Evolution and the Blank Slate

Evolution and the Blank Slate

It is sometimes observed that evolution by natural selection resembles learning by trial and error. Mutation is the trial and natural selection is the error correction. It is certainly apt to think of evolution as a kind of learning process: over time organisms “learn” how to adapt better to their environment, changing in the process. The bodies of organisms are the effects of this “learning”—as if they are stores of knowledge about adaptation and survival. Organisms have gone from simple to complex as they have acquired such biological “knowledge”. This raises the question of how much of evolved species is due to the process of natural selection and how much is due to the initial state of that upon which natural selection operates. How much is “learned” and how much is “innate”? That is, what traits of organisms are the result of their evolution over billions of years, going right back to the origins of life, and what traits are the result of the original state of matter on earth before life began?

The way I have set up the issue is intended to recall the debate between rationalists and empiricists about the origin of knowledge. Of all the knowledge we possess, what is owed to experience and what is inborn? I am suggesting an analogous question about biological form: how much is due to the “experience” of evolution by natural selection and how much is due to the “innate” nature of the original material with which natural selection works? What traits owe their existence to natural selection and what traits owe their existence to the prior state of planet Earth? This is an interesting question in theoretical biology, but not one that I have seen asked. We could put the question this way: how much is really new in the evolution of species? Matter clearly had some traits before the evolutionary process got started, and the question is whether the traits now possessed by organisms transcend what was initially present. How much was really “learned” over evolutionary time? To what degree does the final state differ from the initial state, and in what respects? Answering this question will tell us how “creative” the evolutionary process is—how much it adds to what was present at the beginning. The question of knowledge asks how much knowledge is present at the origin of the mind; the question of biological form asks how much of biological form was present at the origin of life (i.e. before it began).

We can envisage two sorts of theory analogous to classical rationalism and empiricism. At one extreme we have the theory that everything of any consequence was present at the beginning—innateness rules. In particular, all the “atoms” of evolution were present at the beginning (compare “simple ideas”) and all that evolution does is combine these basic components into complex physical forms. Evolution produces no new elements, just new combinations of elements. Geological “evolution” produces no new elements when it makes mountains and rivers, just new combinations of old (“innate”) elements; and biological change is the same. This is the analogue of extreme rationalism: everything significant is present in the initial state of matter on earth, with evolution adding little that is new. At the other extreme we have the theory that matter was a blank slate at the beginning: it had no intrinsic structure, or none of any biological relevance, and everything arose from the creative process of natural selection. This is the empiricist view of evolution: we start with nothing, a mere empty receptacle, and everything about organisms is imposed from the outside by “learning”. There are no constraints deriving from the prior state of matter, just unlimited plasticity–pure potential. Natural selection imposes form on this formless substance, converting it into something rich and unprecedented. The blank biological slate makes trials (mutations) and natural selection eliminates errors while retaining successes; and as a result we have complex biological forms. It is not a matter of combination of antecedent elements but of genuine innovation: for instance, there were no eyes in pre-life matter, but evolution has caused eyes to come into being. Matter was not “born” with eyes, but had to acquire them by a process of “learning”. Biological form is thus acquired not innate—a result of billions of years of evolution, not already prefigured in matter before evolution got to work. That is why we don’t find life everywhere, as if matter could produce life by itself; we need evolution by natural selection, not just the bare existence of matter. We need the machinery of replication, selection, DNA, fitness, and all the rest. Left to its own devices matter has no tendency to generate life–as it would have to under the rationalist model.

Both theories seem to have something to be said for them, so we might wish to find some sort of intermediate position. What is interesting is how much the question resembles the old question of innate versus acquired knowledge, with its metaphors and rhetoric—and lack of clarity. Also its difficulty: it is not at all clear what to say about the question. There is certainly an appearance of novelty, but how deep does it go? Is there perhaps more at the origin than we recognize—more to matter than we tend to think? If we have a rich view of matter, the gap between initial state and final state might not be large, while if we have an impoverished view, the gap appears enormous. Suppose you are an adherent of panpsychism and natural teleology, with a soupcon of élan vital thrown in: then you will favor the rationalist theory and find little that is radically novel in the products of natural selection–it was all already present at the beginning. But if you view matter as merely “mechanical”, you may be inclined to favor the empiricist position—unless you take a very reductive view of the final product. It is difficult to settle the question without a developed view of both the nature of the initial state and the nature of the final state—how expansive to be about the former and how reductive about the latter. It is exactly the same with the classical debate about knowledge: what is the nature of the initial state of the mind and what is the nature of mature knowledge? The richer the latter the more we need to attribute to the former.

However, we can say two things with certainty: the blank slate model is clearly wrong, and we do know quite a lot about matter in its initial pre-evolutionary state. The metaphor of the blank slate does not apply to inanimate matter because matter obviously has a rich inner structure—there is nothing “blank” or “empty” about it. Without rehearsing all of physics, we can confidently report that matter has mass, extension, motion, electric charge, the power of agglomeration, and many other traits. So we know quite a lot about the “input” to the evolutionary process. This knowledge is not controversial or dependent on theoretical commitments: everyone can agree that matter has the properties listed, including the evolutionary empiricist. We needn’t beef it up with panpsychism, teleology, and the élan vital to know it is not a blank slate—we can already see that it has a substantive inner nature. It is not some property-neutral substratum that awaits the imprint of natural selection before it acquires any inherent structure. And because of this we can immediately see that the rationalist position must have a lot going for it: for clearly many of the traits of organisms derive directly from the antecedent traits of matter. Mass, extension, shape, material agglomeration, electricity, and motion—all these properties of organisms derive from the initial condition of matter not from the process of evolution. None of these traits were caused to exist by mutation and natural selection; rather, they are what natural selection had to work on—what was already present in matter.    [1] Thus it appears that most of the traits of organisms are innate in matter—native to matter. No one could suppose that these properties are the product of evolution and did not pre-exist it. Matter did not “learn” to have these traits; it was “born” with them, and evolution simply took them over. The pre-life earth was not a blank slate but a full plate, off which evolution dined. The initial state already included much of the final state.

It might be said that while it is true that bodily extension is derived from a prior property of matter, the particular shapes that animal bodies assume are the product of evolution. These variegated shapes were not present in matter before the evolutionary process got to work: there were no peacock shaped rock formations before life began, waiting to be converted into living peacocks. This point should certainly be conceded, but how much of a dent does it put in the rationalist’s position? For all such novel shapes are really the combination of geometrical properties already found in inanimate nature. The rationalist about knowledge will concede that many ideas are not innate, but insist that those that are not are mere combinations of those that are. Similarly, all the shapes of organisms are iterations and variations of shapes found elsewhere in nature—consider the symmetrical shapes of crystals. True, natural selection forms new shapes, but it does so by combining old shapes; it is not as if a radically new kind of geometry was inaugurated by the evolutionary process. The giraffe’s neck is a famous product of natural selection, but matter can form itself into long objects too—so there is nothing unprecedented there. Certainly the blank slate metaphor finds no support from cases like this: shape-wise the slate is rather full. Isn’t all this just a manifestation of the power of matter to arrange itself into arbitrarily many forms? But then, biological form is derivative from non-biological form, and hence “innate” in matter.

Is there any trait of organisms that is not innate in matter? The structure of DNA surely is: it is just molecular combination—improbable perhaps, but grounded in inanimate molecular reality. All the elements are present in pre-life matter, though the combination is novel (and hard to explain). What about mind? This seems more promising: surely evolution is responsible for the production of mind—there was no mind on earth before life began. So mind is not innate in the initial state of matter. But isn’t it? Some believe in proto-mental properties of matter; if that is right, then mind was innate in matter—there at the start, inborn. But even without that view it is plausible that matter had the potential for mind from the start; it wasn’t injected from the outside into something inherently unable to produce mind. We don’t know how mind arose in the course of evolution, so we don’t know whether it was somehow present at the beginning in implicit form (like innate ideas according to rationalists). So we still don’t have a clear example of a genuinely acquired characteristic of living forms. What looks at first like novelty is apt to reduce to combination or implicit presence at the outset. It is hard to find the analogue of the empiricist’s claim that the blank slate can be filled with a brand new idea of red upon exposure to red objects. Nothing about organisms seems like a radical innovation not prefigured in the nature of matter: life emerges from matter producing new properties, but there is no evidence of life introducing totally novel elements with no preparation or precursor. Empiricism about life has some truth to it, to be sure, but rationalism expresses the deeper truth—organisms are built from matter and only from matter. That is, bodily forms are expressions of innate properties of matter—properties that were present at the outset. How, indeed, could that not be so, given that mutation and natural selection are just causal processes operating on chemical elements originating in far-flung stars? There is nothing magical about these processes, nothing capable of infusing organisms with properties not anticipated in the primordial soup. There is plenty of scope for combination and reorganization, but we won’t find some new primitive ingredient added to what was there at the beginning. Thus the rationalist position is fundamentally correct.    [2]

Colin McGinn     

    [1] With respect to the eye, the rationalist position will be that while there were no eyes dotted around the landscape before life evolved (and for a long time after) the basic mechanism of the eye was present in matter. This is because an eye is a light-sensitive mechanism—something that responds differentially to light. But matter responds differentially to light in all sorts of ways, absorbing, reflecting or refracting it, so the physical basis for the eye existed in matter before life came along to make use of it. Evolution did not invent light and receptivity to light when it produced eyes; it drew upon what was already there. 

    [2] In the case of psychological empiricism there is a potential source of novel material not anticipated in the original constitution of the individual, namely sense experience (however wobbly this theory may be). There is somewhere external to the mind from which it might derive knowledge, thus supplementing its initial state, i.e. the perceived environment. But in the case of evolution there is no such external source: there is just matter and its machinations, pushed and pulled by mutation and natural selection. Where could this process derive novel materials? Not from outside matter! If it could tap into some supernatural reality, that would afford a potential source of novelty, but without that it is not going to get much beyond the material world from which life arose. The rationalist position is really the only game in town.