Chapter 4

Eugene Lashchyk, Scientific Revolutions, 1969

(C) Science Under Perpetual Revolutions

Another popular view of science has been put forth by P. K. Feyerabend.{17} Whereas the inductivist and the textbook views of science tended to stress the continuity in the development of science, Feyerabend presents science as a discontinuous endeavor. One of his central theses is that the terms of any two successive theories, within the same domain, will have undergone a change in meaning. Meaning transformations are intended to include not merely the theoretical but also the observational vocabulary. Such transformation in the meaning of scientific terms is the main reason why, according to Feyerabend, successive scientific theories are incompatible and incommensurable.{18}
On this point Feyerabend's views are to some extent in agreement with those of Kuhn. Both Kuhn and Feyerabend also claim that there is no pure observation language in terms of which we can arrive at a neutral and pre-theoretical description of nature. Thus, both Feyerabend and Kuhn hold that descriptions of facts are theory or paradigm laden.{19} But there are few philosophers at present who do not hold this thesis in one form or another.{20}
There are differences between Kuhn and Feyerabend's writings that are much more important and far reaching.{21}
1. Feyerabend holds, following Karl Popper, that theories and hypotheses are to be "abandoned as soon as a test does not produce the predicted result."{22} Kuhn criticizes both Popper's and Feyerabend's views on falsification{23} claiming that every theory is apparently falsified many times and yet scientists usually do not abandon a theory for this reason. Apparent counter examples (anomalies) or unsolved problems face most scientific theories at all times, but this fact only stimulates the scientist to work harder to remove such troublesome cases, among other ways by rechecking the experimental data.
2. Feyerabend holds, following Karl Popper, that philosophers can and should prescribe normative rules for scientists to follow even though these rules are not generalizations from scientific practice. (An example might be Theoretical Pluralism which Feyerabend claims is one rule which scientists should follow even though many have not followed it.)
Kuhn, on the other hand, would say that in the first place there is no scientific method peculiar to all the sciences and, secondly, if any method is evident it can be only stated on the basis of our observation of actual scientific practice.
A possible candidate for a kind of method in the sciences might be Kuhn's claim as to the developmental stages in the history of science:
(a) A number of theories compete with each other. One of these theories is selected because it shows great promise.
(b) This is the normal science stage -- a single theory dominating the field.
(c) With the proliferation of anomalies the paradigm enters the crisis and proliferation of theories stage.
(d) The crisis period ends with the revolutionary overthrow of the old theory and gradual entrenchment of the new theory.
This so-called method is common to other fields like art and music, besides science and therefore cannot strictly be called a scientific method.
Along these same lines, it can be said that Thomas Kuhn pays much closer attention to the actual operations of scientists and thus is for the most part descriptive. Feyerabend, on the other hand, is for the most part interested in prescribing how scientists ought to be conducting scientific research.
3. Feyerabend holds that the existence of "partly overlapping, mutually inconsistent and yet empirically adequate theories is not only possible but also required."{24} Thus scientists should have as part of their method development of theories inconsistent with each other.
Kuhn, on the other hand, views the simultaneous existence and belief in inconsistent theories in the same scientific specialty a temporary phenomenon. Existence of a plurality of theories which are inconsistent is a temporary stage which occurs during crisis periods in the life of a scientific community. Such a period has normally been followed by the triumph of one of these theories and what then follows is a very active period of scientific work dealing with perfecting the statement of the theory as well as improving theory-nature fit.
By far the most important difference between Kuhn and Feyerabend is number 3 above, the doctrine of theoretical pluralism. This normative rule of "Theoretical Pluralism," as Feyerabend calls it, goes to the very core of the difference between the two positions and because of this I would like to discuss it in much greater detail.
My aim in this chapter, if the reader recalls, is to discuss the various proposals for a definition and adequate description of the nature or essence of science. I think, I have shown that some versions of the inductivist and/or Baconian views of science are inadequate. Furthermore, I have argued that the textbook view of science has led many men astray particularly when it comes to the question of the correct description of the stages in the development of a science. It remains to be shown that Feyerabend's view, in my opinion, would place science in a constant state of crisis. I find that unacceptable.

The Partial True in Feyerabend's Philosophy of Theoretical Pluralism

Feyerabend argues that theoretical pluralism ought to be the posture of scientists if they are interested in providing theories which can have justifiable claims to objective knowledge.
He tries to show that theoretical pluralism ought to be not only the stance of a scientist but also the position of an adequate empiricism. Thus he states:

You can be a good empiricist only if you are prepared to work with many alternative theories rather than with a single point of view and "experience". This plurality of theories must not be regarded as a preliminary stage of knowledge which will at some time in the future be replaced by the one true theory. Theoretical pluralism is assumed to be an essential feature of all knowledge that claims to be objective.{25}

Feyerabend's reasons in support of theoretical pluralism can be broken down into the following:

1. It is an antidote against dogmatism -- the theological element.{26}
2. The view of T.P. is compatible with a humanitarian outlook.{27}
3. The worship of facts can be avoided with strict adherence to T.P.{28}
4. T.P. reduces tendency to create ad hoc hypotheses in order to handle falsifying instances of hypotheses and theories.{29}
5. T.P., which encourages the creation and development of many theories, will enable the scientist to criticize more adequately an accepted theory from the various viewpoints thus developed.{30}
6. T.P. enables scientists to uncover new and unsuspected phenomena or "facts."{31}
7. History of science supports T.P. rather than monism or dualism.{32}

Analyses of Above Arguments

1. In support of T.P. Feyerabend says that: "Also, the use of such a class (mutually incompatible and factually adequate theories) rather than of a single theory is a most potent antidote against dogmatism.{33}
That T.P. is a good antidote against dogmatism cannot be denied. What can be questioned, however, is whether the antidote, besides removing the illness in question, does not result in the overall weakness of the individual under treatment. There are times when insisting that one is right is foolhardy, particularly when evidence to the contrary is overwhelming. A classical example of such a foolhardy stance seems to be Priestley's insistence to the very end of his life of the correctness of the phlogiston theory.
But I think it fairly obvious that there also have been times when scientists and others have been stubbornly upholding a theory, again in the face of powerful evidence to the contrary, and it turned out that they were right. Their theory proved to be, in the long run, the correct theory -- the theory which best explained all the available data, etc. I take this to be a case when a person is truly courageous even though at first he appears foolhardy. A good example illustrating this case is Copernicus's insistence on the correctness of the heliocentric theory of the heavens. If Copernicus had capitulated in the face of criticisms presented by the followers of Ptolemy's theory, then we might not have seen the day when Galileo and Newton were able to reply to the arguments which ridiculed the heliocentric theory. It can be said, therefore, that dogmatism at times fosters the progress of science and at times obstructs it. So, we shouldn't discourage dogmatism. The partial truth in the dogmatic stance is that it doesn't make sense to opt for change in a conceptual scheme just for the sake of change. Conceptual schemes are tools which enable us to map our world and thus make life in it bearable because it is to some extent comprehensible.
Should we look for alternative theories when our present theory in some scientific specialty serves us well? There are enough unsettled beliefs which need to be settled without questioning those beliefs which are soundly established and provide us with workable hypotheses. To question all theories and hypotheses at all times by constantly developing alternatives is a procedure which goes contrary to a very basic demand of learning theory of all living organisms.
Alternatives should be searched for only in those cases where the dominant hypotheses and theories have continually been faced with problems which could not be solved or where counter examples have accumulated and continued attempts at removing them have resulted in failure. No precise rules can be given here, but let it suffice to say that since we cannot foresee which theories and hypotheses will prove fruitful in the future we cannot say in advance which dogmatic stances of scientists will prove to be useful and which will be detrimental. T.P. assumes that all dogmatic stances are bad and therefore alternatives should always be entertained seriously. I would say that alternatives should be entertained and developed only when the present theory or hypothesis is in trouble or has led us astray. What this means is difficult to spell out precisely but more will be said about it at a later time.
For the moment we can say that we do rot need an antidote against dogmatism per se because some instances of dogmatic stances have proved useful for science. Unless we can come up with an antidote for dogmatism which picks out the bad cases of dogmatism, it is better to leave science and scientists to themselves, at least until a more fruitful proposal is at hand.{34}

2. T. P. is Compatible with a Humanitarian Outlook.

Feyerabend states:

Unanimity of opinion may be fitting for a church, for the frightened victim of some (ancient or modern) myth, or for the weak and willing followers of some tyrant; variety of opinion is a feature necessary for objective knowledge; and a method that encourages variety is also the only method: that is compatible with a humanitarian outlook. To the extent to which the consistency conditions (and as will emerge, the condition of meaning invariance) delimits variety, it contains a theological element (which lies, of course, in the worship of "facts" so characteristic for nearly all empiricism).{35}

Since there are few cases in which we can be absolutely certain that we are right or that our belief is true (only in cases of tautology cal statements), we have to keep the possibility open that we might be mistaken. This is particularly true in scientific matters. It seems to me that such a respect for the opinions of others can be fostered among individuals in society without adopting T.P. as an epistemological position. T.P. is one epistemological position among many and people who adopt T.P. must in any case respect the views of others. A position of relativism can be also held dogmatically, and as a result no one view can claim to be uniquely consistent with a humanitarian outlook.

3. Worship of Facts Can Be Avoided

Feyerabend states:

It follows that, to the extent to which both principles . . . delimit variety and demand future theories to be consistent with theories already in existence, they contain a theological element which lies, of course, in the worship of "facts" so characteristic for nearly all empiricism.{36}

I am not disputing that the worship of facts will be undermined by the adoption of T.P., but that this is a good reason for the adoption of T.P. is doubtful.
It has become commonplace these days to claim that there is no independent class of facts which every adequate scientific theory has to take into account, or that there is no pure observation language and as a result any description will presuppose some theoretical point of view. But if the above premise be granted, it doesn't follow that experience or observation should not be esteemed or that, because there is a theoretical framework presupposed in observation, relevant "facts" should not be collected, categorized, and measured with precision. Adopting T.P. might undermine the worship or esteem of "facts," but talk which uses the term "fact" is loaded with ambiguity. There are at least two senses of "facts" currently employed in discussions:
(F1.) One group speaks of facts as if they were out there in the world ready to be plucked and collected. Under this view, there is talk of "brute facts" independent of any theoretical framework or of a pure observation language which can be utilized to report the given without any theoretical commitments.
(F2.) The other group, to which I belong, would like to deny that it is possible to provide descriptions of phenomena which are theoretically neutral. Under this view, to speak of facts is to speak of them as colored by some theoretical framework. But this need not lead to complete chaos in our epistemology. For control of theoretical frameworks is nevertheless possible, even though descriptions of "facts" are theory-laden.
We only have to see how data of "facts" in sense F2 have influenced scientists in modifying or completely rejecting a theory because that theory could not adequately account for the available facts (F2) or data. (See for example why chemists rejected phlogiston theory and adopted the oxygen theory of Lavoisier.){37}
Let us not therefore throw out the baby with the bath water. Adoption of T.P. will, I think, undermine not only our worship of facts in sense F1, but it might also undermine due respect for theory-laden date or "facts" in sense F2.

4. T.P. Reduces Invention of Ad Hoc Hypotheses

He argues that T.P. will reduce the tendency to invent ad hoc hypotheses which are designed to patch up loopholes in theories. This is not an obvious claim to make concerning T.P. Why should this be so? If I understand Feyerabend correctly, his line of reasoning seems to proceed in the following way. Let us imagine a very simplified case of an anomalous example for a law-like statement S1, "All copper conducts electricity," (x) (Cx --> Ex). Let us further imagine that a metal bar has been found in the U.S. resembling copper in all respects except that it does not conduct electricity, (Ca & Ea). Given such a situation, a scientist has at least two alternatives. (1) He can argue that this is not a case of copper because all copper conducts electricity by definition. Such a move preserves the law, S1, intact. (2) If he argues that this metal is copper, then this is a falsifying example of law S1. Normally this requires a radical modification of the law, but the following modification could be proposed for S1. "All copper conducts electricity in Canada, but not in the U.S." This modification could be said to be ad hoc, assuming that no theoretically sound reason was found for making such an arbitrary modification.
How would T.P. guard against making such ad hoc hypotheses? It seems to me that a scientist who subscribed to the philosophy of T.P. would be more receptive to alternative theories in which such an anomalous example would be explained, without making such specious restrictions as geographic location on properties of the electrical conductivity of metals.
The result of such a search for new theories and hypotheses whenever a candidate for an anomalous example comes up would mean that the scientist would be inclined to choose the second of the two alternatives described above. In other words, T.P. inclines the scientist to prefer theoretical revision to careful rechecking of his experimental data. This does not appear to be always advantageous in view of the fact that the history of science is full of apparent counter examples which were later removed after a more careful scrutiny of the experimental data had been made. Moreover, theory creation is a very time consuming activity and one should take recourse to such activity only as a last resort.
Forgetting such practical considerations, it can be said that even though T.P. reduces the tendency to create ad hoc hypotheses, it increases the possibility of creating a multiplicity of ad hoc theories. What does this mean? If one adopts T.P. as a methodological principle, there is a real possibility that scientists will not spend enough time and energy determining the adequacy of these theories. They will not push any one theory to its limits in order to see its real capabilities because they will not be seriously committed to any one theory and as a result there will be a tendency to drop a theory at the slightest sign of trouble. It is doubtful that scientists would spend years trying to solve some anomalies confronting a theory. They would, if committed to T.P., spend their time inventing new theories. The constant proliferation of new theories would lead to chaos and uncertainty. These theories would be ad hoc because there would be no initial theoretical justification of the creation of these radically new theories. Such a situation does arise at times in the history of science. To see this one needs to distinguish between basic or framework theories and theories of limited scope.
Whenever a basic theory is unable to handle a growing number of anomalous examples, and some of these might even involve what scientists have called crucial experiments which deal a death blow to the basic theory, such a series of results puts the basic theory in crisis. During this period, scientists consciously strive to develop alternative theories which will remove the anomalies that led the old theory into crisis. The adoption of the attitude of T.P. during such crisis periods is not only to be tolerated but it is an essential stage in the development of science. Creation of new and apparently ad hoc theories in such periods is then to be admissible.
But if T.P. is adopted as a principle for every stage of scientific development, this will lead to ad hoc theory creation at all times. Such a result is undesirable. When theories of limited scope need to be created because the older theory is in a state of crisis it must be emphasized that there ought to be a serious attempt made to create an alternative theory which is compatible with the framework or basic theory. Such a demand is dictated by the ideal of the unity of all science. Furthermore, there seems to be no obvious advantage to questioning all aspects of our conceptual systems at all times. Life is too short. Questioning some sub-theory of our conceptual system when it performs its service well is a wasteful procedure.{38} I think that alternatives should be developed only when a sensible need can clearly be shown to exist. Such needs for alternative theories, be they theories with narrow scopes of application or with universal application and, therefore, basic theories, arise when clear counter examples have been developed against a law or a theory has been dealt a death blow.

5. Empirical Adequacy of a Theory Can Be Claimed Only After It Has Been Confronted with Alternatives

Feyerabend's most important argument in support of T.P. consists in showing that the creation of a plurality of alternative theories will provide the scientist with a more adequate criticism of the dominant theory than just the comparison of the dominant theory with the facts.
In Feyerabend's own words:

The function of such concrete alternatives is however, this: they provide means of criticizing the accepted theory in a manner which goes beyond the criticism provided by the comparison of that theory with the facts: however closely a theory seems to reflect the facts, however universal its use, and however necessary its existence seems to be to those speaking the corresponding idiom, its factual adequacy can be asserted only after it has been confronted with alternatives whose invention and detailed development must therefore precede any final assertion of practical success and factual adequacy. This then is the methodological justification of the plurality of theories.{39}

The methodological justification of T.P. then seems to be the claim that factual adequacy of a theory cannot be claimed before the theory has been confronted with alternative theories. But such a claim is not intuitively obvious. A legitimate question remains: Why should it be true that a theory can be said to be adequate only after it has been confronted with alternative theories?
In order to find Feyerabend's above claim more plausible, it is necessary to point out that he is also committed to the thesis that facts are theory-laden. Because of this thesis, the classical conception of justifying a theory by showing that it conforms to the facts is no longer found to be sufficient, for the facts justifying a theory are also the ones that are ruled relevant and important by a theory. With such an addition, Feyerabend's argument becomes plausible and would be sound if one could not explicate adequately how theory- laden facts can be used to support the plausibility of a scientific theory. I believe that this can be done as I have shown previously in Chapter I, section (B). Let it suffice to say that at least a necessary condition of a satisfactory scientific theory is that it must explain most of the available phenomena. The fact that phenomena cannot be described independently of some conceptual framework or some set of categories does not invalidate the claim. One reason in support of this claim is that it is possible to formulate a plurality of specialized theories or belief systems under one basic conceptual scheme.{40}
A theory must explain available data or the factual base. Theories can be judged as adequate or not depending on how well they explain the relevant data. Why then does Feyerabend stress the creation of theoretical alternatives for the basis of comparison? I think that the prime reason is Feyerabend's belief that facts are theory-laden and consistency of a set of facts S1 with a theory T1 is no surprise and does not lend very strong evidence in support of T1. Only if T1 could exhibit how well it handles other sets of facts, say S2 which were discovered with the help of T2 would we then have stronger confirmation of T1. The rationale behind Feyerabend's T.P. seems to be the belief that the adequacy of a theory cannot be shown unless the theory in question, i.e. T1, is confronted with other sets of facts unearthed with the help of other theories T2, T3, . . ., Tn. The theory which best explains all the various sets of factual data S1, S2, S3 . . ., Sn will be declared a winner and only then can be declared the best theory from among the competing theories. The argument has some intuitive plausibility. One question that needs to be answered is whether this plausibility can be sustained in the context of Feyerabend's system.
I think that there is some doubt that the argument has been plausibly stated by Feyerabend. If facts are theory-laden and if theories which are inconsistent can unearth new facts, then such facts can, according to Feyerabend, either further confirm the dominant theory or weaken and possibly refute it. But how is this possible? Feyerabend, as we know, is committed to the thesis that theories (in this global sense){41} are incompatible and incommensurable. But if theories are incompatible and incommensurable, wouldn't the facts discovered with the help of these theories also have these properties? I think that Feyerabend is committed to the view that such various sets of facts would be incommensurable since such sets of facts are couched in their appropriate theoretical terminologies. But if such is the case, then a fact from set1 can't disconfirm any theory except T1. Other theories such as T2, T3, . . ., Tn. since incommensurable, cannot have a fact from T1 contradict them. Feyerabend cannot have it both ways. He cannot hold that theories are incommensurable and also that facts which were uncovered with the aid of a particular theory can contradict other theories. His conceptual system does not permit such a move. Arguments in support of T.P. which depend on factual adequacy thus lose their plausibility.
6. The sixth reason, to the effect that T.P. enables scientists to uncover new and unsuspected phenomena suffers for lack of a good example. He has yet to produce a clear example of the discovery of novel phenomena which were made possible by the development of alternative theories.

7. History of Science Supports T.P.

Let us turn next to an analysis of the seventh reason for the adoption of T.P., namely, that the history of science supports T.P. rather than monism or my dual aspect view, as will be elaborated later. Feyerabend states:

Now instead of further engaging in abstract argument, I would invite these opponents of pluralism to compare what I have said with the actual history of thought. Are not the history of heliocentrism and the history of atomizing splendid examples of the way in which antediluvian, preposterous ideas, ideas which are regarded with contempt and are ridiculed by the learned and by the common folk alike, may yet be turned against "modern" views and even succeed in overthrowing them?{42}

The examples from the history of science which Feyerabend cites above support the view that there are times when proliferation of theories is an essential part of scientific activity. Development of Copernican heliocentric theory as an alternative to Ptolemy's dominant geocentric astronomy clearly led to progress. But this is not the point of contention. What Feyerabend needs, to support his historical argument for T.P., is a more detailed historical analysis which would show that creation of alternatives was begun not because the older and more entrenched theory was in trouble, but because creation of alternatives is an activity in which scientists have been engaged throughout the history of science, even when the particular dominant theory was not facing a serious crisis. Concerning the example of Copernicus, Kuhn has shown that the ptolemaic earth-centered astronomy was approaching a crisis stage.{43} Ptolemaic theory did not agree with the observations of the astronomers of the time and, in order to bring the theory into a closer fit with the observations, ad hoc assumptions like the invention of complicated epicycles were necessary. Astronomers sensed that everything was not in order. Copernicus had enough foresight to realize that another approach which placed the sun as the center of the system enabled the explanation of this movement of the planets while making fewer theoretical assumptions. It reduced the number of ad hoc assumptions, in particular the number of epicycles, as well as the number of revolving spheres in which planets were presumably imbedded.
There is nothing that Feyerabend cites as historical support of T.P. which goes contrary to my modified kind of T.P.; i.e., the dual nature conception of sciences. I am not denying that there are times when proliferation of theoretical frameworks is essential for progress. All that I am contending is that this revolutionary process should continue no matter what the state of our dominant theory happens to be. My view opts for T.P. only when the dominant theory is in trouble. Feyerabend argues that the adoption of T.P. will lead to rejection of the dominant theory. Proliferation of theories, according to Feyerabend, leads to trouble for the dominant theory. The historical examples cited do not support the stronger theses that (1) T.P. should always be the position held, nor (2) that the creation of alternatives always precedes trouble for the basic theory. The historical examples only support the weaker claim with which I am in full agreement, that there are times in the development of a science that T.P. should be adopted. These periods in the development of science are broken up by periods of normal science when alternatives are not developed, but when there is feverish work to elaborate the dominant theory.
Feyerabend's reasons thus, do not compel us to adopt T.P. unqualified. Feyerabend has not shown that theoretical proliferation is essential to creating a crisis for the dominant theory. Putting it bluntly, it is not true that "Proliferation precedes trouble."{44} I would like to hold with Thomas Kuhn that trouble precedes proliferation. Crises arise from the internal workings of normal scientific research. It is the attempts to solve theoretical puzzles within normal science that have usually led theories into a crisis stage. A crisis arises when enough such puzzles remain unresolved after repeated attempts to solve them by the best scientific minds. An accumulation of such anomalies leads the dominant theory into the crisis stage. At times the particular theory that happens to be in crisis has served scientists well for hundreds of years. It still remains true to say that the research activity which the theory in question deemed appropriate was responsible for the crisis.
The above description I believe to be the case and not the description that Feyerabend provides. Theoretical pluralism is an acceptable position but only during such crisis periods and not for all theories at all times. This is the partial truth in Feyerabend's position; and I would like to suggest that this is the reason which has led some philosophers to adopt Feyerabend's view. The position is intellectually enticing because of the partial truth implicit in it. Once the reasonable part of T.P. is extricated and the unreasonable part exhibited, we need no longer call ourselves theoretical pluralists but scientific dualists. For the correct view, I believe, is the view which proposes that we strive for a unified scientific theory of nature and not for a splintered picture composed of incompatible theories. Such a candidate for a unified theory must be developed on various fronts. Theoreticians must develop the formulation of the theory so that it meets the highest mathematical standards of the time. Experimentalists must strive for greater precision in the statement of the factual base of the theory as well as the creative invention of new experimental confirmation of the dominant theory. When such work on a theory leads to anomalies -- discovery of phenomena or facts which contradict the theory rather than confirm it -- then the theory enters a crisis stage and the second stage of an adequate science emerges. During this stage there is the proliferation of theories. From these, one has to be chosen which will solve the anomalies that resulted in the downfall of the previous theory; and so the dual process continues. It is not true to say that science is undergoing perpetual revolutions. The truth lies somewhere in between the revolutionary view such as Popper's and Feyerabend's and the view that there is never a replacement of one scientific theory by another.

Table of Contents -- Next

Notes

{17} P. K. Feyerabend, "How to Be a Good Empiricist -- A Plea for Tolerance in Matters Epistemological," contained in Philosophy of Science, the Delaware Seminar, Vol. II, 1962-63, edited by Barnard Baumrin, pp. 3-41. Henceforth this article will be referred to as H.G.E. [Back]
{18} Feyerabend states:
"In section 5 it was shown that the 'inertial law' (8) of the impetus theory is incommensurable with Newtonian Physics in the sense that the main concept of the former, viz., the concept of impetus, can neither be defined on the basis of the primitive description terms of the latter, nor related to them via a correct empirical statement." ( Paul Feyerabend "Explanation, Reduction and Empiricism" in Minnesota Studies in the Philosophy of Science, Vol. III, Scientific Explanation, Space and Time ed. by Herbert Feigl and Grover Maxwell (Minneapolis: University of Minnesota Press, 1962), pp. 28-97, particularly p. 74. This article will be referred to as E.R.E. [Back]
{19} "Not only is the description of every single fact dependent on some theory . . ." ("Problems of Empiricism" in Beyond the Edge of Certainty, edited by R. Colodny (Englewood Cliffs, N.J. Prentice Hall, 1965), pp. 145-260, particularly p. 176. [Back]
{20} Among those who also subscribe to the claim that descriptions of facts are theory-laden see I. Scheffler, Science and Subjectivity, Chapter II; N. Hanson, Patterns of Discovery, Chapters 1 and 2. [Back]
{21} Shapere seems to overlook these differences. See D. Shapere, "Meaning and Scientific Change," pp. 66-67. [Back]
{22} E. R. E., p. 29. [Back]
{23} T. Kuhn, besides criticizing falsification of theories in SSR, does so explicitly in an article on Karl Popper's philosophy of science to be included in the forthcoming Schilpp volume on Karl Popper in the Library of Living Philosophers, "Logic of Discovery or Psychology of Research" (manuscript). [Back]
{24} E.R.E., p. 67. In describing the history of the philosophy of Theoretical Pluralism (from now referred to as "T.P.") Feyerabend states:
"The principle of proliferation is an essential part both of Popper's critical rationalism and of the philosophy of possiblism as it has been developed by Professor Arne Naess. It has been used, more or less consciously, by many scientists before. Boltzmann is one example (cf. the article Boltzmann in the International Encyclopedia of Philosophy, (ed.) Paul Edwards) shows us that the Presocratics were the first thinkers to proceed in accordance with it. (Paul Feyerabend "Reply to Criticism: Comments on Smart, Sellars, and Putnam in Boston Studies in the Philosophy of Science, Vol. II, edited by Robert S. Cohen and Marx W. Wartofsky (New York: Humanities Press, l965), pp. 223-261, particularly p. 253. From now on the above article will be referred to as "Comments". [Back]
{25} H.G.E., p. 6. [Back]
{26} Ibid., p. 30. See also E.R.R., p. 71. [Back]
{27} H.G.E., p. 30. [Back]
{28} Ibid., p. 30, Also E.R.E., p. 71. [Back]
{29} E.R.E., p. 49. [Back]
{30} H.G.E., p. 35 See also "Comments", p. 224. [Back]
{31} H.G.E., p. 22. [Back]
{32} E.R.E., p. 69. [Back]
{33} E.R.E., p. 69. [Back]
{34} By useful cases of dogmatism I mean those cases when a person insists that his beliefs are correct in the face of what appears to be overwhelming evidence against them. The main type of justification or explanation that will be admissible for this type of dogmatism will be the claim that in time evidence can be compiled which will justify the belief in some hypotheses or theories. Scientific progress would be greatly hampered should we not admit such kinds of dogmatisms. It should be mentioned that should such evidence not arrive, there will come a point where such dogmatism becomes useless and bad. See T. S. Kuhn "The Function of Dogmatism in Scientific Research" in Symposium on the History of Science, University of Oxford, July 9-15, 1961, edited by A. C. Crombie (Heineman Etucatienal Books, Ltd.). [Back]
{35} H.G.E., p. 30. [Back]
{36} E.R.E., p. 71. See also H.G.E., p. 30. [Back]
{37} Geerge Gale, "Phlogiston Revisited: Explanatory Models and Conceptual Change," Chemistry, XLI, 4 (April, 1968), 16-20.
See also J. B. Conant (ed.), "The Overthrow of the Phlogiston Theory" reprinted in Harvard Case Histories in Experimental Science, Vol. I (Cambridge, Massachusetts: Harvard University Press, 1957), pp. 65-115. [Back]
{38} SSR, p. 76. [Back]
{39} H.G.E., p. 6. In the following passage Feyerabend brings out another aspect of this argument. He states:
"One need not mention the psychological advantages flowing from the use of alternatives, the fact, that is, that a mind which is immersed in the contemplation of a single theory may not even notice its most striking weaknesses." ("Comments", p. 224).
One may retort to Feyerabend's argument above by paraphrasing it in the following way: A mind not immersed in the contemplation of a single theory may never appreciate the most important strengths of the theory and possibly never solve any problems or puzzles either. [Back]
{40} Israel Scheffler, Science and Subjectivity, Chapter II. [Back]
{41} For some time philosophers like Dudley Shapere and others speculated about what Feyerabend meant by theory. Ever since the publication of "Comments" we need wonder no longer. For by theory Feyerabend included just about everything.
"When speaking of theories I shall include myths, political ideas, religious systems, and I shall demand that a point of view so named be applicable to at least some aspects of everything there is. The general theory of relativity is a theory in this sense, 'all ravens are black' is not. . . .
I prefer this 'accordion' use of the term because it provides a single name for problems which in my presentation are intimately related."("Comments", p. 252) [Back]
{42} "Comments", p. 252. [Back]
{43} Thomas Kuhn, The Copernican Revolution: Planetary Astronomy in the Development of Western Thought (Cambridge, Massachusetts 1957). [Back]
{44} "Comments", p. 255. [Back]

Table of Contents -- Next