Evolution: Chance or Teleology?
Roderick T. Long
[Harvard astrophysics professor David Layzer was a major influence on my intellectual development; I took two year-long interdisciplinary courses from him, “Science A-18: Space, Time, and Motion” in my freshman year and “Science A-22: Chance, Necessity, and Order” in my sophomore year. We wrote two papers a week! It was in his courses rather than in my philosophy courses that I first read such authors as Hume, Poincaré, Russell, Wittgenstein, Popper; I was also introduced to the ideas of Jacques Monod and Ulrich Neisser (and the latter in turn led me to James J. Gibson). Layzer probably bears as much responsibility as any of my philosophy professors for reinforcing my interest in philosophy. (Layzer was also responsible for my winning the Conant Prize that year.)
Layzer’s area of specialisation was spontaneous order, and in particular the connections between biological and astronomical order; when I later read Spencer I realised that Layzer was (perhaps unwittingly) laying modern foundations for many aspects of Spencer’s theories that even his admirers tend to dismiss as outdated.
The following paper was written for the “Chance, Necessity, and Order” course in the spring of 1983; I was 19. Much of what I say in the paper is rather inadequate summary of Layzer’s theories. (For more on Layzer’s views see here, here, here, and here.) Another major influence of course is Aristotle.
Ayn Rand’s influence is also pervasive in this paper, from my use of the objective-subjective-intrinsic trichotomy to my discussion of causality; but I wasn’t quite brave enough to cite her.
With regard to the Randian trichotomy, I now think it was a blunder on my part to treat the subjective, objective, and intrinsic approaches to probability as three different “theories” about something called probability, with two of them being wrong and one right. It would have been better to speak of three different ways, equally valid, of using probability-talk. (Given that I wasn’t a determinist, it was inconsistent for me to reject the notion of intrinsic probabilities.)
I wish I’d known then, as I know now, that Darwin himself described his theory of natural selection as a vindication, rather than a refutation, of Aristotelean teleology. (Though I no longer think that Aristotelean teleology depends on natural selection for its vindication – but that’s another story.)]
The notion of viewing the universe in terms of the actual and the potential goes back at least as far as Aristotle, who defined matter as a spectrum of potential states, only one of which was actualized at any given moment. Attempts to banish the potential from scientific explanation have failed; even the modern theory of events, which acknowledges only the actual and conceives of causality as a relation between independent momentary time-slices of the universe, has been forced to admit such concepts as instantaneous velocity and potential energy – or, in quantum physics, fields of probability – into its history-less split seconds. There is no way to avoid the realisation that a thing’s possibilities are as fully real as its present state, that they exist prior to observation, and consequently that the sequence of time is not a spontaneous Herakleitean flux but an unfolding of nature’s latent properties.
Once one grasps the notion of a system’s having a manifold of possibilities, the question arises as to which of these possibilities is actualised. Knowledge is not free; energy must be expended to obtain it; it is in this context that the idea of probability must be understood.
Traditionally, there have been two main views concerning the nature of probability: the subjective and the intrinsic. The subjective view holds that probability is merely the product of the imperfection of our knowledge. If we say an event is due to chance, we mean that we don’t know what caused it; if we say an event will “probably” happen, we mean we have some information indicating that it will happen, but not enough to make us certain of it. The intrinsic view, on the contrary, maintains that probability is independent of our knowledge, that chance is an actual metaphysical constituent of the universe.
Both views are right as far as they go and wrong as far as they stop. To the subjective school, probability is an attribute of the observer; to the intrinsic school, it is an attribute of the observed. But the true essence of probability lies in the relation of the observed to the observer. Imagine a Maxwell’s Demon trying to create an “improbable” situation: a room with all fast-moving molecules on one side and all slow-moving on the other. Why is such a situation improbable? Here we are brought back to our actual-potential distinction: the situation above described is one of many possible states the room and its molecules can be in. It is easy to see that there are many more possible states which satisfy the requirement “not-fast-on-one-side-and-slow-on-the-other” than there are states fulfilling the conditions our Demon is trying to bring about. That is, our desired state is statistically outnumbered, and therefore less “probable.” (How this criterion of improbability is a relation between observed and observer will be shown in a moment.)
Following the classic Maxwell’s Demon scenario, our Demon is set to opening and closing a slit in a wall separating the two sides of the room, opening the door for hot going right and cold going left, and closing it for the reverse case. If all or most molecules approach the slit, then the room will end up hot on one side and cold on the other, and our Demon will have created our improbable situation.
But how does the Demon know which molecules approaching are hot, and which are cold (i.e., fast or slow moving)? He must measure their speed, and to do so he must expend energy; knowledge is costly. As Szilard has proven, the energy he must expend is equal to the degree of information he receives. The information exists as chemical order in the observed; but it exists as knowledge only in the observer. Probability is thus intimately related to knowledge; but some facts are harder to learn about than others because they are more orderly, i.e., statistically outnumbered as explained above.
One very important consequence follows from this conception of information-as-available-energy: the Szilardian rule that information must always be paid for, that no increase in order can occur without an equal or greater decrease in free energy (assuming a closed system). It apparently follows that the universe can only become more disorderly, not less.
Yet a seeming contradiction of this principle stares us blatantly in the face: the evolution of life, and the improbable situations (like rockets to the moon) that life can create. The evolution of life is generally ascribed to “chance” but we have seen that chance leads to less orderly situations, not more orderly ones.
Some might respond that our dilemma is superficial only, and arises from our anthropomorphic notion of progress. In contrast to Theodosius Dobzhansky’s description in “Chance and Creativity in Evolution”:
Seen in retrospect, evolution as a whole doubtless had a general direction, from simple to complex, from dependence on to relative independence form the environment, to greater and greater autonomy of individuals, greater and greater development of sense organs and nervous systems conveying and processing information about the state of the organism’s surroundings, and finally greater and greater consciousness.
– a skeptic might respond that no organism or species represents a “higher” stage of evolution than any other. But this is surely wrong; a human brain (and the situations such a brain can bring about in the environment) is much more orderly and “unlikely” than an amœba. And any form of life maintains a disequilibrium with its environment which warrants explanation. How does chance create order?
The answer, as Dobzhansky has explained, is that it isn’t. Chance creates the opportunity to evolve; random mutations, “errors” in the process of transmitting genetic information, create fruitful and unfruitful changes alike (with unfruitful most likely outnumbering fruitful), but the unfruitful are weeded out by natural selection. Thus, it is natural selection that generates order; natural selection, as Dobzhansky says, is the “antichance factor” in evolution.
Natural selection actually transmits “information” into the organism:
Natural selection constitutes a bond between the gene pool of a species and the environment. It may be compared to a servomechanism in a cybernetic system formed by the species and its environment. Somewhat metaphorically, it can be said that the information about the states of the environment is passed to and stored in the gene pool as a whole and in particular genes.
Dobzhansky is right; except that it is more than a “metaphor,” it is an accurate description of the process involved.
The notion that evolution is the product of chance, rather than of natural law (the law of natural selection) constitutes a major stumbling block in understanding the progression toward order. Aristotle, the classic opponent of chance as a primary explanatory factor in natural science, once defended teleology by the following argument: every step in the building of a house is determined, not by chance, but by the ultimate telos or end – the final, completed house. But if nature, not man, were building the house, it would have to go through exactly the same steps, and the nature of those steps would still be that of goal-directed actions. So, in the “making” of an adult chicken, all the steps in the development of the embryo are not random, but are directed toward the goal of full chickenhood.
Aristotelean teleology (or what is construed as Aristotelean teleology) is traditionally employed as an opposing foil by modern evolutionary theorists. But what Aristotle actually said is perfectly in accord with Dobzhansky’s rejection of chance as the author of order:
Viewed in the perspective of time, the process cannot meaningfully be ascribed to the play of chance, any more than the construction of the Parthenon or of the Empire State Building could be ascribed to chance agglomeration of pieces of marble or of concrete. What is fundamental in all these cases is that the construction process was meaningful. The meaning, the internal teleology, is imposed upon the evolutionary process by the blind and dumb engineer, natural selection. The ‘meaning’ in living creatures is as simple as it is basic – it is life instead of death.
The teleology of life is internal, not external; unlike the teleology of a house or other artifact, whose order is imposed from without, the teleology of a living being derives primarily from inner chemical reactions.
Who pays for this orgy of orderliness? If the organism acquires teleological information from its environment in the very makeup of its body, it must pay with a corresponding decline in free energy. But if such energy always declined, how would the organism (or gene pool, in the long run) survive? Where does the new supply of free energy come from?
An engine, in order to run, must have both a source and a sink of free energy; work is extracted from thedescent of the free energy from the source to the sink. The source is the sun; free energy light quanta come from the sun and are stored in the chemical bonds of plants; these bonds are broken, either by the plants themselves or by their predators, and free energy is released. The disequilibrium between the sun and the earth is thus the source of biological order.
But whence arises this disequilibrium? Since entropy growth is a universal law, astrophysical order requires as much explanation as biological order.
There are two traditional explanations for the existence of astrophysical order. One states that the universe is presently in macroscopic thermodynamic equilibrium, and that present antientropic situations are mere statistical fluctuations. But this Boltzmannian explanation presupposes an infinite universe – a notion which is both metaphysically untenable and relativistically unlikely. The other common explanation is that the universe began with an initial high degree of order and has been running down ever since; all present order is merely left over from the beginning. But David Layzer, in his article “The Arrow of Time,” proposes a more plausible hypothesis: the universe began with thermodynamic equilibrium, and order subsequently evolved as the universe expanded. This solution implies that the Second Law of Thermodynamics is not applicable to the universe as a whole; order and entropy are generated simultaneously, because the number of possible states does not remain constant, but increases. Thus, the existence of biological evolution can be understood only in terms of the distinction of actual and potential states of a system or substance.
[2010 note: I can tell that the ending is rushed; I must have been running up against a deadline as usual. But what I meant was that since the growth of entropy in a system involves matter distributing itself ever more evenly among the possible states of that system, it follows that if the system is expanding (as the universe is), the number of possible states can increase faster than the rate at which matter is filling them, so that while entropy is increasing, the gap between states filled and states fillable – i.e. order – may increase still faster. In other words, all the Second Law of Thermodynamics predicts is that in the contest between entropy and order, the amount of territory conquered by the forces of entropy will always increase – not that the percentage of territory will necessarily do so. Of course, given a fixed territory, an increase in amount means an increase in percentage (and a corresponding loss for the forces of order); but in an expanding territory we no longer have a zero-sum game, and it is possible for entropy’s domain to increase even as its share of total territory available decreases. And this, Layzer suggests, explains the growth of order in the universe: the universe expands more quickly than its matter can spread out, so we get “clumps,” i.e., stars, and the ongoing temperature disequilibrium between stars and nonstars allows energy to keep flowing from the former to the latter, generating work. For more details, see the links at the top of the page.]
Back to Juvenilia and Other Offenses