Kineman, J. 1997. Theory of Autevolution
Epistemology – I
A proposed synthesis
The following discussion presents a simplified attempt at epistemological
synthesis. The next section (Epistemology II) offers guidelines
for evaluating the foundations of new theory (worldviews).
Epistemology
The epistemological model presented here
(figure 1) is constructed to model
the relationship between theory development, transitions to new
founding assumptions (in either the gradualist sense of Toulmin
or the revolutionary sense of Kuhn as described by Suppe, 1977),
and processes of integration. The model attempts to combine the
postpositivist tradition represented by logical empiricism, the
instrumentalist view of independent theories, the realist view
of seeking fundamental laws of nature, and the historicist tradition,
which deals with the context for science in terms of paradigms
of scientific theory, or worldviews. The dominant philosophy of science
in recent years has been historical realism, which “recognizes
an historically shifting yet relatively theory-neutral empirical
basis for theory confirmation” (Goldberg, 1990).
What is most important, however, is how the various epistemic
processes interrelate and the results they produce.
The search for reality
The model presented here recognizes the existence and value of
instrumental theories, but views the search for “real”
elements of theory as a more robust pursuit. This form of realism
(i.e., Platonic as opposed to existential) maintains that there
is a hidden reality “out there,” and is intent on representing
it from basic principles. Theory then becomes a parsimonious
attempt to represent natural law that is assumed to be universal,
consistent, and unique (qualities that are also sought for the
foundations of worldviews). Even though it is recognized that
theory is never perfect, this philosophy maintains that closer
approximations to reality are always possible, and that a unique,
ultimate reality exists (Rohrlich, 1989).
Thompson (1989), in contrast,
argues for adopting a formal instrumentalist epistemology that
accepts current (separated) ecological and evolutionary theory
structures as part of a “family of interacting theories.”
This “semantic” construction of theory abandons the
idea of constructing theory in terms of elements with “real”
meaning, and instead constructs abstract models, according reality
only to what can be observed or measured.
Thus instrumentalism represents a practical
view of theories as they tend to form, whereas the realism
described above provides a goal where their development may be
more productively aimed. Each of these views actually contains
elements of the other, and therefore they are not so much opposing
views, as different aspects of the growth of knowledge (this relationship
is represented in the model). The view presented here is that
the goal of realism forces basic assumptions and theory structures
to be questioned and modified (by increasing their predictive
scope and thereby uncovering contradictions), and thus advances
science through revolutions to new and more inclusive paradigms.
Theoretical pluralities inherent in instrumentalism, on the other
hand, may be balanced by attempts at theory integration or the
development of interdisciplinary approaches. Such integration
may reveal important contradictions in the foundations of theory,
thus opening the way for new ideas, in a similar manner as revolutions
in theory.
Growth of knowledge
Figure 1 presents a model of the
growth and evolution of knowledge facilitated by the identification
of theoretical crises, combining philosophies of science. In this
figure, scientific investigation (phase A) considers theory development
within established paradigms, as a combination of phenomenal and
causal studies which ultimately result in paradox due to limited
assumptions (phase B). The transition phase (phase C) shows a
process of crisis resolution, which necessitates a move from one
theoretical basis to another (hopefully, but not necessarily more
inclusive) view. Finally the figure shows theoretical pluralities
(phase D), which may become linked or unified (phase E).
Phase A depicts a relationship between two methodological extremes,
shown as the study and description of phenomena (on the left),
and the search for explanations in terms of causal processes (on
the right). This relationship includes
the more restrictive positivist
ideal of logical empiricism (i.e., hypothetico-deductive or strong
inference methodology). In practice these methods often interrelate,
giving a variety of operational approaches. On the one extreme,
the essence of logical empiricism (Popper, 1959
and 1965) can be described
as disproof of reasonable alternatives (hypotheses) in relation
to observation and existing knowledge (i.e., which must necessarily
be used when designing experiments and interpreting data). Within
the established paradigm, new ideas that are contradicted by experiment
(or prior knowledge) are rejected, and those that cannot be rejected
(or as some philosophers argue, are confirmed) are added to the
body of knowledge, which thus grows by accumulating consistent
concepts. This quest for theory that is logically self-consistent
and consistent with both observation and experiment is shown as
the operating procedure or routine of science.
Yet philosophers have discovered the impossibility of defining
an absolute set of knowledge. Historicism therefore recognizes
that a body of knowledge that is built upon previous learning
must be relative to a particular set of assumptions. These
assumptions form a worldview that is defined by historical scientific
development, cultural influences, and current philosophy. Furthermore,
as described by Kuhn (1970)
and his followers, there have repeatedly been scientific revolutions.
This is also shown in figure 1,
where the emphasis on empirical and theoretical studies in the
primary cycle (phase A) leads to paradigm shifts in phase C. In
this view, theory growth can become punctuated when contradictions
in theory are seen as representing a philosophical and scientific
crisis. This assumes, of course, a strong epistemological motivation
to resolve apparent paradox, as is most prevalent in the concept
of realism described here. Whereas apparent paradoxes (for example,
resulting from causal explanations that are accorded some sense
of reality) are crisis inducing within the old theoretical context
(phase A), they are also the fuel for new worldviews (phase C).
This combines philosophical traditions into a simple model
of “punctuated equilibrium,” between stable and transitory
phases where opposite criteria are employed in the testing or
selection process in these two phases. The epistemology in
phase A operates on the assumption that two alternative explanations
cannot both be correct. Phase C, however, operates on the assumption
that two apparently paradoxical elements of theory that have otherwise
been confirmed, must both be correct. (See Crisis Resolution)
Empirical methods of science thus operate within established thought
structures (i.e. paradigms and worldviews), testing well-posed
hypotheses to build a theory, but not testing the structure itself.
As an example, empirical science does not test the idea of force
directly: Under appropriate circumstances, it is equally valid
to view gravity as a force or as an artifact of curved space,
and the merits of each view is an independent issue from how they
are formalized. Empirical testing is concerned with how forces
act (e.g. f = ma) or interrelate (e.g. the equivalence
of acceleration and gravitational force). The paradigm itself
is judged first on whether or not it can be formalized, then on
its overall predictive value and applicability to new phenomena.
Therefore, the means for evaluating the foundations of theory
are quite different from the means for testing hypotheses about
its operation. Both can be scientific if they are not
confused with each other; but confusion between these two levels
of thought is common and characteristic of statements (and criticisms)
about Gaia and other theories of self-directed evolution.
Still, the above synthesis is an incomplete picture, especially
in terms of the debate between realism and instrumentalism (the
idea that theory need not be based on a common reality), and the
obvious existence of theoretical pluralism. The realist view works
well in classical physics, which is more easily referenced to
basic axioms, however other branches of science have not been
so blessed. Instrumentalism is thus represented in several places
in the model. In the form of purely phenomenal or descriptive
theories (such as Ptolemy’s model of a geocentric solar system,
as described, for example, by Rohrlich, 1989)
it might be seen as a hybrid of the two extremes in phase A (i.e.,
a theory which does not seek basic laws, but yet has predictive
value based on the regularity of phenomena). Although logical
contradictions can be identified within such theories, it is undetermined
whether they will become crisis inducing or remain unchallenged.
A more formal instrumentalism appears in phase E
(right side), where theoretical pluralities are treated as a “family
of interacting theories” (Thompson, 1989).
As shown, it is questionable whether or not this approach will
similarly preserve theoretical contradictions, or will eventually
resolve them through some form of synthesis.
Evaluating worldviews
The immediate concern, in regard to autevolution, is to determine
by what means the worldview portion of this model (i.e. new founding
assumptions) can be evaluated. As mentioned earlier, the basis
for theory cannot be evaluated in the same way as the content
of theory (i.e. by logical empiricism and hypothesis testing).
There are, however, specific criteria that can be employed for
evaluating worldview assumptions. Five such criteria are discussed
in the next section.
Revised and reprinted from: Kineman, John Jay. 1991. “Gaia:
hypothesis or worldview?” Paper delivered at the American
Geophysical Union annual Chapman Conference, panel on epistemology,
March, 1988, San Diego, California. Chapter 7 In: Schneider, S.
H., and P. J. Boston (eds). 1991. Scientists on Gaia. Cambridge,
MA: MIT Press. 433p.
Please cite as: Kineman, John Jay. 1997. “Toward a special
and general theory of autevolution.” Boulder: Bear Mountain
Institute. HTML publication on the World Wide Web.
All rights reserved
Please address comments to: John Jay Kineman