| |
Herbert
Simon is the principal figure considered in this
chapter. This
chapter’s material is presented in reverse
chronological order, and the exposition therefore
starts with the work of Paul Thagard, who follows
Simon’s cognitive-psychology orientation for his
computational philosophy of science investigations. Thagard’s philosophy of science is rich, and lends itself
to exposition in terms of the four basic topics in
philosophy of science.
But before considering Thagard’s treatment
of the four topics, consider firstly his
psychologistic views on the nature of philosophy of
science and the semantics of conceptual change in
scientific revolutions.
Thagard’s Psychologistic
Computational Philosophy of Science
Thagard is
a Professor of Philosophy at the University of
Waterloo since 1992, and is also Adjunct Professor
of Psychology and Computer Science, Director of his
Computational Epistemology Laboratory, and Director
of the Cognitive Science Program. He has been an
associate professor of philosophy at University of
Michigan, Detroit, where he was associated with
their Cognitive Sciences Program, and a Senior
Research Cognitive Scientist at Princeton
University. He
is a graduate of the University of Saskatchewan,
Cambridge, Toronto (Ph.D. in philosophy, 1977) and
the University of Michigan (M.S. in computer
science, 1985).
Computational
philosophy of science has become the new frontier in
philosophy of science in recent years, and it
portends to become essential to and definitive of
twenty-first century philosophy of science.
There are many philosophers now jumping on
the bandwagon by writing about the computational
approach in philosophy of science, but only authors
who have actually designed, written and exhibited
such computer systems are considered in this chapter
of this book. Thagard
is one of the handful of academic philosophers of
science, who has the requisite technical skills to
make such contributions, and has demonstrated them
by actually writing such systems. His work is also selected because in the closing decades of
the twentieth century he is one of the movement’s
most prolific authors and most inventive academic
philosophers of science.
Thagard
follows the artificial-intelligence approach and
psychological interpretation of the AI systems
previously proposed by Herbert Simon, who is one of
the founding fathers of artificial intelligence.
In his Computational
Philosophy of Science (1988) Thagard explicitly
proposes a concept of philosophy of science that
views the subject as a type of cognitive psychology.
The linguistic-analysis tradition in
philosophy had achieved ascendancy in
twentieth-century philosophy of science.
The analysis of language has been
characterized by a nominalist view, also often
called “extensionalism or the “referential
theory of meaning.”
The nominalist view proposes a two-level
semantics, which recognizes only the linguistic
symbol, such as word and sentence, and the objects
or individual entities they reference.
It recognizes no third level consisting of
the idea, concept, “intension” (as opposed to
extension), proposition, or any other mental reality
mediating between linguistic signs and nonlinguistic
objects. The
two-level semantics is the view typically held by
the Positivist philosophers, who rejected mentalism
in psychology and preferred behaviorism.
Thagard explicitly rejects the behavioristic
approach in psychology and prefers cognitive
psychology, which recognizes mediating mental
realities. The
two-level semantics is the view that is also
characteristic of philosophers who accepted the
Russellian predicate calculus.
This calculus of symbolic logic contains a
notational convention that uses quantification to
express existence claims.
It therefore fabricates a nominalist newspeak
in which predicate terms are semantically vacuous,
unless they are placed in the range of quantifiers,
such that they reference some kind of entities,
called either “mental entities” or Platonic
“abstract entities.”
The philosopher Nelson Goodman for example
divides all philosophers into nominalists and
Platonists. Not
surprisingly the Russellian symbolic logic was
adopted by the Logical Positivists.
Oddly Thagard does not reject the Russellian
symbolic logic, although it is not clear that he
recognizes the ontological implications of its
notational conventions.
His turn away from linguistic analysis and
toward psychologism has been motivated by
recognition of the mentalistic semantical level.
Like Simon, Thagard wants to admit the
existence of the mental semantical level, so that he
can investigate concepts by viewing computer systems
as analogs for the mental realities, and then
hypothesize about the human cognitive processes of
scientists on the basis of the computer system
designs and procedures.
He refers to this new discipline as
“computational philosophy of science”, the name
that will probably become the conventional one for
this area specialty.
And he defines computational philosophy of
science as an attempt to understand the structure
and growth of scientific knowledge in terms of
computational and psychological structures with the
aim of offering new accounts both of the nature of
theories and explanations and of the processes
underlying their development.
Thagard distinguishes computational
philosophy of science from cognitive psychology by
the former’s normative perspective.
In
his Mind:
Introduction to Cognitive Science (1996),
intended as an undergraduate textbook, he states
that the central hypothesis of cognitive science is
that thinking can best be understood in terms both
of representational structures in the mind and of
computational procedures that operate on those
structures. He labels this central hypothesis with
the acronym “CRUM”, by which he means
“Computational Representational Understanding of
Mind.” He
says that this hypothesis assumes that the mind has
mental representations analogous to data structures
and computational procedures analogous to
algorithms, such that computer programs using
algorithms applied to data structures can model the
mind and its processes.
His
How Scientists
Explain Disease (1999) reveals some evolution in
his thinking, although this book reports no new
computer-system contribution to computational
philosophy of science.
In the book he examines the development of
the bacteriological explanation for peptic ulcers. He finds that collaboration, communication, consensus, and
funding are important for research, and he uses the
investigation to propose an integration of
psychological and sociological perspectives for a
better understanding of scientific rationality.
He also states that principles of rationality
are not to be derived a priori, but should develop
in interaction with increasing understanding of
human cognitive and social processes.
Thagard’s computational philosophy of
science addresses the topics of discovery,
criticism, explanation, and the aim of science. He has created several computer systems for computational
philosophy of science, none of which produce
mathematically expressed theories.
And all of his systems have been applied to
the reconstruction of past episodes in the history
of science. None
of his systems have been applied to the contemporary
state of any science, either to propose any new
scientific theory or to forecast the resolution of
any current scientific theory-choice issue.
Thagard on Conceptual Change,
Scientific Revolutions, and System PI
Thagard’s semantical views are set forth in the
opening chapters of his Conceptual Revolutions (1992).
He says that previous work on scientific
discovery, such as Scientific
Discovery; Computational Explorations of the
Creative Process by Langley, Simon, Bradshaw,
and Zytkow in 1987 has neglected conceptual change. (This important 1987 work is discussed below in the sections
reporting on the views and systems developed by
Simon and his colleagues.)
Thagard proposes both a general semantical
thesis about conceptual change in science and a
thesis specifically about theoretical terms.
His general thesis is that (1) scientific
revolutions involve transformations in conceptual
and propositional systems, (2) kind-hierarchies and
part-hierarchies structure conceptual systems, and
(3) relations of explanatory coherence structure
propositional systems.
His theory of explanatory coherence is his
philosophy of scientific criticism, which is
described separately below.
Consider firstly his general semantical
thesis.
Thagard
opposes his psychological account of conceptual
change to the view that the development of
scientific knowledge can be fully understood in
terms of belief revision, the prevailing view in
analytic philosophy.
He says that his view is that concepts are
mental representations that are largely learned and
are open, i.e. not defined in terms of necessary and
sufficient conditions.
He maintains that a cognitive-psychology
account of concepts and their organization or
structure in hierarchies shows how a theory of
conceptual change can involve much more than belief
revision. He
notes that such hierarchies are important in WORDNET, an electronic lexical reference system.
Thagard states that an understanding of
conceptual revolutions requires seeing how concepts
can fit together into conceptual systems and seeing
what is involved in the revolutionary replacement of
such systems. He
says conceptual systems consist of concepts
organized into kind-hierarchies and part-hierarchies
linked to one another by rules. This idea suggests the ancient tree-hierarchical arrangement
proposed by the third-century logician Porphyry,
which Umberto Eco says in his Semiotics
and Philosophy of Language is a “disguised
encyclopedia.”
It is not clear why Thagard believes that
these structures cannot be expressed in language and
explained by belief revision, unless he mistakenly
associates belief revision with the nominalism of
analytic philosophy.
Thagard maintains that a conceptual system can be
analyzed as a computational network of nodes with
each node corresponding to a concept, and each line
in the network corresponding to a link between
concepts. The
most dramatic changes involve the addition of new
concepts and especially new rule-links and
kind-links, where the new concepts and links replace
ones from the old network. Thagard calls the two
most severe types of conceptual change “branch
jumping” and “tree switching”, and says that
neither can be accounted for by belief revision.
Branch jumping is a reorganization of
hierarchies by shifting a concept from one branch of
a hierarchical tree to another, and it is
exemplified by the Copernican revolution in
astronomy, where the earth was reclassified as a
kind of planet instead an object sui
generis. Tree
switching is the most dramatic change, and consists
of reorganization by changing the organizing
principle of a hierarchical tree, and it is
exemplified by Darwin’s reclassification of human
as animal while changing the meaning of
classification to a historical one.
He also says that adopting a new conceptual
system is more “holistic” than piecemeal belief
revision. Historically
the term “holistic” was opposed to any analysis,
but clearly Thagard is not opposed to analysis;
“systematic” would be a better term in his
context.
In
his Computational
Philosophy of Science Thagard references Willard
Van Quine’s statements that science is a web of
belief, a connected fabric of sentences that faces
the tribunal of sense experience collectively, all
susceptible to revision and adjustment like the
planks of a ship.
He agrees with Quine, but adds that Quine
does not go far enough.
Thagard advocates a more procedural viewpoint
and the abandonment of the fabric-of-sentences
metaphor in favor of more complex cognitive
structures and operations.
He concludes that the web of beliefs does not
consist of beliefs, but rather consists of rules,
concepts, and problem solutions, and the procedures
for using them.
By way of commentary, it may be said that
Thagard’s theory of conceptual change is a theory
of conceptual organization rather than a theory of
meaning description enabled by accepting a defining
role for beliefs.
A belief is any unit of language that may be
true or false, and that is accepted as true for any
reason including notably reasons acceptable in
science. And
once accepted as true, the meaning of its subject
term is defined in part by the meaning associated
with the descriptive predicate in the believed
statement thereby offering a partial meaning
description of the subject term.
Thus belief revision occasions a change in
definition, and thereby both produces and describes
conceptual change including revolutionary change in
science. It may be added that kind-hierarchies and
part-hierarchies can be expressed linguistically in
statements believed to be true, as even ancient
logicians had recognized.
In Conceptual Revolutions Thagard maintains that continuity is
maintained through the conceptual change by the
survival of links to other concepts, and he
explicitly rejects Kuhn’s thesis that scientific
revolutions are world changes.
He says that old and new theories have links
to concepts not contained in the affected theories,
and he cites by way of example that while Priestly
and Lavoisier had very different conceptual systems
describing combustion, there was an enormous amount
on which they agreed concerning many experimental
techniques and findings.
He also says that he agrees with Hanson’s
thesis that observations are theory-laden, but he
maintains that they are not theory-determined.
He says that the key question is whether
proponents of successive theories can agree on what
counts as data, and that the doctrine that
observation is theory-laden might be taken to count
against such agreement, but that the doctrine only
undermines the Positivist thesis that there is a
neutral observation language sharable by competing
theories. He
states that his own position requires only that the
proponents of different theories be able to
appreciate each other’s experiments.
This view contrasts slightly with his earlier
statement in his Computational
Philosophy of Science, where he said that
observation is inferential.
He says that observation might be influenced
by theory, but that the inferential processes in
observation are not so loose as to allow us to make
any observations we want.
He adds that there are few cases of
disagreement about scientific observations, because
all humans operate with the same sort of
stimulus-driven inference mechanisms.
This statement is not enlightening, since
Thagard does not describe this inferential process
he claims occurs in observation.
It should be commented that in both his
earlier and later statements Thagard has finessed
the vexing problem of meaning variance that arises
due to the theory-laden nature of observation
language. Without
a theory of meaning description he cannot
characterize the concepts in language used for
observation, and thus cannot explain how descriptive
terms can be theory-laden.
Since beliefs can function as partial
definitions, they are both empirical and analytical
statements that enable analysis of the composition
in the concept or meaning associated with a
descriptive term.
Beliefs thereby reveal the meaning components
defined in terms of a theory that make the meaning
theory-laden due to the context supplied by the
theory. And
they also reveal the meaning components defined in
terms of the observation and experimental results
that are not in the theory, and that supply the
descriptive language needed for independent
empirical testing.
Thus Thagard is correct in saying that
continuity is maintained through the conceptual
change by the survival of links to other concepts,
i.e. the nontheory concepts, but he does not explain
how it occurs.
It occurs because the links to those other
concepts constitutes the linguistic context that is
believed to be true, that occurs in the language
used to report observation, and that supplies the
components to the meaning complex that are
unaffected by theory change.
Consider
next Thagard’s thesis specific to theoretical
terms. Both
Thagard and Simon accept the ideas of theoretical
and observation terms, and both use the distinction
in some of their computer systems.
In these systems the theoretical terms are
those developed by a system and the observation term
are those inputted to the system.
But in both their literatures the distinction
between theoretical and observation terms has a
philosophical significance apart from their roles in
their systems. Thagard says that new theoretical concepts arise by
conceptual combination, and that new theoretical
hypotheses, i.e. propositions containing theoretical
terms, arise by abduction.
Abduction including analogy is his philosophy
of scientific discovery, which is described
separately below.
Thagard’s belief in theoretical terms
suggests a residual Positivism in his philosophy of
science. But
he attempts to distance himself from the
Positivists’ foundations-of-science agenda and
their naturalistic philosophy of the philosophy of
the semantics of language.
But he rejects assuming a strict or absolute
distinction between theoretical and observable
entities, and says that what counts as observable
can change with technological advances.
Therefore Thagard does not have the
Positivists’ problem with the meaningfulness of
theoretical terms.
But he retains the distinction thus modified,
because believes that science has concepts intended
to refer to a host of postulated entities and has
propositions containing these theoretical concepts
that make such references.
These propositions have concepts that refer
to nonobservable entities, and these propositions
cannot be derived by empirical generalization due to
the unavailability of any observed instances from
which to generalize. He subscribes to the semantical
thesis that all descriptive terms - observational
terms as well as theoretical terms - acquire their
meanings from their functional role in thinking.
Thus instead of a naturalistic semantics, he
admits to a relativistic semantics.
However, while Thagard subscribes to a
relativistic theory of semantics, he does not
recognize the contemporary Pragmatist view that a
relativistic semantical view implies a relativistic
ontology, which in turn implies that all entities
are theoretical entities.
For example Quine calls relativistic
ontological determination “ontological
relativity”, and says that all entities are
“posits” whether microphysical or macrophysical.
From the vantage of the contemporary
Pragmatist philosophy of language the philosophical
distinction between theoretical and observation
terms is anachronistic. Thagard could retire these linguistic anachronisms
“theoretical” and “observational” as
needless paleo-Positivist fossils, if instead he
used the terms “endogenous” and “exogenous”
respectively, which are used by contemporary
modelers to distinguish between the descriptive
terms developed by a system and those inputted to
it.
Thagard collaboratively with Keith J. Holyoak
developed an artificial-intelligence system called PI
(an acronym meaning “Process of Induction”),
which among other capabilities creates theoretical
terms by conceptual combination. In view of the
above discussion it may be said that in the
expository language used in science all descriptive
terms - not just Thagard’s theoretical terms -
have associated with them concepts which are
combinations of other concepts functioning as
semantic values structured by the set of beliefs in
which they occur.
Thagard’s system PI
system is described in “Discovering the Wave
Theory of Sound: Inductive Inference in the Context
of Problem Solving” in IJCAI
Proceedings (1985) and in his Computational
Philosophy of Science.
PI
is written in the LISP
computer programming language.
In a simulation of the discovery of the wave
theory of sound, PI
created the theoretical concept of sound wave by
combining the concepts of sound and wave.
The sound wave is not observable, while
instances of water waves and sound have been
observed. In
PI the
combination is triggered when two active concepts
have instances in common.
However, most combinations of concepts of
observables are uninteresting, but PI only forms permanent combinations when the constituent concepts
produce differing expectations, as determined by the
rules for them in PI.
In such cases PI
reconciles the conflict in the direction of one of
the two donor concepts.
In the case of sound wave the conflict is
that water waves are observed in a two-dimensional
water surface, while sound is perceived in
three-dimensional space.
In PI
the rule that sound spreads spherically is stronger
than the rule that waves spread in a single plane.
Strength is a parameter developed in the operation
of the system.
Thus the combination of the three-dimensional
wave is formed.
The meaningfulness of this theoretical term
is unproblematic for Thagard, due to his
functionalist view of semantics, which gives the
theoretical term its meaning by the rules, concepts,
and messages in PI.
|
|
