Evolutionary Theory
An Esalen Invitational Conference
October 5 to 10, 2003

Ontological Emergence and the Failure of Reductionism
David Deamer and Stuart Kauffman

Schematic drawing of the head-to-toe arrangement of hydro- phobic and phillic lipids in a lipid vesicle.

Deamer’s own research with the organic compounds in the Murchison meteor (which hit the earth in 1969 in Australia) led to the initial discovery and later replication of this amazing process in a scientific laboratory setting. Deamer pointed out, however, that the spontaneous origin of these permeable sac-like vesicles is not a miraculous exception to the second law of thermodynamic entropy. Rather, as they spontaneously organize into bi-layer membranes the lipid molecules are moving into the lowest energy state possible (in accord with entropy). A small amount of spontaneous order (the membranous lipid sac) is being exchanged for a larger amount of disorder (decreased water structure). In a broad sense, self-organization at this scale can be understood as a complementary force to the second law of thermodynamic entropy. Deamer noted that a necessary factor for lipid sac self-organization is the presence of an energy throughput or gradient. This energetic imbalance (non-equilibrium state) in turn leads to the self-organization of the lipid sacs, which channel the energy flow more efficiently. Deamer and other researchers have demonstrated in their labs that these lipid vesicles can capture catalysts (enzymes) and genetic materials (DNA and RNA strands). The more difficult steps of: 1) replicating the polymeraze and template and 2) creating a cellular metabolism have not yet been achieved under lab conditions.

In the second part of his presentation Deamer discussed the attempt to precisely and quantitatively measure biocomplexity. In previous conferences in this Esalen series there has been discussion about the rise of complexity in the universe, yet there were no quantitative assessments provided for what is meant by "greater complexity." Deamer has begun what ultimately will be an immense task of providing such quantitative assessment. For example, he noted that there are 2.36 million proteins in one bacterial cell. So, to measure the quantitative complexity in a neucleated cell, a plant, or a small animal is a gargantuan task for teams of scientists. Some participants responded to Deamer by suggesting that an even larger task to follow the quantitative assessment of biocomplexity is to then look at the growing number of relationships (relational complexity) among parts in a bio-complex system.

When reflecting on the conceptual usage of the term "emergence," Deamer said that the emergence of life from non-life should be considered a rigorous case of ontological emergence (or strong emergence), in which something truly novel did come into the universe. Life cannot be explained reductively by physics alone and thus should not be considered a case of mere epistemological (or descriptive, or weak) emergence.

On Tuesday afternoon, complexity theorist Stuart Kauffman built on some of the features in Deamer’s presentation by discussing a number of cases that he thinks fit the category of ontological emergence—in which explanatory reductionism (reduction to the irreversible laws of physics) simply fails as an explanatory paradigm. Kauffman began with the theme of "broken symmetries," in which something new comes into the universe by breaking an originally symmetrical and stable relationship. He cited some simple examples for which standard physics has done only a meager job of explaining, such as the falling of a wooden pole and the magnetization of a ferromagnet. Kauffman mentioned how the application of the Turing model as early as 1954 paved the way for attempts by contemporary complexity theorists to explain various dissipative structures in nature, such as Bénard cells. Kauffman’s own work has looked at how some of the same core principles of complex behavior apply in a variety of different cases (such as how animals get stripes and spots on their hides). One of the core ideas is that when the symmetry of a system is broken and it is displaced from equilibrium, minor fluctuations in the system can then amplify and express as macro patterning (hexagonal shapes in Bénard cells or stripes and spots on animals hides). Kauffman remarked that even if some of these cases can be explained reductionistically, something new is still coming into existence in the universe.

Drawing on the work of physicist Lee Smolin, Kauffman then made a few remarks on galaxy formation and their homeostatic self-regulation. He noted that galaxies display self-sustaining dynamics, in which individual stars and supernovas come and go over large stretches of time, while the whole of the galaxy endures. Homeostatic temperature gradients are present in galaxies, because stars continuously form in cold clouds and radiate heat outward. Spiral galaxies (unlike ellipticals) maintain an ongoing homeostatic regulation between star formation and supernova explosions. But Kauffman is not sure whether this process in a galaxy constitutes a regular work-cycle.

Next, Kauffman turned to his theory concerning the collective autocatalysis of life. The core idea Kauffman has developed over several years of research is that the ratio of total molecular reactions to total molecules will increase over time, eventually leading to the spontaneous formation of giant webs of molecular reactions (reaction graphs). This collective auto-catalytic set is called a first-order phase transition. According to Kauffman, it is an example of strong or ontological emergence: a collective phenomenon (a chemical reaction graph) emerges from a random chemical soup. (To date, this collective auto-catalytic emergent effect has been demonstrated in sequences of proteins by Reza Ghadiri.)

In response to Kauffman’s description, some of the conference participants pressed for greater clarity on the definition of ontological emergence. Kauffman suggested that striped chemical patterns should be considered a case of ontological emergence. The transition from something spatially homogeneous and lacking pattern to something robustly displaying pattern is ontologically new. For example, the chemical system discovered by Russian scientists Beloussov and Zhabotinsky (the BZ reaction) spontaneously generates time-varying patterns (spiral waves) without instructions from the genes. The BZ pattern is a new relational pattern irreducible to chemistry alone. Kauffman pointed out that until such spontaneously organized systems were discovered, no one knew that simple chemical systems could give rise to novel patterns.

To help clarify some of the muddle over the term "emergence," Evan Thompson mentioned an article by Michael Silberstein ("The Search for Ontological Emergence," Philosophical Quarterly, 1999), in which Silberstein claims that ontological emergence entails an irreducible relational holism. An ontologically emergent phenomenon cannot be analyzed into "element A and B" and the extrinsic relationships between them. Instead, in ontological emergence, A and B are definable only by virtue of the fact that they are related to each other, and the relations that define the system are not analyzable in terms of the intrinsic properties of the constituent parts. Silberstein cites quantum entanglement as the strongest example, because the wholeness of the entangled system cannot be analyzed reductively to the separate electrons without violating Einstein’s special theory of relativity. (If entanglement is reductively explained by the forces transmitted between parts, then a signal moving faster than the speed of light must carry the information between the parts.) If we preserve special relativity, then there must be an irreducible relational holism at the quantum level (what philosophers call "mereological emergence").

Terrence Deacon also commented that in his own three-tiered categorical scheme of emergent dispositions, he is not interested in the claim of whether something is definitively "new." Instead, he is focused on how particular causal architectures and topologies can been shown to come forth hierarchically from one another. The "newness issue" is for Deacon a contentious debate that can distract us from becoming clearer about the precise details involved in emergent causal architectures.

Kauffman concluded by citing a few examples that highlight the failure of strong reductionism. First, he claimed that it is impossible to pre-specify the entire list of possible chemical systems derivable or evolve-able from the chemistry of the early earth. The evolution of new chemical species is another example of symmetry breaking. Kauffman speculates in his book Investigations (Oxford, 2000) that there is a chemical "adjacent possible" (or space of new possibilities) into which the earth’s chemistry has consistently expanded. The inability to pre-specify all possible chemistries could be a mathematical statement (a version of Gödel’s theorem) or an empirical statement based on observation of the history of life on our planet.

Second, Kauffman stated that as new environments arose in the history of the earth, new functions became possible in those environments—and there was, and is, no way to know all those possible functions until they arise. All new functions are environment-specific. For example, at some point in the history of squirrels a baby squirrel was born with flabby skin. One day that flabby squirrel spontaneously realized a new function: the ability to use the extra skin to fly from branch to branch. A new function emerged that could not have been pre-specified. In short, Kauffman suggested that our evolving biosphere seems to be inherently creative, and the limits of this creativity are unknown in advance. This means that everything in the universe cannot be explained reductively by physics alone. Nor, Phil Clayton added, do all functions exist in some disembodied Platonic space of latent potentials.

Terrence Deacon commented that a key question to address is whether there is a definitive way to describe all pre-adaptations (called "exaptations" by Stephen J. Gould) in advance. Most (or all?) explanation of a function is post hoc, happening after it has been recognized as a function. Deacon emphasized that function is always about a relationship to what it is not: a function is always constituted by its relationship to things in the environment. There are no functions in isolation. (See Deacon’s "third-order dispositions" for more on this.)

Kauffman said that if we cannot pre-specify all possible functions, then the universe and biosphere are non-algorithmic. The biosphere keeps evolving through the use of unanticipated pre-adaptations/exaptations. The epistemological strategy of reductionism to the laws of physics simply does not work to describe such robust biological creativity.

Telmo Pievani pointed out a further extension of Kauffman’s argument. He suggested that the adjacent possible may be even larger than Kauffman thinks because if we take into account all the effects of every function, then there is an infinite possibility space of those effects. Along with the propagation of new functions, there is a continuous propagation of their effects as well. These too are not derivable algorithmically.

Further Questions for Emergence and the Failure of Reductionism

Ontological emergence: is something only new if it has fundamentally new parts that comprise it? Or is something new when it has new patterns of relationship among existing parts?

What conceptual and/or paradigmatic shifts are needed for GUTs theory (grand unified theories in physics) to incorporate emergence, pre-adapatations, and creativity? Can physics pre-specify all possibilities?

Philosophers like Michael Silberstein have pointed out the failure of inter-theoretic reductionism (e.g., reductively explaining irreversible thermodynamics with reversible statistical mechanics). What theoretical or conceptual shifts are needed to address the noticeable divide between reversible physics and irreversible biology?

The notion of a law works well for reversible phenomena. Could the concept of "constraint" suffice for the term "law" in the irreversible world of biology and evolution?