Convenient fictions or theoretical entities in Bateson and Pearson
Pearson distinguishes the real from unreal entities in a footnote:
The division between the real and the unreal, and again between the real and ideal, is less distinct than many may think. For example, the planet Neptune passed from the ideal to the real, but the atom is still ideal. The ideal passes into the real when its perceptual equivalent is found, but the unreal can never become real. Thus the concepts of the metaphysicians, Kant’s thing in itself or Clifford’s mind stuff, are in my sense of the words unreal (not ideal), they cannot become immediate sense-impressions, but the physical hypotheses as to the nature of matter are ideal (not unreal), for they do not lie absolutely outside the field of possible sense-impressions. (Pearson 1911, p. 41fn1)
This footnote seems to set out Pearson’s radical positivism, but the concepts do require some unpacking. He indicates that for something to be real it must be immediately perceivable by our sensory modalities. Ideal things are not perceivable by our sensory modalities, but we may infer they exist from the conjunction of really existing things. For instance, it was inferred that Neptune, the eighth planet, existed from certain regularities observed in the other planets. Neptune was an ideal thing until it was verified to exist by the use of a powerful telescope. The unreal can never become real since they cannot be verified perceptually. Pearson relegates Kant’s noumena and Clifford’s mind stuff to such a realm. For nothing will ever instantiate either concept.
This is in consonance with what Pearson said about statistical methods to describe heritable variation. Pearson assumes that units of heredity are unreal things. They do not exist, and nothing will ever affirm their existence. It would seem that Pearson should agree to the existence of Mendelian factors because of Bateson, Tschermak and DeVries’s experiments, but, curiously enough, he does not. Pearson forthrightly rejects the existence of Mendelian factors. The peculiarity of Pearson’s position is this: why should he say that an atom is an ideal thing, which means it is not accessible to our sensory modalities now but may be accessible later, while he says that a Mendelian factors is an unreal thing, which means it cannot be accessible to our sensory modalities at all? To answer this question we must uncover Pearson’s metaphysical views. His metaphysical views should give us some information about his ontological commitment to the existence of atoms and Mendelian factors.
First, we must remember that according to Pearson a scientist cannot explain phenomena. Scientists can only describe the flow of phenomena. One criticism of this view is that it leaves room for multiple interpretations of data. There is a possibility of equally valid alternative descriptions. Pearson makes this clear by saying, “For me there is no absolute truth in scientific knowledge… [it] provides conceptual models of more or less descriptive exactness of our sensations of phenomena” (Pearson 1930, p. 288). Contemporary scientists would find such a claim absurd, but what is more unusual about this statement is its likeness to a metaphysical claim. To claim that there is no absolute truth in scientific knowledge is not to deny that there is no absolute truth simpliciter. Also, the claim itself is metaphysical since it is talking about what scientific knowledge aims. The claim leaves one wondering why Pearson went to such great lengths to abandon Mendelism. If there is no absolute truth in scientific knowledge, then it may be the case that Mendelism is more exact than biometry. What are the grounds for Pearson’s rejection of Mendelism if all truth claims in scientific knowledge are equally valid? He seems no more able to reject Mendelism than he may reject the claims of astrology. Perhaps he does not commit himself to such a view. He may only affirm that empirically equivalent theories are equally acceptable.
Second, Pearson may have abandoned the notion of a Mendelian factor because he could not infer a cause from an effect. He rejected the rigidity of causation prominent in the writings of Kant – who is known for his claim that, for a man to make sense of his existence, his experience must conform to the category of cause and effect. Pearson wished to deny the category of cause and effect (Kant 1929, pp. 102-296. For a good summary of Kant’s category of causation, see Strawson 1967). Pearson wanted a broader category of ‘association’ to replace the category of ‘causation’. The category of association was less strict than causation. Pearson says, “Our rational being requires for his active existence a certain degree of sameness in his perceptions, he does not require for conduct absolute sameness” (Pearson 1911, p. 153). This is an abstract way of saying that there is a constant conjunction between two events, but that nothing naturally required the two events to occur sequentially at all times. It is merely the case that they usually occur, one after the other. Nothing, however, forbids one from not following from the other at some other time. There is a possibility that what almost always happens to occur as the result of one event may not occur in the future after the same event. Pearson went to the extreme and applied this category of association to science:
The conclusions of the physicist and the chemist are based on average experiences, no two of which exactly agree; at best they are routines of perception which have a certain variability. This variability they may attribute to errors of observations, to impurities in their specimens, to the physical factors of the environment, but it none the less exists and, when it is removed by a process of averaging, we pass at once from the perceptual to the conceptual, and construct a model universe, not the real universe. (Pearson 1911, p. 154)
In this passage, Pearson asserts that he wants to avoid speculating about unobserved phenomena in the world. He does not want to construct a conceptual world even though one is tempted to do so from certain observable facts. There is nothing, in his mind, that justifies constructing the conceptual world. He puts matters in a slightly different way later:
It is this conception of correlation between two occurrences embracing all relationships from absolute independence to complete dependence, which is the wider category by which we have to replace the old idea of causation. Everything in the universe occurs but once, there is no absolute sameness of repetition. Individual phenomena can only be classified, and our problem turns on how far a group or class of like, but not absolutely same things which we term ‘causes’ will be accompanied or followed by another group or class of like, but not absolutely same things which we term ‘effects’. (Pearson 1911, p. 174)
It is possible for one event that usually follows some prior event does not follow from that prior event in the future. “No phenomena are causal; all phenomena are contingent” (Pearson 1911, p. 174). In other words, we cannot speculate about the cause of some event because that cause may not actually be the cause of the event. It could have been something else. This type of association theory forbids us to speculate about the possibility of Mendelian factors at work in the variation of character traits. The Mendelian form of genetic atomism was thereby rendered unsatisfactory as a form of phenomenal description, so Pearson remained faithful to his own statistical methods of description.
Finally, Pearson rejected Mendelian factors because of his overwhelmingly conservative metaphysical and epistemological views. On the one hand, Pearson was neither interested in absolute truth nor in explaining why certain phenomena occurred. This was clearly a metaphysical view of the world. To hold that there is no absolute truth in scientific knowledge is a metaphysical claim. There are incompatible but equally valid accounts of scientific knowledge. In some sense, I believe that Pearson may agree with what today is termed pluralism. His text may seem relativisitic but that is hard to believe since he must have thought that there are some truths in order to describe th occurrence of certain events. See Michael P. Lynch, Truth in Context (1998). Such a view amounts to a metaphysical account. We cannot describe something without ontologically committing to it. Moreover, Pearson disliked explanations. He thought the main purpose of science was to describe the flow of phenomena, sense-impressions, and deduce certain observed regularities from these phenomena. In some sense, however, the descriptions of phenomena amount to explanations. For instance, one can describe the occurrence of certain events. These events occur in a certain sequential order discovered to recur several times in a like manner. The description of these events could amount to an explanation, one that includes the possibility of prediction and understanding. Perhaps Pearson has something else in mind when he declares the object of science to be purely descriptive, but he never enlightens us as to what that objective is. Even the most mundane of descriptions could be as explanatory as the most fully developed explanation. It remains to be seen how Pearson defines the term ‘explanation’. Pearson’s causal picture of the world shows that he could not infer the existence of theoretical entities. Mendelian factors could not be observed, so they must not exist. And, simply because we observe variation in organisms does not mean that there is some unobservable entity causing this variation to occur. (Could we use the category of association to establish the existence of factors at least for part of the time?) Regardless, Pearson, in the end, refuses to acknowledge the existence of unobservables because the are perceptually unavailable.
Bateson did not forthrightly reject the existence of unobservable entities. Bateson’s acceptance of something similar to the principle of instrumentalism allows him to infer that such things exist even though they are not accessible by our sensory modalities. In this post, I will try to clarify these philosophical presuppositions. The philosophical presuppositions of Bateson are not easily disseminated. Unlike Pearson, Bateson never wrote a treatise about his philosophical views. Most of what has to be said here comes from careful reading of Bateson’s texts before the discovery of Mendel’s paper.
Bateson was concerned with interpreting results of breeding experiments in terms of a particulate theory of inheritance. He operated with a kind of theoretical model of the mechanism of inheritance. R.C. Punnett explained Bateson’s methodology best in his book entitled Mendelism:
Bateson saw that if we are ever to answer this question [Why are not intermediates of all sorts more abundantly produced in nature than is actually known to be the case?] we must have more definite knowledge of the nature of variation and of the nature of the hereditary process by which these variations are transmitted. And the best way to obtain that knowledge was to let the dead alone and return to the study of the living. (Punnett 1913, p. 15)
What Punnett tries to make abundantly clear is Bateson’s primary concern. Bateson’s first concern was the breeding experiments. The breeding experiments were supposed to show there was some underlying mechanism of heredity. He was prepared to accept characters as distinct ‘units’ of heredity, but his conservatism prevented him from accepting the existence of such entities sans phrase. (For a discussion of Bateson’s reluctance to accept the existence of unobservable ‘units’ of heredity, see Bowler 1989, pp. 117-118.) The driving force behind Bateson’s acceptance of such entities was to defend discontinuous variation and to prove to Weldon and Pearson that the units of hard heredity were particulate (Depew and Weber 1995, p. 225). Ultimately, Bateson theory supposed that some entity existed, but we could only infer their existence from what is observed in the breeding experiments.
The main characteristic of Bateson’s experiments that is different from that of Pearson, besides the technical matter of the acceptance of discontinuous evolution, was that Bateson did accept the existence of unobservable entities, at least implicitly. Bateson accepted the thesis that there is nothing at the macro-level which is not made up of parts existing at the microlevel. From his breeding experiments, Bateson must have concluded that variation in organisms must have been caused by something. That something must exist since it causes a change in the organism. So, some ‘unit’ of heredity caused the variability in the heritable characteristics.
Some may consider Bateson’s implications unwarranted because of the lack of evidence between the macro- and micro-level. For instance, it may be the case that the micro- and macrolevel work independent of one another, whereby no interaction takes place. Be that as it may, Bateson’s implicit acceptance of these ‘unit’ factors of heredity did lead to important discoveries as well as the birth of Genetics. Bateson would eventually go on to coin the term genetics in a lecture at the International Congress of Botany, “a new and well developed branch of Physiology has been created. To this study we may give the title Genetics” (Bateson 1906, p. 90).