| Resolving a Controversy or Shoring up a Belief System? |
The authors failed to realize that bee language advocates have unwittingly undermined their case these past three decades, due to their repeated attempts to prove their belief system true (sort of like "crying wolf"). A single "proof" should have sufficed - multiple attempts at "proofs" implicitly reveal/admit failures of earlier proofs. The absolute certainty expressed by the authors reminds me of a mid-1960s visit to our campus by Ronald Ribbands, esteemed bee researcher from Britain and author of the influential 1953 volume, The Behaviour and Social Life of Honey Bees. He spent several days with us while we conducted some of our double-controlled experiments (as summarized in Ch. 9 of our 1990 Columbia University Press book - Anatomy of a Controversy). Ribbands observed the conduct of our experiments with great interest, saw no problems with their design or execution, but could not accommodate those results within his grasp of reality, results that contradicted implicit expectations of the bee language hypothesis. We debated the implications but could not agree on interpretation of those clear-cut results. Finally I thought to ask, "Do you think it conceivable that honey bees do not have a "language"? He responded immediately, "No, that is not possible." We had a great time together during the remainder of our visit, because we understood that we "lived in different worlds" on the matter. Neither of us had a question about the evidence but differed on our attitude toward that evidence. It eventually became obvious that, by the time of the Ribbands' visit to our campus, the bee language hypothesis had already evolved into dogma - a strong paradigm hold or "fact," as it were. A decade ago one of the "robot bee" researchers from Germany visited my office. He had brought a number of reprints that we went over together. As we looked at one reprint, I pointed out that the results in one of his tables of data did not conform to expectations of the language hypothesis. My comment fell on deaf ears; instead, he said, "Here, let me show you some other results." His attitude remained dismissive when I showed him results obtained by both bee language advocates and by us that clearly did not fit the language hypothesis. In 1935 Ludwik Fleck published a book, Genesis and Development of a Scientific Fact (later translated into English). In that text he used the phrase, "thought collective" to denote a community of scientists who mutually exchange ideas or maintain intellectual interaction. Such collectives then become intellectual prisons of sorts, because, in Fleck's words, "The individual within the collective is, never, or hardly ever, conscious of the prevailing thought style, which almost always exerts an absolutely compulsive force upon his thinking and with which it is not possible to be at variance." Later, Thomas Kuhn (in his 1962 book, The Structure of Scientific Revolutions) expanded upon Fleck's ideas with his use of the terms paradigm hold and paradigm shifts. That is, scientists unwittingly become locked into a prevailing theory to such an extent that they can no longer recognize other alternatives. As Fleck emphasized, social constraints also play a large role in that fixation to prevailing thought. Fortunately, we had become aware of Kuhn's writing at the time we were first gaining evidence at variance with the dance language hypothesis and had begun to appreciate the rough road ahead if we persisted with our "heretical" objections to that hypothesis. At the same time we also began to appreciate that this controversy had the potential to become a classic case of how science proceeds - as against notions many scientists hold about how they conduct their research. However, we had no idea of the intensity of opposition we would face, even from such esteemed thinkers as E.O. Wilson (in 1969 and 1972) and Richard Dawkins (in 1970). Robert Park wrote in his 2000 book, Voodoo Science, "The more persuasive the evidence against a belief, the more virtuous it is deemed to persist in it." That is why those who endorse prevailing thought (dogma) gain accolades from the scientific community. In 1975 Paul Feyerabend assessed the situation as follows: "Popular science books spread the basic postulates of the theory; applications are made in distant fields, money is given to the orthodox and is withheld from the rebels. More than ever the theory seems to possess tremendous empirical support." It seems that the authors of the recent ABJ two-part series had begun (and had conducted) their review of the literature under the dance language paradigm hold. They summarized all the positive evidence they could glean from various contributions but failed to provide a balanced summary of all other available evidence. In doing so, they failed to see the forest for the trees; namely, any true resolution must be inclusive of all available evidence, not merely an advocacy of a particular hypothesis. Science philosopher Karl Popper addressed that point: "It is easy to obtain confirmations, or verifications, for nearly every theory - if we look for confirmations." Popper also wrote, "A theory which is not refutable by any conceivable event is non-scientific. Irrefutability is not a virtue of a theory (as people often think) but a vice." The authors' choice of the word "resolution" was thus indeed unfortunate, as were their phrases, "the dance language" and "functionality of the dance language." Those expressions reveal a deep-seated bias, an eagerness on their part to establish bee language as "irrefutable truth," never again to be open to challenge. In that approach, they followed in the same vein as that of sociologist Eileen Crist ("Can an Insect Speak? - The Case of the Honeybee Dance Language"). Her paper was written under the tutelage of James Gould and Donald Griffin, both long term committed advocates of bee language. The authors' use of the word "resolution" would imply that we now have a theory of honey bee recruitment to food sources, a theory that accommodates all available results (the meaning of the word), but that is clearly not the case. That word also suggests that one must no longer challenge "prevailing" theory - one must believe in bee language as fact - that a mere insect can possess such an extra natural ability. Notice that the authors used the term, "the dance language" throughout their presentation (as was true in the Crist article). Note here, then, an important distinction - "the dance maneuver" is fact (it does occur), but "the dance language" is interpretation (for which there is much supportive evidence but also much compelling negative evidence). Furthermore, one cannot merely dismiss or ignore negative evidence not in keeping with a hypothesis. Neither does an "overwhelming body of positive evidence" make a hypothesis irrefutable, especially when a large body of negative evidence remains unexplained. The term, "functionality of the dance language," is by far the most interesting aspect of their presentation. Whenever I have given a talk on our findings about the overwhelming importance of odor during recruitment of naïve bees to food sources (the Occam's razor approach), the first question raised during the discussion period has nearly always been, "But, then, why do bees dance?" The authors apparently now seem convinced that they have resolved that teleological question. It would appear, in their minds, that the bee dance maneuver has distance and direction information in it "in order to send hive-mates to food source locations in the field." It is here that one can recognize a striking parallel between the bee language/odor-search controversy and the "intelligent design"/evolution controversy. Both are characteristic of belief systems - once something is known to be true, that idea is no longer open to question. That is, intelligent design people (creationists) claim that life is too complicated to have arisen by normal chemical/biological processes. Bee language advocates claim that the dance maneuver is too complicated to not have some "purpose" and could not be a functional-less activity (presumably "intelligent design" or its equivalent). The dance language hypothesis rests upon an assumption that honey bees have special capabilities above and beyond those possessed by all other insects. The odor-search hypothesis of von Frisch and later re-formulated by us requires no special equipment on the part of bees. If we apply Occam's razor to the question, there is no contest. Is, then, the controversy "finally laid to rest" or "resolved" as Crist and the current authors have claimed? No, although a bee language belief system remains firmly entrenched in some quarters (as elucidated by Feyerabend, above), an accumulation of negative evidence from many experiments continues to erode unquestioning faith in bee language. By contrast, attention to the overriding importance of odor during recruitment of bees to food sources should eventually yield rich rewards not anticipated by bee language advocates - rewards that have remained elusive under the dance language paradigm these past several decades. Adrian M. Wenner |