I. Introduction: Aristotle and Complexity
I wish to suggest in what follows that it may be fruitful to undertake a close reading of Aristotle’s biological and other philosophical writings, with a view to seeing what aspects of them are or could be congruent with certain recent developments in modern science that are often grouped together under the heading of “the new science” or complexity theory; that is, the branches of science that focus on themes such as self-organization, spontaneous pattern formation, dissipative systems, and morphogenesis. Obviously, I cannot hope to touch on all aspects of this program here, so I’d like instead to offer a case study as an example of the potential fruitfulness of this approach. I’ll therefore concentrate of Aristotle’s theory of final causation and teleology, sketch the history of its interpretation and eventual rejection, and then provide a few examples of its rehabilitation in some recent theorists of science.
I.1. Plato and Aristotle
Almost everyone is, I take it, familiar with Raphael’s painting The School of Athens in the Stanza della Segnatura at the Vatican, which gives a kind of a class picture of the most eminent representatives of Greek philosophy, as the artist imagined them in the early sixteenth century.
At the center stand Plato and Aristotle: to the left, Plato, grey-bearded and venerable, holds a copy of the Timaeus and points toward the heavens, while on the right a youthful Aristotle, looking tanned and fit, holds a copy of the Nicomachean Ethics and holds out his right hand palm downward, in a gesture that could be interpreted as meaning “Whoa there, Plato! You’re going just a bit too far! Let’s keep our feet on the ground here!”
This painting gives a good summation of the image of the two philosophers that has largely prevailed down through the ages until today. Plato, represented by the cosmogony of the Timaeus, has his head in the clouds: he is the archetypal idealist, concerned primarily, if not exclusively, with intelligible forms at the expense of the material world in which he lives. Aristotle, in contrast to his teacher, is the paradigmatic empiricist: preoccupied by things of this world, he is the first genuinely “down-to-earth” philosopher. In the words of the theoretical biologist Harold J. Morowitz,
[M]uch of the intellectual history of the past 2400 years depends on a difference of outlook between Plato and Aristotle. Plato, profoundly influenced by mathematics, thought of the world in terms of ideal and abstract perfect forms that lay behind the sloppiness of sensory experience. Aristotle the biologist . . . thought of the world in terms of experimental categories and inductive generalizations. At its extreme, Platonism finds reality by deriving the observed world from abstract ideas. Aristotelianism at its extreme finds reality by deriving the essences from the observed world. (152)
One is struck by the resemblance of the conflict between Plato and Aristotle and the one that has taken place since the seventeenth century in the development of modern science, between physics on the one hand and biology and chemistry on the other. Physics, since Galileo the most prestigious scientific discipline, deals with the behaviour of abstract, idealized entities in quantitative-mathematical terms, and if such sciences as biochemistry have experienced an increase in visible success (and hence in prestige and funding) over the past half-century or so, it is, some would argue, because it has turned away from qualitative studies such as embryology and morphology, to concentrate on such quantifiable fields as genetics. David BerlinskiFootnote 1 has described the contrast between modern mathematical physics and the life sciences in terms of a conflict between two attitudes. In such disciplines as calculus, adequacy of description is traded for depth of insight, paying the price of neglecting certain aspects of experience in order to obtain the theoretical and predictive benefits of a world recreated in terms of numbers. The life sciences, in contrast, trade depth for adequacy of description, an attitude whose origins “lie in the immemorial animal empiricism that informs our unforced and natural account of the real world.” Math and physics rejoice in theories, writes Berlinsky, while biology and chemistry delight in the accumulation of facts.
I.2. Aristotle on Nature and Natural Things
For Aristotle, nature (phusis) is that which has the principle of its motion within itself, as an internal source of its change or stability.Footnote 2 Things that are “physical” (phusikos) or natural are distinguished from all others by three main features: they can move themselves, they are substances (that is, they have independent existence), and they consist of form within matter.Footnote 3 The phusis of an object or organism, for instance, can be considered as consisting in the sets of motions or changes it undergoes if it is not impeded by external forces. In this sense, an organism’s nature is thus its path toward nature; in other words, an organism’s nature just is its own process of self-realization and -perfection. By extension, phusis can serve to designate Aristotle’s view of the workings of the world, a world organized primarily according to the principles of biology.Footnote 4
Aristotle’s world was one of purposeful striving, in which each of the four elements of the sublunary realm – earth, water, air, and fire – strove to reach its “proper place”: matter strove for form, potentiality for actuality, and the celestial bodies strove to imitate the perfection of the immaterial Prime Mover. This realm of constant change from generation to corruption stood in contrast to the supralunary world, where the divine celestial bodies moved according to the perfect, immutable circular motion characteristic of the fifth element. It is as if Aristotle retained the idealizing, mathematical approach of his teacher Plato only for the celestial domain. In contrast, to study the natural or phenomenal world that surrounds us (which Plato barely thought worthy of study), Aristotle was forced to practically invent the disciplines of physics, biology, and chemistry, among others.
This striving, or goal-oriented behaviour, thus characterizes all of nature (both animate and inanimate) according to the Stagirite, but is most clearly obvious in the case of the living organism, which, in the words of the classical scholar and professor of zoology D’Arcy Wentworth Thompson, “is the telos, or final cause, of its own processes of generation and development.” Aristotle was particularly impressed by the way embryos develop according to a plan that seems prearranged but is not imposed from without, and he coined the term entelechy to designate this phenomenon of immanent, organic purposefulness.Footnote 5
I.3. Aristotle vs. Plato in the History of Ideas
Historically, Platonism, and particularly the cosmology of the Timaeus as transmitted by such Latin-speaking authors as Cicero, Chalcidius, Macrobius, and Boethius, played an important role in the scientific and philosophical thought of the so-called Dark Ages, the period when most of Aristotle’s works, and especially those dealing with natural science, were lost to the Greekless West. In the meantime, Aristotle’s views on nature were exerting a formative influence on the thought of his Eastern followers, from the Greco-Roman neoplatonists of Late Antiquity, through Syriac translations made by Nestorian Christians and Arabic translations made from the Greek and Syriac by Christian, Muslim, and Harranian scholars. Finally, beginning in the twelfth century, Jewish, Christian, and Muslim scholars working together in southern Italy and at Toledo in Spain began the great movement of translation from Arabic to Latin that was to revolutionize medieval Scholastic thought, restoring a large part of the Corpus Aristotelicum to the West, together with many of its ancient and Arabic commentaries.Footnote 6
II. Aristotle in the History of Science
II.1. Coveney-Highfield, The Arrow of Time
Standard histories of ideas have long maintained that Aristotelian thought, in the form of medieval Scholasticism, stifled the development of science throughout the Middle Ages until Galileo and then, definitively, Newton, ushered in the modern era by mathematizing physics and liberating science from the fetters of the Aristotelian world view. As a picturesque historical relic, this world view can henceforth be of interest only to over-leisured antiquaries.
Typical of the kind of account one often finds in popular scientific literature is the following passage, from Peter Coveney and Roger Highfield’s Reference Coveney and Highfield1990 work The Arrow of Time:Footnote 7
[W]ith the benefits of hindsight, we can see that Aristotle set back the development of science even more than Plato. This father of Western philosophy preferred a teleological chain of explanation to a causal one. Teleology is concerned with the search for purpose in the phenomena of the world. Whereas we would explain the existence of the humpback whale by invoking a causal argument – Darwin’s theory of evolution – a teleological argument would ascribe it to the action of a beneficent creator (God) for the benefit of mankind.
The authors go on to render Aristotelianism responsible for Catholicism’s rejection of the heliocentric viewpoint first set forth by Aristarchus of Samos – they do not mention the name of PtolemyFootnote 8 – and to speak of “Aristotelian pseudo-science”Footnote 9 and even the “festering corpse of Aristotelian thought” (p. 63).Footnote 10
Several things are interesting about the passage cited above. First, teleological explanation is contrasted with causal explanation, although, as we shall see in a few moments, for Aristotle a teleological explanation is precisely one that corresponds to the fourth and most fundamental kind of cause, the final cause.
Next, Coveney and Highfield simply equate teleology Aristotelian-style with “the action of a beneficent creator (God),” even though both the concept of “creator” and that of an anthropomorphic or benevolent God are completely foreign to Aristotle’s thought. One would not suspect, from these lines, that there is a distinction to be made between immanent and transcendent teleology, and that while the former is applicable to Aristotelian final causation, the latter is not. This point has recently been emphasized by T. L. Short, who writes:Footnote 11
Aristotelian teleology . . . is “internal” or “immanent.” An organism’s ends are not given to it by an external agent. It is that of the organism itself, but not in the vitalistic sense that there is an agency within the organism that holds this end in some sort of quasi-conscious intention. The end is “internal” to the organism because it is the organism’s form.
We shall shortly return to the question of teleology and final causation in Aristotle. In the meantime, let’s continue to follow Coveney and Highfield’s exposition:
Thanks to the support of Thomas Aquinas (1225-1274), who worked with manuscripts preserved by the Arabs, Aristotle’s model came to be cherished by the Catholic Church because it made man the centre of the universe.Footnote 12
The image of Thomas working with Arabic manuscripts – or perhaps Greek manuscripts preserved by the Arabs? – may raise a smile: in fact he worked mainly on the basis of Latin translations made from the Greek by his associate William of Moerbeke, archbishop of Corinth.Footnote 13 The explanation of Aristotle’s popularity in the Church, for its part, is simplistic, not to say simple-minded: one fails to see in what relevant sense man is the “center of the universe” for Aristotle.Footnote 14 For the Stagirite, man does indeed inhabit the Earth, which is in one sense at the centre of the universe. Yet in a more important sense, the Earth is at the bottom of the universe: the element earth, heavy and sluggish, collects at the centre because it is the farthest from incorporeality and the lowest in the hierarchy of the four elements.Footnote 15 In fact, when Aristotle comes to draw an analogy between the modes of action of sublunary beings and the motions of the heavenly bodies, the Earth, precisely because of its immobility, corresponds to plants, which occupy the lowest rank in the scale of living beings since they are unable fully to achieve the good.Footnote 16 Closer to the mark are Schneider and Sagan in their 2005 Into the Cool: “Aristotle’s thoughts on purpose, absorbed into church doctrine by Saint Thomas Aquinas and others, became conflated with ideas of a purposeful creator [emphasis mine].”Footnote 17
Finally, one may wonder in what sense Darwin’s theory of evolution could “explain” the existence of the humpback whale. But that’s another story, which would take us too far afield here and into current debates on evolutionary theory.Footnote 18
II.2. Enter Galileo
According to the standard scenario canonized by such historians of science as Alexandre Koyré, both Renaissance thought and the advent of Modern Science were made possible primarily by a return to Platonism, in the person of such cultural heroes as Galileo. The victory of Platonism over Aristotelianism meant the replacement of a hierarchized, qualitatively differentiated but closed cosmos by one that is infinite, homogeneous, quantitative, and uniquely describable by mathematics. In the words of René Thom,Footnote 19 “The world of Galileo is a world of motion, but in which generation and corruption have no place; hence the almost complete disappearance of considerations of form from modern physics.”
The triumph of Galilean thought meant that the primary model for understanding the world was no longer, as it had been for Aristotle, that of the organism, but rather that of the machine.Footnote 20
II.3. Physics vs. the Life Sciences
I mentioned above the contemporary tension between biology and physics, which seems in some respects to echo and continue the contrast between Platonic and Aristotelian ways of approaching the world.
One point that may serve to illustrate the difference between the approaches of physics and biology is their attitude toward irreversible and dissipativeFootnote 21 processes. Since the development of thermodynamics in the early nineteenth century, dissipation, or the inevitable loss of energy in the form of heat during mechanical processes, has been associated with waste and the increase of entropy, so interpreted as increasing disorder or randomness. Thus, the Second Law of Thermodynamics, which states that entropy must always increase in a closed system, was initially interpreted as meaning that the universe is inevitably heading to a state of thermal equilibrium, or “heat death.”Footnote 22 Since Darwin, in contrast, biology has associated evolution, another dissipative process, with an increase not in disorder but in complexity.Footnote 23 It was to reconcile this tension, between the inevitable increase in disorder apparently mandated by the Second Law of Thermodynamics and life’s tendency to increase order and complexity, that in 1943 the quantum physicist Erwin Schrödinger delivered the lectures later published as his classic work What is Life? Footnote 24
Once again, it is hard not to think here of the ancient conflict between Plato, the star-gazing proto-physicist, and Aristotle, the empiricist founder of biology. But such trenchant contrasts often tend to have fuzzy edges when one takes a closer look at them. Thus, some modern historians of philosophy have recently taken up a tendency, which flourished in Late Antique Neoplatonism, to downplay the differences and emphasize the similarities of some aspects of the thought of Plato and Aristotle.Footnote 25 Similarly, in the work of Ilya Prigogine, René Thom, Stanley Rosen, Stuart Kauffmann, and others, recent years have witnessed the beginnings of fruitful interactions, if not a full-fledged reconciliation, between mathematical physics and the life sciences.
But we’re getting ahead of ourselves here. Let’s pause to refresh our memory with regard to the role of teleology and final causation in Aristotle. This in turn will require us to remind ourselves of the overall structure of his so-called theory of the four causes.
III. Aristotle, Causes, and Teleology
III.1. What Is Teleology, Anyhow?
For Aristotle, a causal investigation is an attempt to answer the question “Why?”, precisely the question that some historians claim modern science has tended to avoid since Newton.Footnote 26 A “why-question” is a request for an explanation, so that the four Aristotelian causes can best be considered as four kinds of explanation.Footnote 27 Teleological explanation is an explanation in terms of final causation, and the Swedish scholar Ingemar Düring maintains that Aristotle’s entire doctrine of the four causes is merely a thematization of the philosophy of the Telos.Footnote 28 We have seen earlier that while final causes and teleology may be features of systems implying conscious design on the part of some supernatural entity, this does not need to be the case, and it was not the case for Aristotle.
For Aristotle, the final cause appears as the culmination of the system of four causes, usually enumerated in the order material, formal, efficient, and final. These causes can be viewed as four different but complementary ways of answering questions about why things are as they are,Footnote 29 or as four types of explanation that are not mutually exclusive, but complementary.
The locus classicus for Aristotle’s four-cause theory is a text found both in the Physics and in the Metaphysics:Footnote 30
* The works of Aristotle, translated into English under the editorship of William David Ross. Above from vol. 8, Metaphysica (Oxford: Clarendon Press, 1928).
Needless to say, Aristotle’s own exposition of the four-cause system has given rise to volumes worth of commentary and to conflicting interpretations,Footnote 31 so here I’ll limit myself to citing a few illustrations from the secondary literature.
The four types of causes can be roughly summarized as in the following table:
If we take a house as an example, we could say that the material cause is the bricks of which it is made, the formal cause its blueprint, the efficient cause is the mason’s knowledge and skill,Footnote 32 and the final cause the need for human shelter. The ecologist Robert Ulanowicz uses the example of a battle:Footnote 33 here, the material cause consists in weapons, transportation, and communications technology; the formal cause is the battle plans and positions of attacking and retreating troops, the lay of the land, weather conditions, and so forth; the efficient cause is the officers and soldiers who use such technology; and the final cause is the need for housing, political conflicts, economic malaise, and so forth.
In an influential paper first published in 1976, Allan Gotthelf has argued persuasively that Aristotle’s theory of final causation can best be explained in terms of irreducibility.Footnote 34 The basic concepts of Aristotelian explanation of natural things, says Gotthelf, are natures (phuseis) and tendencies or potentialities (dunameis), and to explain, for Aristotle, is just to identify the natures and potentials at work in a given process. Now, some such processes, such as heating and cooling of a liquid, can be perfectly well explained as the actualization of potentials, both active and passive, of the elementary constituents – that is, the matter – of the liquid in question. There is thus no sense in invoking the notion of final causation in such cases. Similarly, as Aristotle explains in the Parts of Animals 4.9, 685b12-16,Footnote 35 the fact that octopuses have suckers is to be explained by final causation, since these animals need them to be the kind of animals they are. Whether they have one or two rows of suckers, however, is due not to teleology but to material necessity (i.e., their relative sizes). In Aristotle’s view, however, other processes, such as the development of an embryo into a mature animal form, cannot be so explained: no combination of realizations or actualizations of the collective or individual potentials of its elementary constituents can suffice to account for such development. In order to explain them, therefore, Aristotle postulates a cause that is identical with the realization of an organism’s potential for form.
III.2. What’s Wrong with Teleology?
One of the aspects of Aristotelian thought that has attracted most scorn since the time of Galileo and Descartes is his theory of teleology and final causation; that is, his view that certain aspects of nature can be viewed as goal-directed, and that at least one way to understand natural processes is from their goals. For reasons that might justify a sociological enquiry, scholars often use sexist imagery to express their distrust of these ideas. Thus, Francis Bacon proclaims that “Research into final causes, like a virgin dedicated to God, is barren and produces nothing,”Footnote 36 while teleology has been described as a “maiden without whom no biologist can live, yet he is ashamed to show himself in public with her.”Footnote 37 Bertrand RussellFootnote 38 omits the sexual overtones, but still insists that those who formulate scientific explanations must be ever vigilant against the encroachments of teleological thinking into their accounts of nature. This is perhaps not surprising when we recall that Russell elsewhere suggests doing away with the notion of cause altogether.
Richard Sorabji enumerates a number of grounds on which final causation or teleological explanation has been held to be unpalatable. They include the views (1) that teleological explanation is supposedly incompatible with necessitating causes; (2) that teleological explanation is supposed not to be genuinely explanatory; (3) that teleological explanation must presuppose the involvement of a conscious agent; (4) that teleological explanation is reducible to something else; (5) that nothing distinctly teleological is involved in the explanation of the presence of a biological organ; and (6) that teleological explanation can be rendered otiose by purely causal explanations. Sorabji provides persuasive arguments against all these views, which I shall not rehearse here. Yet perhaps the most important objection to teleological explanation has been pointed out by, among others, Robert Rosen, who argues that while material, formal, and efficent causes all entail their effect in one way or another, final causation seems to be entailed by its effect, as though the future were somehow able to influence the present.Footnote 39 We’ll return to this view in a moment.
The objections against final causation or teleological explanation have varied throughout history. What most of these objections seem to have in common is that such causes or explanations are viewed as too metaphysical, with all the worries about verifiability, fuzziness, anthropomorphism, and so forth that this term of opprobrium implies. In recent times, the strongest motif for unease is surely that introducing final causation or teleology into the scientific study of nature would be to open the door to creationism, intelligent design, and other kinds of scarcely veiled theism. I’ve already suggested that this is not necessarily the case: Aristotle’s system itself was not creationist, nor was it intelligently designed by any anthropomorphic deity. What’s more, we’ll see a bit later on that Aristotelian ideas on teleology and final causation may be useful for a view of the origin of order that is opposed to creationism or intelligent design: the view that many phenomena manifesting order, including life itself, can arise spontaneously by processes of self-organization.
IV. The Return of the Stagirite?
We have seen that the modern consensus seems to be that progress in the history of science is more or less equivalent to a liberation from Aristotelianism. In the course of the past generation or so, however, some dissenting voices have been heard, which suggest that some aspects of the Aristotelian philosophy of nature may have been swept too quickly into the dustbin of history, and that Aristotle, having been kicked out the front door, may be sneaking back in through the back window. This phenomenon in turn is linked to changes that have taken place within contemporary science itself, which has led some theorists to question the adequacy of the Galilean-Newtonian paradigm.Footnote 40
IV.1. Aristotle and the “New Science”
Science, some contemporary theorists have argued, has changed since the time of Galileo and Newton. In the place of the deterministic world of Laplace, where an omniscient genius, armed with a knowledge of the position and rate of change of every particle contained in the universe at a given moment, could use Newton’s laws to determine every past and future state of the universe, some have begun to favour a different model of the universe, one that emphasizes its complex and sometimes chaotic nature. On this view, the world of Newtonian dynamics, with its time-reversible laws that enable us to calculate the orbit of the moon several hundred years from now and to retrodict it several hundred years in the past, has come to be recognized as a special, exceptional case,Footnote 41 or an idealization, of the laws of nature. To quote Morowitz again,Footnote 42
The stark fact that keeps emerging with great clarity from post-Newtonian science is that physics is the only field in which the Platonic model holds, and it doesn’t even work for all of physics. In other areas the evidence leads to laws and generalizations, but the great complexity of the consequences of the laws makes it impossible to be predictive in a Platonic sense. . . . All systems, except for those covered by classical physics, are complex and require an Aristotelian approach to their understanding [emphasis mine].
In this new paradigm, Nature is seen as often functioning in a non-linear and therefore largely unpredictable way, ruled by probability rather than by determinism. Instead of working mechanistically like an immense clock wound up by a geometer God, some contemporary theoreticians of science see Nature as marked by instabilities and bifurcations that lend it a fundamental spontaneity and creativity.
If indeed it is legitimate to speak of a “new science” that has emerged over the past generation or so, it might be said to include at least the following aspects: an emphasis on unpredictability, complexity, self-organization, and emergence in both physics and biology, and a concomitant interest in questions of form, function, and morphology in the study of evolutionary biology. These recent concerns have led to a reawakening, in certain theoretical circles, of an interest in the work of several renegade theorists whose work had previously been marginalized: from Goethe and Geoffroy St.-Hilaire to D’Arcy Thompson, by way of the morphogenetic work of Alan Turing, to contemporary theorists of complexity such as Thom, Prigogine, and Rosen. The suggestion I’d like to make is that Aristotelian thought fits in well with the work of these neglected scientists, whose time may once again have come to be taken seriously. Already in 1966, well before the study of complexity and its key concepts became fashionable, Ingemar Düring had pointed out two key aspects of Aristotle’s thought that render it congenial to contemporary thought on complexity: his emphasis on irreversible processes in physics, and his doctrine that the whole is often more than its parts.Footnote 43
IV.2. Causation and Finality in Three Recent Theorists: Thom, Prigogine, and Rosen
Despite the general rejection since Galileo of the Aristotelian theory of causation, a key component of which is teleology or final causes or explanations, some twentieth-century philosophers have argued for its continued pertinence: one thinks especially of Bergson, Whitehead, and the German theorist of biology Hans Jonas. Recently, however, contemporary theorists of science have also begun to argue for a rehabilitation of the Aristotelian theory of causation. Three recent theorists of science, all of whom died within the past decade or so (Robert Rosen died in 1998, René Thom in 2002, Ilya Prigogine in 2003), devoted a good part of their careers to arguing that it may be time to rehabilitate some aspects of the Aristotelian four-cause system, and especially final causation, abandoned as a result of the dominance of the Newtonian paradigm. They argue that this paradigm, largely adequate for the idealized world of classical mechanics, has proved to be less appropriate for complex phenomena at the borders between physics, biology, and chemistry. First and foremost among these pheneomena is, as we shall see a bit later, life itself.
IV.3. René Thom
The French mathematician René Thom was the originator of catastrophe theory in the 1970s. In his paper “Aristotle and the Emergence of Modern Science: The Galilean Break,” published in 1991, Thom wrote as follows:Footnote 44
It is common to say that the Galilean epistemological break has brought about a radical progress in science, annihilating the fundamental concepts of Aristotelian physics. Here again, however, we must see the effect of this brutal transformation rather in the manner of a scientific revolution in the sense of Kuhn, that is, as a change in paradigms: the problems solved by the old theory cease to be the object of interest and disappear from the field of speculation. The new theory brings out new problems, which it can solve more or less well, but above all it leads to the concealment of the entire old set of problems,Footnote 45 which nevertheless continues its subterranean course in the guise of new techniques and new formalisms. I believe . . . that in the past ten years or so, we have seen the reappearance in modern science of a number of themes and methods that are close to the Aristotelian doctrine, and for my part I salute this kind of resurgence.
While recognizing the value and important contributions of the Galilean paradigm, Thom believes it has led not so much to the solution as to the neglect of at least some of the problems addressed by the Aristotelian approach. In this regard, he goes on to say, the greatest difference between ancient and modern science has been the impoverishment of the Aristotelian concept of causation. Of the four Aristotelian causes, post-Galilean science recognizes only material and efficientFootnote 46 but not final causality. This situation has, according to Thom, begun to change: approaches such as Prigogine’s theory of dissipative structures, which we shall look at in more detail in a moment, and Thom’s own catastrophe theory recognize that “a form can be defined by mathematical algorithms acting by themselves on a competent substrate.” Thom sees signs of the use of this hylemorphic model, in which matter is “as it were captured by the form,” in such disciplines as embryology with its morphogenetic fields.Footnote 47 Even Aristotle’s division of the world into a sublunary realm of change and a celestial world of perfect, invariable circular motion – another view supposedly rendered obsolete by Galileo – can, according to Thom, be justified by today’s science. This distinction can be seen as referring to two types of time. There is a stationary, reversible, quasi-cyclic time which corresponds to reversible Hamiltonian mechanics and quantum formalism, and then there is an irreversible time, acting unidirectionally on the world of generation and corruption: this, says Thom, corresponds to the second principle of thermodynamics and the inevitable increase in entropy.
IV.4. Robert Rosen
The Aristotelian theory of causation occupies a special place in the thought of Robert Rosen, the New York-born theoretical biologist who spent a good part of his teaching and research career at Dalhousie University. According to Rosen, science since Newton has concentrated on what Rosen calls “What? questions.” If given a clock to study, for instance, the post-Newtonian scientist will tend to try to figure out what it is, i.e., how it works. The Aristotelian scientist, in contrast, is primarily concerned with the question “why?”, and would focus on trying to understand how it came to be a clock.Footnote 48 For Rosen, “Aristotle correctly identified science with the study of ‘the why of things,’ and scientific explanation with the elucidation of causal sequences.”Footnote 49 In this context, Rosen offers a unique interpretation of the Aristotelian four-cause system.Footnote 50 Using the example of the analysis of a theorem P, Rosen identifies its material cause with the axioms of a system by which P is entailed; the formal cause with its proof, or the list of inferential rules that actually entail P, that is, its algorithm or program; and the efficient cause with its rules of production, or the rules of inference governing the system.
Note the absence of the final cause in this enumeration. Rosen gives an interesting explanation of this absence in the context of the Newtonian paradigm.
Let us assume, says Rosen, that we are in the presence of an effect, and we choose to ask why it has occurred. The three first Aristotelian answers we can give, those that specify its material, efficient, and formal causation, will all entail the effect in question. Yet final causation is different, for the final cause of an effect is defined in terms of something entailed by the effect. In this sense, Rosen argues, finality is expressible in the Newtonian paradigm only in terms of future states. In this paradigm, however, entailment is embodied in the recursiveness of state transition sequences: the only thing a state can entail is a subsequent state, just as a state can be entailed only by a preceding state. Yet final causation seems to involves the future acting on the present: it follows that final causation is excluded in the Newtonian paradigm.
Rosen’s own attempt to resolve this paradox involves the interpretation of final causality as a function, but the details of his proof, obtained through recourse to category theory, are too abstruse to be recounted in detail here.
IV.5. Ilya Prigogine
Another of the leading proponents of this new view of nature was the Russian-born chemist and physicist Ilya Prigogine, who worked in Belgium and the USA and won the Nobel Prize for chemistry in 1977. At the conclusion of a work written with the Belgian philosopher Isabelle Stengers in 1986, Prigogine wrote:
The thermodynamics of irreversible processes has discovered that the flux that traverses certain physico-chemical systems and takes them farther from equilibrium, may nourish phenomena of spontaneous self-organization, of symmetry-breaking, and evolutions toward increasing complexity and diversity. Where the general laws of thermodynamics cease, the constructive role of irreversibility may be revealed. This is the domain in which things are born and die, or are transformed in a singular history woven by random fluctuations and the necessity of laws.
We are henceforth closer to that nature about which . . . the Presocratics wondered,Footnote 51 as well as to that sublunary nature whose powers of growth and corruption Aristotle described, whose intelligibility he claimed was inseparable from its uncertainty. The paths of nature cannot be foreseen with certainty, the share of accident is irreducible in it, and much more decisive even than Aristotle understood it to be. Bifurcating nature is that in which tiny differences, insignificant fluctuations, can, if they occur in opportune circumstances, invade the entire system, generating a new functional regime. (pp. 360-1)
Earlier in the same work (p. 232), Prigogine had emphasized that “my theory of dissipative structures brings us back to a conception closer to that of Aristotle.” For Prigogine and Stengers, the aspects that modern science would do well to recuperate from Aristotelian thought include the rehabilitation of qualitative diversity, and the reality of irreversible processes and physical time as a whole.Footnote 52 To the extent that recent scientific developments have begun to privilege the reality of the arrow of time, in contrast to classical physics with its emphasis upon reversible dynamic processes as the only ones worthy of study, we may speak, with the Belgian scholar Laurence Bouquiaux, of “the rediscovery of time and change as the passage from a “Platonizing” or “Parmenidean” vision to a more “Aristotelian” vision . . . the idea that change possesses a reality, and even that it can constitute the object of a science” (p. 75).
V. Aristotle, Creativity, and Entropy
V.1. The Creativity of Dissipative Systems
Since the codification of the theory of thermodynamics in the mid-nineteenth century, entropy, which is, properly speaking, the measure of energy differentials in a system,Footnote 53 has been viewed as a measure of disorder, decay, and destruction. However, the gradual shift of interest from closed to open systems – i.e., those which exchange matter and energy with their environment – has led to a re-evaluation of the creative potential of entopic processes. Here, far from being the sign of the world’s inevitable tendency for organization and structure to decay and collapse, entropy may become a mark of spontaneous, self-organizing creations of structure.
V.2. Bénard Cells
A widely cited example of this re-evaluation of the power of entropy is the phenomenon of Rayleigh-Bénard instability,Footnote 54 first discovered by Henri Bénard in 1900 and then elaborated upon by Lord Rayleigh. Here, a flat pan containing a millimetre-deep layer of silicone oil, sometimes sprinkled with aluminum dust, is heated uniformly across the bottom. What initially takes place is conduction: the hot fluid at the bottom of the layer expands and becomes rarefied, whereupon it tries to rise through the colder, denser fluid above it. As the heat beneath the fluid is increased while the upper layer is kept cool, the gradient or temperature difference is increased until, at a specific critical point, a pattern of hexagonal cells, each about as large across as the silicone oil is deep, spontaneously forms. This change marks the transition from conduction to convection.
At relatively low temperatures, when the gradient or difference between hot and cold is still low, transmission of heat takes place by conduction along a curve with a straight slope, i.e., in a linear process.Footnote 55 Here, energy transfer takes place by direct interaction from molecule to molecule. As the applied heat, and therefore the gradient across the system, increases, a critical point is reached, the system suddenly becomes unstable, the fluid’s molecular action becomes coherent, and convection ensues.
This experiment provides a good illustration of a fundamental division in thermodynamics. When it first developed as a science in the early-to-mid nineteenth century, thermodynamics concentrated on the state of equilibrium, at which gradients have been eliminated and no further change takes place. In the late nineteenth and twentieth centuries, thanks to the work of scholars such as Josiah Willard Gibbs and Lars Onsager, attention shifted to non-equilibrium phenomena, in which a system that interacts with its environment is maintained away from equilibrium by exchanging matter and energy with its surroundings. Yet there is a further distinction to be made within non-equilibrium thermodynamics: when a system is close to equilibrium, it behaves in a linear fashion; here, there is a directly proportional relationship between two quantities. This linear regime is a steady state, characterized by minimum dissipation and entropy production,Footnote 56 and here, the system’s behaviour can be understood as the sum of its parts.Footnote 57 When the system gets further away from equilibrium, by contrast, its behaviour suddenly becomes non-linear. Here, small changes in the initial conditions can give rise to disproportionately large effects, and the system’s behaviour must be understood as being more than the sum of its parts. This implies the system’s instability and unpredictability: owing to small but inevitable imprecisions in measuring the system’s initial conditions, and the tendency for small anomalies to be rapidly magnified, the system is said to show deterministic chaos, meaning that its evolution can be predicted only for its initial stages of development.
In the case of the Bénard cells, we have moved, by the simple means of the application of heat, from a homogeneous system in which no up or down, right or left, or before and after would be perceptible to a microscopic observer, to a differentiated system which allows such distinctions; in other words, we have here a good illustration of the phenomenon of symmetry breaking. In the words of Nicolis and Prigogine,Footnote 58 “[S]ymmetry breaking brings us from a static, geometrical view of space to an ‘Aristotelian’ view in which space is shaped or defined by the functions going on in the system.”
With the formation of the patterns of Bénard cells, we also witness the formation of correlations, or statistically reproducible relations between distant parts of the system. When heat is applied to a system, its randomness or disorder is usually increased as the motion of its constituent molecules becomes more and more violent. In the Bénard instability, by contrast, molecules are interacting over a distance of about a millimeter, the usual measurement of a Bénard cell, whereas under normal circumstances, closer to equilibrium, they interact over a scale of about one angstrom, or 10-8 centimetres. Despite the fact that each separate molecule is characterized by random thermal motion, here we have some 1020 molecules spontaneously acting in conjunction to form each Bénard cell, thus providing a clear example of the emergence of complex behaviour. All this takes place while entropy production – usually, as we have seen, considered as a measure of disorder or disorganization – is increasing at a furious rate.Footnote 59
Such experiments illustrate a number of phenomena that are central to what has been called the “new science” of complexity, including self-organization and symmetry-breaking. They have also been interpreted as indicating a new understanding of the role of entropy, and hence of the second law of thermodynamics. Textbooks usually present entropy as a measure of disorder or decreasing organization, and since the second law states that entropy always increases in a closed system, this law has often been taken to decree that the universe is inevitably headed toward a state of heat death or thermal equilibrium, at which no motion or change will take place, meaning the end of life as we know it. Yet experiments such as that of the Bénard cells, the Belusov-Zhabotinsky experiment that reveals the existence of chemical clocks, and a plethora of others show that entropy is not necessarily synonymous with disorder. Under the right conditions, it can give rise to structure and self-organization, rather than destroying them.
VI. The Home Stretch
VI.1. Back to the Four Causes
Bénard convection can thus be explained in a fairly adequate way according to the first three causes of the Aristotelian paradigm: we have described the material cause by indicating the materials of the experiment, and we have described the efficient cause by giving an account of the method, or the initial conditions. We could easily go on to give the formal cause of our phenomenon by explaining how it obeys the laws of thermodynamics. What we have not yet discussed, however, is the final cause: why (dioti) does Bénard convection occur, or in other words, what is it for (to hou heneka)?
One answer that has been suggested is the following: Bénard cells form because they help increase the rate of heat transfer, and therefore of gradient destruction. As Schneider and Kay have shown, the heat transfer rate under convection, with the formation of Bénard cells, increases as a linear function of the temperature gradient. Interestingly, this increase in gradient destruction is accompanied by a parallel increase, albeit non-linear, in entropy production.
Schneider and Kay interpret the data as meaning that Bénard cells form in order to increase heat transfer and thus gradient destruction in the most efficient way possible. They explain this phenomenon as an instance of their “revised second law of thermodynamics,” according to which “[t]he thermodynamic system which governs the behaviour of systems is that, as they are moved away from equilibrium, they will use all avenues available to counter the applied gradients. As the applied gradients increase so does the system’s ability to oppose further movement from equilibrium” (p. 6).
For these authors, the revised second law holds not only for inanimate chemical systems like those displaying Bénard convection, but also for ecological systems and for all other dissipative systems that are maintained in a far-from-equilibrium state by external influxes of energy. In fact, they hold that the phenomenon of life itself is a far-from-equilibrium dissipative structure that maintains its local level of organization at the expense of producing entropy in the larger system of which it is a part. Living systems, they suggest, are “dynamic dissipative systems with encoded memories . . . that allow the dissipative processes to continue.” In short, “Living systems are sophisticated mini-tornadoes, with a memory (its DNA), whose Aristotelian “final cause” may be the second law of thermodynamics” (p. 15).
VI.2. Conclusion
We began with the widespread view that Aristotelian teleology and final causation must be outdated and wrong, primarily because it implies an intelligent, benevolent, and anthropomorphic Creator. We have seen, I hope, that this view is wrong: the idea of such a Creator, absent from Aristotle, was imported into Aristotelian thought in late Greco-Roman Antiquity, whence it flourished first in medieval Jewish and Islamic thought, before finding a congenial home in Latin Scholasticism.
Through our final example of the Bénard instability, I hope we have glimpsed something more. The Aristotelian four-cause system in general, and final causation in particular, can, at least in the view of some contemporary scientists, be used fruitfully to describe the paradigmatic instance of self-organization and spontaneous formation of hexagonal cells in a chemical dissipative system maintained far from equilibrium. We have also seen, moreover, that at least some thinkers are willing to go so far as to define life itself as a dissipative system whose final cause is the degradation of gradients and the production of entropy in accordance with the Second Law of Thermodynamics.
Clearly, if some contemporary philosophers of science avoid final causation like the plague, it is at least partly because they are afraid that if they allow it, they are opening the door to Intelligent Design and Creationism, doctrines that are indeed inimical to science.
I would argue that this need not be the case. On the contrary, final causation may be an element in a scenario that eliminates the need for Intelligent Design by any kind of anthropomorphic Creator, just as there was no need for them within Aristotle’s original system.
Indeed, how did Life first arise on Earth? Surely not, as the Creationists rightly argue, because amino acids swimming around in the pre-biotic chemical soup just happened to bang together in a way that happened to give rise to macromolecules capable of self-replication. As Kauffman and others have shown, when one does the math, such a scenario is wildly unlikely. If, however, as Coveney and Highfield have written,
there was a suitable feedback mechanism operating in the pre-biotic soup which could account for non-linearities, then the general conditions were ripe for self-organisation . . . if a molecule in that soup catalysed its own production, then non-linear feedback . . . could emerge. This could break the spatial homogeneity of the medium and trigger patterns and rhythms possibly along lines broadly similar to those proposed by Turing in 1952. (p. 217)
In other words, given the right initial conditions of a system maintained far-from-equilibrium by suitable constraints, life may have arisen as a result of processes of self-organization similar to those we have observed in the formation of Bénard cells. The formation of life from inanimate nature may thus be not only non-miraculous, but virtually inevitable. And once such proto-living formations had emerged, Darwinian natural selection, henceforth working on a vastly more limited range of possibilities, might be sufficient to explain the development of life as we know it.
Whether life just is, as Schneider, Kay, and Sagan propose, a system whose goal or final cause is to dissipate gradients, or else, as Kauffmann would have it, is simply an almost inevitable phenomenon arising from enzymatic autocatalysis, the result is the same: an anthropomorphic Ur-Creator or Intelligent Designer becomes redundant. Life is a process that explains itself,Footnote 60 and final causation may still have a role to play in this process.Footnote 61
