The Emergence of 'Emergence':
Now What?

Peter A. Corning
Institute for the Study of Complex Systems, USA


It seems that emergence has finally emerged: It's the buzzword of the hour among complexity theorists. However, there is also a serious problem. Despite the recent proliferation of writings on the subject, it is still not clear what the term denotes or, more important, how emergence emerges. One difficulty is that the term is frequently used (as I have here to illustrate the point) as a synonym for “appearance” or “growth,” as distinct from a parts-whole relationship. Thus, one of the dictionaries I consulted defined the term strictly in perceptual terms and gave as an example “the sun emerged from behind a cloud.” Even the Oxford English Dictionary, which offered four alternative definitions, gives precedence to the version that would include a submarine that submerges and then re-emerges.

It is not surprising, then, that the overwhelming majority (close to 100 percent) of the new journal articles on “emergence” and “emergent” that are identified each week by my computer search service involve such subjects as the emergence of democracy in Russia, the emergence of soccer as a school sport in the US, the emergence of the internet, the emergence of mad cow disease, and the like. I have deliberately played on this conflation of meanings above, but even avowed complexity theorists commonly use the term (perhaps unwittingly) in both ways. For instance, the subtitle of Mitchell Waldrop's book Complexity (1992) is The Emerging Science at the Edge of Order and Chaos.

Unfortunately, some theorists seem to take the position that emergence does not exist if it is not perceived; it must be apparent to an observer. But what is a “whole”—how do you know it when you see it, or don't see it? And is the mere perception of a whole—a “gestalt” experience—sufficient, or even necessary? Complexity theorist John Casti (1997) associates emergence with dynamic systems whose behavior arises from the interaction among their parts and cannot be predicted from knowledge about the parts in isolation. “The whole is bigger than the sum of its parts,” echoed Michael Lissack in the inaugural issue (1999) of Emergence. John Holland (1998), by contrast, describes emergence in reductionist terms as “much coming from little” and imposes the criterion that it must be the product of self-organization, not centralized control. Indeed, Holland tacitly contradicts Casti's criterion that the behavior of the whole is irreducible and unpredictable.

Perhaps the most elaborate recent definition of emergence was provided by Jeffrey Goldstein (1999). To him, emergence refers to “the arising of novel and coherent structures, patterns and properties during the process of self-organization in complex systems.” The common characteristics are: radical novelty (features not previously observed in the system); coherence or correlation (meaning integrated wholes that maintain themselves over some period of time); a global or macro “level” (i.e., there is some property of “wholeness”); it is the product of a dynamical process (it evolves); and it is “ostensive”—it can be perceived. For good measure, Goldstein throws in “supervenience.”

Goldstein's definition is hardly the last word on this subject, however. One indication of the ambiguous status that the term emergence currently holds in complexity science is the discordant dialog that occurred in an online (internet) discussion of the topic hosted by the New England Complex Systems Institute (NECSI) during December 2000 and January 2001. Some participants claimed that emergence has to do with concepts and perceptions—when an observer recognizes a “pattern.” Others claimed that perception is irrelevant; emergence can occur when nobody is there to observe it. Some saw the mind as an emergent result of neural activity, while others disagreed. One found emergence in language, where meaning emerges from combinations of letters and words. Another called society an emergent that is in turn composed of emergent collections of cells. Still another identified emergence with what happens when water boils and turns to steam—something new in the macro world emerges from the micro world. Temperature and pressure were also called emergents—macro-level averages of some quantity present in micro-level phenomena.

In addition, emergence was likened to a dynamical attractor, or some “deep structure”—a pre-existing potentiality. However, another participant objected that dynamical attractors are mathematical constructs—they say nothing about the underlying forces. To one participant, emergence involves rule-governed creativity based on finite sets of elements and rules of combination. But others disagreed—if the properties of the whole can be calculated from the parts and their interactions, it is not emergence; emergence does not have logical properties; it cannot be deduced (predicted). Another participant replied maybe not, but once observed future predictions are possible if it is a deterministic process. In support of this viewpoint, another discussant asserted that a “very simple example” is water, and its properties should in principle be calculable by detailed quantum-level analysis. However, a discussant familiar with quantum theory disagreed. Given the vast number of “choices” (states) that are accessible at the quantum level, one would, in effect, have to read downward from H20 to make the right choice. Yet another discussant pointed out that quantum states are always greatly affected by the boundary conditions—the environment.

In short, contradictory opinions abound. There is no universally acknowledged definition of emergence, nor even a consensus about such obvious examples as water. And if emergence cannot be defined in concrete terms—so that you will know it when you see it—how can it be measured or explained? As Jeffrey Goldstein noted, “emergence functions not so much as an explanation but rather as a descriptive term pointing to the patterns, structures or properties that are exhibited on the macro-scale” (Goldstein, 1999: 58). And Michael Lissack acknowledged that

it is less than an organized, rigorous theory than a collection of ideas that have in common the notion that within dynamic patterns there may be underlying simplicity that can, in part, be discovered through large quantities of computer power ... and through analytical, logical and conceptual developments. (Lissack, 1999: 112)


How can we sort all of this out? The place to start, I believe, is with the more inclusive (and more firmly established) concept of “synergy.” I have treated this concept in depth elsewhere (Corning, 1983, 1995, 1996, 1997, 1998a, 1998b, 2003), so here I will be brief. (See also the two volumes on the evolution of complexity by Maynard Smith & Szathmáry, 1995, 1999.) Broadly defined, synergy refers to the combined (cooperative) effects that are produced by two or more particles, elements, parts or organisms—effects that are not otherwise attainable. In this definition, synergy is not “more” than the sum of the parts, merely different. It is associated with cooperative effects of all kinds, from water molecules to laser beams, pack-hunting predators, and human technologies.

Furthermore, there are many different types of synergy. One important category involves what can be called “functional complementarities,” effects produced by new combinations of different parts. Water is an obvious example, but so is sodium chloride—ordinary table salt. NaCl is composed of two elements that are toxic to humans by themselves, but, when they are combined, the resulting new substance is positively beneficial (in moderate amounts). Another commonplace example is Velcro, where the two opposing strips, one with many small hooks and the other with loops, are able to create a secure bond with one another.

Another important form of synergy—in living organisms and complex social organizations alike—involves the division of labor (or what could perhaps more felicitously be called a “combination of labor”). Anabaena provides an unusual example. Anabaena is a cyanobacterium that engages in both photosynthesis and nitrogen fixing. However, these two processes are chemically incompatible. So Anabaena has evolved a way of compartmentalizing these two functions. The nitrogen fixing is done in separate heterocysts, and the products are then passed through filaments to other cells (Shapiro, 1988). In more complex organisms—from eukaryotic protists to social mammals—specialization among the parts produces many advantages for various “wholes.”

Likewise, there are many different kinds of “symbiosis” between two or more different species in the natural world that involve a division/combination of labor. Thus, virtually all species of ruminants, including some 2,000 termites, 10,000 wood-boring beetles, and 200 Artiodactyla (deer, camels, antelope, etc.) are absolutely dependent on the services provided by endosymbiotic bacteria, protoctists, or fungi for the breakdown of the cellulose in plants into usable cellulases (Price, 1991). Then there are the nitrogen-fixing microbes—Rhizobia—that provide an indispensable chemical for plants (and their predators, including humans), in return for vital energy resources.

Still another form of synergy involves what I refer to as a “synergy of scale”—an aggregation of interchangeable, like-kind parts that produce unique cooperative effects (say a river, or a sand pile). Indeed, many synergies of scale produce yet another form of synergy commonly known as “threshold effects” (say a flood, or an avalanche). An elegant example involves the Volvocales, a primitive order of marine algae that form colonies of different sizes, from a handful of cells to quasi-organisms with several dozens to hundreds of functionally integrated cells. As it happens, Volvocales are subject to predation from filter feeders, and a detailed study some years ago by the biologist Graham Bell (1985) documented that Volvox, the largest of the Volvocale species, is virtually immune to filter feeders. The reason, as it turned out, was that there is an upper limit to the prey size that the filter feeders can consume. In a similar vein, in the orb web spider, Metabus gravidus, 15-20 females are able to produce a synergy of scale when they band together to build a giant collective web that can span a stream where their prey are especially abundant (Maynard Smith, 1982).

These and many other forms of synergy—such as joint environmental conditioning, information sharing and joint decision making, animal-tool “symbioses,” gestalt effects, cost and risk sharing, convergent effects, augmentation or facilitation (e.g., catalysts), and others—are discussed in various recent and forthcoming publications by this author cited above.

It should also be stressed that, far from being vague or ephemeral, synergistic effects are, as a rule, very concrete and eminently measurable. To cite one of the many examples in the publications cited above, during the bitterly cold Antarctic winter emperor penguins (Aptenodytes forsteri) huddle together in dense colonies, sometimes numbering 10,000 or more, for months at a time. In so doing, they are able to share precious body heat and provide insulation for one another. A careful study of this collective behavior many years ago showed that these animals were thereby able to reduce their individual energy expenditures by up to 50 percent (Le Maho, 1977).

Similarly, in a comparative study of reproduction among southern sea lions (Otaria byronia) during a single breeding season, it was documented that only one of 143 pups born to gregarious group-living females died before the end of the season, compared to a 60 percent mortality rate among solitary mating pairs. The main reasons were that pups in colonies were protected from harassment and infanticide by subordinate males and were far less likely to become separated from their mothers and die of starvation (Campagna et al., 1992). In short, functional synergies are the source of many “economies” in the natural world.

A crucial corollary of this point is that the synergistic effects produced by “wholes” provide a definitive answer to the charge that wholes are merely “epiphenomena”—nothing more than an expression of their parts. In a nutshell, a whole exists when it acts like a whole, when it produces combined effects that the parts cannot produce alone. Moreover, the synergies produced by wholes provide a key to understanding “why” complex systems have evolved. (We will return to this crucial point shortly.) And if there is any doubt about the matter, one can test for the presence of synergy by removing an important part and observing the consequences—a test first suggested by Aristotle in the Metaphysics (Book H, 1043b-4a), to my astonishment. I call it “synergy minus one.” As a thought experiment, imagine the consequences if you were to remove the gut symbionts from a ruminant animal. Or imagine the consequences for an automobile of removing, say, a wheel, or the fuel supply, or the ignition key, or the driver for that matter. Of course, there are also a great many cases where the removal of a single part may only attenuate the synergy; you may have to remove more than one part to destroy the synergy completely. (Call it synergy minus n.) Thus, if you take away a chrome strip from a car, it may only affect the sale price.


Accordingly, some of the confusion surrounding the term “emergence” might be reduced (if not dissolved) by limiting its scope. Rather than using it loosely as a synonym for synergy, or gestalt effects, or perceptions, and so on, I would propose that emergent phenomena be defined as a “subset” of the vast (and still expanding) universe of cooperative interactions that produce synergistic effects of various kinds, both in nature and in human societies. The definition I suggest in fact accords closely with its original meaning, back in the nineteenth century.

The philosopher David Blitz in his definitive history of emergence entitled, appropriately enough, Emergent Evolution: Qualitative Novelty and the Levels of Reality (1992) reports that the term “emergent” was coined by the pioneer psychologist G. H. Lewes in his multivolume Problems of Life and Mind (1874-9). Like many post-Darwinian scientists of that period, Lewes viewed the evolution of the human mind as a formidable conundrum. Some evolutionists, like Alfred Russel Wallace (the co-discoverer of natural selection), opted for a dualistic explanation. The mind is the product of a supernatural agency, he claimed. But Lewes, following the lead of the philosopher John Stuart Mill, argued that, on the contrary, certain phenomena in nature produce what he called “qualitative novelty”—material changes that cannot be expressed in simple quantitative terms; they are emergents rather than resultants. To quote Lewes:

Every resultant is either a sum or a difference of the cooperant forces; their sum, when their directions are the same—their difference, when their directions are contrary. Further, every resultant is clearly traceable in its components, because these are homogeneous and commensurable... It is otherwise with emergents, when, instead of adding measurable motion to measurable motion, or things of one kind to other individuals of their kind, there is a co-operation of things of unlike kinds ... The emergent is unlike its components in so far as these are incommensurable, and it cannot be reduced to their sum or their difference (Lewes, 1874-9: 413).

Years earlier, John Stuart Mill had used the example of water to illustrate essentially the same idea: “The chemical combination of two substances produces, as is well known, a third substance with properties different from those of either of the two substances separately, or of both of them taken together” (1872 [1843]: 371). However, Mill himself had an illustrious predecessor. In fact, both Mill and Lewes were resurrecting an argument that Aristotle had made more than 2,000 years earlier in a philosophical treatise, later renamed the Metaphysics, about the significance of “wholes” in the natural world. Aristotle wrote: “The whole is something over and above its parts, and not just the sum of them all.” (Book H, 1045: 8-10). So the ontological distinction between parts and wholes was not exactly a new idea in the nineteenth century. The difference was that the late Victorian theorists framed the parts-wholes relationship within the context of the theory of evolution and the challenge of accounting for biological complexity.

The basic quandary for holistic theorists of that era was that evolutionary theory as formulated by Darwin did not allow for radically new phenomena in nature, like the human mind (supposedly). As every first-year biology student these days knows, Darwin was a convinced gradualist who frequently quoted the popular canon of his day, natura non facit saltum—nature does not make leaps. (The phrase appears no fewer than five times in The Origin of Species.) Indeed, Darwin rejected the very idea of sharp discontinuities in nature. In The Origin (1968 [1859]) he emphasized what he called the “Law of Continuity,” and repeatedly stressed the incremental nature of evolutionary change, which he termed “descent with modification.” Darwin believed that this principle applied as well to the evolution of the “mind.” In The Descent of Man, he asserted that the difference between the human mind and that of “lower” animals was “one of degree and not of kind” (1874 [1871] I: 70).

Many theorists of that era viewed Darwin's explanation as unsatisfactory, or at least incomplete, and emergent evolution theory was advanced as a way to reconcile Darwin's gradualism with the appearance of “qualitative novelties” and, equally important, with Herbert Spencer's notion (following Lamarck) of an inherent, energy-driven trend in evolution toward new levels of organization. Emergent evolution had several prominent adherents, but the leading theorist of this school was the comparative psychologist and prolific writer Conwy Lloyd Morgan, who ultimately published three volumes on the subject, Emergent Evolution (1923), Life, Mind and Spirit (1926) and The Emergence of Novelty (1933). (Other theorists in this vein included Samuel Alexander, Roy Wood Sellars, C. D. Broad, Jan Smuts, Arthur Lovejoy, and W. M. Wheeler. Jan Smuts, a one-time prime minister of South Africa, deserves special note because his volume, Holism and Evolution (1926), advanced the concept of “holistic selection”—the idea that wholes of various kinds might be units of selection in nature. It was a prescient precursor to such later concepts as David Sloan Wilson's “trait group selection,” John Maynard Smith's “synergistic selection” and my synergism hypothesis; see below.)

The main tenets of Lloyd Morgan's paradigm will sound familiar to modern-day holists: Quantitative, incremental changes can lead to qualitative changes that are different from, and irreducible to, their parts. By their very nature, moreover, such wholes are unpredictable. Although higher-level, emergent phenomena may arise from lower-level parts and their actions, there may also be “return action,” or what Lloyd Morgan was the first to call “supervenience” (meaning “downward causation” in today's parlance). However, most importantly, Lloyd Morgan argued that the evolutionary process has an underlying “progressive” tendency, because emergent phenomena lead in due course to new levels of reality.

While present-day theorists as a rule avoid the value-laden term “progress,” many would agree with the underlying premise. New levels of reality have indeed arisen in the course of the evolutionary process. Accordingly, I propose to adopt the original meaning of emergence. In this definition, emergence would be confined to those synergistic wholes that are composed of things of “unlike kind” (following Lewes' original definition). It would also be limited to “qualitative novelties” (after both Lewes and Lloyd Morgan); that is, unique synergistic effects that are generated by functional complementarities, or a combination of labor. In this more limited definition, all emergent phenomena produce synergistic effects, but many synergies do not entail emergence. In other words, emergent effects would be associated specifically with contexts in which constituent parts with different properties are modified, reshaped, or transformed by their participation in the whole.

In this definition, water and table salt are unambiguous examples of emergent phenomena. And so is the human body. Its 10 trillion or so cells are specialized into some 250 different cell types that perform a vast array of important functions in relation to the operation of the whole. Indeed, in biological systems (and in technological systems like automobiles, for that matter) the properties of the parts are very often shaped by their functions for the whole. On the other hand, in accordance with the Lewes/Morgan definition, a sand pile or a river would not be viewed as an emergent phenomenon. If you've seen one water molecule you've seen them all. It is simply a synergy of scale.

Must the synergies be perceived/observed in order to qualify as emergent effects, as some theorists claim? Most emphatically not in the Lewes/Morgan definition. The synergies associated with emergence are real and measurable, even if nobody is there to observe them. And what about the claim that emergent effects can only be the result of “self-organization”? Is this a requirement? Again, emphatically not. Self-organization is another academic buzzword that is often used rather uncritically. But, as John Maynard Smith (1999) points out, there is a fundamental distinction between self-organizing processes (or, more precisely, what should be called “self-ordering” processes) and wholes that are products of functional organization (as in organ systems). Living systems and human organizations are largely shaped by “instructions” (functional information) and by cybernetic control processes. They are not, for the most part, self-ordered; they are predominantly organized by processes that are “purposeful” (teleonomic) in nature and that rely on “control information.” The role of teleonomy and cybernetic control information in biological evolution is discussed in some depth by this author and a colleague in a number of recent publications (Corning, 1995, 2001; Corning & Kline, 1998a, 1998b).

Consider this example. A modern automobile consists of some 15-20,000 parts (depending on the car and how you count). If all of these parts were to be thrown together in one great “heap” (a favorite word of Aristotle), they could be described as “ordered” in the sense that they are not randomly distributed across the face of the earth (or the universe, for that matter). Nevertheless, they do not constitute a car. They become an “organized,” emergent phenomenon—a useable “whole”—only when the parts are assembled in a very precise (purposeful) way. As a disorganized heap, they are indeed nothing more than the sum of the parts. But when they are properly organized, they produce a type of synergy that the parts alone cannot.

In this light, let us return briefly to the NECSI internet discussion. As defined here, emergence has nothing to do with concepts, or patterns, or appearances (despite the conflated usage of the term in everyday language). The mind is indeed an emergent phenomenon in the Lewes/Lloyd Morgan definition, but steam is not. Some emergent phenomena may be rule governed, but this is not a prerequisite; many of them are also instruction governed. A water molecule is an emergent phenomenon, but the debate over whether or not the whole can be predicted from the properties of the parts in fact misses the point. Wholes produce unique combined effects, but many of these effects may be codetermined by the context and the interactions between the whole and its environment(s). In fact, many of the “properties” of the whole may arise from such interactions. This is preeminently the case with living systems.

We can use water, the paradigmatic example of emergence, to illustrate this. The basic atomic properties of water have been understood for almost two centuries, thanks to John Dalton. At the micro level, we can understand how the constituent atoms of hydrogen and oxygen are linked together by their covalent bonds. We also know that quantum theory is required to explain some of the remarkable energetic properties of water. But the properties of water also entail numerous macro-level physical principles related to the chemistry, statics, dynamics, and thermodynamics of water.

For instance, additional principles of chemistry are needed to account for the state changes that produce water from its constituent gases and, under appropriate conditions, the changes that can reverse the process. Still other principles are required to account for the macroscopic properties of water as a liquid medium: its compressibility, surface tension, cohesion, adhesion, and capillarity. Thermodynamic principles are needed to understand the dynamics of temperature changes in water. Static principles relating to density and specific gravity must be invoked to account for, say, the buoyancy of a rowboat. Hydraulics are needed to understand how water reacts to a force exerted on it. Dynamics, and Newton's laws, are relevant for understanding the tidal action of water in large bodies, while hydrodynamics is required to explain the behavior of water flowing through a pipe, or in a river bed. Here Bernoulli's principle also becomes germane. By the same token, at the most inclusive geophysical level, the problem of understanding the role of water in world climate patterns presents a formidable research challenge that has necessitated multileveled, multidisciplinary modeling efforts (Goldberg, 1994). In sum, the properties of an emergent phenomenon like water, or proteins, or people, may be codetermined by the context(s).


Although reductionism will no doubt continue to play a vital role in helping us to understand “how” such emergent phenomena as organized systems work in nature, a number of theorists, including this author, have argued that a multileveled “selectionist” approach is necessary for answering the “why” question: Why have emergent, complex (living) systems evolved over time? David Sloan Wilson (1980; also Wilson & Sober, 1994) speaks of “trait group selection.” John Maynard Smith (1982, 1983, 1989) utilizes the concept of “synergistic selection.” I refer to it as “holistic Darwinism” (Corning, 1997, in press).

Holistic Darwinism, and the multileveled approach to complexity, are based on the cardinal fact that the material world is organized hierarchically (some prefer novelist Arthur Koestler's term “holarchy”). What the reductionist claims overlook is the fact that new principles, and emergent new capabilities, arise at each new “level” of organization in nature. (Again, our water example provides an illustration.) A one-level model of the universe based, say, on quantum mechanics and the actions of quarks and leptons, or energy flows, or whatever, is therefore totally insufficient. This point was argued with great clarity and erudition many years ago in a landmark essay by the biologist Paul Weiss entitled “The living system: Determinism stratified.” “Organisms are not just heaps of molecules,” Weiss pointed out (1969: 42). They organize and shape the interactions of lower-level “subsystems” (downward causation), just as the genes, organelles, tissues, and organs shape the behavior of the system as a whole (upward causation). Furthermore, one cannot make sense of the parts, or their interactions, without reference to the combined effects (the synergies) that they produce.

Two important articles published four years apart in the journal Science advanced similar arguments. In “Life's irreducible structure” (1968), chemist Michael Polanyi pointed out that each level in the hierarchy of nature involves “boundary conditions” that impose more or less stringent constraints on lower-level phenomena, and that each level operates under its own, irreducible principles and laws. Polanyi's argument was seconded and augmented by the Nobel physicist Phillip Anderson in a 1972 Science article called “More is different”:

The ability to reduce everything to simple fundamental laws does not imply the ability to start from those laws and reconstruct the universe ... The constructionist hypothesis breaks down when confronted with the twin difficulties of scale and complexity ... At each level of complexity entirely new properties appear ... Psychology is not applied biology, nor is biology applied chemistry ... We can now see that the whole becomes not merely more but very different from the sum of its parts.

Accordingly, emergent phenomena in the natural world involve multilevel systems that interact with both lower- and higher-level systems—or “inner” and “outer” environments, in biologist Julian Huxley's characterization. Furthermore, these emergent systems in turn exert causal influences both upward and downward; not to mention horizontally. (If determinism is stratified, it is also very often “networked.”) The search for universal “laws” of emergence is destined to fall short of its goal because there is no conceivable way that a set of simple laws, or one-level determinants, could encompass this multilayered “holarchy” and its inescapably historical aspect.

This conclusion, and the fundamental distinction that was drawn above between emergent phenomena that are self-ordered and the many products of “purposeful” organization (functional design), also have important theoretical implications, I contend. Indeed, this distinction goes directly to the heart of the long-standing reductionist-holist debate about the properties of “wholes” (and how to explain them) tracing back to the nineteenth century, and it poses a direct challenge to the contemporary search for “laws” of emergence and complexity in evolution.


One alternative to a law-driven theory of complexity in evolution is what I call the “synergism hypothesis.” This theory is discussed in detail in the publications that were cited earlier, so I will again be brief.

In a nutshell, the core hypothesis is that synergistic effects of various kinds have played a major causal role in the evolutionary process generally and in the evolution of cooperation and complexity in particular. Although this may sound like a contradiction of Darwinian natural selection theory, in fact the opposite is true. It is, rather, a matter of viewing the same phenomena from a different perspective—a shift of focus from the role of the genes to the role of the “phenotype” (the organism itself in a given environment). What is often downplayed in the gene-centered, neo-Darwinian paradigm is the fact that it is actually the phenotype that is differentially “selected.”

Moreover, natural selection does not in fact do anything. It is often portrayed as a “mechanism,” or personified as a causal agency that is out there in the environment somewhere. The practice started with Darwin, who wrote in The Origin that “natural selection is daily and hourly scrutinizing throughout the world, every variation, even the slightest; rejecting that which is bad, preserving and adding up all that is good; silently and insensibly working” (1968 [1859]: 133). (In a later edition, Darwin preceded this passage with the phrase: “It may be metaphorically said...”) In reality, the differential “selection” of a trait, or an adaptation, is a consequence of the functional effects that it produces in relation to the survival and reproductive success of a given organism in a given environment. It is these functional effects that are ultimately responsible for the transgenerational continuities and changes in nature.

Another way of putting it is that, in evolutionary processes, causation is iterative; effects are also causes. This is equally true of the synergistic effects produced by emergent systems. In other words, emergence itself (as I have defined it) has been the underlying cause of the evolution of emergent phenomena in biological evolution; it is the synergies produced by organized systems that are the key. To be sure, a change in any one of the parts may affect the synergies produced by the whole, for better or worse. A mutation associated with a particular trait might become “the difference that makes a difference” (to use Gregory Bateson's mantra), but the parts are interdependent and must ultimately work together as a team. That is the very definition of a biological “whole.” (A point often overlooked in the debate is that a particular trait may affect differential reproductive success, but it is still the whole organism that must survive and reproduce.) Furthermore, natural selection is not only a process that “weeds out” what doesn't work. It also weeds in what does work; both aspects are equally important. In other words, evolution is both a trial-and-error and a trial-and-success process (as paleontologist George Gaylord Simpson put it).

The synergism hypothesis can also be characterized as, essentially, an economic (or better said, bioeconomic) theory of complexity; it is the functional “payoffs” produced by synergistic phenomena that have been responsible for the “progressive” complexification of living systems (and human societies as well). Natural selection is essentially indifferent to whether or not a trait is self-ordered by some law-like process or is functionally organized by the genes (or by cultural influences for that matter). No trait is exempt from being “tested” in relation to its functional consequences (if any) for survival and reproduction. To assume otherwise would be Panglossian in the extreme; it would assume away the contingent nature of life—and evolution.

Consider three brief examples of “synergistic selection,” among the many contained in the writings by this author that were cited earlier. The first example is the eukaryotic cell, a triumph of both specialization (a division/combination of labor) and symbiogenesis, or a merger among previously independent organisms. Eukaryotes may grow to several thousand times the size of their bacterial ancestors, and this giant step in evolution was made possible in part because the eukaryotes' abundant endosymbionts—the mitochondria and chloroplasts (in plant cells)—are able to produce some 15-20 times more energy than a typical bacterium, while the machinery of respiration in eukaryotes is able to make much more efficient use of this energy. In short, emergence often “pays” in evolutionary terms; though of course not always.

A second example is lichen, a symbiotic partnership involving various kinds of green algae, or cyanobacteria, and fungi. (There are more than 20,000 different lichen species.) The algae or cyanobacteria are photosynthesizers. They provide energy-capturing services, while the fungi bring surface-gripping and water-storage capabilities to the relationship—talents that are especially useful in the barren, harsh environments that lichens are legendary for “pioneering.” How do we know that this is an emergent, synergistic system? Because the “team” can do what neither partner can do alone. There happen to be asymbiotic forms of various lichen partners that lack their joint capabilities and are far less efficient at energy capture (Raven, 1992).

A third example, close to home, is humankind. Much has been made of the role of bipedalism, tools, our large brains, language, and other supposed “prime movers” in human evolution. However, the fact is that there was no prime mover. Our evolutionary success was the result of a synergistic nexus of all of these capabilities and more—most especially our ability to exploit the potential for synergies in social organization in relation to self-defense, food acquisition, information sharing, and an ever-expanding division of labor. How do we know that human evolution involved a synergistic “package”? Just apply Aristotle's test. Imagine the consequences for evolving hominids if one could magically take away our bipedalism, our dextrous hands, our large brains, our tools, our social cooperation, or our language skills. (For a more detailed rendering of the “synergistic ape” scenario in human evolution, see Corning, 1997, 2003.)

In sum, the synergism hypothesis offers a functional (economic) explanation for the evolution of emergent, complex systems in nature. Moreover, it is fully consistent with Darwin's theory, and with the growing research literature on the evolution of biological systems at various levels of organization, not to mention the “major transitions” that are the particular focus of Maynard Smith and Szathmáry's work in this area (cited earlier). It does not deny self-ordering, even “law-like” processes in nature (many of these have been documented and appreciated for generations). But it does make natural selection the ultimate arbiter in biological evolution—the “supreme court”—and some laws do not pass the test.


Nobody can gainsay the fact that a great deal has been learned about how nature and living systems work through the use of reductionist methods in science, and surely there is much more to come. There may indeed be many law-like patterns at different levels and in different domains of the natural world. Nevertheless, the water example illustrates why there are ultimate limits to reductionism, and why holistic, systems approaches (and even systems-environment approaches) are also essential for understanding “organized” biological wholes.

Although many other theorists in recent years have embraced a multilevel paradigm, many more, unfortunately, remain dogmatic reductionists. This prompted two distinguished physicists, the Nobel Laureate Robert B. Laughlin and David Pines, to publish a frontal assault on reductionism in the Proceedings of the National Academy of Science (2000) under the title “The theory of everything.” In their words:

The Theory of Everything is not even remotely a theory of everything ... The fact that the essential role played by higher organizing principles in determining emergent behavior continues to be disavowed by so many physical scientists is a poignant comment on the nature of modern science ... The central task of theoretical physics in our time is no longer to write down the ultimate equations but rather to catalogue and understand emergent behavior in its many guises, including potentially life itself. We call this physics of the next century the study of complex adaptive matter. For better or worse we are now witnessing a transition from the science of the past, so intimately linked to reductionism, to the study of complex adaptive matter, firmly based in experiment, with its hope for providing a jumping-off point for new discoveries, new concepts, and new wisdom.

It is time to put an end to the vexatious debate over “holism” versus “reductionism” (in reality, both are necessary but insufficient) and shift our focus to “architectonics”—the study of how the world has been “built up” from the novel cooperative effects produced by many interacting parts. We need to focus more intently on the joint effects produced by the relationships that arise between things, or organisms. If reductionism is necessary for understanding how the “parts” work and how they interact, holism is equally necessary for understanding “why” living systems have evolved, and what effects they produce as wholes.

As I suggest elsewhere (Corning, 2003), the universe can be portrayed as a vast superstructure of synergies, a many-leveled “magic castle” in which the synergies produced at one level serve as the building blocks for the next. Moreover, unpredictable new forms of synergy, and even new principles, emerge at each new level of organization. In this paradigm, reductionism—including the search for some law, or laws governing the evolution of life—is inapposite. What is required instead is a science that recognizes the causal role of emergence—and history.


The author would like to acknowledge the resourceful and diligent research assistance and insightful comments by Zachary Montz and the bibliographic assistance of Pamela Albert. Any errors are, of course, my responsibility. Also helpful was the online discussion of emergence hosted by the New England Complex Systems Institute (NECSI) during December 2000 and January 2001. The contributions of its various (here unnamed) participants were valuable and much appreciated, even though I have disagreed with many of them. I also gratefully acknowledge Stanley Salthe's thoughtful reading and provocative comments, although we agreed to disagree on some major issues. Geoffrey Hodgson, an economist with a keen interest in emergence, was also most helpful with comments and suggestions.


Anderson, Phillip W. (1972) “'More is different': Broken symmetry and the nature of the hierarchical structure of science,” Science, 177: 393-6.

Bell, Graham (1985) “Origin and early evolution of germ cells as illustrated by the Volvocales,” in H. O. Halvorson and A. Monroy (eds), Origin and Evolution of Sex, New York: Alan R. Liss: 221-56.

Blitz, David (1992) Emergent Evolution: Qualitative Novelty and the Levels of Reality, Dordrecht, Netherlands: Kluwer Academic Publishers.

Campagna, Claudio, Bisioli, Claudio, Quintana, Flavio, Perez, Fabian, & Vila, Alejandro (1992) “Group breeding in sea lions: Pups survive better in colonies,” Animal Behavior, 43: 541-8.

Casti, John L. (1997) Would-be Worlds: How Simulation is Changing the Frontiers of Science, New York: John Wiley

Corning, Peter A. (1983) The Synergism Hypothesis: A Theory of Progressive Evolution, New York: McGraw-Hill.

Corning, Peter A. (1995) “Synergy and self-organization in the evolution of complex systems,” Systems Research, 12: 89-121.

Corning, Peter A. (1996) “The co-operative gene: On the role of synergy in evolution,” Evolutionary Theory, 11: 183-207.

Corning, Peter A. (1997) “Holistic Darwinism: 'Synergistic selection' and the evolutionary process,” Journal of Social and Evolutionary Systems, 20(4): 363-400.

Corning, Peter A. (1998a) “Synergy: Another idea whose time has come?”, Journal of Social and Evolutionary Systems, 21(1): 1-6.

Corning, Peter A. (1998b) “'The synergism hypothesis': On the concept of synergy and its role in the evolution of complex systems,” Journal of Social Evolutionary Systems, 21(2): 133-72.

Corning, Peter A. (2001) “'Control information': The missing element in Norbert Wiener's cybernetic paradigm,” Kybernetes, 30(9/10): 1272-88.

Corning, Peter A. (2003) Nature's Magic: Synergy in Evolution and the Fate of Humankind, Cambridge, UK: Cambridge University Press.

Corning, Peter A. (in press) Holistic Darwinism: Synergy, Cybernetics and the Bioeconomics of Evolution, Westport, CT: Greenwood Press.

Corning, Peter A. & Kline, Stephen J. (1998a) “Thermodynamics, information and life revisited, Part I: To be or entropy,” Systems Research and Behavioral Science, 15: 273-95.

Corning, Peter A. & Kline, Stephen J. (1998b) “Thermodynamics, information and life revisited, Part II: Thermoeconomics and control information,” Systems Research and Behavioral Science, 15: 453-82.

Darwin, Charles R. (1874 [1871]) The Descent of Man, and Selection in Relation to Sex, New York: A.L. Burt.

Darwin, Charles R. (1968 [1859]) On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life, Baltimore, MD: Penguin.

Goldberg, Larry P (1994) “Interconnectedness in nature and cooperation in science: The case of climate and climate modeling,” rev. doctoral dissertation, Boulder, CO: Philosophy Department, University of Colorado.

Goldstein, Jeffrey (1999) “Emergence as a construct: History and issues,” Emergence, 1(1): 49-72.

Holland, John H. (1998) Emergence: From Chaos to Order, Reading, MA: Addison-Wesley (Helix Books).

Laughlin, Robert B. & Pines, David (2000) “The theory of everything,” Proceedings of the National Academy of Sciences, 97: 28-31.

Le Maho, Yvonne (1977) “The emperor penguin: A strategy to live and breed in the cold,” American Scientist, 65: 680-93.

Lewes, George Henry (1874-9) Problems of Life and Mind, London: Truebner.

Lissack, Michael R. (1999) “Complexity: The science, its vocabulary, and its relations to organizations,” Emergence, 1(1): 110-25.

Lloyd Morgan, Conwy (1923) Emergent Evolution, New York: Henry Holt.

Lloyd Morgan, Conwy (1926) Life, Mind and Spirit, London: Williams and Norgate.

Lloyd Morgan, Conwy (1933) The Emergence of Novelty, New York: Henry Holt.

Maynard Smith, John (1982) “The evolution of social behavior: A classification of models,” in King's College Sociobiology Group (ed.), Current Problems in Sociobiology, Cambridge, UK: Cambridge University Press: 28-44.

Maynard Smith, John (1983) “Models of evolution,” Proceedings of the Royal Society of London (B), 219: 315-25.

Maynard Smith, John (1989) Evolutionary Genetics, Oxford, UK: Oxford University Press.

Maynard Smith, John (1999) Shaping Life: Genes, Embryos and Evolution, New Haven, CT: Yale University Press.

Maynard Smith, John & Szathmary, Eörs (1995) The Major Transitions in Evolution, Oxford, UK: Freeman Press.

Maynard Smith, John & Szathmary, Eörs (1999) The Origins of Life: From the Birth of Life to the Origin of Language, Oxford, UK: Oxford University Press.

Mill, J. S. (1872 [1843]) A System of Logic Ratiocinative and Inductive, London: John W. Parker and Son.

Polanyi, Michael (1968) “Life's irreducible structure,” Science, 160: 1308-12.

Price, Peter W. (1991) “The web of life: Development over 3.8 billion years of trophic relationship,” in L. Margulis & R. Fester (eds), Symbiosis as a Source of Evolutionary Innovation, Cambridge, MA: MIT Press: 262-72.

Raven, John A. (1992) “Energy and nutrient acquisition by autotrophic symbioses and their asymbiotic ancestors,” Symbiosis, 14: 33-60.

Shapiro, James A. (1988) “Bacteria as multicellular organisms,” Scientific American, 258(6): 82-9.

Smuts, Jan C. (1926) Holism and Evolution, New York: Macmillan.

Waldrop, Mitchell M. (1992) Complexity: The Emerging Science at the Edge of Order and Chaos, New York: Touchstone (Simon & Schuster).

Weiss, Paul A. (1969) “The living system: Determinism stratified,” in A. Koestler & J. R. Smythies (eds), Beyond Reductionism, London: Hutchinson.

Wilson, David S. (1980) The Natural Selection of Populations and Communities, Menlo Park, CA: Benjamin/Cummings.

Wilson, David S. & Sober, Elliott (1994) “Reintroducing group selection to the human behavioral sciences,” Behavioral and Brain Sciences, 17: 585-608.