the teleological language which Darwinians avail themselves of without restraint in explaining particular adaptations, into the non-teleological language which their own theory of adaptation requires. But they have never paid, or even tried to pay, this debt. (DF 191)

Comments

[n--blog]

Mary Midgley famously took Dawkins to task over his slippery semantics in her article 'Gene-Juggling':

http://www.royalinstitutephilosophy.org/articles/article.php?id=14

Dawkins does the same thing with the words 'selfish' and 'immortal' in reference to genes that he does with 'design', allowing him to make all these startling claims whilst retaining the ability to translate them into trivialities if anyone calls him on them.

[Bill Vallicella]

Thanks for reminding me of that article.

[Joseph A.]

Edward Feser's "The Last Superstition" also gets into this, though most of his focus is on Dennett's handling of design and natural selection rather than Dawkins'.

I think the confusion underscores a major and embarrassing truth - evolution itself (particularly the scientific theories and experiments, rather than any metaphysical and philosophical presuppositions) doesn't do much to disprove a d/Designer's reality or necessity. But it's easy to use evolution or reference to science in general to muddy the waters on the topic. For every Midgley who points out that Dawkins is playing fast and loose with key ideas, there's ten or a hundred people who get confused but just nod their heads. He must be correct - he's such a good writer, after all.

[Bill Vallicella]

Dawkins is an ideologue first and foremost. That is what is so offensive: the scientistic ideology. A chief purpose of serious philosophy is to expose this sort of rubbish for what it is.

[Court]

I have often thought that the popularizers of evolution often use slippery language to explain the process of selection. They often seem to imply that there is design in process, when of course there is none. But this stems from a sloppy use of explanatory language, as in Dawkins above, not from problems with the theory of evolution itself. Teleological notions slip into the sentences of scientists explaining evolution to us mortals in everyday language.

The problem, it seems to me, is that in using everyday language, scientists are dealing with an archaic artifact of past cultural accretions, as opposed to the clean scientific language with which they speak to each other in scientific journals and the like. It is an odd thing to do to create an English sentence wherein there is no teleology whatsoever. "Land mammals evolved from sea-dwelling creatures." This is an odd sentence, if a true one. The mind searches for its purpose. In ordinary understanding, the agent and teleology thereof must be in the verb somewhere, as in "Bob hit a football over the fence." (Also a somewhat odd sentence, but easily explicable.) So we impute this responsibility to the verb "evolved".

The very fabric of our language requires us to speak in terms of intention, purpose, agency, etc. Scientists attempting to speak to laypersons must adopt this language, the same one they themselves think in. So naturally confusion seeps in. I know a number of educated, intelligent people, who nonetheless think that the "purpose" of evolution is to create beings of greater and greater complexity. Of course no such purpose exists. Evolution is just a process that perhaps tends certain ways, but never on purpose.

By way of comparison, almost no one in Japan has problems understanding evolution. I submit that this is at least in part owing to the language. Japanese is far less strict about agency, subject-verb agreement, pronouns, and clauses of causation. "Land mammals evolved from sea-dwelling creatures" is easily rendered in Japanese, with no particular oddness. There is no particular requirement in Japanese sentence for the agent or cause or purpose to be particularly clear. Since evolution lacks all of these, it is that much easier to understand the concept in everyday language. (I would also add in Japan it was recognized long ago there that things happen for no reason other than that they happen the way they happened - maddeningly circular to us perhaps, but also a pretty good rough understanding of the process of evolution.)

You are right, Dr. Vallicella, to call Dawkins et al out for lack of clarity, but let us be clear ourselves: you are not calling the veracity of evolutionary theory itself into question. You are just calling for better communication. Which I agree should be provided.

[Brodie Bortignon]

For the purposes of popular exposition, using teleological language to describe non-teleological processes isn't bad in itself; deployed with care, it's a decent pedagogical technique. Problems arise when the teleological implications of certain metaphors (such as the selfish gene) are wielded inconsistently--one moment implying metaphor, the next reality. It then becomes difficult to know what is and is not the proper place for teleological thinking. In something like the Dawkinsian metaphysic, where teleology in all its forms is ruled out a priori, such inconsistency wreaks havoc. Notions of 'selfishness' and 'design' are no longer innocent terms of art, teaching aids that serve an introductory purpose and are then discarded when the real science begins, because the very distinction between the metaphors and the science is hopelessly vague due to systematic imprecision. Teleological metaphors go from innocent, potentially useful terms of art to non-metaphorical explanatory principles. A perfect example of teleological metaphors reified in service of an explicitly non-teleological metaphysic is the notion of a meme, which, being essentially a little packet of intentionality, is about as teleological a thing as you can get. Yet Dawkins seems to think it's somehow consistent with his thoroughly mechanistic, anti-teleological worldview.

How can teleology so deeply permeate a theory one of whose principle and most controversial claims is its utter lack of teleology? Part of the answer, I think, is that people like Dawkins (and legions of other popularisers, some of them working scientists) don't really understand what teleology is, and what a denial of teleology actually implies. If teleology doesn't exist then neither does God, at least classically construed: a conclusion Dawkins is more than happy to adopt. But without teleology there can be no genuine intentionality (among other things). By his invention of 'memetics', and his constant equivocation on words like 'design', Dawkins seems to have some inkling that his wholesale denial of teleology leads to certain evident absurdities, such as there being no such thing as intentionality at any explanatory level. Claim that God doesn't exist and you aren't saying anything particularly controversial or unexpected, at least in academia; but start claiming that nobody ever intends anything and you start to look embarrassingly extreme, like a over-enthusiastic Victorian kook. That the dominant varieties of evolutionary theory up until mid-last-century were explicitly teleological (evolution is the mechanism of progress) seems to have made it particularly hard for philosophical illiterates like Dawkins to break out of the teleological thinking they so vigorously deny yet so criminally and opportunistically assume.

It's one of the many reasons I barely read pop-science anymore. There is so little science and so much crappy philosophising. And I don't think it's ever going to go away. Grand, shocking theories sell books, no matter how bone-headed and simplistic they may be. Careful discussion and sober, tentative conclusions don't stir the interest of a public who read pop-science for gee-whiz factoids to shock their friends. Chesterton had more insightful things to say about science in his 'Ethics of Elfland' than almost the entire output of modern popular science put together. Contempt for philosophy is the death of good science, and Dawkins is a perfect example.

[Joseph A.]

The idea that evolution just happens and that "no such purpose exists" for this or that evolutionary development - or even for evolution as a whole - is an idea that necessarily has one stepping right outside the bounds of science and right into philosophy. And philosophy is a fine and grand thing. But so is science - and the limits of both should be recognized and respected.

[Joe]

Bill, as always, thanks for the blog!

Stephen Jay Gould:

The Darwinian principle of natural selection yields temporal change—"evolution" in the biological definition—by a twofold process of generating copious and undirected variation within a population, and then passing only a biased (selected) portion of this variation to the next generation. In this manner, the variation within a population at any moment can be converted into differences in mean values (such as average size or average braininess) among successive populations through time.
"What Does the Dreaded 'E' Word Mean Anyway?;" p. 246; I HAVE LANDED (2002) Harmony Books, New York

I personally have never understood why being gay can be claimed to be a result of evolution -- any "gay" genetic material wouldn’t survive in a population as it is housed in a non reproducing body.

[Teague Tubach]

I'm not totally clear. You quoted Dawkins: "His hypothesis was that living watches were literally designed and built by a master watchmaker. Our modern hypothesis is that the job was done in gradual evolutionary stages by natural selection." Did you think that "the job" in the second sentence meant "design"?

[David Brightly]

Bill,

In this subject the word 'design' has caused far more trouble than it's worth. A search through the text of the "Origin" at http://darwin-online.org.uk for the word 'design' produces 9 hits. A search for the word 'form' produces 650 hits. Darwin uses 'form' three times on the last page: http://www.flickr.com/photos/cpurrin1/73530276/ . My suggestion is that if we want an alternative to 'design' we should return to Darwin's own word. It has the advantage of few if any teleological connotations. The Himalayas were formed by tectonic forces. There is no implication that there need be a 'former'.

[Bill Vallicella]

David,

I take it you agree with me that if x is designed, then there is a designer, and that it is silly to speak of natural selection, or Mother Nature, or evolution as the designer.

But 'form' is not an alternative to 'design' since 'No form without a former' is false whereas 'No design without a designer' is true.

[Bill Vallicella]

Teague,

>>Did you think that "the job" in the second sentence meant "design"?<< Yes, think I made that clear.

[David Brightly]

I agree entirely, Bill. I urge that we simply drop the word 'design' in this context. When we need a word for 'design' (noun) meaning plan, shape, structure, etc, we just use 'form'. And we avoid like the plague locutions involving 'design' as a verb, such as 'natural selection designs new species'. It's sloppy and leads to trouble. As has 'selfish'.

[Bill Vallicella]

David,

Part of the problem here is that 'design' is ambiguous. As you say it could mean shape, structure, pattern. "Look at the interesting design on the back of that spider; it resembles an hour glass." In this sentence, 'design' does not imply a designer. But in the other, the relevant, sense there is no design without a designer.

'Plan,' however, would seem to imply a planner.

[Malcolm Pollack]

Again: the problem is simply that there isn't a suitable word in English for a process that creates superbly "designed" things (I even have to use the word here myself, because there really isn't another), but does so in the absence of a conscious designer. This is due to nothing more than an old word's being irrevocably bound to a pre-Darwinian understanding of how the world works.

*Obviously* things like kidneys, wings, eyes, brains -- in other words, the bodies of living things -- are astonishing examples of what we would surely call "design" or "engineering" if they were cooked up by a human designer. *Obviously* there is something remarkable going on here, something that we ought to be able to discuss without getting caught up in niggling terminological disputes every time we try. So what do we call it? We have two simple choices. Either we:

1) expand the definition of "design" to include what evolution does, or;

2) we come up with a new word.

I despair of ever successfully modifying the definition of "design", so I'm all for 2). I suggested "freesign" in here a year or two ago, but I never even liked the coinage that much myself. Surely we can clear up this mess, though; it is a useless distraction for us to bicker about it so.

[Malcolm Pollack]

Amendment: not a "useless" distraction - it's good to get areas of confusion and murk out into the light - but now that we've identified the problem, it would be good to be able to correct it and move on.

[bobkoepp]

I've never liked "design talk" in the context of natural selectionist explanations, and I think we do have a ready to hand substitute, namely, 'adaptation' and its cognates. People have even offered refinements, such as 'exaptation', to cover special cases.

[M Temples]

I recently purchased a some tiles made from "Kaibab Wonder Stone." The sandstone is strikingly patterned in various. These colors make designs, and thus I can say they are designed; and even that they were designed (in the sense of patterns) by geological processes. We can say that a thing is designed without also implying that it was an intentional design. In that same sense, evolution produces design without intent. I don't think the use of the word design in the sense of pattern is an equivocation, though restricting its use to intentional design may be somewhat pedantic.

[bobkoepp]

Are there any "takers" out there for my suggestion that we might avoid some wrangling if we speak of adaptations instead of designs? Note that there is nothing at all queer about saying that wings are adaptations for flying, eyes are adapted for seeing, hearts for propelling blood, etc. The use of 'for' in these examples doesn't seem to imply anything about goal directedness, but does point in the direction of features of wings, eyes, hearts, etc., that we think were relevant to their evolution.