Teleological Explanations in Biology

Biologists tend to use teleological language in explaining the origin and evolution of living organisms and their characteristics. As John Reiss has pointed out (Reis, J. Not by Design: Retiring Darwin's Watchmaker, University of California Press, 2009), this entails the idea that evolution is necessarily a teleological process. This entails the idea that evolution is not a "natural" process, like gravity or oxidation, and that therefore there is some "non-natural" component (i.e. "magic") in biology that fundamentally distinguishes it from the other natural sciences.

Evolutionary biologists like Richard Dawkins try to make this distinction when referring to the problem of human free will (see "Let's all stop beating Basil's car"), but unless they are careful about how they talk about evolution (especially natural selection) they revert to the same teleological descriptions and explanations that Reiss so decries. What is the problem, here?

I believe that the underlying problem is the tendency by most evolutionary biologists to think of natural selection as a "force" or "mechanism". As John Endler has pointed out (Endler, J. Natural Selection in the Wild, Princeton University Press, 1986), natural selection is not a "force" or "mechanism", it is an outcome. To be precise, it is the outcome of four separate, but related processes:

• Variety: structural and functional differences between individuals in populations,

• Heredity : the inheritance of structures and functions from parents to offspring (either genetically or epigenetically),

• Fecundity : the ability to reproduce, especially (but not necessarily) at a rate that exceeds replacement, and

• Demography : some individuals survive and reproduce more often than others.

As a result of these four processes, the heritable characteristics of some individuals become more common in populations over time.

Notice that the same list of processes can be used to explain non-adaptive evolutionary change (e.g. genetic drift). Also notice that the only source of new phenotypic variations is what I have called the "engines of variation": all of those processes that produce heritable phenotypic changes in phylogenetic lines of organisms in populations. There are at least fifty such processes (you can see a summary list here). While it is the case that "random mutation" is included in this list, there are many other processes in this list that do not involve "mutation" (in the genetic sense) and which also are not "random" (at least insofar as that term is often used).

Is there a real distinction between non-teleological and teleological processes, or are all processes either teleological or not? If all processes (i.e. changes over time) are teleological, as asserted by Aristotle (and some of the commentators), then there is no point in talking about it. However, if some processes are teleological and some are not (as most people, including presumably most of the commentators here, now believe), then the question becomes "how can one distinguish between teleological and non-teleological processes, and what explains the differences between them?"

In his comprehensive analysis of teleology, Andrew Woodfield (Woodfield, A. Teleology, Cambridge University Press, 1976) pointed out that all teleological descriptions can be reformulated to conform to the linguistic formula " x happens in order to/for y outcome." He also asserted that such linguistic formulations describe metaphysically real processes. That is, some processes are genuinely teleological – they involve pre-existing designs or plans that cause processes to tend toward particular outcomes, regardless of perturbations or outside interference – while other processes only seem teleological – they involve laws of nature, such as gravity, that result in particular outcomes, without responding actively to perturbations or outside interference.

This distinction is essential when considering whether "genuine" teleology exists. To be precise, teleological descriptions sound "reasonable" when they are applied to genuinely teleological processes, but sound ridiculous if they are applied to non-teleological processes. For example, does it sound reasonable to say that when you drop a rock, it falls "in order to" reach the ground? By contrast, does it sound reasonable to say that birds have wings "in order to" fly? Is there a difference between the "reasonableness" of the first teleological explanation and the second?

When I pose this question to my students, almost all of them say yes: the first is ridiculous and the second isn't. I then point out that this implies that the origin of the wings of birds therefore seems to be the result of a teleological process. I then point out (reprising Aristotle) that there are at least four ways of explaining the presence of wings:
• "this bird has wings because it is composed of materials that are assembled and operated as wings" (Aristotle's "material" cause);
• "this bird has wings because its parents had wings" (Aristotle's "efficient" cause);
• "this bird has wings because birds have wings" (Aristotle's "formal" cause); and
• "this bird has wings in order to fly" (Aristotle's "final" cause).

Since at least the 17th century (and mostly because of Newton), natural scientists have stopped using formal or final causes to explain natural phenomena...except in biology. This was first pointed out by Colin Pittendrigh (Pittendrigh, C. S. Behavior and Evolution) (ed. by A. Rose and G. G. Simpson), Yale University Press, 1958), who coined the term "teleonomy" to refer to the kind of teleological phenomena observed in biological processes. Francisco Ayala modified and extended Pittendrigh's analysis (Ayala, F. J. 'Teleological Explanations in Evolutionary Biology', Philosophy of Science, vol. 37 pp. 1-15, 1970). Ernst Mayr finally sorted the whole thing out in 1974 in "Teleological and teleonomic: A new analysis" (Boston Studies in the Philosophy of Science, vol. XIV, pp. 91 -117). According to Mayr, the difference between the "behavior" of dropped rocks and genuinely purposeful processes is the presence or operation of a pre-existing information-containing program in the latter. Rocks do not fall because there is an encoded program in nature that makes them fall. They fall because there is a force (i.e. a law of nature) that causes them to fall. However, a bird has wings because there is a program encoded within its genome which, as the result of interactions between the "phenome" of the bird and its environment, causes the construction and operation of wings.

To say that natural selection is teleological would therefore require that there be a pre-existing encoded program somewhere that would cause natural selection to bring about its effects. This is ridiculous for at least two reasons:
• there is no such program as far as we can tell (where would it be encoded?), and
• this would require that natural selection be a process in and of itself, rather than the outcome of the four processes listed above.

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

As always, comments, criticisms, and suggestions are warmly welcomed!

--Allen

Teleology vs Teleonomy: Can a "Program" Exist Before Its "Programmer"?

This post is a follow-up to the previous post on the subject of the "randomness" in the processes that generate the variation that is necessary for biological evolution.

Can a "program" exist before its "programmer" (and therefore bring it into being)? This seems to be the core of the disagreement between ID supporters and mainstream scientists. The former (which include Charles Darwin's very close friend, Asa Gray) advocate the idea that the variation upon which natural selection and other evolutionary processes work is neither random nor unintentional. The latter (which include Darwin and his intellectual heirs) do not disagree with the idea that such variation is not "random". What they disagree with is the idea that there is some "intention" or "plan" guiding the variation that occurs, so that certain outcomes (including, but not limited to, the origin of humans) are more likely than others.

To answer the question that stands at the head of this post, I think it's essential to emphasize (as I did in the original blogpost) that the terms “foresighted” and “goal-oriented” are not equivalent, nor are the processes to which they are applied. As I have pointed out in many posts, there is no inherent contradiction between a process being purely "natural" (i.e. the result of the operation of purely natural processes) and being "goal-oriented".

Ernst Mayr (surely no advocate of "intelligent design") argued forcefully (and, in my opinion, convincingly) that biological organisms are indeed "goal-oriented". That is, their genomes provide a program, the function of which is to bring about a particular state of affairs: the survival and reproduction of the organism via its interactions with its environment.

The origin of the genome (i.e. the "program" itself) is an entirely different situation, however. Ever since Darwin it has been a standard assumption that the evolutionary processes by which the genetic "programs" that direct the assembly and operation of living organisms are not goal-oriented. These evolutionary processes – natural selection, sexual selection, founder effects, genetic bottlenecks, neutral "drift" in deep evolutionary time, exaptation, heterochronic development, changes in homeotic development, interspecific competition, species-level selection, serial endosymbiosis, convergence/divergence, hybridization, phylogenetic fusion, background and mass extinction/adaptive radiation, and internal variance – do not require any kind of "goal orientation" to produce the living entities and processes we observe around and within us. And, since such processes do not require goal-orientation or intentionality, these are not included in evolutionary explanations. Some, but not all, evolutionary biologists extend this idea to the assumption that goal-orientation or intentionality do not exist in nature, in the absence of pre-existing genomic "programs").

The main reason for this assumption has been that it is extremely problematic to agree on how one would go about showing that the aforementioned evolutionary processes have indeed been goal-oriented. The most serious objection to this idea is that there seems to be nowhere for such a "directing agency" to exist in material form, nor any natural means by which its goals could be impressed upon physical organisms.

The genomes of organisms are physical/chemical "stuff", which is translated via physical/chemical "machinery" into biological entities and processes. That is, there is a physical/chemical "vehicle" in which the information for assembling and operating organisms is carried and expressed.

The same would not the case for the putative source of the "evolutionary program" which might direct the evolution of the genomes of living organisms. Since such an "evolutionary program" would cause the evolution of the "genomic programs" which direct the assembly and operation of living organisms, such a program would necessarily have to exist before the origin and evolution of biological genomes, as it would be necessary for it to do so to direct their coming into being.

This presents two serious problems:

• By what mechanism(s) would such an "evolutionary program" cause "genomic programs" to come into existence, and

• Precisely where in the physical universe would such a pre-existing "evolutionary program" itself exist?

We seem to have two direct logical contradictions in terms:

• How can a non-natural "evolutionary program" cause a natural "genomic program" to come into existence, and

• How can a programmer pre-exist the program which brings itself into existence?

There is a proposed answer to these two questions, but one which most ID supporters seem loathe to invoke:

• That the "pre-existing program" that directs the evolution of the genomic programs of living organisms is woven into the structure of physical reality itself.

This is the line of inquiry pursued by Ilya Prigogine and Stuart Kauffman (among others), but which is rejected out-of-hand by nearly all ID supporters (most notably Michael Behe, William Dembski, and Phillip Johnson), who prefer a purely "supernatural" source for the "pre-existing program" by which evolution has been directed).

As always, comments, criticisms, and suggestions are warmly welcomed!

--Allen

Teleological and Teleonomic: A New(er) Analysis

AUTHOR: Ernst Mayr

SOURCE: Boston Studies in the Philosophy of Science, Volume XIV, pages 91 -117

COMMENTARY: Allen MacNeill

On April 17, 2006 04:35 PM David B. Benson wrote in the Panda's Thumb (scroll way down):

[Can you] explain why you state that ant colonies, cities, and other emergent organizations have purpose? In particular, what is wrong with mere teleomentalism, that an ascription of purposeful behavior is only metaphor?

Ernst Mayr [1] distinquished between two different kinds of natural processes that appear to be “goal directed”:

Teleomatic processes: Processes that simply follow natural laws, i.e. lead to a result consequential to concomitant physical forces, and the reaching of their end state is not controlled by a built-in program. The law of gravity and the second law of thermodynamics are among the natural laws which most frequently govern teleomatic processes. Examples include the cooling to ambient temperature of a red hot bar of iron and the falling of a rock to the ground.

Teleonomic processes: Processes that owe their goal-directedness to the operation of a program. The term teleonomic implies goal-direction. This, in turn, implies a dynamic process rather than a static condition, as represented by a system. Examples include the development of an adult organism from a fertilized zygote and the building of a dam by beavers.

Mayr argues very strongly that the common use of teleological language by biologists is legitimate because it recognizes the goal-directedness of biological processes. He also stresses that, although many biological processes (such as ontogeny) are clearly goal-directed, they owe their goal-directedness to the operation of programs, such as the genetic program encoded in the DNA. He concludes that although such programs themselves are goal-directed (i.e. purposeful), the process by which such programs have come into being – evolution by natural selection – is NOT itself goal directed.

[ I would state this slightly differently from Mayr: that there is no observable evidence that the evolutionary processes by which such programs come into being are goal-directed (i.e. “designed” or “purposeful”). Therefore, although such purposes may exist, they are invisible to us on principle and therefore irrelevent to scientific explanations of natural phenomena.]

Mayr concludes:

• The use of so-called teleological language by biologists is legitimate; it neither implies a rejection of physico-chemical explanation nor does it imply non-causal explanation

• At the same time, it is illegitimate to describe evolutionary processes or trends as goal-directed (teleological). Selection [reifies] past phenomena (mutation, recombination, etc.), but does NOT plan for the future, at least not in any specific way [as far as we can tell]

• Processes (behavior) whose goal-directedness is controlled by a program may be referred to as teleonomic

• Processes which reach an end state caused by natural laws (e.g. gravity, second law of thermodynamics) but not by a program may be designated as teleomatic

• Programs [of the type described above] are in part or entirely the product of natural selection

• Teleonomic (i.e. programmed) behavior occurs only in organisms (and man-made machines) and constitutes a clear-cut difference between the levels of complexity in living and in inanimate nature [i.e. they are “emergent properties” of living systems, not present in the non-living materials of which living organisms or their artifacts are composed]

• Teleonomic explanations are strictly causal and mechanistic. They give no comfort to adherents of vitalistic concepts [including supporters of “intelligent design,” if such supporters believe that the kinds of programs desctibed above come into existence as the result of a purposeful process]

• The heuristic value of the teleological Fragestellung makes it a powerful tool in biological analysis, from the study of the structural configuration of macromolecules up to the study of cooperative behavior in social systems.

COMMENTARY:

I agree with Mayr on virtually every point. In other of his publications, Mayr argues strongly for the idea that biological systems exhibit “emergent properties,” and that this is one of the primary differences between biology and the other natural sciences, such as physics. At the same time (and contra some supporters of “emergent properties,” such as Andrew North Whitehead and Teilhard de Chardin), Mayr argues very strongly for the naturalist position that such properties are well within both the darwinian paradigm and the tradition of naturalistic explanation that underlies the natural sciences.

Given the foregoing, I therefore believe that the kind of teleonomy exhibited by ants and ant colonies is indeed a natural property of such biological systems, and not just a “teleomentalism” (i.e. a semantic distinction rooted in human language and cognition, having no actual reality in nature). Furthermore, it seems to me that the kinds of advances that we have seen since Mayr wrote his paper – developments in artificial intelligence and the programming of cybernetic “expert systems”, advances in genetics, and especially a much deeper understanding of the processes of evolutionary development – lend support to Mayr’s analysis, and that using Mayr’s theoretical framework can not only assist people working in the aforementioned fields, it can also lead to some clarity in understanding the origin and evolution of purpose in nature.

This analysis leaves us with the following problem: Is the term "teleology" an umbrella term that subsumes both teleomatic and teleonomic processes, and if so, what term is most appropriate for the kinds of unambiguously goal-directed behavior exhibited by humans (and our artifacts, such as heat-seeking missiles)? My preference is to reserve the term "teleology" for the latter (i.e. clearly goal-seeking processes initiated and controlled by rational entities, such as ourselves), and apply the terms teleonomy and teleomatic the way Mayr suggests in his article.

I believe that these distinctions clearly and unambiguously distinguish between natural processes that appear to be (or even are) goal-directed, such as ontology, and processes that are not, such as natural selection and other evolutionary mechanisms.

And yes, before an ID supporter points out the obvious, the foregoing analysis doesn't eliminate the problem of evolution versus ID, since ID theorists can still argue that God (remember, I asserted a while back that I will use the proper/role name "God" instead of "the intelligent designer"...it requires fewer keystrokes and is more honest IMHO) both "designed" and "guided" the processes by which Mayr's "programs" bring about teleonomic goal-seeking behavior.

But that discussion will have to wait for another post...

REFERENCES CITED:

[1] (Mayr, E. (1974) “Teleological and Teleonomic: A New Analysis.” Boston Studies in the Philosophy of Science, Volume XIV, pages 91 -117),