ok for my previous post - the essay i dunno the picture i wanted to attach isnt attaching first my comp was not letting me post now its not letting me attach pictures nonetheless. this is another very intresting article i found. it is not my journal entry it is just an intresting article related to the movie what the bleep do we know...

The rationality of science
The image that the scientific community likes to project itself is that of rationality par excellence. The scientific community sees itself as the very paradigm of rationality. It is claimed that the scientific method, provides a logic of justification which enables us to choose between competing theories objectively. This view of science perhaps originated in the recent success of physics. However this model of science has come under attack from philosophers of science, especially Khun and Feyerabend.
Models for the explanation of scientific change ( say from Newtonian mechanics to relativistic mechanics) can be divided into two classes. 1) rational model and 2) non rational model. Scientific endeavor can be rational only if scientific change can be explained rationally. Rationalists ( Popper, Lakatos) disagree among themselves about the goals of science ( explanatory or predictive?) and the principles of comparing theories.
Science will be a rational activity if a theory A is selected or preferred over B, solely for internal reasons- i.e. they relate only to the features of the theories and available evidence. By contrast non rationalists attribute scientific change (or non change) to psychological or sociological factors- factors not related to the theories themselves but to the scientists. These may be their propagandizing abilities or the social or political climate of the time. These explanations would be called external forces. An example of lack of change explained by external factors would be the acceptance of Newtonian Corpuscular theory of light over Young's wave model. Because of the hero worshipping of Newton at that time and due to certain mannerisms of Young and subsequent character assassination of him in Edinburgh Review, the better Young theory was not accepted at that time.
According to Khun and Feyerabend theories are incommensurable (nothing in common). For them the same words in competing theories have different meaning. For Khun the shift in meaning is so extreme that the concepts of one theory cannot be expressed in terms of the concepts of another theory. Fayerabend says that, proponents of incommensurable theories differently constitute the facts. In other words there are no facts independent of theories i.e. facts are theory dependent. This means that two competing theories cannot be rationally compared. If this assertion is true, how do we choose between theories? Is scientific activity and its method irrational?
Meaning of scientific terms
The meaning of a theoretical term in any theory is a function of the role that term plays in the theory. This role is demonstrated by 2 sets of postulates. In one set the term is linked to other theoretical terms in the theory( e'n is negatively charged). In the other set the term is linked to observational terms( e'n will be deflected by a magnetic field). These postulate sets are said to define the meaning of the term electron. Any change in these sets of postulates will ipso facto change the meaning of the theoretical term electron. Some took these meaning postulates to be true analytically. But the extreme positivist position that for any term to have a meaning, the meaning postulate should have a good amount of empirical content, made the scientists to abandon the meaning postulate approach and to replace it with the holistic conception for the meaning of a theoretical term. The meaning of a theoretical term was said to be determined by the entire set of sentences within the theory containing the terms. Hence if Newton and Einstein talk about mass, they will not disagree, but this is not because they agree. They are simply equivocating. On this account, the Newtonian mass is invariant and the relativist mass is not invariant are not logically incompatible as the meaning of mass is not constant across the theories.
This approach to the meaning of a term applies only to theoretical terms, for the observational terms were directly related to experience, and hence their meaning were invariant. Therefore the observational vocabulary is said to be theory free. That is Newton and Einstein could agree or disagree at purely observational level. This means that the 2 theories could be compared at the observational level. Thus if Einstein's theory entails an observation statement O and Newton's theory entailed the negation of O, then the 2 theories are genuinely in conflict.
But there was a complication. Popper, Khun, Feyerabend and others said that " all observation is theory laden". Under this thesis, there is no difference between theoretical terms and observational terms-both are defined in terms of the theory. Under this circumstance Newton and Einstein cannot even compare their observational terms. They mean different things by the term say mass.
Hence under the thesis that when a theory changes, the meaning of all terms changes, 2 theories will always be incommensurable and there cannot be any rational basis for choosing between theories.The realist view of science is that theories are true or false in virtue of how the world is and that the goal of science is to discover explanatory truths about the world. The problem with the above view is that there is a strong inductive evidence for a pessimistic induction, i.e. all physics theories have had their heyday and have been eventually rejected as false. So we can be almost certain that most of the current scientific theories would be discovered to be false say within 200 years. In other words we can never attain absolute truth about the real world. If this is the case then what is the good of science? May be the goal is getting nearer to the truth, i.e. saying that while the current theories are strictly speaking false, but yet we are getting nearer the truth. Popper calls this verisimilitude.
Observation, Theory, and Truth
Theorizing ( Einstein's relativity) leads to general statements and observation ( Eddington photographing field stars in the Hyades group during the solar eclipse of 1919) leads to particular statements. These different type of activities may go on simultaneously.
O/T dichotomy:
It appears that most of the observational statements contain some theoretical terms. For ex. a scientist on seeing a track on a photographic plate exposed in a cloud chamber may report the presence of an electron. But the absence of an O/T dichotomy can be problematical for the progress of science. For if any observational statement presuppose some theory, then how do we get evidence on the basis of observation for any theory? We will be accused of pulling ourselves up with our own theoretical boot straps. According to Newton Smith, there are no purely observational statements. O terms are also theory laden in the sense that no matter how corroborated by many observers, it may be revised by appealing to theory. He says that O, T statements are not different in kind but only in degree. He says that we must have confidence in low level O statements because they have served their purpose well, otherwise we will not be still existing to make judgments. Our confidence in these low level O statements are enough for us to choose between the theories.
Realism Vs Instrumentalism
According to realism scientific propositions are true so long as they correspond with reality. In other words true means true in virtue of how the world is independently of ourselves. For instrumentalists , scientific theories or hypotheses need not be true or even probable, so long as the theory provides a calculus consistent with the observation.( The church was not against Galileo using the heliocentric theory to calculate the orbits of the planets, but did not want him to propound that the heliocentric theory was true, i.e. the earth was really moving around the sun.) For instrumentalists facts are irrelevant- so long as it works.
Instrumentalists assume that theories have the status of instruments or tools or calculating devices in relation to observational statements. "........no question of the truth or reference of the theories themselves arise" (Hesse) The pessimistic induction does not worry an instrumentalist because he does not aim at anything more than an increase in predictive power. But science aim at more than prediction. We also want to explain. For instrumentalists theories are devoid of meaning and hence cannot be used to explain. They regard only singular statements involving only O terms as true or false. Since T terms are devoid of meaning, any sentence containing a T term cannot be true or false. So for an instrumentalist O/T dichotomy exists. If theoretical terms do not exist, then all entities postulated by theories do not exist. This implies that all theories are false. Instrumentalists use logic to predict O statements from theoretical entities. For example, if A and B is true then A is true. How can instrumentalists use standard logic based on the notion of truth preserving rules of inference if they do not attribute any truth value to their premises?
Relativism:
For the relativist, what is true depends on part or entirely on the social perspective of the scientist or on the theory. Accordingly, as time passes, or as society changes or from theory to theory what is true changes and not merely what is taken to be true. Is it possible that sentence "S" is true in @ and false in #. For ex. the earth was flat was true say 1000 years ago (@) and false now (#). This is not possible because the truth conditions for both statements should be the same- i.e. the conditions under which the proposition "earth is flat" should be the same as the conditions under which the proposition " earth is not flat" are true. To specify the truth conditions for a statement is to state what will make it true and what will make it false. If the truth conditions differ, then they are saying different things and hence they do not express the same proposition. Therefore a proposition p contained in a statement A in @ and by a statement B in # cannot be true in one case and false in another, provided the truth conditions are the same. For ex. if a sentence A in my language translates a sentence B in another language, I cannot accept that, these two sentences A and B differ in truth value unless I doubt the translation.
But we can certainly find cases where a sentence S of @ is translated into B of # (our language) and we are absolutely certain that B is true but speakers of @ claim that S is false. Given that the translation is correct then either they (@) are correct or we are correct. We cannot conclude under these circumstance that what is false for them is true for us. We can only say that we differ as to what is true. If this situation is widespread (specially about a lot of low level empirical statements) then we have to question the translation schema. This does not mean relativism is true. Surely many of our beliefs have varied from age to age, culture to culture and theory to theory. This cannot make us accept relativism. The fact that we can identify these diversities means we can also identify sameness across these boundaries i.e. sameness of truth conditions across these perspectives. Therefore the acknowledgment of this plurality of diverse belief systems across different perspectives means the acceptance of the invariance of truth across these perspectives.
Realism and Ontology:
According to realists, a theory must have an associated model in reality. But there are theories such as quantum mechanics and relativity which do not fit this picture. In these theories mathematical models have replaced descriptions of physical reality. How then can realists claim that a statement is true or false in virtue of how the world is? How will he know the reality if it is modeled mathematically, which is abstract? There is no reason to think that unfamiliar theoretical concepts should have an analogy with familiar objects. For there is no reason to think a priori that the unfamiliar must be like the familiar. Items at the subatomic level is a clear example of this class. Hence it may not always be possible to give a realist construction to all theories. (quantum mechanics is one example) Theories which have a realist construction are nice and easy to teach, but we must remember that it is not possible to have such theories in all domains of investigations. All we can say is that evidence of truth (or approximate truth) is evidence of whatever has to exist for the theory to be true. (not an analogical requirement that fits reality)
Popper: The falsificationist:
Truth is the aim of science for Popper. But we do not have any criterion for truth and hence we will never know when we have reached truth. Popper also is not prepared to accept the truth of most low level observational reports. He rejects the use of all inductive argumentation in the pursuit of truth. A scientist who has shown that when 1000 randomly selected sodium samples will give a yellow flame, he has no business to claim that the hypothesis that " a sodium salt will turn a flame yellow" is probably true. For him only deductive argument is permissible. So if just one piece of sodium salt fails to turn the flame yellow, Popper will throw out or reject the sodium hypothesis. The method of science according to Popper is not evidence gathering but rejection and falsification. This inability to attain truth leads Popper to the attainment of better approximation to truth or higher degree of verisimilitude. But as we shall see there are problems with Popper's concept of verisimilitude. How can he choose or what criteria would he use to choose a theory with a higher verisimilitude? Popper claims that a theory that is more corroborated will have higher verisimilitude. But to prove the above thesis we need inductive argument. But Popper has rejected all inductive arguments. Hence he has no basis to correlate corroboration with verisimilitude. What of his refutation? we will see that this is not how progress in science is accounted for. Refutation and rejection of a hypothesis is not the regular practice in a scientific community.
Popper's aim in science is the discovery of explanatory truths. He is also committed to the concept of verisimilitude according to which the history of a mature science is made up of a sequence of theories which are getting nearer to truth. But Popper repeatedly declares that there is no criterion of truth. We may posses true theories but we are unable to show that they are true. This means that there is no finite procedure which we could follow that would provide us with certain knowledge of the truth of a scientific theory. Popper's denial of a criterion of truth depends on his attack on O/T dichotomy. For him all observation is theory dependent. Hence he cannot believe in the truth of the statements the positivists would call observational. For him even a statement like "here is a glass of water" cannot be verified by any observational experience. In accepting this observational statement we are implicitly assuming some theory. On the other hand Popper wants us to accept what he calls basic statements. There are existential statements such as "there is a raven in that tree". Only theories that would negate a basic statement are scientific according to Popper.
We should examine why Popper rejects instances of generalizations(particular O statements) as providing positive evidence for these generalizations.
According to Bayer's formula where is the probability of the generalization g being true given evidence e and is the probability of e given g. Given that the evidence is an instance of generalization, = 1 and assuming that we are absolutely confident of the truth of e, . Therefore since we have , we have .
Hence no matter how many instances of generalizations we observe the probability of the generalization will remain where we started namely zero. Popper to overcome this difficulty completely rejects the inductive evidence and adopts a deductive strategy for scientific methodology. We have noted Popper's correct rejection of the view that observations could provide a criterion of truth and we have also seen his reasons for rejecting this view. Popper's central thesis is his assumption that a) David Hume has shown that no inductive argument is ever acceptable and that it is impossible to justify a law by observation or experiment since a universal law transcends experience and b) the principle of empiricism which asserts that in science only observation and experiments may decide upon the acceptance or rejection of scientific statements. Does not the views a) and b) clash with each other? No says Popper, because a law or theory is only accepted tentatively and so long as a theory stands up to the severest tests it is accepted and if it does not it is rejected. It is never inferred from empirical evidence. How correct or rational is this argument?
If all inductive arguments are precluded what is the method of science? According to Popper science progresses by bold conjectures and eliminating the false ones by rejecting them using empirical or experimental evidence. But is this the methodology of science? Lakatos once asked "do you know a scientist who wants to falsify his theory?" Popper's goal is truth, but he does not define the content of truth. He does not accept empirical evidence as evidence for truth. How then can we rationally pursue this goal (truth) when we have no criteria for recognizing that goal? Popper tries to overcome this problem by revising his goal. The goal is not truth itself but increasing verisimilitude. It means given a pair of rival theories, we ought to adopt the one that is nearer to truth than the other. But then how do we determine this?
The number of assertions of any theory is infinite. Also it will contain an infinite number of true premises and an infinite number of false assertions. Hence we cannot choose between theories (depending on greater verisimilitude) in terms of the number of truths and the number of falsehoods generated by the theory. What then is the rational guide in making judgments of verisimilitude? Popper says that we are to be guided by the degree of corroboration of a theory, where corroboration is "a concise report .......with respect to the way it solves its problems, its degree of testability, the severity of the tests it had undergone, and the way it had stood up to these tests." But for Popper corroboration is not forward looking, "Obviously the degree of corroboration at the time t ....says nothing about the future- for example about the degree of corroboration at a time later than t." Suppose there are two theories A and B and a test falsifies A and not B and if this test is the only test to date, does it mean that A has less verisimilitude than B? Popper does not give a positive answer and so within the Popperian framework corroboration is no indication of verisimilitude.
So far we have seen that in a Popperian deductivist falsification framework there is no correlation between corroboration and verisimilitude. Secondly , the Popperian view of science depends on basic statements, the acceptance of which cannot be rationally justified and thirdly the rejection of a theory because it is in conflict with a basic statement is not the actual practice of the scientific community
On the question of theories with greater verisimilitude, we may increase verisimilitude if the increase in truth content is not offset by an increase in falsity content. We subject the new stronger theory to tests, if we fail to refute it, or if a refutation of the new theory is also a refutation of the old theory, we conjecture says Popper that the new theory has a greater verisimilitude than the older one. This is a correct conclusion, but Popper cannot use this argument as it is a strong inductive argument. That is after sometime we conduct say 100 appropriately designed tests, all of which the theory has passed, we infer inductively that it is likely that the theory will pass other tests too. Popper is arguing an inductive correlation between corroboration and greater verisimilitude which runs like this:
1. Current theories have more verisimilitude than previous theories. (inductive)
2. Current theories have greater degree of corroboration than previous theories.
3. Thus corroboration is a sign of verisimilitude.
Accepting the above argument means accepting induction in the scientific method, the rejection of which is unique in Popper's methodology. Hence Popper the rejecter of inductivism cannot prove his point without using inductive arguments.
For Popper a theory is scientific if and only if it is falsifiable. That means it must entail some basic statement which could turn out to be false, thus falsifying the theory. The question is do theories entail basic statements? The answer is no. In order to derive a testable prediction from a theory we need to specify initial conditions together with a host of auxiliary hypothesis. Suppose we predict that P = ma for certain initial conditions and we did not get the predicted result, have we falsified the theory P = ma ? Certainly not. We have an anomaly ( to use Khun's term). We might have used defective measuring instruments, may be the initial conditions were not correctly specified or some extraneous factors might have intervened. Whatever it is we cannot conclude without further ado that our theory is false. It will be very difficult to falsify a theory that has something going for it. The discovery of a falsity content is not sufficient ground for rejecting a theory, as Feyerabend says "all theories are born falsified".
According to Popper criteria of refutation has to be laid down before hand. Will scientists ever lay down in advance what would lead them to reject or abandon their theories? Do physicists ever lay down conditions for rejecting quantum mechanics? As far as we know this is never done. The question whether to reject a theory in the face of an anomaly has to be decided on the basis of the availability of an alternate theory and on the basis of positive evidence for the theory in question. Hence to stress the goal of attempted refutation as strongly as Popper does is to give a distorted picture of science and in fact fails to fit the actual practice in science. Not withstanding these failures we have to give credit to Popper for his important insight that the goal of science is to increase verisimilitude.
Khun and the image of science:
According to Khun, between the ideology of science and the realities of scientific practice there falls a vast shadow. Khun denies that there is any ideal way with which actual practice could be compared. Khun explains his model in terms of his notion of a paradigm. Critics have accused that Khun uses this word so loosely that it nullifies his work. One critic claims 22 different meanings of the word paradigm in Khun's book " The structure of Scientific Revolution". By the word paradigm Khun means the common factors which are shared by a group of scientists who are working on a common subject. There are:
1) The basic theoretical assumptions held in common which are deployed without question. For ex. nuclear physicists might accept the quantum theory without question.
2) Agreement on models and analogies which might guide them in their research. For ex. treating heat as molecular motion.
3) Agreement on what constitutes a nice problem and what constitutes its solution, and consensus on the significant unsolved problems in a particular field.
The above notions are included in the word paradigm which Khun now prefers to call disciplinary matrix. A paradigm directs our attention not only to the theory but also to the wider range of beliefs, attitudes and procedures of the scientific community.
Normal science and Scientific revolutions:
A period of time during which a particular scientific community shares a paradigm is called normal science by Khun. During this time the community devotes its time to solving puzzles defined by this paradigm. During this time the faith in the accepted theories are so strong that anomalies are not treated as refutation but as problems to be solved. If a theory fails a test it is the scientist who is at fault not the theory. Just like a student in a lab test, when he gets a contradictory result, it is either his equipment or his readings that are at fault, not the theory. In time there may be many unsolved puzzles and working within a paradigm might be a constrain. When this happens alternative theoretical postulates might be articulated. Khun calls this scientific revolution during which time competing paradigms face each other and scientists are left without any principles to choose between them. Khun says that " as in political revolution, so in paradigm choice - there is no standard higher than the assent of the relevant community". This assent according to Khun is obtained by the propagandizing ability of the proponents of the rival paradigm. For according to him the " normal scientific paradigm that emerges from a scientific revolutions is not only incompatible but often incommensurable with that which has gone before".
Khun therefore presents an irrational picture of scientific progress where theory selection is explained by external forces such as propaganda, as competing theories are incommensurable. Khun gives Newtonian and Relativistic mechanics as examples of incommensurable theories. He likens the transition between paradigms to a religious conversion or a gestalt shift rather than to a rule governed investigation which justifies one paradigm against the other. Scientists decide which is a better theory by learning the tricks of the trade of the scientific fraternity. This view emerges from Khun's skepticism about induction, i. e we cannot accept an empirical claim as true evidence about the world. Regarding the selection of theories Khun remarks that " what better criterion than the decision of the scientific group could there be?" As his critics point out this seems to make theory choice a matter of mob psychology.

Paradigm shift and change in world view:
Khun argues that transition from one paradigm to another involves a change in world view. He says that scientists with a new paradigm sees differently from the way he had seen before. If Khun means by this, just a change in our basic beliefs / or attitudes, then it is trivial. For example when we adopt quantum mechanics, the belief in determinism has to be abandoned. But Khun means more than this. For him a paradigm shift may influence not only how we describe what we see, but will also bring about change in the objects we see. Such a radical change in view is not substantiated by our experience. For example, we studied Newtonian mechanics in school and when we adopt to the Relativistic mechanics, we do not alter our concept of reality or the world radically. The analogy of duck- rabbit given by Khun is too far fetched. He says that you can see only the duck or the rabbit but cannot see both simultaneously. What he wants us to believe by this analogy is that the two theories are incommensurable and depict different world views.
Do paradigm changes, change the way we look at the world, or does it bring about changes in the world itself? Khun says that " practicing in different worlds the two group of scientists see different things when they look from the same point in the same direction." Does this mean that the contents of the world change at the behest of our theories? Again quoting Khun: " This is not to say that they can see anything they please. Both are looking at the world and what they look at has not changed. But in some areas they see different things and they see them in different relations one to the other." This means that Khun is just saying what is obvious, that any theory change will make us look at different aspects of the world. The world itself does not change in response to the theoretical changes.
From Khun's writing it is clear that he is not a realist. He says that we need not assume the existence of an objective true account of nature to explain the success of science. There are no theory independent ways to describe reality. The notion of a match between the ontology of the theory and its real counterpart in nature is an illusion for Khan. Finally Khun behaves as a non rationalist when he says that his five values guide theory choice but these values cannot be justified. Later he says that these 5 values can be justified as the criteria to be used in achieving progress in science which is a rationalist behavior. Confused? Yes we are. Whatever Khun may be he is not a realist. For him progress in science is not related to truth or verisimilitude but to increase in problem solving ability- an instrumentalist perhaps!
Feyerabend, Against method:
According to Feyerabend there is no such thing as the scientific method. Science is just one tradition. The other traditions, he wants to see given equal treatment are astrology, traditional medicine and even witchcraft. He wants science to be removed from its privileged position in order to create a society in which all traditions are given equal access to power and education.
Most of the philosophers of science would agree that scientific discovery is a lawless province of intuition, inspiration, luck or unlucky hunches and guess work, but justification of scientific theories and theory selection is a legitimate rational and objective activity. But Feyerabend denies that there is a distinction between the discovery of scientific theories and their justification. In fact he denies that there is something called the scientific method.
He denies that there are rules which guide scientists in theory selection and says that adopting such rules and methodology would only impede scientific progress. In this respect Feyerabend is more radical than Khun, for Khun holds the view that there are rules commonly held by the scientific community (though unjustified), even though its application is problematic. Feyerabend rejects such attempts to compare theories objectively in terms of its content or verisimilitude on the grounds that opposing theories are incommensurable. He believes that scientific change can only be explained in terms of subjective preferences, propaganda, political beliefs etc.
Feyerabend’s attack on method is mainly based on the fact that throughout the history of scientific development, such methods or rules have been violated with impunity.
“The idea of a method that contains firm, unchanging and absolutely binding principles for conducting the business of science meets considerable difficulty when confronted with the results of historical research. We find then, that there is not a single rule however plausible, and however firmly grounded in epistemology, that is not violated at some time or other. It becomes evident that such violations are not accidental events, they are not results of insufficient knowledge or of inattention which might have been avoided.
Feyerabend hopes to undermine the rationalist’s faith in all rules of evidence by showing that any such rule has an equally plausible counter-rule the use of which would give opposite results. The first rule he considers is the consistency condition “which demands that new hypotheses agree with accepted theories”. According to Feyerabend we should proliferate theories at odds with accepted theories in order to improve our chances of discovering facts relevant to assessing the acceptability of the original theories: “variety of opinion is necessary for objective knowledge”. This is similar to Mill’s argument for freedom of opinion. Since we are infallible we should recognize that theories incompatible with our currently accepted theory may be true and consequently we should permit other theories to flourish. Even if our current theory is true, proliferation of rivals should be permitted and indeed encouraged. For we shall not fully appreciate the grounds in favor of our own theory unless we are forced to deal with objections to it. Mill also noted that a theory incompatible with our current theory while in error may contain “a portion of truth”, and since the general or prevailing opinion on any subject is rarely or never the whole truth, it is only by the collection of adverse opinions that the remainder of the truth has any chance of being supplied. Feyerabend has contended that he has not shown that proliferation should be encouraged but only that the rationalist cannot exclude it.
No doubt in the long run scientific progress requires that the scientific community contain some lunatics who receive some support. But can science flourish if every one was all the time trying to develop his own totally unique theory.
Inductive reasoning is such that something can be a good supporting reason for a hypotheses even though that hypotheses tuns out to be wrong. Thus it cannot count against a particular inductive rule that it has on some occasion led us to hold false beliefs. To abandon such inductive rule, it should have led us wrong more often than right. The way Feyerabend regards counter examples to a principle of comparison, indicates that he erroneously assumes that the rationalist is committed to believing in exceptions algorithmic principles of comparison. Of course there cannot be such an algorithm in science.
Even Feyerabend claims that he has no knowledge of what is good and bad in science. His theses is that anarchism helps to achieve progress in any of the senses one cares to choose. Even a law and order science will succeed only if anarchist moves are occasionally allowed to take place.
Feyerabend castigates the idea that there are objective facts to which we have access through observation and experiment.
“It is this historico-physiological character of evidence, the fact that it does not merely describe some objective state of affairs but also expresses some subjective, mythical and long forgotten views concerning the state of affairs, that forces us to take a fresh look at methodology......A straight forward and unqualified judgment of theories by “facts” is bound to eliminate ideas simply because they do not fit into the frame work of some older cosmology.”
Here Feyerabend is rejecting the idea that experience provides unproblematic evidence for the assessment of theories. For him the development of theories incompatible with our considered judgments about the observational facts will assist us in helping to improve those judgments through exposing untenable assumptions which may be implicit in these judgments.
Feyerabend like Khun believes in the incommensurability of competing theories. In such cases none of the so called rational methods, scientists use can be applied.
“We have a point of view (theory, framework, cosmos, mode of representation) whose elements (concepts ,facts, pictures) are built up in accordance with certain principles of construction....Let us call such principles universal principles of the theory in question. Suspending universal principles means suspending all facts and all concepts. Finally let us call a discovery, or a statement, or an attitude incommensurable with cosmos (theory) if it suspends some of its universal principles.
For example Newtonian mechanics is incommensurable with relativistic mechanics.
Finally for Feyerabend, science is but one ideology among many. The only constrain he advocates for an ideology is that it should be coherent.
“There is only one task we can legitimately demand of a theory, and it is that it should give us a correct account of the world, ie. of the totality of facts as constituted by its own basic concepts.”
Feyerabend thinks that we are attaching too much priority to science. He says that “the apostles of science were the more determined conquerors” and that they “materially suppressed the bearers of alternative cultures”.
• Popper's aim in science is the discovery of explanatory truths.
• He rejects the use of all inductive argumentation in the pursuit of truth.
• For him only deductive argument is permissible.
• The method of science according to Popper is not evidence gathering but rejection and falsification.
• Only theories that would negate a basic statement are scientific according to Popper.
• For Popper the goal is not truth itself but increasing verisimilitude.
• For Popper a theory is scientific if and only if it is falsifiable.
• We have to give credit to Popper for his important insight that the goal of science is to increase verisimilitude.
• According to Khun a paradigm or diciplinary matrix directs our attention not only to the theory but also to the wider range of beliefs, attitudes and procedures of the scientific community.
• A period of time during which a particular scientific community shares a paradigm is called normal science by Khun.
• According to Khun the " normal scientific paradigm that emerges from a scientific revolutions is not only incompatible but often incommensurable with that which has gone before".
• He likens the transition between paradigms to a religious conversion or a gestalt shift.
• For him a paradigm shift may influence not only how we describe what we see, but will also bring about change in the objects we see.
• Feyerabend denies that there is a distinction between the discovery of scientific theories and their justification. In fact he denies that there is something called the scientific method.
• He believes that scientific change can only be explained in terms of subjective preferences, propaganda, political beliefs etc.
• Feyerabend castigates the idea that there are objective facts to which we have access through observation and experiment.

Reference: Rationality of Science-Newton Smith