1. Introduction
The multiple realizability thesis is the claim that the same mental or psychological property can be realized by different physical properties. Since Putnam (Reference Putnam, Capitan and Merrill1967) and Fodor (Reference Fodor1975) proposed the multiple realizability thesis and argued for the irreducibility of psychological properties based upon this thesis, subsequent debates surrounding multiple realizability have often concerned whether this alleged implication holds true.Footnote 1
Philosophers however, have not paid much attention to the epistemological issue of whether there is any evidence for the multiple realizability thesis. The thesis is implied by a functionalist conception of the psychological—the idea that a psychological property is construed in terms of its causal roles abstracted from physical properties—and the main reason epistemological issues are neglected is that lots of philosophers think that the multiple realizability thesis is a priori true.Footnote 2
In their paper “Multiple Realizability Revisited: Linking Cognitive and Neural States” (in this journal), Bechtel and Mundale discuss an important but less frequently addressed issue: Is there any evidence for or against the multiple realizability thesis? Bechtel and Mundale argue that a careful investigation of neuroscientific practices provides us a reason to doubt the multiple realizability thesis. Multiple realizability is not plausible, they contend, since a biological aspect of the brain that is relevant to neuropsychological state individuation provides evidence against multiple realizability. I argue that Bechtel and Mundale's argument against multiple realizability is flawed, but claim that it invites an interesting possibility for testing multiple realizability. I suggest a way in which neuroscientific findings could have evidential impact on multiple realizability.
2. Testing Multiple Realizability: Homology or Homoplasy?
A frequently cited example in philosophical discussions of the multiple realizability thesis is that the same mental property, say pain, (probably) has different physical realizations in mammals, reptiles, and mollusca; another is that the same mental property is subserved by different physical properties even within a single species or individual. In discussing the identity thesis that mental states are physical states of the brain, Putnam says,
Even though octopus and mammal are examples of parallel (rather than sequential) evolution, for example, virtually identical structures (physically speaking) have evolved in the eye of the octopus and in the eye of the mammal, notwithstanding the fact that this organ has evolved from different kinds of cells in the two cases. Thus it is at least possible that parallel evolution, all over the universe, might always lead to one and the same physical “correlate” of pain. But this is certainly an ambitious hypothesis (Putnam Reference Putnam, Capitan and Merrill1967, 44).
The multiple realizability thesis does not imply that terrestrial life forms must have different physical realizations of a mental property if they instantiate the same mental property. However, the negation of the multiple realizability thesis, the identity thesis, implies that no terrestrial organism can have different physical realizations of one and the same mental property.Footnote 3 If both octopus and mammal have pain but with different physical realizations and if the same mental property is subserved by different physical properties even within a single species or individual, shouldn't we conclude that the multiple realizability thesis is confirmed and the identity thesis is rejected?
Bechtel and Mundale (Reference Bechtel and Mundale1999) claim that these alleged biological examples, even if they were right, should not be taken as evidence for multiple realizability. The reason, according to them, is that the relevant criteria for individuation of brain states as practiced in neurobiological and cognitive neuroscientific research are not “physical” as intended in these examples. The individuation of brains in terms of physical-chemical properties is an artificially “fine-grained” philosopher's invention which does not find its place in actual neuroscientific practice.Footnote 4 Bechtel and Mundale point out that when neuroscientists individuate brain states, both reference to psychological functions or processes and comparisons across species are made. Bechtel and Mundale say,
One might think, at first glance, the ability to make comparison across species actually depends upon multiple realizability. In fact, it is the very similarity (or more precisely, homology) of brain structures which permits us to generalize across certain species… . in the context of neuroscientific research, they are not multiply realized. (Bechtel and Mundale Reference Bechtel and Mundale1999, 177–178, italics added).
Neuroscientists assume that multiple realizability is false in their practice and find that psychological traits are realized similarly in homologous brain structures. I take it that Bechtel and Mundale, at least implicitly, claim that once the relevant criteria for brain state individuation based upon this homologous similarity are properly identified, we will have reason to believe that the multiple realizability thesis is false.
Homologies are (physiological, morphological, behavioral, psychological, etc.) characters shared between species that are inherited from their common ancestor. On the other hand, a character shared between species is a homoplasy if it is not inherited from a common ancestor, but arises from distinct origins. For instance, bird wings and human arms are homologous with each other. Both evolved from the forelimb of a common ancestor who lived around 300 million year ago. In contrast, although bird wings and bat wings resemble each other, they evolved separately and so their wings are homoplasious. The claim that some characters shared between birds and bats are homoplasious does not imply that all characters are.
It is worthy noticing that testing multiple realizability requires at least two independent lines of evidence. Suppose that a psychological property is realized both in a terrestrial organism and in an alien organism. If the psychological property has the same physical realization in these organisms, then these organisms provide evidence against multiple realizability. If each has a different physical realization, then they provide evidence for multiple realizability. In contrast, if the psychological property is realized by the same physical property in two individual organisms of the same species, this fails to provide relevant evidence for or against multiple realizability.Footnote 5 Regardless of the truth of the multiple realizability thesis, different species will tend to exhibit the same character if it is an inherited character that was present in their common ancestor.Footnote 6 When the similarities are homologies, one can't rule out the hypothesis that the species are similar simply because they share a common ancestor and there is no need to invoke the hypothesis that there is only one possible realization in this instance. It provides neither evidence for nor evidence against the multiple realizability thesis that brain structures are homologous.
Just as the wings of birds and bats are constructed from different materials and supported by different limb digits, some psychological functions or processes might be “constructed” from different “material” and supported by different structures. The wings of birds and bats are homoplasious since they evolved separately. If brains had evolved several times in different lineages and still had the same physical-cum-structural properties, then we would have evidence against the multiple realizability thesis. For it is not likely (although not impossible) that brains with separate origins would share the same structural properties if they could be multiply realized. On the other hand, if structural properties corresponding to psychological properties cannot be multiply realized, then brains must share the same structural properties even if they are homoplasious. If brains arose repeatedly by convergence, this would provide evidence against multiple realizability. It might be weak evidence since all present and past terrestrial organisms are carbon-based, DNA-based, etc. and biological conditions on Earth might make the brain the only realization that is available. Nevertheless convergence would provide evidence. It is homoplasies, not homologies, that can provide evidence for or evidence against multiple realizability.
3. Conclusion
I agree with Bechtel and Mundale that the way neuroscientists individuate brain states casts doubt on the belief that the multiple realizability thesis is a priori true. This invites us to consider an important question that has been neglected by philosophers: the testability of multiple realizability. I argue, contrary to Bechtel and Mundale's suggestion, that the fact that no psychological properties are multiply realized in brains is neither surprising nor relevant to testing multiple realizability if one is looking at homologous traits of brains. In order to test multiple realizability of psychological properties in terrestrial organisms, we should look at homoplasious traits of brains. The multiple realizability thesis is not a priori true. Homoplasious traits of brains can provide evidence for or evidence against multiple realizability.
