( Tansu KÜÇÜKÖNCÜ ( in Turkish alphabet ) )
------------------------------------------------
PERCEPTION, THEORY and COMMITMENT
The New Philosophy of Science
( Harold I. BROWN , 1977 )
Book review and criticism by Tansu KUCUKONCU
Two central problems of theory of knowledge: problems of meaning and of truth.
The empricist approach to these problems received its classical form in the work of D. Hume.
All the perceptions of the mind resolve themselves into two distinct kinds, which he called IMPRESSIONS and IDEAS.
Impressions are the immediate objects of awareness that weexperience when we perceive or introspect. Ideas are the objects we are aware of in all mental activities other than perception and introspection.
Ideas:
1.simple : copies of impressions which remain in the mind
2.complex: ideas that the imagination creates by combining simple ideas
The basic unit of meaning, for Hume, is the term. And a term is meaningful only if there is an idea which corresponds to it. An individual can learn a term the meaning of a term only if he has experinced the impressions necessary for the formation of the corresponding idea. And any term which is supposed to refer to an object beyond the limits of possible experince is a mere meaningless term or sound. The range of meningful language ids limited to the range of possible experince.
Meaningful propositions:
1.relations of ideas
2.matters of fact
The only world that can be known is the world of impressions and every impression is ontologically distinst from every other impression.
This causes some problems. Suppose that observed that a particular set of impressions always occur together. But it can not be quarateed that those will occur together in the future. This make it difficult to make universal propositions.
Historically, one o fthe most important objections to empricism has come from the philosophy of mathematics. It's not experince that we have come to know that 2+3=5, etc.
That all pure mathematics deals exclusively with concepts definable in terms of a very small number of fundamental logical concepts, and that all its propositions are deducible from a very small number of fundamental logical principles. (Whitehead and Russell)
Propositions:
1.elementary (atomic): either true or false
2.molecular: constructed out of elemantary propositions by means ofoperators.
The essential property of implication for the mathematician and philosopher of mathematics is that anything implied by a true proposition must be true.
Logicism maintains that mathematics is logic and thus that mathematics is true in just those ways in which logic is true. This does not solve the problem of the nature of mathematical truth, but only transforms it into the problem of the nature of logical truth.
Truth tables (Wittgenstein):
1.true for some values of argument and false for the others
2.false .. .. .. .. .. .. true .. .. ..
3.true .. all values of the arguments (tautology)
Hilbert's formalism:
Logicist holds that pure mathematics and logic are true. Formalist .. .. they are neither true nor false, being merely rule-governed games with symbols.
All arguments must consist of the manipulation of symbols inaccordance with precise rules.
Positivism: a form of strict empricism. The positivist maintains that only those knowledge claims which are founded directly on experience are genuine.
Modern logical positivism: For the logical positivist there are two forms of research which yield knowledge: emprical research, which is the task of the various sciences, and logical analysis of science, which is the task of philosophy.
The central doctrine of logical positivism is the verificationtheory of meaning, the thesis that a contingent proposition ismeaningful if and only if it can be emprically verified.
For Hume: the basic elements of experince are impressions
For Wittgenstein: the basic units of experience are facts: not just qualities such as 'red', but rather 'that there is red at a given time and place.
Purported propositions:
1.formal propositions, tautologies and contradictions
2.atomic
3.molecular
4.pseudo: mere meaningless combinations of sounds or signs without cognitive content.
First three of are them are meaningful and their truth values can be determined by means of observation and logic.
It can be understood as a more moderate version of logical positivism.
The central difficulty for logical positivism as a philosophy of science is that scientific laws which are formulated as universal propsositions cannot be conclusively vreified by any finite set of observation statements.
Carnap proposed replacing the notion of verification with the notion of a 'gradually increasing confirmation', and takes rthe notion of an 'observable predicate' as fundamental,defining 'confirmable sentence' in terms of this notion.
Two of the central problems of logical empricist philosophy of science:
1.analysis of the confirmation relation that is to hold between a scientific law and the observation statements which confirm ordisconfirm it, and the analysis of how scientific terms get their meaning.
2.some terms of theoretical physics, which cannot be observable.
Nicod: The question of whether a particular observation statement confirms a given hypothesis should depend only on the content of the hypothesis, not on its formulation. Hempel: The hypothesis in question is 'All ravens are black', thediscovery of any instance of an object which is not black and not a raven, such as a yellow pencil, would serve to confirm our hypothesis.
He argues that the appearence of paradoxicality derives from themistaken assumption that the universal statement 'All ravens are black' is only about ravens.
His second argument: In so far as we are attempting to the logic of confirmation, we are only concerned to analyze the relation between an hypothesis and specified body of evidence, so that in any particular case we must adopt the fiction that the piece of evidence in question is all the information we have. Brown: In planning an experiment to test an hypothesis, or deciding whether a particular body of information is relevant to the truth of some hypothesis, the scientist does not pretend that the experimental result in question is the only piece of evidence he has; rather he brings to bear every bit of evidence that he can muster inorder to draw his conclusions.
Logical calculi, like mathematical calculi, can be constructed independently of the examination of any body of knowledge or experience, but the fact that a calculus exists is not a sufficient reason for assuming that it is an appropriate tool for the analysis of any particular area of human experience.
Philosophy of science could not proceed just by collecting information about what scientists do.
In the case of earlier versions of logical empricism there is, to begin with, the empricist assumption that all scientific knowledge consists of generalization from experience.
The continuing failure of the logical empricists to reach agreement on a resolution of the paradoxes of confirmation provides an important reason for looking seriously at alternative approaches to the philosophy of science. Hempel: assumes that if an hypothesis is confirmed by a given observation statement, the addition of further observation statements cannot negate that confirmation relation.
Hempel: If an observation report confirms a hypothesis H, then it also confirms every consequence of H.
Goodman: While ocnfirmation is indeed a relation between evidence and hypothesis, this does not mean that our definition of this relation must refer to nothing other than such evidence and hypothesis.
There is another vast range of additional information that we can legitimately take into account: information about the history of past projections and their successes and failures.
Contemporary empricists generally agrre that a central characteristic of succesful scientific theories is that it is possible to predict new phenomena on the basis of them, and to use theories which were originally proposed in oredr to account for other phenomena that were not considered in their construction.
A theory which was postulated in order to account for a specific set of phenomena and which could account for no other phenomena would not generally be considered a significant scientific achievement. Russel: But on his approcah, no theory would be extendable for every time a new kind of data were taken into account, we would have to add this data into the definitions of our terms and we would thus be redefining the theoretical concepts rather than bringing a new area ofexperience under an old concept. Bridgman: In general, we mean by any concept nothing more than a set of operations; the concept is synonymous with the corresponding set of operations.
It is the empricist thesis that only terms which are defined by reference to observables have cognitive meaning that led to the attempt to define disposition terms by means of observables, and it's the empricist requirement that such definitions be formulated in the symbolism of principia logic that led to the rejection of the proposed definition and the need for further research.
Reduction sentences introduce terms only for the specified test conditions; they do not provide general definitions for this terms. Carnap: Rather than defining theoretical terms by reference to observables, theoretical terms are given a partial interpretation for a particular set of experiments and a particular set of possible outcomes of these experiments.
The terms introduced by reduction sentences are 'reduced in a certain sense' to observables.
Reduction sentences only introduce terms for a specific set of specified experimental conditions. Hempel: Reduction sentences only define the terms that they introduce for a specified set of circumstances, it remains possible to extend theterm to new circumstances by the introduction of new reduction sentences for those new circumstances. Brown: From analytic statements we can only deduce further analytic statements. Carnap: But, as he recognized, if reduction sentences are accepted as a form of definition, the general thesis that all definitions areanalytic must be given up, for any reduction pair has a definite emprical content.
For many empricists the sole function of science is to find relations between observation statements; theoretical statements function only as intermediaries in this process. Thus if a way can be found to formulate all connections between observables without having to make use of theoretical terms, theoretical terms will have been shown to be unnecessary and the problem of analyzing their emprical meaning is eliminated. Craig: What his theorem provides is a general method of eliminating a selected group of terms from a formalized system without changing the contenet of the system. In order to apply the method it is first necessary that we have an effective criterion for distinguishing the 'essential' expressions of the system from the 'auxilary' expressions. Taking the 'content' of the system to be identical with the class of essential expressions. He gives a method for constructing a new axiomatized system which includes all the essential expressions and none of the auxilaries.
The most important defect: It can only be applied to completed deductive systems.
Campbell: Viewing a scientific theory as an axiomatized formal system, a distinction is made between two parts of the formal system: the body of theoretical propositions which is formulated solely in the theoretical vocabulary, and a set of 'correspondence rules' which connect functions consructed out of the theoretical terms with observations terms.
What is important about this analysis in the context of our present concerns is that the terms appering in the theoretical propositions alone, independently of the correspondence rules, aretaken as having no emprical meaning; emprical meaning is conferred on these terms when they are connected with experience by means of thecorrespondence rules.
These terms will get their emprical meaning by virtue of their occurence in formulas in which other terms which occur in corresppondence rules also occur.
Correspondence rules are affected by emprical data and they cannotbe taken to be either analytic statements or conventional rules.
Hempel and Oppenheim: The defect in te method that they construct forusing any law to explain any event is seen by them equivqlent to eliminating the predictive force of the proposed explanation and the anomaly is resolved by restoring this predictive force.
An explanation of an particular event is not fully adequate unless its explanans could have served as a basis for predicting the event in question and conversely. Hempel: For the systematic purposes of scientific explanation, reliance on anologies is thus inessential and can always be dispensed with.
For him, models and analogies may play an important heuristic role in the process of constructing theories and a pragmatic role in helping us to understand theories, but they have no relevance to studies of the logical structure of explanation. Scriven: The scientific explanation represent a refinement on the ordinary explanation. It is the understanding which is the essential part of an explanation.
Unlike the deductive explanation, a statistical explanation does not show that given the premises, the phenomenon to be explained necessarily occurs, but only that it is highly probable. Hempel: In the application o finductive logic to a given knowledge situation, the total evidence available must be taken as the basis for determining the degree of confirmation. Nagel: The normal expansion of some body of theory are shown to be derivable from that theory when suitably specialized. Reichenbach: Physical theories give an account of the observational knowledge of their time; they cannot to be eternal truths. Hempel and Oppenheim: The sentences constituting the explanans must be true.
It is not sufficient that the explanans be higly confirmed. Hempel: He has given up the notion that true explanations are achievable in science and the consequent notion that science is capable of establishing the final truth of propositions. Nagel: The requirement that the premises in a satisfactory explanation muts be true seems inescapable.
But at the same time he denies that these premises must be known to be true. Current science includes a fair number of acceptable scientific explanations.
But a an acceptable scientific explanation must have true premises so that he must hold that a fair number of contemporary scientific theories are indeed true.
We do know that many currently accepted scientific explanationsare satisfactory and thus have true premises, but we do not know whichscientific claims fall into this class.
Popper: His central thesis is that there is no process of induction bywhich scientific theories are confirmed and thus no role in the philosophy of science for a theory of confirmation as understood by logical empricists.
Popper: The criterion of demarcation that Popper finds implicit in thework Of the positivists is one which might be called 'verificationosm'.
If one of these deduced observation statements is shown by experience to be false, then the universal proposition in question in question is false.
A proposition is scientific only i fit can be falsified by experience.
Pseudo-scientific theories such as astrology often make correctpredictions, but they are so formulated as to be capable of evading any falsification and are thu snot scientific.
Scientific theories must not only be emprically falsiable, but a scientific claim must be rejected as it has encountered a single falsifying instance.
If observation shows that the predicted effect is definitely absent, then the theory is simply refuted.
It must be agreed which observable situations mean that theory is refuted.
Science progresses as a result of scientists making bold cojectures which go beyond the available data; having made his conjectures, the scientists prime concern in testing his theories is not to attemp to prove them true, but to attemt to refute them.
The discovery of instances which are in accordence with the predictions of a theory neither confirm the theory nor confer a degree of probability on it, but they are not totally irrelavent to the evaluation of the theory; under certai circumstances they serve as corroborating instances. A theory is corroborated whenever it possibly have refuted the theory fails to refute it.
We cannot define a numerically calculable degree of corroboration, but can speak only roughlyin terms of positive degrees of corroboration, negative degrees of corroboration, and so forth. Popper: There is no final refutation of a theory.
Basic statements are accepted or rejected as a reuslt of experience, but it is logically impossible to prove or disprove any statement.
Falsification takes place only after scientists agree to accept a basic statement as adequately corroborated.
Every test of a theory must stop at some basic statement or other which we decide to accept.
Falsification o fa theory requires the acceptance of a basic statement which cotradicts it. This condition is necessary, but not sufficient for falsification.
No actual falsification can occur until after a falsifying hypothesis has been corroborated.
Every case of falsification requires a prior corroboration and in no particular case can a falsification be any stronger or more final than a corroboration.
The judgement of the scientific community plays a much greater role than does the application of formal rule and effective criteria, and in which theory and observation are much more nearly co-equal partners in the construction of science.
The system of postulates which make up a theory 'floats' or 'hovers' freely above the plane of emprical facts, but it's the emprical facts, which are known independently of any theory, that quarantee the objectivity of science.
In response to the view that perception provides us with pure facts, it is argued that the knowledge, believes and theories we already hold play a fundamental role in determining what we perceive. Hanson: There is more to seeing than meets the eye. Duhem: An observer who lacks the relevant knowledge will not gain the same information by watching the experiment as will a trained physicist and there is thus an important respect in which the layman and the physicist see different things when they observe the same experiment.
These differences are of fundamental importance for understanding the nature of significant perception and thus for understanding the way in which perception can contribute to knowledge.
If our knowledge and believes play a central role in determining what we perceive, then the scientific theories that a scientist holds should play the same sort of role in determinig what he observes in the course of his research. Hanson: Scientific observation should be theory-laden. Kuhn: If we can succed in stripping off the identification, we make a transition from theory-laden to non-theory-laden perception, from 'seeing as' to 'seeing' and thus succeed in observing the object itself.
Concept-free seeing is not the criterion for what is really there, but rather our independent knowledge of what is really there tells us what to look for. Hanson: In every case in which I see is the case I gain information as a result of my seeing, but what information I gain depends not only on what visual events takes place in my eyes, nerves and brain, but also on what information I bring with me.
Every instance in which I gain information as a result of seeing is an instance in which I see that something is the case and the more I already know about the situation in question the more I can learn.
Observation only occurs in those cases in which we make use of our physical senses. Hanson: ~ The meaning of the observed situation that becomes a part of our knowledge and the objects of significant perception are thus meanings. Brown: Meanings are the objects of significant perception. Sense-data cannot be the primary objects of our knowledge.
In my everyday experience I continually select out as meaningful only a very small number of sensations which appear in my visual field, my auditory field, etc. and it's only to these sensations that I pay attention.
The scientist does not record everything he observes but rather only those things which the theories he accepts indicate are significant.
The important case in which a scientist identifies a phenomenon as anomalous or problematic he is clearly obserrving its meaning in terms of the theories he holds, for if he had no beliefs about what ought to occur in the situation in question, no occurance could be perceived as problematic.
Popper: Empricism requires that theories be tested by deducing from the theory consequences which can be compared with the results of observation.
Experience can motivate a decision to accept or reject a singular statement but can never prove or disprove any singular statement, and thus that all singular statements are accepted or rejected by convention. Brown: A single perceiver can see a given object in different ways and two different observers can simultaneously see the same object differently.
For scientific theories to be emprically testable, the percepts against which they are tested must be independent of the theory. A theory which creates its own data can never be refuted by this data.
If observation is theory-laden there is no adequate reason for accepting one theory rather than another. Kuhn: The theory-ladenness of perception omplies relativism is only plausible if one accepts the prior presupposition that only the observation of theory-free data can give us a reason for accepting one theory rather than another. Kuhn: Describes normal science as research done in accordance with a 'paradigm'. Brown: In normal science, an accepted fundamental theory serves to organize and structure scientific research.
Kant: Knowledge is only possible in so far as we have experience, but we must distinguish between experience and sensations.
Experience is also constituted of a combination of form and content, with sensation providing the content of experience while the mind itself provides the form. There are two ways in which the minds provides this form: via the sense of via the concepts of the understanding. There are two forms of sense: space and time.
For him, the fact that the objects we experience are found in space and time is a fact about the structure of the human mind and nota fact about the mind-independent structure of reality.
We cannot know a priori what the particular cause of any given event is, this part of the content of experience and must be discovered by emprical research.
'Every event has a cause', and it is this proposition that we will consider further in oreder to illustrate the role of presuppositions in scientific research.
For him, the task of the scientist is to find the causes of events and a scientific problem is solved only when a cause for the event in question has been found. If scientists are unable to find the cause of an event, it is always the scientists and not the principle of causality that is at fault. Collingwood: For him, as for Kant, all knowing requires presuppositions, but they change throughout the course of human history.
His basic thesis is that every meaningful propositions is the answer to some question and we can only understand the meaning of a proposition if we know what question it is intended to answer.
Because questions have presuppositions which generate them, it is always possible to take one of two attitudes toward a question: we can answer it, or we can reject the presupposition and thereby reject the question.
The presuppositions those which are propositions and themselves answers to questions, are all relative presuppositions. But at the root of each sequence of questions and answers, there is an absolute presupposition which is not the answer to any questio, which is not therefore, a proposition, and which is neither true nor false. Rather an absolute presupposition is something like a methodological principle to be judged by what he calls its 'logical efficacy'. Thus absolute presuppositions are the foundation of all intellectual activity: but unlike Kant, for him what absolute presuppositions are hekd are the charecteristic of a given era and charge in the course of history although he gives no analysis of how change takes place beyond saying that absolute presuppositions develop 'strains' which lead to their collapse.
Each era in the history of science has been charecterized by somefundamental conception of what nature is.
His thesis that absolute presuppositions are neither true nor false and that they are taken to be neither true nor false by by those who propound them is most doubtful.
It is not clear whether each discipline has its own absolute presuppositions or whether they are charecteristic of all thought in an era, nor even whether the answer to this question may not be different with respect to different eras in the history of thought. Brown: Perhaps the most striking feature of presuppositions is that they do not fit into the customary dichotomy between analytic and emprical propositions.
A working scientist who seeks the cause of some phenomenon does not have the option of deciding that he is not dealing with an event and thus eliminating his problem when he finds that he can discover no cause.
When a scientist has difficulty finding phenomena that his presuppositions tell him must be present he will often carry out emprical research to seek this phenomena. This suggests two more arguments against the thesis that presuppositions can be expressed in analytic propositions. First, one of the central charecteristics of an analytic proposition is that no counter-instance is logically possible. Second, a research project is often neede to eliminate a scientific scandal and show that the accepted presupposition is justified. Although presuppositions rae protected propositions which are not given up lightly at the first sig of a counter-instance, presuppositions do change and a presistent failure to account for an anomaly might well lead to changes. Therefore it is often necessary to carry out emprical research in order to defend a presupposition, but such research is never necessary for the defense of an analytic proposition.
Many of scientific propositions have been taken tp be necessary, eternal a priori truths, but some of them nonetheless been abondened, it should by now be clear that every scientific proposition is subject to possible revision.
From the point of view of logic alone, a counter-instance only tells us that there is something wrong in the accepted structure, it tells us nothing about where the problem lies. Duhem and Quine: We never test an isolated scientific proposition, but only the entire body of science
Even though there are no a priori propositions in science, not all propositions are treated as testable emprical hypothesis. It is only because a large body of knowledge is taken as paradigmic that we canisolate individual propositions for purposes of testing, and what conclusions we draw from a particular test depends on what propositionsare taken as paradigmic.
A scientist who does research within the framework of an accepted theory has to learn hi strade, this involves more than learning a set of paradigmic propositions, and the trained scientist knows a good deal more than he could state as a set of propositions.
The information can only be passed on to the learner by taking himthrough a course of problem solving which will lead him to discover the necessary techniques for himself.
The student is being initiated into a normal tradition: he is learning how a particular style of science is done, how one solves physical problems, and eventually, what the unsolved problems are.
Whenever the structure of the phycical world fail to mesh, anomalies will appear and although many anomalous event may eventualy be interpreted in terms of an accepted theory, it is the recalcitrant anomalies that eventually lead to the overthrow of one theory and its replacement by another, i.e. to scientific revolutions.
Copernicus' attemp to deal with an astronomical problem had theeffect of undermining the foundation of the accepted physics so that the new astronomy required to construction of a new physics.
The thesis that meaning of scientific concepts changes as a resultof a scientific revolution has been regarded by many logical empricists. Israel Scheffler: No significant conceptual change takes place unless the extension of the concept has changed. Brown: Besides concepts with change in the course of a scientific revolution, there are others which are dropped altogether. Hempel: The concepts of science are the knots in a network of systematic interrelationships in which the laws and theoretical principles form the threads.
The meaning of theoretical terms cannot be completely specified byreference to an antecedently available vocabulary, but he concludes from this that the problem of the meaning of theoretical terms 'does not exist'.
A scientific concept is a knot ina web; the strands in the web are the propositions that make up the theory; the meaning of a concept is its location in the web. Thus the meaning of a concept is determined by the strands that come into this knot, by the other knots the one in question is connected to, and by further connections of these other knots.
A concept is not a simple which is grasped entirely or not at all, but rather a complex which can be learned a bit at a time. Brown: Just as tehere can be no significant observation without theories, so there can be no scientific theory that is not used to organize some area of experience.
Concepts, propositions and observations are the elements from which scientific theories are built. It is impoosible to introduce any of these elements without introducing the other two, and impossible to begin learning one of these aspects of a theory without beginning to learn the others.
Relativity theory is a generalization of Newtonian theory are based on the formal and quantitative relations between the two theories, not on their historical relation, and on formal and quantitative grounds there is no reason for maintaining that relativity theory is a generalization of Newtonian theory any more than of Mewtonian or Pewtonian theory. None of this is to be taken as denying the obvious fact that relativity theory grew out of Newtonian theory, not out of its successes, and the process which yielded the new physics was not a process of generalization, but rather a process of adopting new presuppositions about the structure of physical reality and transforming the basic concepts with which scientists think about and deal with the physical world.
The notion of a scientific revolution is a philosophical, not a scientific, notion. Just as scientists make use of data supplied by observation and experiment in constructing their theories, so the philosopher of science makes use of the data supplied by the history of science.
A philosophy of science is a theory of the same general kind as a scientific theory.
The philosopher is concerned only with logical questions and that such questions arise only after a scientific theory has been formulated; the process by which a scientist happens to think up a particular theory is of no concern to the logician or philosopher, even though it may be of conciderable interest to the psycologist or sociologist. Reichenbach: A sharp line can be drawn between the discovery and the testing of scientific theories; and that is only with respect to the testing of theories that we can speak of a logic at all, for logic has nothing to say about discovery. Brown: The context of justification is thus part of the context of discovery and no sharp line can be drawn between discovery and justification.
Two parts of the discovery process:
1.logical part (justification)
2.creative part (non-logical)
Popper:There is no such thing as a logical method of having new ideas, or a logical reconstruction of this process. Every discovery contains an 'irrational' or a 'creative' part. Plato: Dialectical logic applies to atte[ts to answer a question, but questions only arise in the context of presuppositions. In the context of a particular set of presuppositions many answers to a question are possible, teher is no effective procedure for determinig which answer ought to be put forth at a given point, although the detailed structure of the argument will often be highly suggestive, and many self-consistent proposals will be ruled out by the presuppositions. Brown: Dialectical logic does not deal with the relations between isolated or relatively isolated propositios, but with the role of propositions and questions in so far as they are parts of structured systems of presuppositions and problems. It does not provide a set of formal rules for analyzing the relationships between statements, as does deductive logic. It is a content logic, not a formal one.
What the concept of a dialectical logic provides is a tool for examining the structure of research in terms of the historical context.
A deductive logic offers only an instrument for the rational recvonstruction of completed research programs. The concept of dialectic can provide an instrument for analyzing both the relations between successive theories and the actual research process because it is concerned with analyzin scientific thought in terms of the intellectual tools used by the scientist.
The study of scientific discovery examines scientific research from the point of view of the practicing researcher; the study of scientific development looks back over the history of science and examines the relations between successive theories.
Scientists sought to use the accepted theory in an area to which it ought to apply and found a conflict between theory and observation. Again the response was to offer an hypothesis, one completely in accord with the accepted theory.
While in some cases a fundamental discovery involves a wholly new hypothesis is by no means a necessary feature of a scientific revolution. Poincare: Absolute space , absolute time, even Euclidian geometry, are not conditions to be imposed on mechanics; one can express the facts connecting them in terms of non-Euclidian space.
No one has seriously maintained that new theories are deduced from old ones, and the view that new theories are generalizations of old theories willnot stand up.
A dialectical change from one theory to another is a reorganization of the strands of the theoretical web, along with the removal of some strands and the addition of others; this reorganizationaccounts for the changes in the menaing of scientific concepts and observations associated with a scientific revolution, since both the concepts and the data of a theory derive their meaning from their location in the theoretical web. But theory change takes place within definite problem situations.
The thesis that if we are to have a rational basis for choosing between rival theories at a perticular moment in the history of science, we must have some standart to appeal to which is accepted by proponenets of both theories, and the thesis that if scientific change is to be rationalthere must be some eternal standart against which we can compare any theories.
The appeal to observations and to emprical laws cannot be made independently of theory.
The choice between scientific theories does not takes place by appeal to eternal standarts, established by philosophers, but rather by appeal to scientific standarts which are provided by the theories involved.
It being assumed that beliefs can be either true or false while knowledge can only be true.
There is no clear, simple relation between the results of experiment or observation and scientific theories. Aristotle: Science is deductive demonstration of necessary truths from premises which are themselves both necessarily true and knowmn to be true. Brown: His proposal is to take man of practical wisdom as a model of the maker of crucial scientific decisions which cannot be made by appeal to an algorithm, and he offes the making of these decisions as a model of rational thougth. It is the trained scientist who must make these decisions, and it is the scientists, not the rules they wield, that provide the locus of scientific rationality.
It is the ability to decide how an exceptional case should be handled that is charecteristic of rationality.
The results of logic and experiment must themselves be evaluated. It is the task of the skilled scientist to carry out this evaluation, and such evaluations furnish paradigm cases of rationality.
Rather than declare the process of scientific discovery ot be irrational, we should consider the scientist actively seeking the solution of a problem as providing another paradigm case of rational thought.
A theory arived at rationally is not sufficient to make it a part of the body of science; that requires not an individual but a group decision. No thesis becomes a part of the body of scientific knowledge unless it has been put before and accepted by the community of scientists who make up the relevant discipline. Kuhn: It is the consensus of the workers in a discipline that determines what constitutes knowledge in this discipline, but the group may later discover that it made a mistake. The group is no more infallible than the individual. Brown: The model of the man of pracytical wisdom applies to group as well as to individual decisions: these are judgements made on the basis of information and experience, but without the benefit of necessary truths or algorithmic procedures which can quarantee a decision immune from being overtuned by a future research.
The scientific community has been notebly self-correcting.
Scientific knowledge at any period consists of a number of elements: the fundamental theories which guide research, and with them the body of laws, fundamental constants, and observations that are of particular significance in the light of the guiding theory.
Scientific knowledge in any era is what the scientists actively take as such, and the scientific knowledge of one era may be rejected as error in the next. but the rejection of previously accepted claims will itself be made on the basis of the currently accepted views, which are themselves falible.
Unless scientists have an effective method for determinig once and for all which propositions are true, we cannot datermine which part of currently accepted science is indeed knowledge, nor even whether there exists any scientific knowledge at all.
We free ourselves from the belief that science can establish final truths, and accept instead that the best that science can hope to attain is tentative rational consensus on the basis of available evidence.
Truth1: 'Absolute' or 'ultimate' sense of the term, in hwich that a theory is true to assert that it gives a correct description of an aspect of reality. Dentes the goal for which scientists strive in constructing theories, but it has no relevance for the evaluation of theories since theories provide the only access we have to reality.
Truth2: Whatever is known must be true. Any proposition which is a part of the body of scientific knowledge is a true2 proposition.
A proposition is false1 if and only if it does not provide an adequate desciption of reality; it is false2 then when it is rejected by the current consensus.
Scientists are trying to understand a reality which is objective in the sense that it exists independently of their theories.
A consensus can be influenced or even forced by social, economic or political factors, e.g. the dominance of a particular ondividual or school, the availability of funds for a particular kind of research, or the prohibition of some kinds of research by a government or a powerful religious organization.
It would seem impossible to distinguish a ligitimately established consensus from an illegitimate one.
