Synergy and Self-organization in the Evolution of Complex Systems

Synergy of various kinds has played a significant creative role in evolution; it has been a prodigious source of evolutionary novelty. Elsewhere it has been proposed that the functional (selective) advantages associated with various forms of synergistic phenomena have been an important cause of the "progressive" evolution of complex systems over time. Underlying the many specific steps in the complexification process, a common functional principle has been operative. Recent mathematical modelling work in biology, utilizing a new generation of nonlinear dynamical systems models, has resulted in a radically different hypothesis. It has been asserted that "spontaneous," autocatalytic processes, which are held to be inherent properties of living matter itself, may be responsible for much of the order found in nature and that natural selection is merely a supporting actor. A new "physics of biology" is envisioned in which emerging natural laws of organization will be recognized as being responsible both for driving the evolutionary process and for truncating the role of natural selection. This article describes these two paradigms in some detail and discusses the possible relationship between them. Their relevance to the process of human evolution is also briefly discussed. SYNERGY VS. SELF ORGANIZATION It has always seemed to me ironic that we are surrounded and sustained by synergistic phenomena -combined (or "co-operative") effects that can only be produced by two or more component parts, elements or individuals -yet we do not, most of us, seem to appreciate its importance; we take its routine miracles for granted. Nor do evolutionists, for the most part, seem to recognize the important causal role of synergy in the evolutionary process, despite the fact that we depend upon it in a myriad of ways for our survival and reproductive success, and so do all other living things. Synergy is literally everywhere around us, and within us; it is unavoidable. Here are just a few examples: • About 2,000 separate enzymes are required to catalyze a metabolic web. But if you were to remove one of the more critical of these enzymes, say the hexokinase that facilitates glycolysis, the process would not go forward. • Water has a unique set of emergent, combinatorial properties that are radically different from those of its two constituent gases. But if you simply mix the two gases together without a catalyst like platinum, you will not get the synergy. • Our written language, with well over 300,000 words, is based on various combinations of the same 26 letters. Thus, the letters "o," "p" and "t" can be used to make "top", "pot", "opt" and "p.t.o." (paid time off). But if you remove the vowel, there will be no "pattern recognition" in the reader's mind. • The humble clay brick can be used to make a great variety of useful structures -houses, walls, factories, jails, roads, watchtowers, fortifications, even kilns for making more bricks. Truly a synergistic technology. But without mortar and human effort (and a plan), you will have only a pile of bricks. • A modern automobile is composed of roughly 15,000 precisely-designed parts, which are derived from some 60 different materials. But if a wheel is removed, this incredible machine will be immobilized. • The African honey guide is a bird with a peculiar taste for bees' wax, a substance that is more difficult to digest even than cellulose. Moreover, in order to obtain bees' wax, the honey guide must first locate a hive then attract the attention of and enlist a co-conspirator, the African badger (ratel)(Mellivora capensis). The reason is that the ratel has the ability to attack and dismember the hive, after which it will reward itself by eating the honey while leaving the wax. However, this unusual example of co-operative predation between two different species in fact depends upon a third, unobtrusive co-conspirator. It happens that honey guides cannot digest bees' wax. They are aided by a parasitic gut bacterium which produces an enzyme that can break down wax molecules. So this improbable but synergistic feeding relationship is really triangular. And, needless to say, the system would not work if any of the partners, for whatever reason, withdrew (Bonner, 1988). • Economist Adam Smith's classic description in The Wealth of Nations (1776) of an eighteenth century pin factory is often cited as a paradigm example of the "division of labor." Smith observed that 10 laborers, by dividing up the various tasks associated with making pins, were able collectively to produce about 48,000 pins per day. However, Smith opined that if each laborer were to work alone, doing all of the tasks independently, it was unlikely that on any given day the factory would be able to produce even a single pin per man. THE UBIQUITY OF SYNERGY Synergy is clearly not a peripheral phenomenon associated only with drug interactions or corporate mergers. Though it often travels in disguise, synergy can be found in the subject-matter of most, if not all of the academic disciplines. In physics, it is associated with the behavior of atoms and subatomic particles, as well as with superconductivity, synchronous light emissions (lasers) and such esoteric molecular phenomena as scale effects -the "broken symmetries" highlighted in physicist Perry Anderson's classic article "More is Different" (1972). Indeed, the periodic table of elements is a monument to the many forms of synergy that are responsible both for the naturally occurring stable elements and for the more unstable or even transitory creations of modern physics; various combinations of atomic building-blocks produce substances with very different "emergent" properties. Even the "chaotic" phenomena which have been the subject of intensive research by physicists and mathematicians in recent years exhibit many forms of synergy. Biochemistry and molecular biology are also rife with synergy. Living matter (at least in the form we know it) is composed mainly of a few key constituents -carbon, oxygen, hydrogen, nitrogen and energy. In various configurations these constituent parts have produced a wondrous array of emergent products, perhaps 10-20 million different species -nobody really knows. By the same token, as we all know, the DNA that is used to write the genetic code consists of only four nucleotide "letters." With this modest alphabet, evolution has been able to fashion a human genome with perhaps 100,000 genes. During ontogeny, our genome is able co-operatively to fabricate an incredibly intricate emergent product composed of an estimated 500 trillion cells of about 250 different types. Many individual organisms, from bacteria to humans, also engage in internal or external symbiosis -synergistic relationships with "dissimilar" organisms -a subject that will be discussed in more detail below. Sociobiologists, likewise, have documented numerous behavioral synergies among members of the same species, from co-operative foraging and hunting activities to co-operative defense, reproduction, environmental conditioning and even thermoregulation. (More also about sociobiology below.) In the social sciences, synergy can be found in many of the phenomena studied by economists -from market dynamics (demand-supply relationships) to economies of scale, the division of labor and, of course, the influence of technology. Psychologists also deal with synergistic effects, ranging from gestalt phenomena to social facilitation, group "syntality," mob psychology and cult behavior. And political scientists observe synergistic effects in voting processes, interest group activity, coalition behavior, and a host of organizational phenomena, among other things. The computer sciences are also grounded in synergy. There is, for example, the microscopic complexity of Intel's Pentium microprocessor, which embodies the equivalent of 3.1 million transistors in a substrate that is about 2.17 inches square (it varies with the temperature). There is also the current generation of word processing software, which utilizes -synergistically -about two million separate lines of programming code, or instructions. Or consider the multi-leveled synergy that occurs when a computer and its software are combined. We know that the result is synergistic because we also know what happens when the two are not combined, or when the computer and software are incompatible. Similarly, massively parallel computers, which in effect exploit the synergies associated with a division of labor and hierarchical control, offer performance improvements that are many orders of magnitude greater than what can be achieved by conventional sequential processing technology. THE SYNERGISM HYPOTHESIS What is the principle underlying such mundane forms of magic? It is not magic at all, of course, but a fundamental characteristic of the material world that things in various combinations, sometimes with others of like kind and sometimes with very different kinds of things, are prodigious generators of novelty. And these novel cooperative effects have over the past 3.5 billion years or so produced at every level of life distinct, irreducible "higher levels" of causation and action whose constituent "parts" have been extravagantly favored by natural selection. Furthermore, in many instances these emergent wholes have themselves become parts of yet another new level of combined effects, as synergy begat more synergy. The formal hypothesis is that synergistic effects of various kinds have been a major source of creativity in evolution (see Corning, 1983); the synergism hypothesis asserts that it was the functional (selective) advantages associated with various forms of synergy that facilitated the evolution of complex, functionally-organized biological and social systems. In other words, underlying each of the many particular steps in the complexification process, a common functional principle has been at work. THE SELF-ORGANIZATION PARADIGM The recently developed theories of self-organization would seem to be orthogonal to this functionalist, selectionist theory. Mathematical modelling work in biophysics, utilizing a new generation of non-linear partial differential equations, has produced a radically different hypothesis about the sources of biological order. As articulated by Stuart Kauffman in an important new synthesis (Kauffman, 1993), much of the order found in nature may be "spontaneous" and autocatalytic -a product of the generic properties of living matter itself. Kauffman envisions a new physics of biology in which the emerging natural laws of organization will be recognized as being responsible both for driving the process and for truncating the role of natural selection. Natural selection in Kauffman's paradigm is viewed as a supporting actor. This article will explore the relationship between synergy and self-organization in some detail in the hope of shedding additional light on how complex systems have evolved and how they may be expected to continue doing so over the course of time. The relevance of these two major theoretical paradigms to the process of human evolution will also be briefly discussed. THE EVOLUTION OF COMPLEXITY AS A THEORETICAL CHALLENGE Complexity seems of late to have become a buzzword. There have even been popular books chronicling the research and theory that have burgeoned in this category (Lewin, 1992; Waldrop, 1992). Nevertheless, the underlying theoretical challenge is not new. Attempts to explain the origins and evolution of living systems can be traced back at least to the Old Testament. Even a concept as fashionable as autocatalysis can be found in the writings of Aristotle, the first great biologist, who developed what has become an enduring theme in western natural science. Aristotle postulated an intrinsic directionality, or unfolding process in nature (orthogenesis), which was distilled in his concept of physis. Aristotle also inspired the concept of an ascending ladder of perfection, or hierarchy, that later came to be associated with the Latin term scala naturae (Granger, 1985; Lovejoy, 1936). At the beginning of the 19th Century, the French naturalist Jean Baptiste de Lamarck postulated a "natural tendency" toward continuous developmental progress in nature, energized by what he called the "power of life" (Lamarck, 1963 [1809]). Likened by Lamarck to a watch spring, it involved the idea that living matter has an inherent developmental energy. Orthogenetic theories of evolution reached an apogee of sorts during the 19th century with the multi-volume, multi-disciplinary magnum effort of Herbert Spencer, who was considered by many contemporaries to be the preeminent thinker of his era. Spencer formulated an ambitious "Universal Law of Evolution" that spanned physics, biology, psychology, sociology and ethics. In effect, Spencer deduced society from energy by positing a cosmic progression from energy to matter, to life, to mind, to society and, finally, to complex civilization. "From the earliest traceable cosmical changes down to the latest results of civilization," he wrote in "The Development Hypothesis" (Spencer, 1892 [1852]), "we shall find that the transformation of the homogeneous into the heterogeneous is that in which progress essentially consists." Among other things, Spencer maintained that homogeneous systems are less stable than those that are more differentiated and complex. (It is worth noting that, while Spencer viewed this progression as "spontaneous" in origin, he also believed that it was sustained by the fact that more complex forms are functionally "advantageous.") There have been many less imposing vitalistic and orthogenetic theories since Spencer's day, ranging from Henri Bergson's élan vital to Hans Driesch's Entelechie, Pierre Tielhard de Chardin's Omega point, Pierre Grassé's idiomorphon, and Jean Piaget's savoir faire. However, in this century Darwin's theory of natural selection has cast a long shadow over various autocatalytic theories. Darwin seemed to be rebutting Lamarck and Spencer directly when he wrote in The Origin of Species "I believe in no fixed law of development." And again: "I believe...in no law of necessary development" (1968 [1859]:p.318,348]. One of the formulators of the so-called modern synthesis (or sometimes neo-Darwinian synthesis), the late Theodosius Dobzhansky, put the matter succinctly: "Natural selection has no plan, no foresight, no intention" (1975:p.377). A striking illustration is the eye, that revered object of 19th century natural theology. We now know that the eye did not unfold deterministically or arise fullblown. It developed independently on perhaps 40 different occasions in evolutionary history, utilizing at least three different functional principles -the pinhole, the lens and multiple tubes. Nor do all the eyes of a similar type work in the same manner. While the evidence for natural selection as a directive agency in evolution is overwhelming, many theorists over the years have felt that the neo-Darwinian synthesis is inadequate, by itself, to account for the undeniable "progressive" trend from the primordial chemical soup to simple one-celled prokaryotes, eukaryotes and, ultimately, to large, complex, socially-organized mammals. The evolution of complexity has seemed to require something more than "random" point mutations in a genetic "beanbag" (to use Ernst Mayr's felicitous caricature). The long-term trend toward greater complexity (in tandem with the many examples of stasis) seems to suggest the presence of some additional mechanism or mechanisms. Some years ago, the late C.H. Waddington articulated these doubts with characteristic bluntness: "The whole real guts of evolution -which is how do you come to have horses and tigers and things -is outside the mathematical theory" (quoted in Rosen 1978:p.371). WHY COMPLEXITY? AND WHAT IS IT? More broadly, the question is: Why does complexity exist? Why have various parts aggregated over time into larger, more complex wholes? And why have many wholes differentiated into various specialized parts? For that matter, what is complexity? And, in the context of modern biology, what are wholes, and parts? The accumulating data on mutualism, parasitism, colonialism, social organization, coevolution and the dynamics of ecosystems have revealed many nuanced interdependencies and have blurred the supposedly sharp demarcation lines among various biological units. Physicist Larry Smarr (1985) has pointed out that complexity is in reality a multidimensional, multi-disciplinary concept; there is no one right way to define and measure it. A mathematician might define it in terms of the number of degrees of freedom in computational operations. A physicist might be concerned with the number and frequency of interactions in a system of interacting gas molecules. The systems theorists of the 1960s were fond of using the rubric (suggested independently by mathematicians Alexei Kolmogorov and Gregory Chaitin) of "algorithmic complexity" the size of the smallest mathematical description of system behavior. Social scientist Herbert Simon (1965) advocated the use of a hierarchical measure -the number of successive levels of hierarchical structuring in a system, or what biologist G. Ledyard Stebbins (1969) characterized as "relational order." For obvious reasons, biologists have traditionally preferred such biologically-relevant measures as the number of parts (say cells), or types of parts (cell types), or the number of interdependencies among various parts. In recent years there have also been a number of efforts to define complexity in relation to thermodynamics, entropy and information (see especially Wicken (1987); Haken (1988); Brooks and Wiley (1988); Weber et al., (1988) and Salthe (1993)). John Tyler Bonner, in his recent book on the evolution of complexity (1988) suggests that biological (and by extension social) complexity should also be defined in terms of the functional nature of living systems. What is most salient about biological systems is not just the number of parts, or even the number of interconnections among the parts, Bonner argues, but the division of labor (and the combining of capabilities) which result; these are the distinctive hallmarks of biological complexity. In other words, biological complexity should be associated with the functional synergies that it produces. In recent years there has also been increasing acceptance of the views of biologists Ludwig von Bertalanffy (1950, 1967), W. Ross Ashby (1952, 1956), C.H. Waddington (1962, 1968), Paul Weiss (1971) and others that biological complexity is characterized by cybernetic properties; it is not just ordered but also organized (see also Norbert Weiner, 1948; William T. Powers, 1973 and James G. Miller, 1978). That is, biological (and social) systems are distinctive in being goal-oriented (or teleonomic), hierarchically-organized and self-regulating (they display processes of feedback control), as well as being uniquely self-developing and self-determining. The physical chemist Engelbert Broda (1975) stressed the functional imperatives: "The more the division of labor was developed [in evolution], the more important became intercellular and interorganismal communication and control. Hence, for an understanding of more complicated systems, thermodynamics and kinetics must increasingly be supplemented by cybernetics, by applied systems analysis." In hindsight, Broda might have added molecular (morphogenetic) and intracellular communication and control to the list of biological processes with cybernetic properties. One other distinctive feature of complex living systems is that they cannot be fully understood, nor their evolution and operational characteristics fully explained, by an exclusive focus either on the system as a "whole" or on the component "parts". Though the long-standing "holism-reductionism" debate in the sciences still lingers, chemist Michael Polanyi's classic article "Life's Irreducible Structure" (1968) remains the definitive peace-making effort. Polanyi noted that, in the process of constructing a complex living system, the causal dynamics are in fact multi-levelled. On the one hand, the properties of the whole are constrained and shaped by the properties of the parts, which in turn are constrained and shaped by the lower-level properties of their constituent raw materials, and by the laws of physics and chemistry. To a devout reductionist, this is a truism which modern science daily reconfirms. On the other hand, the extreme reductionist argument that an understanding of the parts fully explains the whole leads to what C.F.A. Pantin called the "analytic fallacy." Polanyi pointed out that a whole also represents a distinct, irreducible level of causation which "harnesses," constrains and shapes lower level parts and which may in fact determine their fates. In effect, wholes may become both vessels and selective fields for the parts -and may even come to exercise hierarchical, cybernetic control over the parts. Moreover, wholes can do things that the parts cannot do alone. An automobile cannot be fully understood or its operation explained by separate descriptions of how each part works in isolation. Not only is the design of each part affected by its role and relationship to the whole but its performance and functional consequences may only be comprehensible in terms of its interaction with other parts and the whole. (See the discussions in Corning, 1983 and Haken, 1973, 1977, 1983; also, cf., the concept of "interactional complexity" in Wimsatt, 1974.) Thus, an automotive engineer must always look both upward and downward (and horizontally) in the hierarchy of causation when trying to comprehend the operation of any part. And the same applies to the students of living systems. Evolution has produced several emergent levels of wholes and parts. Furthermore, the power and impact of these emergent wholes has greatly expanded over the course of time; complexity has been at once a product of evolution and a cause of evolution (an important point to which we will return below). And yet, the question remains: Why complexity? How do we account for the "progressive" evolution of complex systems? As noted above, two alternative theoretical approaches are currently in contention -the somewhat tattered neo-Darwinian (functionalist) theory and the theory of autonomous self-organization (autocatalysis). Until a few years ago, the neo-Darwinian explanation, while subject to vigorous debate over the details, was essentially uncontested; it was assumed that the trend toward biological complexification was functionally-driven. But the nascent science of complexity has challenged the selectionists' hegemony; non-Darwinian vitalistic/orthogenetic theories -now respectably clothed in a new wardrobe of nonlinear, dynamical systems models -are again in vogue. Let us consider each of these explanatory paradigms in more detail. SELF-ORGANIZATION AND COMPLEXITY "Self-organization" is almost as much of a buzzword these days as "complexity". However, it is not a newly discovered phenomenon. Aristotle enshrined it in his classic metaphor about the growth of acorns into oak trees. The pioneering nineteenth century embryologists, such as Karl Ernst von Baer, also appreciated, and observed, selforganization in the process of morphogenesis. But more important, self-organization is also compatible with Darwin's theory. Modern neo-Darwinians, following the lead of Francisco Ayala (1970), Theodosius Dobzhansky (1974) and Ernst Mayr (1974a,b), have generally associated self-organization with Colin Pittendrigh's term "teleonomy" (evolved purposiveness) and the concept of an internal "program" (Roe and Simpson, 1958). In this formulation, self-organization has been equated with the mechanisms of cybernetic self-regulation and feedback. Self-organization is viewed as being a product of, and subordinate to, natural selection. Darwin also categorically rejected the idea of an inherent energizing or directive force in evolution, as mentioned earlier. However, it is important to note that the theory of evolution via natural selection does not stand or fall on this issue, so long as any autocatalytic processes are (a) of a materialist nature, (b) empirically verifiable and (most important) (c) subject to testing for their functional (fitness) consequences in relation to survival and reproduction.