The inputs and outputs for the discovery system are called “state descriptions.”  A state description exhibits the state of the object language for a scientific profession at a point in time.  This phrase is also borrowed from Carnap, but it has a very different meaning and purpose in Hickey's metatheory.  The state description consists of all the object-language statements that are relevant to a scientific problem, that have explicit or implicit universal logical quantification, and that are believed to be true by at least some members of the cognizant scientific profession.  Problems in basic science are not like engineering problems.  A scientific problem is one that is solved by a new explanation having the status of what Quine calls an “empirically warranted belief”, and the development of the new explanation changes the state description for a profession.   A scientific profession is a set of one or several scientists who are directing their problem-solving research efforts to the same problem, and they typically communicate with one another either informally in conversation or correspondence or formally in a shared published literature.  Hickey's metatheory is not concerned with the psychology of the individual scientist; it is a theory about an individual scientific profession considered as a special language-using community performing a distinctive professional function regulated by shared institutional values.
          One of the functions of the state description is to describe the semantics shared by the cognizant profession at the chosen point in time, or in the case of a cumulative state description, up to the chosen point of time.  On Hickey's metatheory all the universal affirmative statements in a state description having the same descriptive term as a subject term, have predicates that describe the components of the meaning of the common subject term.  In other words meanings have component parts.  For example empirically warranted belief in the statement "all ravens are black" makes the phrase "black raven" semantically redundant, because the concept of blackness is already included as part of the meaning of "raven.”  Belief functions to give the universally affirmative statement a definitional role, but not exhaustively; they resemble what Carnap called “partial definitions.”  Thus a list of all the predicates in such universal affirmations about ravens would describe the parts of the meaning associated with the univocal term "raven.”  Each statement in the state description may be said to be both a synthetic statement, because it is believed to be empirically true, and an analytic statement, because belief in the statement also enables using it for semantical analysis.  Hickey thus joins Quine's rejection of the analytic-synthetic distinction in so far as the distinction amounts to a separation of statements into two dichotomous classes, and he accepts Quine's rejection of an analytic theory of truth.   But Hickey does not reject analyticity as such, and he says that empirical statements believed to be true may be used analytically for semantical analysis.  Statements need not be in categorical form, to be used for semantical analysis.  It is sufficient that they have a descriptive function, that they make universal claims with or without explicit quantifiers, and that they be accepted as true.  Mathematical equations containing descriptive variables are universal statements when their numerical measurement values are not assigned; so long as their equality condition is believed to obtain, they exhibit the components of the meanings associated with their measurement variables.  However, mathematical statements are not part of what Carnap calls the “thing language”, because they do not reference things, i.e. instantiated entities.  They are better described as statements in the measurement language, which make universal claims about measurement instances when variables are used in place of measurement values, and which may be accompanied by other statements in the thing language describing measuring procedures for obtaining numeric measurement values for their variables.  Thus mathematical equations and inequalities statements are universal but reference measurement instances instead of instantiated entities.  And they are particularly quantified when the values of their variables are specified by numeric values obtained by measurements.
          The idea that meanings have parts is not commonplace in philosophy.  As it happens the piecemeal nature of meanings has recently been proposed in neurology.  In "Stroke Patients Yield Clues to Brain's Ability to Create Language" the Wall Street Journal (12 Oct. 1993) reports Dartmouth College neurologist Dr. Alfonso Caramazza as saying that the meaning of a word must be stored "piecemeal" in the brain.  For example the meaning of "lemon" is the sum of many attributes that the brain has filed away separately, and cerebral strokes have been observed to damage an area of the brain where just one of the attributes is stored.  The article notes that neurologists are presently using positron emission tomography (PET) to locate such storage areas physically in the brain.  But Hickey's thesis is not a neurological or a physiological theory; it is a thesis in philosophy of language based on such obvious evidence as the fact that lexical entries in a dictionary display the parts of meanings, because the dictionary definitions are believed to be true.  In Hickey’s semantical thesis the “parts” of the meaning associated with a descriptive term are those features of the world that a language is capable of distinguishing at a given point in time.  The smallest distinguishable features are almost never isomorphic to the descriptive terms of the language, such that there are no “primitive terms.”  The smallest distinguishable features are smaller than the meanings of terms, because they are components of the meaning complexes consisting of them.  Hickey calls these more elementary parts “semantic values.”  The meanings associated with the terms are structured composites of these semantic values.  There is no reason to believe that there are not always features and aspects of reality that either are not presently distinguished or are currently distinguished but not recognized in the descriptive vocabulary of a language.  Adults recognize this growth in the discriminating ability in the semantics of the discourse of children, but such growth is not limited to children.  Bilingual speakers perceive differences not adequately translated between languages.  The student biologist learns to discriminate many features of animals noted by the professional biologist but unobserved by the layman adult.  In fact such semantical refinement often occurs at the moving frontiers of human knowledge including notably scientific knowledge.  Examples may include new phenomena revealed by new observation instruments, such as the optical or electron microscope, the optical or radio telescope, or the X-ray photograph. And when such refinement occurs, the transition to new knowledge may be called “semantic incommensurability”, by which is here meant the introduction of new semantic values that make prior language incapable of describing the new ontology.  On Hickey’s view and contrary to Kuhn’s later view incommensurability is not merely a restructuring of available semantic values, which he calls “taxonomic categories” or “lexicon.”  Furthermore incommensurability does not imply complete discontinuity; it occasions only partial discontinuity, since continuity is supplied by the existing semantic values constituting the other parts of the meanings associated with the affected terms.
      A state description may be called a static (or synchronic) analysis, since it pertains only to a point in time.  A metatheory is a description of the transition from one state description to another in a science, and may be called a dynamic diachronic analysis.  In computational philosophy of science a metatheory does not describe institutional change, but rather describes changes within the institution of science.  Both empirical testing and theory development produce new state descriptions.  A discovery system is a computer system that generates a new state description from the object-language input in the initial state description.  The output is a terminal state description, which contains new theories.  The comparative examination of a semantical change resulting from the transition to a new state description may be called a comparative static diachronic analysis.  Such a semantical comparison between an initial and a terminal state description for the same scientific problem reveals which parts in the meanings of the descriptive terms in the statements remain unchanged, and which parts have changed due to the change in beliefs.  Each descriptive term exhibits semantical continuity and semantical discontinuity.  Thus Hickey finds it unnecessary to accept either the wholistic doctrine of "paradigm shift" advocated by Kuhn and Feyerabend or the wholistic gestalt theory of meaning used by Kuhn and Hanson.

Hickey’s Functional Analysis

      With these linguistic-analysis basics in mind turn now to the four basic topics common to modern philosophies of science, beginning with the aim of science.  Hickey's statement of the aim of science could be called a "rationality postulate" for science, if that phrase is not taken to mean that the goal statement is an incorrigible dogma, as it is in neoclassical economics.  His thesis of the aim of science is an empirical hypothesis about the regulating institutional value system for empirical science that is responsible for scientific progress, and is based in the most noteworthy achievements in the history of modern science.  The current statement of the aim of science is a statement in the contemporary Pragmatist philosophy of science, the philosophy of science that has evolved from philosophers’ examination of the institutional evolution of twentieth-century physics, and that articulates its institutional views and values.  At the opening of the twentieth century the prevailing institutional views and values were those of the Positivists.  The lenses of Pragmatism enable the contemporary philosopher to recognize the dysfunctional effects of both Positivism and Romanticism, especially in current research practices in the behavioral and social sciences, even though the researchers in these sciences are oblivious to their institutional retardation.  The contemporary Pragmatist statement of the aim of science may be expressed as follows:

Scientists aim to construct explanations by developing theories that satisfy the most critically empirical tests that can be applied at the current time.

Historically scientists have accomplished great achievements with other aims in mind, and then later in retrospect the criteria they had actually employed are seen to be different.  Newton for example denied that he created hypotheses, notwithstanding the hypothetical character of the laws of motion and gravitation.  Thus the aim of science can also be re-expressed without referring to a conscious aim:

Science achieves explanations by developing theories that satisfy the most critically empirical tests that can be applied at the current time.

The meaning of these statements is explained by the other three topics in the Pragmatist philosophy of science - discovery (i.e. theory construction), criticism (i.e. theory evaluation), and explanation - and it is based on the distinctively Pragmatist concept of "theory.”
          Basic research regulated by this institutionalized aim encounters various constraints, which scientists view as impediments to be overcome.  But there is one constraint which is a voluntary constraint that scientists do not view as an impediment to be overcome, but rather view with an attitude of obligation and respect as integral to the aim of science.  That constraint is the empirical constraint; it is like a moral constraint in which scientists are socialized by their professional education, and which is reinforced by social controls that condition professional recognition upon conformity to its regulation of research practices.  The operation of this institutional constraint is considered below in the discussion of scientific criticism.  All other constraints are impediments to be overcome.  Simon's "computation constraint" is an example of such an impeding constraint, as are other more circumstantial conditions such as limited financial resources.  Furthermore there are two constraining impediments that are more than merely circumstantial; Hickey calls these the "cognition constraint" and the "communication constraint.”  These two are distinctive in that they are semantical constraints, which are integral to language and therefore to the final product of basic research science.  And they are operative in basic research, because basic scientific research in a science depends upon the existing articulate beliefs in the current state of the science.  The mechanical production of a new state description will produce a greater or lesser semantical change, depending on how radically or moderately the terminal state description revises the beliefs included in the initial state description.  This change is a restructuring of the semantic values available in the initial state description, the input object language.  And because there are no semantic values in the outputted theories that were not already in the input language, there can be no semantic incommensurability. 
          All resistance to learning involved in assimilating the outputted theories is due merely to change of psychological habit. Beliefs are the glue that bonds and structures the components of the meaning for each univocal descriptive term or for each of the meanings associated with an equivocal descriptive term.  When a scientist develops a new theory that contradicts many previously held beliefs, he disassociates some of the component parts - semantic values - of the meanings of the descriptive terms and he re-associates and restructures those components according to the theses of a new theory.  This semantical dissolution and restructuring has a disorienting and confusing effect due to linguistic habit, which Hanson called "conceptual resistance” that impedes the scientist's theory-developmental work.  This resistance is actually a psychological one, but its impeding effect on reconceptualization is the cognition constraint.  Similarly when the scientist has created a radically new theory, his colleagues to whom he attempts to communicate his new theory also experience the disorientation and confusion as they achieve the dissolution and reconstruction of their semantics associated with habitually familiar terms.  The impediment or psychological resistance that they encounter in this learning experience due to this semantical change is the communication constraint.  A computerized discovery system has no psychological habits and therefore is not impeded by any cognition constraint, but the scientists who read its output may have to overcome a severe communication constraint in their attempt to "communicate" with the machine by assimilating its outputs.
          Consider next the topic of scientific explanation.  An explanation is the fruitful outcome of successful work in accordance with the regulating institutionalized aim of science: it is a theory that has been tested and not falsified.  It is a belief which had been a theory prior to its testing, which has since been tested by the most critically empirical test available at the time, and which has not been falsified by any test to date.  Such a universal statement may also be called a scientific “law”.  A theory in turn is a set of one or several related statements having explicit or implicit universal logical quantification, and that is proposed for testing.  A law a theory that is no longer a theory, because the successful test outcome has removed its exceptionally hypothetical status relative to all other empirical universal statements accepted for the test.  The law statement operates in an explanation as a premise in a deduction that concludes either to descriptions of particular events or to other universal statements.  Such derived universal statements must be tested, and prior to testing they may reduce the law back to the status of a theory by enabling the further testing.  The law could later be used as a test-design statement for defining another problem in another state description.  Laws and therefore explanations with laws are not permanent.  New test designs that resolve the vagueness in the semantics of current test-design statements thereby enabling improved measurements or observational techniques, will occasion reversion of the explanatory law to the status of a theory for retesting, and the result could be falsification.  This definition of "theory" as universal statements proposed for testing and this distinction between theory and law are based in a Pragmatic definition of “theory” in terms of the use or function of theory in critical testing under the institutional regulation of the aim of science. 
          This contemporary Pragmatist definition of theory language in terms of its function is opposed to the earlier Romanticist and Positivist definitions in terms of some preferred semantics and ontology.   The Romanticist definition still prevails in many social sciences, where "theory" language is has a semantics that describes an ontology consisting of mental states. The Positivist view defines "theory" in contrast to "observation" language, which is also alleged to have a specific semantical content such as sense perceptions, phenomena, or sense data.  Both the Romanticist and Positivist views of the aim of science represent anachronistic institutional views and values for empirical science.  Hickey believes that philosophers who continue to accept the Positivist philosophy have done so because they have not rejected the semantical distinction between observation and theory language. The contemporary Pragmatist may define "observation" language in contrast to theory in the sense that the universally quantified test-design statements supply the vocabulary and semantics for observational description.  This observation vocabulary conceptualizes and articulates the perceptual experiences involved in the test situation including the report of the test outcome.  But the Pragmatist distinction is not a semantical distinction; Pragmatists do not recognize any inherently observational semantics.  The Pragmatist distinction is based on the strategic functions of the test-design and theory statements in the empirical test - which can and do change.
          Theory language is defined pragmatically, but theories are individuated semantically.  Theories are individuated for either of two reasons: Firstly two theory expressions that address different problems are different theories in the sense that they are different theories relative to different scientific professions.  What is theory for one profession is not so for some other; theories are proposed solutions for a problem, and each profession is defined in terms of the scientific problem that it is addressing.  What is conventionally called “a science” is actually many scientific professions.  Secondly two theory expressions that address the same problem but make contrary claims are different theories, and they are different theories for the same scientific profession.  This is similar to Popper’s criterion for identifying different theories in crucial experiments.  He says that they address the same problem in the sense that they share the same test-design statements, the language that characterizes the problematic phenomenon that the tested theories propose to explain.  In an active science there may be many alternative theories in the same state description, since for each theory there need be only one member of the profession who has sufficient confidence in the theory to propose it for testing.  In practice there are typically one or several minority views and a majority view, with the minority supporting the new upstart such as Einstein’s theory in 1919 at the time of Eddington’s eclipse experiment.
          The topic of explanation pertains to the synchronic perspective, since it depends on the state of beliefs and test outcomes at a point in time.  The topics of criticism and discovery pertain to the diachronic perspective, since they involve change in scientific beliefs between two points in time.  These two functions are performed by the research scientist within the regulating institutional matrix defined by the aim of science, and they are the practices that have the effect of changing the state of the object language and therefore its state description.  Both of these functions are operative in a discovery system, which produces scientific change.  In the dynamic perspective consider firstly scientific criticism by empirical testing. Contemporary Pragmatist philosophy of science admits only empirical criteria for scientific criticism, and it excludes prior ontological criteria, including those required by the Positivist and Romanticist philosophies.  Hickey thus joins Quine's rejection of any metaphysics or "first philosophy", which would impose any nonempirical criteria for scientific criticism.  And he therefore also joins Quine's doctrines of ontological relativity and scientific realism, which were practiced by Galileo, Einstein and Heisenberg (but not Bohr), when these historic physicists affirmed that the real world is as their empirically tested and nonfalsified theories describe it. Hickey maintains a relativistic thesis of the semantics and ontology of language implied by the artifactual character of meaning, but he does not maintain a relativist thesis of truth, save for the banal fact that truth is a property of statements and is therefore relative to what is said. 

Pages [<< Previous 10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] 
NOTE: Pages do not corresponds with the actual pages from the book