INTEGRAL WORLD: EXPLORING THEORIES OF EVERYTHING
An independent forum for a critical discussion of the integral philosophy of Ken Wilber
Publication dates of essays (month/year) can be found under "Essays".
SEE MORE ESSAYS WRITTEN BY ANDY SMITH
IS THE CURE WORSE
THAN THE DISEASE?
A Reply to Sheldrake
Andrew P. Smith
He accuses Carroll and other mainstream scientists of hand-waving, but what is Sheldrake doing when he postulates these fields, for which there is no direct evidence?
Frank Visser recently discussed [in "Rupert Sheldrake and the Evo-Devo Revolution"] the discovery of homeobox and other regulator genes, which control the spatial and temporal expression of protein-coding genes, in a manner that developmental biologists like Sean Carroll believe can finally explain the great variation in the shape and form of organisms. He contrasted this view with that of Rupert Sheldrake, known for his original and highly provocative hypothesis of morphogenetic resonance. Sheldrake believes that shape and form can only be understood through the influence of invisible morphogenetic fields, associated with all organisms throughout all of evolutionary history. These fields carry the crucial information needed to specify form, which is transferred to developing organisms in the process of resonance.
Sheldrake has kindly taken the time to reply to Frank [in "Morphogenetic Fields"], and further explain his views and contrast them with Carroll's. Here I make some fairly brief comments on Sheldrake's response.
A difference that doesn't make a difference
I begin by noting that Sheldrake actually presents two different arguments against the mainstream scientific view. I think it's important to distinguish these arguments clearly, because Sheldrake himself does not, and the reader can easily conclude that they are the same argument. They most definitely are not.
Here is the first:
Although hox genes play a role in the determination of the body plan, they cannot themselves explain the shape of the organisms. Since these genes are so similar in fruit flies and in us, they cannot explain the differences between flies and humans.
According to this argument, the problem with Carroll's view is simply that the homeobox genes are basically identical in very different organisms, such as fruit flies and our own species. This argument very strongly implies that if they were not so similar, there would be no problem.
But then, without warning the reader that this is what he is doing, Sheldrake changes the argument:
Regulatory genes by definition regulate other genes, and hence regulate the formation of proteins. That's what coding genes do: they code for the sequence of amino acids in protein molecules. They do not code for the shape of a wing, or an eye, or a leg. They just enable organisms to make particular proteins. So even if the dark matter of the DNA contains complex regulatory sequences, all they will enable the developing organism to do is to make particular proteins in particular sequences.
Sheldrake is now making a very different claim. Whereas initially he was arguing that homeobox genes can't account for the diversity of organisms because they are essentially the same in every organism—to repeat, an argument that implies that if they were different in different organisms, they could explain these differences—now he is saying that even if such differences did exist they still would not be sufficient, since they would only result in different proteins, which “do not code for the shape of a wing, or an eye, or a leg.”
The two arguments might be thought to be presented as independent lines of evidence against the mainstream view. Homeobox genes can't account for differences in organisms because the genes are too similar in different organisms; and even if they weren't so similar, they still couldn't account for the differences, because they code for proteins, not for form or shape. I think this may be the way Sheldrake is reasoning, though he never says so.
But in fact, the second argument, as Sheldrake goes on to develop it, works against the first:
Homeotic mutations can cause one module to be supplanted by another, or a module to be suppressed or duplicated, and hence lead to changes in form, as in bithorax fruit flies with four wings instead of two. I see these switch genes as affecting the tuning system, causing an nascent embryonic structure to tune into one morphogenetic field rather than another, rather like a mutation in the tuning circuit of a TV receiver causing it to tune into a different TV channel. But changing, suppressing or duplicating existing modules does not account for their appearance in the first place, just as tuning into one TV channel rather than another does not explain the content of the channels themselves.
Sheldrake says, “Homeotic mutations can cause one module to be supplanted by another, or a module to be suppressed or duplicated, and hence lead to changes in form”. Wait a minute! I thought the genes were too similar to account for differences in different organisms. But Sheldrake now seems to be acknowledging that, yes, these genes can result in differences. He goes on to say that “I see these switch genes as affecting the tuning system, causing an nascent embryonic structure to tune into one morphogenetic field rather than another”. How could these genes result in differences in the morphogenetic field that is tuned into, if they are so similar in different organisms? Wouldn't every organism, with the same complement of genes, tune into the same field?
So I think we can reject the first argument completely. The fact that homeobox and other regulatory sequences are so similar in different organisms doesn't mean that they can't account for these differences—any more than the fact that the genes are identical in most cells in the body of any particular organism doesn't mean that these cells can't differentiate into different tissues. Sheldrake is basing his entire stand on the second argument—that genes can't determine form or shape. This is consistent with the view he has held from the publication of his first book on the subject.
Science is modular, too
So how do we judge this second argument? I'll just make two points. First, the notion that genes can't code for shape or form is a little simplistic. Sheldrake says, “That's what coding genes do: they code for the sequence of amino acids in protein molecules. They do not code for the shape of a wing, or an eye, or a leg. They just enable organisms to make particular proteins.” What Sheldrake neglects to point out is that while only the one-dimensional sequence of amino acids in proteins is coded for in the genome, this one dimensional sequence becomes a very specific three dimensional conformation when the newly synthesized peptide folds into a functional protein. In other words, linear amino acid sequence specifies three-dimensional form.
So specific protein conformations are de facto coded for in the genome. Given a particular amino acid sequence, the conformation is a virtual certainty, because of certain physicochemical processes involved in protein folding. And from these three dimensional proteins, much of the three-dimensional form of the cell is built up, through key structural networks within the cell as well as the plasma membrane enclosing it. Interactions among cells, in turn, result in the particular three-dimensional structures of tissues.
I know Sheldrake is aware of all of this. Indeed, he has argued that even protein folding is governed by morphogenetic fields. His argument is fundamentally the same as the one he is making against Carroll—that the process is not fully understood, and that no one has yet specified completely, in physicochemical terms, how the final conformation could result.
It's true that for very long amino acid sequences, the folding process is extraordinarily complex. There may be an enormous number of theoretically possible conformations, and we don't fully understand how one particular state emerges, and quite rapidly. But we do understand the process in general terms. We know the kinds of chemical bonds formed between different amino acids during this folding process, and we know that the final conformation represents a stable state, at which energetic requirements are at a relative minimum. The folding of relatively short, simple sequences is quite well understood, and most scientists believe that there is no reason in principle why the same processes can't account for larger proteins.
In other words, science is doing here what it usually does: breaking the problem down into smaller, more manageable components (just as the modular structure of organisms does, I might add). Going directly from genes all the way to, say, the shape of a wing seems like a stretch. Going from genes to the shape of one cell in a wing is still daunting, but it doesn't seem like quite as much of a stretch. Going from genes to the shape of one protein in one cell of a wing is a still easier problem. Even if Sheldrake disagrees that protein folding can be explained purely in terms of physical forces between amino acid molecules, I think he will have to concede that the problem of explaining the link between genes and body form has been made much more tractable; it has been put on a footing where very well-established physical laws can account for much of the final outcome. So he is in a position where he continues to protest that current science can't get from point A to point B, even as science has definitely moved far beyond A towards B, and is moving along further in that direction with every new advance.
A theory that needs modulation
Sheldrake's argument, it seems to me, is analogous to that used by advocates of Intelligent Design.
The second point is more fundamental. While Sheldrake criticizes Carroll for supposedly not being able to explain form, his view of morphogenetic fields of resonance inherited from the past is also unexplained. He accuses Carroll and other mainstream scientists of hand-waving, but what is Sheldrake doing when he postulates these fields, for which there is no direct evidence? He dismisses Carroll's description of an eventual explanation in terms of “chains of parallel and successive operations that build complexity”, but that is exactly how science has proceeded for several centuries. Sheldrake accepts the existence of genes, proteins and other material components of cells, but we understand these macromolecules precisely as the result of “chains of parallel and successive operations”. So Carroll is at least basing his promissory note, if that is really all that it is, on an approach that even Sheldrake will concede has been highly successful in explaining other phenomena. If Carroll and other scientists are debtors who have yet to pay off what they owe, they at least have a job with an income that was sufficient to pay off many other major debts in the past.
What is Sheldrake basing his morphogenetic fields on, other than that they could explain phenomena that, according to him, can't be explained by Carroll's approach? Not only is there nothing that, in the view of most scientists, would constitute direct evidence for the existence of these fields, but it's very difficult to understand what these fields would consist of, and how they could interact with organisms. He has suggested possible roles of quantum phenomena or postulated superstrings (Sheldrake 2009), but at this point this is sheer speculation.
As apparent from one of the quotes above, Sheldrake is fond of the analogy with a radio or TV, with the organism “tuning in” to the morphogenetic field. But while radio waves carry the information that is transduced into sound or pictures, this information ultimately comes from somewhere else—the radio waves have to be modulated. Given that Sheldrake believes genes can't specify form, the morphogenetic fields would have to contain all the necessary information on form or shape within them from the beginning, a very different situation from radios. But how or why would they have this information even before cells and organisms actually existed? These and other difficulties suggest to me that it's very hard to make the case that he isn't waving his hands much more desperately than Carroll and others are.
Science has already shown how information is stored in organisms, both in the genome and in nervous systems, and it has also shown how it can be transduced into some forms that even Sheldrake would find unproblematic. For example, I don't think he would argue that we need his morphogenetic fields to explain how a certain gene coding for a certain protein results in eye color. Genes produce pigmented proteins, or their absence, which appear as color. This is an extremely simple example, but it is already far beyond morphogenetic resonance in explaining the relationship between information storage in something that exists over long periods of time (genes) and information expression in the body.
Sheldrake's argument, it seems to me, is analogous to that used by advocates of Intelligent Design. They begin by denying that evolution can really explain the appearance of major new types of organisms, then propose as an alternative an intelligent creator, the origin of which has no explanation at all. I don't believe that Sheldrake is promoting ID, but like those who do, he seems to be advocating a cure that most scientists believe is worse than the disease—i.e., a solution to the problem that implies a much larger gap between what we know and what is proposed than the current alternative. And even worse, a gap that does not suggest itself as capable of being crossed by means of shorter, more manageable steps in the manner of traditional science.
Creatures of habit
I have been very critical of Sheldrake here, but I want to close by adding that I find some of his ideas intriguing, the kind of far-out hypotheses from which major breakthroughs very occasionally arise. In fact, his concept of morphogenetic fields is rooted in a much broader and more profound notion that I think may well prove to be correct: that there may not be immutable scientific laws, but rather habits, for want of a better word, that change over time. The belief in immutable laws, as Sheldrake emphasizes in his writings, has been a bedrock of modern science. It performs for science much the same function as the belief in God used to perform for philosophers (and of course still does perform for the world's major religions and their billions of followers): it is an absolute, on which knowledge can be grounded independently of an observing subject.
Postmodernists, however, argue that subjects and objects can't be separated, and reject any absolute. The philosopher Quentin Meillassoux (2008) has pointed out that if we take the postmodern belief seriously, it is inconsistent with the scientific assumption that a world existed prior to human subjects—that we can know anything about what existence was like before our species appeared. As I have discussed previously here ("Do We Need an Absolute?"), Meillassoux's solution to the problem is to argue that scientific laws are contingent, the product of chance. While Meillassoux uses other arguments to suggest that such laws could remain relatively stable over long periods of time—as of course has been observed time and again by science--his work clearly constitutes a major assault on the traditional understanding of scientific laws.
So I support Sheldrake's general challenge to immutability. But this challenge does not necessarily imply his concept of morphogenetic fields, and vice-versa, any more than evolutionary theory presupposes immutability. He is making a considerable leap when he says that because processes in nature are constantly evolving, the principles resulting in these processes must also be evolving. To see the fallacy in that reasoning, it's only necessary to consider cellular automata, which can evolve into incredibly complex forms following just a few simple rules (Wolfram 2002; see also David Lane's Information Field Theory).
I think Sheldrake's morphogenetic fields have to be judged on their own merit, and in my view the theory still fails the tests of necessary and sufficient. He has not shown that morphogenetic resonance is necessary to explain body form, that current mainstream ideas are incapable of doing so. At best, he can argue that Carroll and others have not yet provided a complete picture, but it's surely premature to argue that it is impossible for them ever to do so. And at the same time, we currently lack sufficient evidence and understanding to move beyond speculation in regard to what these fields might be like, and how they could interact with organisms. I know Sheldrake has proposed many tests of his theory, but to the best of my knowledge, he has not and so far cannot provide any tests that would provide us any more insight into the nature of these fields.
1. Sheldrake describes Carroll and like-minded scientists as “materialists”. I prefer the word mainstream, because as far as I can tell, Sheldrake himself is a materialist. That is, since these morphogenetic fields must interact with physical processes, they presumably have a physical basis themselves. Sheldrake (2009) has suggested a number of possible ways this might occur, such as through quantum processes or through postulated extra dimensions provided by string theory. On the other hand, he has also described morphic resonance as a process by which information is transferred to the organism with no energy transfer. As far as I know, there is no scientific precedent for this, and it suggests dualism, with all the standard problems that come with that view.
2. There is an online game that allows anyone to contribute to knowledge in this field. The object of this game is to search through various conformations to find the stablest state of a peptide. See http://fold.it/portal/info/science.
3. Such as information transfer in the absence of energy transfer. See note 1.
4. Sheldrake probably would argue that even in this case, his fields are necessary for the pigmented proteins to adopt the proper conformation. But the point is, there are numerous examples of simple, fairly direct links between genotype and phenotype that underscore how much we understand about the relationship between information at the two levels.
Meillassoux, Q. (2008) After Finitude. An Essay on the Necessity of Contingency (New York: Continuum)
Sheldrake, R. (2009) Morphic Resonance: The Nature of Formative Causation (New York: Park Street Press)
Wolfram, S. (2002) A New Kind of Science (Wolfram Media)