Menu:




Shortcuts:

Selection

System Dynamics





Selectionism
Camilo Olaya

An evolutionary stance is interested in explaining change which is assumed as a principle of nature and living systems, including social systems. Invariances and regularities are rejected as starting points.

How does evolution proceed? It can be characterized as the continuous sequence of two steps: (i) Variations that provide material upon which selections acts on. (ii) Selection due to the elimination of unsuccessful forms by the environment or by internal constraints or internal selection processes. Since the very idea of process implies trans-temporal constancies we can safely say that every particular process is an instantiation of a general, atemporal and generic pattern (Rescher, 2008). The variation + selection combination forms an algorithm with substrate neutrality, that is, its logical structure implies guaranteed results independent of the “materials” that happen to be used to carrying it out (Dennett, 1995). This abstractness provides a generic schema that can be instantiated to give an actual theory (Darden & Cain, 1989) by proper theoretical recontextualization in specific domains (Dopfer, 2005; Hayek, 1942). Selection theory is perhaps the most accepted explanation of generic processes of fit, that is, the explanation of satisfaction relationships in which one thing comes to be adapted to another thing (Bickhard & Campbell, 2003; Cziko, 1995; Darden & Cain, 1989). This statement asserts then that selection defines an abstract form of explanation in which the process itself is the explanans.

I prefer to choose the terms selectionist thought over the more general expression evolutionary thought. "Selection" places the emphasis on the opposition to "instruction"; the latter one has been widely used in evolutionary thought in fields outside biology; such a view holds what can be labelled as "Lamarckian" assumptions regarding evolutionary processes which I find inconvenient not only in biology but in general in any evolutionary conception.


Instruction

In short, I find the pervasive instructionist, i.e. Lamarckian, view largely problematic.  Yet, this "learning by instruction" logic permeates almost any knowledge process we know of. Prominent examples are: (i) The way we understand how we, as human beings, think and learn (any model of learning based on the notion of feedback is usually rooted in such assumptions) including our old theories about the nervous system that still provide the main support for studies on learning, perception, cognition, etc. (ii) The works of Ashby and Wiener that produced several disciplines and areas of inquiry such as cybernetics and also our models for computers (e.g. the "von Neumann machine", etc.). (iii). Our idea of "science by observation", all the restricted empiricism and the disguised positivism that dominates our conception of science, e.g. any inductive method; Humean empiricism; idealism in general; any scientific work that looks for "confirmation" and "verification"; any search for "observed" regularities; usually all the areas concerned with "the problem of the observer"; relativism; etc. (iv) Aggregated approaches and top-down models. (v) Any approach holding that adaption is a process of "correction" according to a changing environment. (vi) Any approach that understands evolution as a mere problem of competition or fitness (e.g. the neglect of reproductive success). (vii) Any Lamarckian-like approach to evolution (i.e. inheritance of acquired traits); this type of studies is very popular in the social sciences. (viii) Any approach that restricts the generation of trials or possible solutions according to diverse criteria such as previous success, a posed goal, etc.  (iv) Any area that holds any form of "transmission" of knowledge. And so on.

Selection

Selectionism is opposed to the logic of instruction depicted above. My research interests are focused on how to understand, design, and  improve knowledge processes following a logic of selection  For instance, a selectionist ground is found on the work of Donald Campbell with a two-step process made of blind-variation and selection mechanisms. Diverse possibilities of research and applications include: computational algorithms, optimization heuristics, artificial intelligence, artificial life, complexity science, computational biology, innovation processes (e.g. product design), individual learning (e.g. our daily lives), education (e.g. classes  at schools and universities, a pensum design, etc.), science (methodology, epistemology), social organizations (enterprises, public agencies, governance structures) and so on.  Instead of a process of correction that depends on observation and external change, the involved organisms should be engaged in a constant process of blind (free) variation. Not only for maximizing chances but also for enhancing autonomy and increasing their own choices and possibilities.


Evolution of knowledge

Social systems can be understood as a special type of systems defined by knowledge evolutionary regimes driven by the social agents that conform the system. I endorse the definition of Campbell (1987): Knowledge refers to any process providing a stored program for organismic adaptation in external environments. Moreover, any gain in the adequacy of such a program is regarded as a gain in knowledge. Therefore, evolution is conceived as a process in which information regarding the environment is literally incorporated in organisms; adaptation is an increment of knowledge which is driven by the logic of selection.

Two major steps can be indentified behind this creative force.

(i) Variation : The process of variation  provides the ”raw material”; this variation should be generated in copious and dependable amounts and should be undirected. For knowledge processes Campbell coined the expression blind variation (which does not mean "random"); the term ”blind” denotes the fact that variations are produced without prior knowledge of which ones, if any, will furnish a selectworthy encounter. Three connotations are paramount: (i) The variations emitted are independent of the environmental conditions of the occasion of their occurrence. (ii) The individual occurrences of trials are uncorrelated with the solution — specific correct trials are no more likely to occur at any point in a series of trials than another correct one, nor than specific incorrect trials. (iii) The rejection of the notion of a "correcting" process between variations, that is, a variation subsequent to an incorrect trial is not a "correction" of an earlier one.

(ii) Selection : The direction of evolutionary change is granted to the second step, selection as such, which works upon variation. Selection occurs due to the elimination of unsuccessful forms by the environment and through a hierarchy of flexible controls (as suggested by Popper and Campbell). The general logic of problem solving is trial and error-elimination through this hierarchical system of flexible controls.  For instance Campbell remarked that general knowledge processes involve several mechanisms at different hierarchical levels of substitute functioning; at each level — a type of template — there is a form of selective retention process; regarding human knowledge, he identified several levels ranging from the most primitive non-mnemonic problem solving layer, through higher levels like habit, instinct, visually and mnemonically supported thought, imitation, to the highest levels such as language, cultural cumulation, and science. Evolving selection criteria are nested in this hierarchy; what are criteria at one level are but `trials' of the criteria of the next higher level. It must be emphasized that within those layers, a trial and error-elimination logic takes places as well.

Thus, variations and selections of knowledge can form a base for studying and designing social systems. A detailed argument can be found in my doctoral thesis.

References

Bickhard, M. H., & Campbell, D. T. (2003). Variations In Variation And Selection:The Ubiquity Of The Variation-And-Selective-Retention Ratchet In Emergent Organizational Complexity. Foundations of Science, 8, 215-282.

Campbell, D. T. (1987). Evolutionary Epistemology. In G. Radnitzky & W.W. Bartley, III (Eds.), Evolutionary Epistemology, Rationality, and the Sociology of Knowledge (pp. 47-89). La Salle, IL, USA: Open Court.

Cziko, G. (1995). Without Miracles. Universal Selection Theory and the Second Darwinian Revolution. Cambridge, MA, USA: MIT Press (A Bradford Book).

Darden, L., & Cain, J. A. (1989). Selection Type Theories. Philosophy of Science, 56, 106-129.

Dennett, D. C. (1995). Darwin's Dangerous Idea. The Sciences (May/June), 34-40.

Dopfer, K. (2005). The Evolutionary Foundations of Economics Cambridge, UK: Cambridge University Press.

Hayek, F. A. v. (1942). Scientism and the study of society (Part I). Economica, New Series, 9(35), 267-291.

Rescher, N. (2008). Process Philosophy. In E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy. Stanford, CA: The Metaphysics Research Lab, Stanford University.