Excursus TEL: Teleology

Anatomy of a Controversy : Pages 362 – 366

“Teleological explanations, then, are essential to biology. They imply that the parts, processes, and behavior patterns of living things are organized so as to attain specific goals, which contribute as a rule to the ultimate goal of reproductive fitness.” —George F. Kneller 1978:146

“The nature of our mind leads us to seek the essense or the why of things….experience soon teaches us that we cannot get beyond the how, i.e., beyond the immediate cause or necessary conditions of phenomena.” —Claude Bernard (1865) 1957:80

“But the greatest fallacy in, or rather the greatest objection to, teleological thinking is in connection with the emotional content, the belief. People get to believing and even to professing the apparent answers thus arrived at, suffering mental constrictions by emotionally closing their minds to any of the further and possibly opposite `answers’ which might otherwise be unearthed by honest effort” —John Steinbeck (1941) 1962:143

Most nonscientists and many contemporary biologists, particularly students of animal behavior, are quite comfortable with Kneller’s position on the issue of teleology. They want to know “why” an animal does this thing or that in a given situation.

For example, one can ask, “Why do geese fly south for the winter?” “To keep warm” is a disarmingly satisfying answer! This explanation of biological phenomena in terms of purposeful or goal directed behavior has wide appeal, but it leads nowhere in scientific investigation. Claude Bernard (see the epigraph) early on recognized that severe limitation in teleological approaches.

We often encountered a teleological pattern of thinking when we offered an “odor search” answer to the question: “How do naive bees find a food source to which they have been recruited?” Our audiences appeared to be uncomfortable, not so much with our answer as with our question! “But why then”, they wanted to know, “do they dance?” Indeed, von Frisch himself objected, as follows: “The reason why these complicated and ingenious behavioural patterns could evolve and be a functionless repertoire remains undiscussed [by Wenner and Wells]” (1973:626).

Martin Lindauer even more forcefully articulated the teleological argument in the opening sentences of his challenge of our work. He wrote: “each morphological structure and behavioral act is associated with a special function” (1971:89). Although his assertion is open to question, it is not necessarily teleological; if true, it could be merely descriptive.

However, Lindauer continued that passage with a teleological nonsequitur: “On this basis alone, it would seem highly unlikely that information contained in the waggle dance of a honeybee is not transmitted to her nest mates” (1971:89). Lindauer implied by that reasoning that design or purpose was inherent in the system and that such purpose would mesh with what human beings would do in like circumstances (an anthropomorphic conclusion).

We later cited Hempel (1966) and Popper (1957) when we replied to Lindauer’s use of: “Aristotelian (Darwinian) teleology as an argument in favour of language.” We did that “in part by discussing the relevant data in terms of philosophy of science which we consider to be more powerful” (Wells and Wenner 1973:175).

We also offered a disproof by counterexample of the notion that “each behavioral act” is necessarily adaptive. We described Steiner’s (1952) use of methyl eugenol as an attractant for oriental fruit flies in Hawaii, and noted:

[Methyl eugenol] is not a component of the natural food of this fly and probably has no nutritional value. Yet male oriental fruit flies are irresistibly attracted to it and “apparently cannot stop feeding when they have free access to it, and they kill themselves with over indulgence” (Wells and Wenner 1973:175)

We could have added that the flash rate of fireflies (Family Lampyridae) contains ambient temperature information; however, that fact does not lead to the conclusion that fireflies are communicating information about temperature to one another.

We further indicated that this example reveals: “a weakness in the teleology argument. The mere presence of a characteristic behavioural pattern in an animal cannot be construed as purposeful, adaptive or `associated with a special function’” (Wells and Wenner 1973:175).

That particular debate might have ended there, but Anthony Ferguson (1975) took exception and came to the defense of von Frisch and Lindauer. Ferguson, in turn, was challenged by Jack Hailman (1977), who was then rebutted by Ferguson (1977). Elaboration follows.

First, Ferguson transformed von Frisch’s “why” into “what for?” (a semantic disguise), posed a question, and gave his own answer: “Question: what is bee dance information for? Answer: To find food more efficiently. Even expressed in these terms, teleology is not involved….[Such] ‘what for?’ questions are merely a convenient verbal shorthand”, Ferguson argued. He then continued, “While the linguistic form…is teleological, its conceptual content is not” (Ferguson 1975:369).

Hailman replied that we (Wells and Wenner 1973) had actually addressed the points raised by Lindauer (1971), since von Frisch’s 1973 article had not yet been published when we wrote our paper. Hailman argued that Lindauer’s position was vulnerable to several epistemological points inherent in our comments, as follows:

  1. Wells and Wenner (1973) are denying that just because information is encoded in dances it must be transmitted, as implied by Lindauer.
  2. [Wells and Wenner] are also arguing that it is teleological to see communicative design “apparent” in dancing.
  3. it is irrelevant whether the teleology is naively Aristotelian or framed in Darwinian language–it is still incorrect to “see” communicative design apparent in dancing (Hailman 1977:187).

Ferguson (1975) continued with a challenge of our “disproof by counterexample.” He argued that, since Steiner’s fruit flies had never encountered methyl eugenol in nature, their reaction to it need not be considered “nonadaptive.” He felt that fruit flies may react to methyl eugenol (as do “mammals to saccharine”) because they perceive it to be similar to a relevant environmental stimulus. Ferguson wrote, “The foundations of the response could thus ultimately be adaptive.”

This last argument by Ferguson, however, fell to the next point in Hailman’s discussion, the citing of a naturally occurring counterexample:

4. Yet Blest (1960) discovered that the postflight rocking movements of saturniid moths also contain information concerning the locus from which they have flown. The adaptive significance, if any, of this behavior is unclear, and no one has suggested that the inherent information is transmitted to other moths” (Hailman 1977:187,188).

Thus, Hailman (1977) refuted Ferguson’s objections largely with arguments we had offered in the first place–but much more clearly stated and with a far better example of nonadaptive (in this case, neutral) behavior.

The remainder of Hailman’s article dealt with two questions: 1) whether von Frisch’s statement (quoted above) really was teleological, and 2) whether we were obliged to offer an alternative answer to Ferguson’s “why” question. He concluded both that “it was” and “we weren’t.” Ferguson’s (1977) rebuttal relied largely on semantic arguments dealing with what we had said (or presumably had meant to say) and included an attempt to trivialize Hailman’s “moth” counterexample.

Apparently neither author succeeded in “converting” the other in the sense of Atkinson (1985). Neither did the inconclusive debates between Lindauer and us, then by Ferguson and Hailman, and later by Rosin (1988a,b) and Walls (1988), below, settle the issues. The conceptual foundations of the disagreement both preceded and survived those published discussions.

As noted by Michael Ruse (1973), teleological explanations are a carry-over from pre-Dawininan times “when the dominant biological paradigm was the Argument from Design (for God’s existence).” Ruse continued:

Thus, [philosopher William] Whewell could write that “each member and organ not merely produces a certain effect [but] was intended to produce the effect. . . . each organ is designed for its appropriate function. . . . each portion of the whole arrangment has its final cause; an end to which it is adapted, and in this end, the reason that it is where and what it is. . . . Here it is clear that things are being explained in terms of what we would call their effects–modern evolutionists can and do do the same thing, . . . because of the quasi-Design effect selection has. (See Young 1971, for a discussion of how pre-Darwininan thought found its way into the Origin and later biological works.) (Ruse 1973:196; see also Whewell 1840, 2:79-80).

James Gould (1975b), writing at about the same time as Ferguson, approached the teleological argument more carefully. “It has been argued that the dance correlations must be useful, or they would not exist”, he wrote (1975b:686). He then suggested: “In the case of the dance language [emphasis ours], the evolutionary argument appears particularly persuasive.”

However, Gould recognized the significance of “disproof by counterexample” and included several relevant instances in addition to those cited by Hailman and by us. By then, however, he had unwittingly implied that “the dance language” was a “fact.”

The pragmatic argument that teleological thinking may actually inhibit research was at the root of Steinbeck’s thinking earlier (see epigraph). Gould also advocated caution “lest we glibly explain away phenomena and inhibit research” and forcefully articulated: “Used as a basis of proof or as an article of faith [the teleological argument] can merely reinforce our predispositions, stifling research by making further work seem unnecessary” (1976:238).

Did any of the above debates and arguments make a lasting difference in the dance language controversy? Perhaps a little, but the teleology argument is still raised by proponents of the dance language hypothesis– by Roland Walls, for instance, who wrote: “One has to assume that in the competitive world of the honey bee there is survival value to all patterned energy expenditure. Karl von Frisch and his colleagues showed that there was meaning and predictability in those gyrations bees do” (1988:576).

In spite of Rosin’s observation (1988b:576) that “this [teleological] argument is probably the one most important cause for the quick success of the ‘dance language’ revolution” (see also our chapter 6), that argument has lost some of its force. Rosin continued:

By now both staunch opponents of the “dance language” hypothesis (Wenner et al. 1967; Wenner 1971; Wells & Wenner 1973; Wenner 1974), as well as staunch supporters (Gould 1975[b], 1976) have, for good reasons, invalidated the Teleological argument, and concluded that the dances may have a yet unknown adaptive value, or no adaptive value at all. (1988b:576,577)

Thus it is that, to this day, the issue of teleology remains an important one in the dance language controversy, in behavioral ecology, and in other important fields of biology. Perhaps that is why Hailman (after expressing confidence that bees had a “language” after all) ended his article with the observation: “the underlying epistemological issues…in some respects transcend in importance [the] scientific [dance language] controversy” (1977:188).

We agree with this last assessment, of course; that is why we wrote this book. We further suspect that the current devotion to teleological explanations by some biologists is but one more episode of a pendulum swing between adoption and rejection of that type of explanation.