20th WCP: Cosmic Teleology and the Crisis of the Sciences

There has been considerable discussion in recent years regarding the emergence of a new scientific paradigm centered on holism and self-organization. This discussion has been motivated by a number of distinct developments in such diverse fields as physical cosmology and nonlinear thermodynamics, all of which seem, however, to point in a common direction. On the formal side there is a push towards the unification of physical theory, while on the substantive side there is an increasing willingness to talk about the universe as a radically interconnected system with a fully intelligible structure, which develops towards ever higher degrees of organization. Proponents of the so-called "anthropic cosmology" even suggest that the universe may be fine-tuned in just such a way as to make possible, or even necessary, the emergence of complex organization, life, and intelligence (Prigogine et al 1977, 1979, 1984, 1988, Bohm 1980, Barrow and Tipler 1986, Gal-Or 1987, Pines 1987, Lerner 1991, Tipler 1994, Harris 1991, 1992).

I would like to argue that talk of a new paradigm is in fact premature. There is, to be sure, powerful evidence that the universe is a unified, self-organizing system, structured to generate life and intelligence-evidence which the mechanistic paradigm which has dominated the sciences since the seventeenth century makes it difficult to interpret and which it has largely chosen to ignore. But most of the "new science" of holism and self-organization is, in fact, little more than an attempt to put new wine in old skins. While attention to previously shunned evidence is to be commended, there has yet to be any real break with the view that the pinnacle of science consists in the mathematical formalism, and there continues to be a deep-seated resistance to the only kind of explanatory strategy which can both unify science and account for the evidence in question. Specifically, I want to argue that science is entering a period of profound crisis, a crisis which is rooted in an inability to theorize "ordering to an end" and which will be resolved only by a return to teleological explanation, albeit a teleology expanded to accommodate the reality of chaos, contradiction, and even disintegration, all of which clearly form part of material reality.

The aim of science is to explain as much as possible using as little as possible-i.e., to reduce the complex diversity of sensory experience to the smallest number of principles possible, and if possible to rise to a single first principle from which, were we to understand it perfectly enough, all particular phenomena could be derived. In the process, we hope to discover what purpose, if any, the universe possesses, and what role we play in the realization of that purpose.

The first great scientific synthesis was developed by Aristotle, who argued that it was, indeed, possible to rise from sensory data to a single first principle-the first unmoved mover. Matter for Aristotle was the possibility of organization, form its actuality. Motion was the gradual realization, over the course of time, of the latent potential of matter as it moved towards the perfection of form. The explanation of particular forms of motion involved an understanding of the matter in question, of the form which was being perfected, and of the efficient cause-that by which the change took place, as well as the end towards which the change was ordered. Ultimately, however, the whole process of change was grounded in the attractive power of the first unmoved mover, the incredible beauty, truth, goodness and integrity of which drew all things to itself (Aristotle, Physics, Metaphysics, Lindberg 1992).

The Aristotelian synthesis collapsed for two reasons. First, it was unable to advance a unified theory of motion. How does one explain teleologically a decaying corpse or a thrown javelin? These processes do not seem in any sense ordered to the perfection of form. Thus the distinction between natural and violent motion. This in turn led to a distinction between the celestial realm, where all motion is natural, and the sublunar realm where both kinds of change occur. Second, Aristotelian science had considerable difficulty coming to terms with the growing evidence that even the heavens were not ordered in the perfect manner required by theory. This was a problem long before Copernicus and Kepler. There are sharp differences between Aristotle's cosmology and the formal, mathematical models of his near-contemporary Eudoxus. Refinement of these models by Ptolemy and others involved a departure from the perfect spherical motion which was central to Aristotle's vision long before Copernicus opted for heliocentrism or Kepler displaced the circle with the ellipse (Murdoch and Sylla 1978, Grant 1978, Pedersen 1978, Lindberg 1992).

There were two ways to resolve this problem. One would have been to generalize the concept of teleology in such a way as to accommodate the reality of violent motion, and to abandon the particular cosmological models developed by Aristotle in order to save the principle of teleological ordering. There were powerful reasons to take just precisely this approach. Aristotle and his interpreters had, after all, already implicitly shown that the only complete explanation is a teleological explanation. This is because a complete explanation must terminate in a principle which (directly or indirectly) explains everything else while being self-explanatory. Such a principle must be necessary, infinite, and perfect (and thus divine), and it must cause exclusively by the attractive power of its own perfection (otherwise it would be in motion itself and would thus require some other explanatory principle, resulting in an infinite regress) (Aristotle, Metaphysics 1071b-1076b, Aquinas, Summa Theologiae I, Q2).

This was not, however, the road taken. Teleology was abandoned altogether, and (though this was never acknowledged, or perhaps, even really recognized) the possibility of a complete explanation along with it. Instead, an attempt was made to develop increasingly general mathematical formalisms which describe motion (now conceived exclusively as change in place). Thus the whole history of mathematical physics, beginning with the special theories of

Galileo and Kepler, up through the "first unification" by Newton, and each of the successive generalizations and unifications: Hamiltonian dynamics, Maxwell's equations, relativity, quantum mechanics, and most recently quantum cosmology.

Mechanistic science has been remarkably successful at this enterprise of producing formal descriptions. And, perhaps because mathematics is ultimately the science of possible beings, it has provided an enormous impetus to the technological development which requires us to imagine new possible systems. But the fact is that it has not been any more successful than was Aristotelian science at producing a unified theory of motion, nor is it able to describe-much less explain-anything like the full range of natural phenomena. Among the internal contradictions of mechanistic science the most important are:

an inability to unify relativity, which depends on the notion of a space-time continuum and a concept of signalling which imply strict causal relations, and quantum mechanics, which theorizes the universe as a discrete order and which calls into question certain aspects of strict causal relatedness (Bohm 1980), and

an inability to unify dynamics (understood to include both relativistic and quantum theory), which treat reversible processes, and thermodynamics, which treats irreversible change (Prigogine et al 1979, 1984).

Above and beyond this, mechanistic science has profound difficulty coming to terms with certain critical bodies of evidence. The standard "Big Bang" cosmology has run into increasing empirical difficulties-the existence of large scale structure which contradicts assumptions of cosmic homogeneity, missing dark matter, stars older than the universe itself is supposed to be, and incorrect predictions regarding the basic ratios of such elements as Deuterium, Helium, and

Lithium (Lerner 1991). More important is the inability of mechanistic science to explain adequately the emergence of complex organization, something which appears to be ruled out by such basic principles of thermodynamics as the Second Law and the Boltzmann Order Principle (Prigogine 1977, 1984). And even where mechanistic science is able to patch together an authentically powerful explanation of a range of natural phenomena, it does so only with theories which ultimately contradict each other. Thus, one of the triumphs of mechanistic science is the explanation of chemical organization using quantum mechanics and thermodynamics-but these two theories have fundamentally different understandings of such basic concepts as "time."

The "new" science, as it is generally called in popularizations, is more or less defined by an attempt to resolve some or all of these difficulties, and thus to unify, or at least significantly advance the unification of, science, while arguing, in one way or another, for the ultimate meaningfulness of the universe. Let us consider briefly three very different variants of this program, and see why it is, in fact, quite impossible to carry out-at least without a more fundamental epistemological rupture and what amounts to a return to the alternative, teleological strategy of explanation.

Of the various attempts to unify science and argue for the ultimate meaningfulness of the universe, the most conservative, in terms of its rigorous adherence to the norms of mechanistic science, (if not in terms of its claims, which are so bold as to be outrageous) is that of Frank Tipler (Tipler 1994). Tipler's strategy is fundamentally to reduce complex organization, life, and intelligence, to "information" which is physically encoded in various systems, and then to theorize the universe, and the fate of intelligent life, in terms of a further generalization of the quantum cosmology developed by Hartle and Hawking. Tipler rejects explicitly the notion of irreversible change, and uses temporal reversibility as a way to mimic, unsuccessfully we will argue, the concept of final cause.

The universe, for Tipler, is a vast information processing system. Matter is the "hardware" component of the system, the laws of nature the "software." Tipler argues that the organization of a system is its negative entropy, or the quantity of information encoded within it. "Life" is simply information encoded in such a way that it is conserved by natural selection. A system is intelligent if it meets the "Turing test," i.e., if a human operator interrogating it cannot distinguish its responses from those of a human being. Tipler defines in rigorous physical terms what it would mean for intelligent life to continue forever (and thus for the universe to be ultimately meaningful), as well as the physical conditions for this happening-something which requires that intelligent life re-organize the entire universe to create sufficient gravitational shear to provide the energy necessary to process an infinite amount of information in the finite time permitted by the closed cosmology which we he favors, so that the universe terminates in a omniscient, omnipotent, and subjectively eternal Omega Point. This re-engineering is to be carried out, or at least initiated by, a fleet of intelligent, self-reproducing automata which will, quite literally, devour the universe.

Tipler acknowledges that such a universe is only one of many possible worlds. The totality of possible worlds is described in terms of quantum cosmology. In quantum cosmology the universe is represented by a wave function &#thorn;(h,F,S), which determines the values of h and F on S, where h is the gravitational and F the nongravitational fields respectively, and S the underlying three-dimensional manifold (Tipler 1994: 174-175). Most formulations of quantum cosmology leave the selection of the underlying three-dimensional manifold S arbitrary. Hartle and Hawking extend the domain of &#thorn; to all possible values of S, but still require h to be space-like-something which contradicts general relativity. Tipler proposes instead to allow the domain of the wave function to include all four-dimensional manifolds which permit a Lorentz metric g. All possible universes exist mathematically; those which permit observers may also be said to exist physically. But the only universes which "ultimately" contain an observer are those which contain an "ultimate observer"-i.e., those which terminate in an Omega Point. Thus only these universe are physically real. One can understand this either as an immanent requirement that the intelligent beings arise and re-engineer the universe-or, since causality is reversible under general relativity and quantum mechanics-as a claim that the Omega Point in fact creates the universe by exercising causality back through time.

There are a number of difficulties with this approach. First of all, it seems very difficult to theorize organization, as Tipler wants to, in terms of only externally related particles. If systems are nothing but aggregates of externally related particles, then organization is nothing more than the order which prevails among those particles-the negative entropy or information content of the system. But negentropic and information theoretical approaches to organization and complexity run into serious problems when we attempt to apply them to biological and social systems.

IBM scientist Charles Bennett (1987) has recently pointed out that the negative entropy theory has limitations even at the physical level. The human body, for example, is intermediate in negentropy between a crystal and a gas, while being more highly organized than either. Similarly, organized objects, "because they are partially constrained and determined by the need to encode coherent function or meaning, contain less information than random sequences of the same length, and this information reflects not their organization but their residual randomness." But if we reject the information theoretical approach to organization, we reject Tipler's definitions of life and intelligence as well.

More broadly, Tipler's approach to unifying science seems simply to gloss over the difficult questions. He makes ample use of thermodynamics, for example, in theorizing the conditions for processing an infinite quantity of information, but does not really make an argument that thermodynamic irreversibility is merely an illusion. Similarly, he does not address the contradiction between the continuous and discrete orders envisioned by relativity and quantum mechanics, but merely crams both orders into a single, fundamentally quantum mechanical formalism-and this in spite of the fact that the real meat of his theory is relativistic. In the end one is forced to conclude that Tipler is trying to make mechanistic science say something it cannot, and ends up transforming the hope of a universe ordered to life and intelligence into a nightmarish vision of a universe consumed by self-reproducing automata and transformed into a gravitational supercomputer.

The doctrine of Israeli physicist Benjamin Gal-Or is rather more promising. Gal-Or brings a "dialectical" understanding of the process of unification, as "a process of criticism wherein lies the path to the principle of all inquiries (Aristotle in Gal-Or 1987: 47)." This dialectic leads Gal-Or to the conclusion that it is necessary to unify theories of reversible and irreversible change first (i.e., dynamics and thermodynamics), before attempting to unify relativity and quantum mechanics (Gal-Or 1987: 29ff, 47-48). He rejects, furthermore, attempts at unification which give a leading role to quantum mechanics and to an information-theoretical understanding of organization. Quantum mechanics and information theory both treat the universe as first and foremost a statistical ensemble. Order and disorder are, in this context, fundamentally subjective concepts, and the expectation of an evolution towards maximum "entropy" or chaos is already given in the statistical underpinnings. Quantum mechanics, furthermore, cannot theorize even this sort of change, since it has an irreducibly reversible understanding of time, and can by made to yield time asymmetries only by imposing unexplained boundary conditions. Because of this, he argues, it must be treated strictly as a local theory and not as the matrix for unification (Gal-Or 1987: 47-48, 261-262, 374).

Gal-Or assigns priority instead to general relativity and to the gravitational processes which it describes. It is gravity which drives cosmic expansion and galaxy and star formation, and thus nucleosynthesis, and the emergence of chemistry, life, and intelligence (Gal-Or 1987: 41-46, 154). Once gravity driven phenomena are taken into account, furthermore, it becomes clear that the direction of evolution is not towards chaos, but rather towards even higher degrees of organization, understood as complexity (an increased diversity of elements) coupled with "centreity"-i.e., the closing of these elements in on themselves (Gal-Or 1987: 382). Gal-Or retheorizes thermodynamics in a way which is free of the "subjectivist" concept of entropy, so that science terminates in a recognition of the ultimate unity and organization of all things-what he calls Hayavism, after the Hebrew word for the whole (Gal-Or 1987: 348ff).

The strengths of this approach notwithstanding, there are problems. First of all, while Gal-Or's critique of quantum mechanics is powerful, he does not show exactly what we should do with it. It is one thing to relegate it to the status of a special theory and quite another thing to unify that special theory with his larger relativistic and thermodynamic framework. Gal-Or's synthesis, furthermore, is dependent on the larger Big-Bang cosmology, the empirical problems of which he seems unaware, or at least chooses not to address. Gal-Or's understanding of organization, finally, is seriously constrained by his insistence on the priority of physical concepts. While gravity can produce an objective structuring, he does not show how it produces purposefulness, nor does he ever really settle the question of the ultimate purposefulness of the universe, remaining caught, as it were, between Aristotle, to whom he aspires, and Spinoza, with whom he is ultimately more comfortable.

The third approach to unification, holism, and emergent organization from within the physical sciences illustrates especially well how substantive rejection of key aspects of atomism and mechanism can exist side by side with an understanding of "unification" which remains wholly within the old paradigm. I would like to discuss two distinct examples of this approach: the work of David Bohm (Bohm 1980), and that of Ilya Prigogine and his students.

Bohm's starting point is a recognition that both of the great theoretical innovations in physical theory of this century, relativity and quantum mechanics, call into question the atomism which had dominated science since at least the time of Newton. Relativity calls this idea into question because the notion of rigid bodies and point particles implies signals faster than light, which relativity forbids; relativistic theory understands the cosmic order rather in terms of events

and processes in a universe which must be regarded as an unbroken whole (Bohm 1980: 123-125). Quantum theory calls atomism into question with its concepts of particle-wave duality and nonlocality (Bohm 1980: 128-129). At the same time, he argues, both relativity and quantum mechanics fail to break decisively with atomism, mutually contradictory elements of which they conserve. The concept of signaling which is central to relativity implies an autonomy between events which quantum mechanics forbids; quantum theory, for its part, assumes the autonomy of quantum states prior to observation in a way which contradicts relativity. Behind this contradiction lies the underlying difference between discrete quantum and continuous relativistic orders (Bohm 1980: 136-137).

Bohm proposes to resolve these difficulties with the notion of an underlying implicate order, various aspects of which different theories "revelate" or make explicit, while concealing others, much as quantum measurement reveals position or momentum, but not both (Bohm 1980: 144ff).

Prigogine, similarly, begins by acknowledging what he regards as the fundamental weakness of mechanistic science-its inability to explain the emergence of complex organization. His early work was centered on remedying this weakness by extending thermodynamics far into the nonlinear, nonequilibrium region and showing that, provided there is sufficient exchange of matter and energy with the environment, fluctuations from equilibrium (disorder) far from being damped may in fact be amplified, leading to large scale organization in space and time (Prigogine 1977: 49-61, Prigogine and Stengers 1984: 140-141). This insight has profound importance for understanding how life could have emerged from inorganic matter, as well as for the description of a wide range of inorganic and organic processes which have hitherto seemed at variance with physical theory. Prigogine's approach has also inspired the first real break with the hegemonic Big-Bang cosmology. Hannes Alfvn and Eric Lerner (Lerner 1991) have argued that the facts of cosmic evolution are most economically explained by a universe infinite in space and time (so that any region, however large, has an environment with which it can exchange matter and energy) over the expanse of which fluctuations (initially electromagnetic, later gravitational) have produced departures from equilibrium, leading to the formation of large scale structures, and eventually of galaxies, stars, etc.

Prigogine (Prigogine 1979) then goes on to unify this expanded thermodynamics with dynamic theory using mathematical formalisms similar to those used in quantum mechanics. He does this by treating entropy and time as noncommuting operators, like position and momentum in quantum mechanics. The result is to unify dynamics and thermodynamics without reducing one to the other. More recently he has used a similar strategy to introduce irreversibility into quantum mechanics itself, using superoperators which are both noncommuting and nondistributive (Prigogine and Petrosky 1988). While the specifics are quite different, and while Prigogine's approach is more conservative, the basic strategy for unification is not unlike Bohm's.

That Bohm sees something fundamental when he talks about the "undivided" nature of the universe is unquestionable. Similarly, the fact that Prigogine has made critical contributions to our understanding of the universe, and more specifically to the defeat of cosmological pessimism and entropism, can hardly be doubted. Precisely because of these contributions, however, the limitations of their strategies stand out clearly. This is especially true of their basically common strategy for unification, which ultimately seems to have little in the way explanatory power. It replaces a larger number of more specific formalisms with a smaller number of more general ones, and even, in the more radical form advocated by Bohm, develops a metaformalism in which all possible formalisms can be embedded. It describes more with less. But Bohm cannot explain physically the relationship between the continuous and discrete orders "revelated" by relativity and quantum mechanics respectively, nor can he reconcile physically the apparent contradiction between the relativistic prohibition on faster than light signalling and quantum nonlocality. Similarly, Prigogine's formalisms do not really tell us how irreversible change emerges from reversible (in this Gal-Or is superior) nor do they show, contra Tipler, that reversibility is an illusion, or to be more precise, an artifact of a certain kind of formal abstraction.

But even at the substantive level there are problems. What is matter if it is not particles? Similarly, Prigogine's theory of self-organization is ultimately descriptive rather than explanatory. It tells us how complex organization is possible, and how it emerges, but not why. We never advance to a principle which can explain why the universe is, and is as it is, and not otherwise. Similarly, as with Gal-Or, we never get an argument regarding the ultimate meaningfulness of the universe. Clearly this is not because either scientist is hostile to the idea. On the contrary, both are clearly friends of progress, human and cosmic, and make it clear that they would welcome a convincing argument that our labors here are not in vain. Rather, it is a result of the limitations of mathematical physics itself, of the tyranny of the formalism and of what the Thomistic tradition calls formal abstraction, which allows us to grasp the structure or order of a system, without telling us what it is, something which depends of what Aquinas called "total abstraction," or why, which depends on "separation" or what we prefer to call "transcendental abstraction (Aquinas, In Boethius De Trinitate Q5,6)." The concepts of essence and of final cause simply have no place within mathematical formalism. And only a scientific strategy which has room for these concepts can generate a complete explanation, telling us what things are and why as well as how, and only a such scientific strategy can approach in an open-ended way the question of the ultimate meaningfulness of the universe.

We are now in a position to suggest a solution to the current crisis in the sciences. Our starting point is the growing evidence for meaning and purpose in the universe (cosmological fine-tuning, self-organization . . .) coupled with the inability of existing theory either to theorize this kind of cosmic teleology or to resolve its own empirical difficulties and internal contradictions. At the same time, it is clear that mathematical physics has produced authentic progress, both theoretical and practical and ought not to be abandoned entirely. Rather, it is time to admit into science, including the physical sciences, thinking which represents both lower and higher degrees of abstraction. Mathematical formalism is an essential step in the struggle for knowledge because it makes possible rigorous definition, from which alone inference is possible. At the same time, it abstracts from the actual organization of systems in a way which may (this must be demonstrated) be seriously distorting our picture of the universe. Thus the residual atomism of physical theories which proclaim the interconnectedness of all things, thus the problem of reversibility in dynamics and entropy in thermodynamics, all of which, I would like to suggest, may be artifacts of abstraction. We must allow the agenda of the sciences-the account of what must be described and explained-to be set at a lower degree of abstraction, that of totalization. Totalization is essentially a taxonomy which grasps the specific differences and the similarities of phenomena and lays the groundwork for formalization. A universal taxonomy would force mathematical descriptions to take into account the full body of phenomena experienced by humanity, while forcing physics to think about such questions as what the underlying interconnectedness of things tells us about the nature of matter itself.

More important, however, is a new openness to transcendental abstraction. This is the kind of abstraction which moves from the judgements of internal consistency to judgements of Being, Beauty, Truth, Goodness, Integrity, etc.

While properly the province of philosophy, this kind of abstraction has a role in the special sciences as well. First, from the methodological side, it alone can provide a criterion for what constitutes truth. Second, on the substantive side, it is the transcendentals (Being, Beauty, True, Good, One) which ultimately define teleological processes. A judgement that the universe is ordered to the generation of increasingly complex levels of organization, would at once be a judgement that it is Good, and offer an insight into what the Good is. Transcendental abstraction is thus an essential component of complete scientific explanation. At the same time, clarity on the transcendental character of the judgement involved will assure the theoretical openness necessary to accommodate evidence of chaos, contradiction, and disintegration. Cosmic teleology does not require (indeed ultimately cannot require) ordering to some finite end or the existence of perfect finite physical structures.

Openness to transcendental abstraction does not, to be sure, by itself settle the question of cosmic teleology. That depends on empirical research. But it does provide us with a way to account for the evidence we have, and continue our search for truth in a spirit of hope and wonder at the beauty, intelligibility, and goodness of the universe, a wonder which is the mother of all scientific exploration.