Turn next to the philosophy of Feyerabend, which is more elaborate and more sophisticated than Kuhn’s.  Feyerabend began with an agenda for modern microphysics: to show how a realistic microphysics is possible.  Initially the conditions that he believed a realist microphysics must satisfy were taken from Popper's philosophy of science, and these conditions are contained in Popper's idea of universalism.  However, there is an ambiguity in Popper's universalism, and that ambiguity was not only brought into Feyerabend's agenda while he had accepted Popper's philosophy, it was also operative in his philosophy after he rejected Popper's philosophy, because he rejected universalism in both senses.  The first meaning of "universal" refers to the greater scope that a new theory should have relative to its predecessors, and the second meaning refers to the logical quantification of general statements.  Feyerabend thought that Bohr's insistence upon the use of classical concepts for observational description in quantum theory experiments makes quantum theory inconsistent with both these meanings of "universal".  Thus his later acceptance of Bohr's interpretation of the quantum theory led him to reject universalism in both of Popper's senses, and consequently to advance his radical historicist philosophy of science.  Quite apart from his acceptance of Bohr's use of classical concepts, Feyerabend had adequate reason to reject universalism in Popper's first sense: if it is not actually logically reductionist, as Feyerabend sometimes says, it does gratuitously require an inclusiveness that demands that a new theory explain the domain of the older one.  Recent developments in string theory and M-theory notwithstanding, there are historic exceptions that invalidate such a demand.  Feyerabend notes explicitly in his Against Method for example that Galileo's theory of motion is less universal than that of the preceding theory, namely Aristotle's doctrine of the four cause, which explained qualitative change as well as mechanical motion.  And it might also be noted that quantum theory is less universal than Newtonian mechanics, which was believed to be applicable to microphysical orders of magnitude.
          But Feyerabend also believes that his Thesis I with its dependence on universal logical quantification cannot be applied to quantum theory due to Bohr's semantical thesis of complementarity, which is duality expressed with classical concepts, and he therefore rejects universalism in the sense of universal logical quantification.  In fact Quantum theory must be universal in this second sense, because its experiments are repeatable.  Feyerabend’s rejection involves a semantical error that is made by many philosophers including the Logical Positivists and the Copenhagen physicists.  That semantical error consists of implicitly regarding the meanings of descriptive terms or variables, or even larger units of language, as unanalyzable wholes.  What is needed in order to see how universal logical quantification is consistent with duality without complementarity, is a semantical metatheory of meaning description which enables an analysis of the semantical composition of the meanings of the descriptive terms.  Consider such an analysis, which may serve as a modification of Feyerabend’s Thesis I:
          Since Hempel’s and Quine's rejections of the analytic-synthetic dichotomy, and notwithstanding the fact that Quine rejected analyticity altogether, the distinction may still be retained as a pragmatic one instead of a semantic one, such that any descriptive universally quantified statement may be viewed as both analytic and synthetic, or what Quine calls an “analytical hypothesis”.  The theories found in physics and in many other sciences use mathematical syntax, where universal quantification is expressed implicitly with numeric variables having no measurement values; the variables await assignment of their measurement values by execution of a measurement procedure or by evaluation in an equation from other variables having measurement values assigned.  Furthermore the universality in mathematical language is claimed only for measurement instances; it makes no ontological reference to entities.  The following analysis applies to mathematically expressed language, but for the sake of simplicity the analysis is here given in terms of categorical statements, because statements have explicit quantifiers.  Imagine a list of statements that are universally quantified affirmations having the same subject term and believed to be true.  The concepts associated with the descriptive terms predicated of the common subject in the several statements constituting the list exhibit a composition or complexity in the meaning of the subject term.  The meaning of the subject term may therefore be said to have parts consisting of the predicate concepts, and the meaning need not be viewed wholistically.  Consider in turn the relations that may obtain among the concepts that are universally predicated in the affirmations having the common subject term.  These predicate terms may or may not be related to each other by other universal statements.  If any of the predicate concepts are related to one another by universally quantified negative statements, then the subject term common to the statements in the list is equivocal, and the predicate concepts related to one another by universal negations are parts of different meanings of the equivocal subject term.  Otherwise the subject term common to the statements in the list is univocal, whether or not the predicate concepts may be related to one another by universally quantified affirmations, and the predicate concepts are different parts of the one meaning of the univocal subject term.
          Terms are either univocal or equivocal; concepts are relatively clear or vague.  All concepts are vague, but vagueness may be reduced in discrete increments by adding information.  Adding universal statements to the list reduces the vagueness in their common subject term by clarifying the meaning of the shared subject term with respect to the added predicate concepts.  Adding universal negations relating concepts predicated of the common subject clarifies the meaning of the subject term by showing equivocation.  Adding universal affirmations relating the concepts predicated of the common subject, clarifies the meaning of the subject term by revealing additional structure in the meaning of the common univocal subject term, and makes a deductive system. 
          Now turn to science.  In all scientific experiments, the relevant descriptive language is dichotomously divided into a set of universal statements that are presumed for testing and another set of universal statements that are proposed for testing.  The former is called test design statements and the latter are called theory statements.  The distinction between them is pragmatic, because it depends upon the functions of the statements in testing.  A given descriptive term occurring in a theory may be viewed as a subject term occurring in a list of universal affirmations with the list dichotomously divided into test design and theory statements.  The subject term is thus common to the test design statements and to the theory statements.  The dual analytic-synthetic nature of the statements makes the common subject term have part of its semantics supplied by the descriptive terms predicated of it by the test design statements, which are presumed true for the test.  And this part of the term's semantics remains unchanged through the test, so long as the distinction between theory and test design statements remains unchanged.
          For each descriptive term common to the test design and the theory, the part of the term's semantics supplied by the test design statements does not change; it supplies semantical continuity.  But the semantics of the descriptive term changes; it is different before and after the test.  Before the execution of a test of the theory, all interested scientists who agree to the test design, must also agree that the universal statements describing the test design are true independently of the theory, such that if the test outcome is an inconsistency between the test design statements and the theory statements, then it is the theory that is to be viewed as falsified.  This independence of test design statements is required for contingency in the test, and it also precludes the test design statements from either implying or denying the theory to be tested or any alternative that addresses the same problem.  Therefore for the cognizant scientific profession the semantical parts defined by the test design statements before test execution must be vague with respect to the theory.  This amounts to saying that the theory does not define any part of the semantics of its constituent terms.  However it may happen that the originating proposer and his supporting advocates of the theory may have such high confidence in their theory that for them the theory may also have come to supply part of the semantics for its constituent terms even before the test.
          After the test is executed in accordance with its test design, the test-design statements and the theory statements are either consistent or inconsistent with one another (after discounting for measurement error not attributable to failure to execute the test in accordance with the agreed test design).  Therefore they either characterize the same observed instances or they do not.  If the test outcome is an inconsistency between the test design statements and the proposed theory, then the theory is falsified.  And since the theory is therefore no longer believed to be true, it cannot contribute to the semantics of its constituent descriptive terms even for the proposer and advocates of the theory.  But if the test outcome is not a falsifying inconsistency between theory and test design statements, then for each common term the semantics contributed by the two sets of statements are parts of one meaning complex of the univocal descriptive term, and they identify the same instances.  Furthermore, the additional characterization supplied by the semantics of the theory statements resolves the vagueness that the meaning of the descriptive term had before the test for those who did not share the confidence had by the theory's proposer and its advocates. However, the original proposer and the supporting advocates of the theory have options if the test outcome was a falsification.  They may choose to reverse the status of the test design statements and theory statements, such that the theory assumes the role of defining the subject of the test, and the test design is rejected as an adequate or appropriate description of the phenomenon under investigation.  This prejudice or tenacity is a strategy that need not be rejected as content decreasing, as Popper would have, but may occasion what Feyerabend calls counterinduction.
          While the vagueness in the concept associated with the common subject term is reduced by a nonfalsifying test outcome, the vagueness in the concepts predicated of the subject term by the two sets of statements are not necessarily resolved in relation to one another merely by the nonfalsifying test outcome.  Any resolution of the vagueness in these predicate concepts requires that additional universal statements furthermore relate them to one another.  Such would be the case were the statements formerly used as independent test design statements augmented such that they could be incorporated into a deductive system and derived from the nonfalsified theory after the test.  The resulting deductive system would then make test design statements logical consequences of the theory, but with the theory tested and not falsified, this loss of independence of the test design statements is no longer important.  This amounts to deriving from the theory a new set of laws applicable to the functioning of the apparatus and physical procedures of an experiment and described by the test design statements.  Such a revision of test design language is possible in the case of relativity theory, but is not possible in the case of quantum theory.  In cases where description of the apparatus and physical procedures in terms of the laws derived from the theory is possible after the nonfalsifying test outcome, the original pretest description by the independent test design must result in what retrospectively may be called errors.  Furthermore for the test to have been valid, these errors must be very small relative to the physical effect that the apparatus is used to produce or detect in the nonfalsifying test of the theory.  The concepts associated with the descriptive terms in the original test design statements were initially viewed as vague relative to the terms in the theory, but may later receive more precise meanings from the definitive role of the nonfalsified theory after the test design statements are made derivable from the theory.  This vagueness means that before the test the concepts associated with the vocabulary used in the test design statements had assumed the semantical status that Heisenberg called "everyday" concepts.
          Feyerabend's Thesis I requires that the test design statements, which describe the macrophysical experimental set up, must be incorporated into a deductive system consisting of the microphysical quantum theory in a manner analogous to the incorporation of Kepler’s empirical laws into Newton’s theory enabled by the approximate nature of Kepler’s laws.  And since this derivative macrophysical description has never been achieved for the quantum theory, he later accepted Bohr's complementary thesis, which is the description of the microphysical phenomena with classical macrophysical concepts.  As it happens, contrary to Bohr's instrumentalist thesis but consistent with Heisenberg's semantical views, the microphysical phenomena can be described with the variables in the mathematical expressions of the quantum theory and without classical concepts.  But there is no quantum description of the functioning of the macrophysical apparatus by means of laws logically derived from the quantum theory.  Thus at the conclusion of the first section of "Trivializing Knowledge" in Farewell to Reason Feyerabend says that though Popper rejects reductionism, the variety of entities Popper admits to be real can be admitted as parts of the same world only if the theories that constitute them can be united in a way precluded by the incommensurability that Feyerabend finds in the relative knowledge in Bohr's complementarity thesis of quantum theory.  He then concludes that science is not a theoretical tradition expressed as deductive systems, as he says Popper assumes, but rather is a historical tradition.
          But contrary to Feyerabend, relativism is not the exclusive alternative to deductivism.  The choice between classical and quantum macrophysical descriptions is a false dichotomy; there is a third alternative.  The universal test design statements, such as those describing the experimental set up, need not say anything about the fundamental constitution of matter; that is what the microphysical theory describes.  The semantics supplied by these test design statements may remain vague about this subject for an indefinite time after the nonfalsifying test outcome, just as they had to before the test was performed and while its outcome was not yet known.  After the test the semantics supplied by the tested and nonfalsified quantum theory provides further resolution of the concepts associated with these terms common to both test design and theory statements.  But the semantics supplied by the macrophysical descriptive terms in the test design statements may retain their vagueness indefinitely until a reductionist macrophysical quantum theory may be developed, if it ever is developed, since the concepts associated with these terms are not unanalyzable wholes, but rather are complexes of semantic values.  If Feyerabend's Thesis I were modified such that after the nonfalsifying test outcome the theory as a set of universal analytic-synthetic statements defines only part of the semantics of its constituent descriptive terms, then such a modified Thesis I becomes applicable to quantum theory.  The application of the modified semantical principle implies that the test-design-defined part of the meaning complex associated with the theory's descriptive term is not properly called "classical", because it makes no microphysical claims.   Before the test it is vague with respect to any microphysical theory, and Heisenberg's term "everyday" is appropriate to describe the vague concepts associated with these terms.  But after the nonfalsifying test outcome is known, the whole meaning complex constituting each concept is more properly called a "quantum" concept, because the quantum theory then resolves vagueness by the addition of the quantum theory-defined meaning parts to the whole meaning complex.  And it is for this reason Heisenberg was able to use quantum concepts when he described the observed free electron in the Wilson cloud chamber, and those quantum concepts were resolved by the context supplied by his matrix mechanics.  He thus reconceptualized his observation language, and practiced what Feyerabend called counterinduction.
          In summary, semantical analysis reveals that duality need not be expressed in classical terms required by Bohr's complementarity principle, because the semantics of the descriptive terms used for observation are not simple, wholistic, or unanalyzable, and because prior to testing the semantics of these terms cannot imply an alternative description to that set forth by the quantum theory, in order for testing to have the contingency that gives it its function as an empirical decision procedure in the practice of scientific criticism.  Therefore Feyerabend was closer to the mark with the first of his two approaches to realism in microphysics set forth in his "Complementarity" (1958), and he might have retained universalism in quantum theory had he ignored the reductionist program of Ludwig, developed a metatheory of semantical description, and appropriately modified Thesis I.  With appropriate modification as described above, the application of Feyerabend's Thesis I  to the quantum theory need not imply historical relativism, the rejection of the validity of universal quantification.  The quantum theory with its quantum postulate, its duality thesis, and its indeterminacy relations has no need for Newtonian semantics, either before, during, or after any empirical test.  It is a universal theory with a univocal descriptive vocabulary, and it is not semantically unique in empirical science due to any internal incommensurability resulting from any need to express duality with complementarity.  Had Feyerabend considered Heisenberg's realistic philosophy of the quantum theory, he would probably not have been driven to advocate his incommensurability and historical relativist theses, in order to implement his realistic agenda for microphysics.  Then instead of speaking of the Galileo-Einstein tradition, he could have referenced the Galileo-Einstein-Heisenberg tradition including Heisenberg's pluralistic thesis.
          Consider further Feyerabend's incommensurability thesis, which is central to his historical relativism.  Rejecting the naturalistic theory of the semantics of language including the language of observational description enables dispensing altogether with classical concepts in quantum theory, and thereby with incommensurability within the quantum theory.  But Feyerabend sees incommensurability in Bohr's complementarity thesis only as a special case, a case that is intrinsic to a single theory due to the use of classical concepts.  Most often Feyerabend treats incommensurability as a relation between successively different theories, and he maintained the existence of incommensurability even before he adopted Bohr's interpretation of quantum theory.  In his earlier statements of the thesis he says that two theories are incommensurable, if they can have no common meaning, because there exists no general concept having an extension including instances described by both theories.  The two theories therefore cannot describe the same subject matter, and therefore are incommensurable.  In Against Method he also referenced Wharf’s thesis of linguistic relativity to explain incommensurability in terms of covert resistances in the grammar of language.  There he maintains that these covert resistances in the grammar of an accepted theory not only lead scientists to oppose the truth of a new theory, but also lead the scientists to oppose the presumption that the new theory is an alternative to the older one.  He considers both the quantum theory and the relativity theory to be incommensurable in relation to their predecessor, Newtonian mechanics.  How­ever, he offers no evidence for his implausible historical thesis that the advocates of Newtonian physics had failed to recognize that either quantum theory or relativity theory is an alternative to Newtonian physics at the time of the initial proposal of these new theories or at any other time.
          Feyerabend furthermore maintains that since incommensurability is due to covert classifications and involves major conceptual changes, it is hardly ever possible to give an explicit definition of it.  He says that the phenomenon must be shown, and that one must be led up to it by being confronted with a variety of instances, so that one can judge for oneself.  Feyerabend's concept of incommensurability thus suffers from the same kind of difficulty as Kuhn's concept of paradigm.  Readers of Feyerabend must rely on his identification of which transitional episodes in the history of science are to be taken as involving incommensurability and which ones do not, just as Kuhn's readers must rely on the latter's identification of which transitional episodes are transitions to a new and incommensurable paradigm, and which ones are merely further articulations of the same paradigm.  Although the two philosophers do not hold exactly the same views on the nature of incommensurability, and while they disagree about Kuhn's thesis of normal science, they both refrain from developing a metatheory of semantical description that would enable their readers to individuate theories and thereby to characterize semantical continuity and discontinuity through scientific change.  Feyerabend's recourse to the Wittgensteinian-like view that incommensurability cannot be defined but can only be shown, may reasonably be regarded as an obscurantist evasion in the absence of such a seman­tical metatheory.

Pages [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
NOTE: Pages do not corresponds with the actual pages from the book