Philosophy of Science & medicine

PHILOSOPHY OF SCIENCE The concepts and theories of every specialised body of knowledge and practices prevalent at any time in a culture are always rooted in the prevalent philosophy. Further, these also reflect the changes, conflicts and compromises in the philosophical paradigms. Therefore, to understand the concepts of modern scientific medicine, it is necessary to explore the developments in western philosophy that influenced and shaped them from time to time. For understanding the concepts of any specialised body of knowledge and practice, it is essential to understand the epistemological and ontological positions implicit or explicit in that system. Empiricism is an epistemological position and realism is an ontological position. In the concepts, theories and practices of modern science, both empiricist and realist trends can be discerned. Since modern medicine is also a scientific discipline, this is true for its concepts, theories and practice also. Before examining the impact of philosophy on modern medicine, it would be worthwhile to discuss briefly the ontological and epistemological questions and problems as they concern the science in general. PHILOSOPHICAL PROBLEMS OF EMPIRICISM AND REALISM Ontology is the theory of being and ontological questions are concerned with existence, reality and the true nature of things. Realism is the ontological position that the external world is real and it ‘exists’ independent of the human observer. This position also takes the view that the ‘really existing’ objects, structures and mechanisms of that world stimulate the sense perceptions of human observer. On the other hand, anti-realist position professes complete agnosticism as regard to ontological questions. Epistemology is the theory of knowledge and epistemological questions are concerned with what can be known about the world and methods of knowledge. Empiricism is the epistemological position that states that ultimately all the knowledge stems from the sense experience. On the other hand, rationalism is the position that, besides sense experience, reason is a source of knowledge. Empiricism on the epistemological level is usually associated with anti-realism at ontological level. On the other hand, justification of realism on the ontological level requires rationalism on the epistemological level. The rise of empiricist philosophy in Europe in 17th and 18th century was result of the changed intellectual climate following the 'scientific' achievements of men like Kepler, Galileo, Harvey, Newton etc. This school of philosophy is largely associated with philosophers like John Locke, George Berkeley and Auguste Comte. In the 20th century, the logical positivists of Vienna Circle and their successors analysed the logical consequences of empiricist thinking. Though there is disagreement on several important points, all the empiricists share the fundamental belief that all knowledge is derived from experience. Empiricists, although accept the laws of logic and mathematics, assert that logical and mathematical deductions are analytic which means that they do not generate new knowledge, but serve to analyse the existing knowledge. Further, anti-realism of empiricists makes it necessary to redefine the concepts like causality and objectivity in such a way that they do not presuppose any reality beyond observation. The ontological scepticism of empiricists, though seemingly counter-intuitive, is quite acceptable in advanced science. Scientists are generally concerned with the study of natural phenomena, which are not directly observable, by the use of ingenious instruments. Medical scientists have postulated the existence of specific receptors on the surface of brain cells based on sophisticated experiments. It is difficult is say whether such entities 'really' exist or they are creations of scientists' imagination that only serve to organise their ideas. Even medical practitioners depend on sense-extending instruments (e.g. X-ray equipment), detecting instruments (e.g. electrocardiograph) and complicated chemical analysis. They create a picture of reality by such indirect observations, which they can not perceive directly. Empiricists do not accept the generative theory of causality that implies that causal relation takes place in the external world independently of our observations. They propose the alternative succession (Humean) theory of causality in which idea of causality is explained in psychological terms. This theory points out that if we observe one event followed by second event several times, our expectation of the occurrence of second event after the occurrence of first event increases. The causal relationship is this expectation, a mental habit, which we erroneously extrapolate to an external world. The succession theory may be refined by analysing in detail the logical relationship between cause and effects. The usefulness of such analysis can be shown in discussion of the causation of disease. From the point of view of empiricism, the laws of nature do not tell us what really goes on in the world. Their function is the 'mnemonic reproduction of facts in the mind' i.e. these laws are only mental constructions that serve to describe as concisely as possible, the observations made. The term objectivity is used by empiricists in a very restricted sense. The term has ontological implications and refers to phenomenon believed to exist independently of the observer. Since empiricist position denies this interpretation, empiricists equate objectivity with inter-subjectivity. They assert that if exactly same observation is made by two different observers then the observation statement is said to be objective meaning that it is inter-subjective, public or verifiable. This assertion lead logical positivists of 20th century to formulate a demarcation criterion, which may serve to distinguish those propositions, which are meaningful from those that are not. They proposed a criterion of verifiability according to which only those statements are meaningful, which in principle could be labelled as verifiable. However, this criterion leads to quite serious consequences and difficulties: 1. According to this criterion, all attempts to discuss moral issues rationally are considered futile. This nihilistic attitude to moral philosophy is called emotivism. 2. The criterion itself is not verifiable and, therefore, must be regarded as meaningless by empiricist standards themselves. Thus, the criterion is self-defeating. 3. The criterion leads to the classical problem of induction since the criterion may well be applicable to singular statements. However, science is generally concerned with general statements, which can not be verified with absolute certainty by experience alone. The empiricism asserts that scientific process starts with observation. However, the leap from experience, which always consists of singular observations, to a 'law of nature' can not be logically defended as it can never be excluded that the next observation will prove the law to be false. This logical problem has always vexed empiricist philosophers and despite much effort by logical positivists in the last century, it has yet not been possible to logically define the jump from singular observation statements to general theory. Bertrand Russell has simply concluded that 'induction is an independent logical principle incapable of being inferred either from experience or from other logical principles, and that without this principle science is impossible'. Karl Popper has been the most influential philosopher of science of this century. He has made very important point that observation is always selective. It needs a chosen object, a definite task, an interest, a point of view, a problem. He asserts that the first step in scientific process is not observation but the generation of hypothesis, which may then be tested critically by observations and experiments. The goal of scientific effort is not verification but falsification of the initial hypothesis. As empiricists use the criterion of verifiability to distinguish between meaningful and non-meaningful statements, Popper used the criterion of falsifiability to distinguish between those theories that fall within province of science and those that may be labelled as pseudoscience. This thought and the criterion of falsifiability has influenced the modern scientific methodology to a very large extent. However, the criterion of falsifiability, despite its importance from the logical point of view, is not easy to handle in practice. Popper himself has pointed out that there is no such thing as pure observation. Therefore, a scientist who conducts an experiment and contradicts a theory can not be sure whether the theory has been falsified or the observation (or the experimental set-up) was at fault. Further, the biological and medical scientists mostly have to subject their observations to a statistical analysis. Here the final analysis depends not only on the observations but also on the convention as regards the choice of statistical tests and the accepted level of significance. Thus, Popper instead of solving the problem of induction, in a sense, has tried to bypass the problem. The scientists undoubtedly in many cases try to falsify hypotheses but in many other cases, they try to reason inductively. Medical research papers also contain numerous statistical calculations that reflect inductive reasoning. Despite the importance of empiricist philosophers' teaching regarding the importance of empirical evidence, it is obvious that from a purely philosophical point of view the radical empiricism leads to a dead end and creates insurmountable problems. It imposes a view of the world that is quite counter-intuitive. The empiricist belief that all knowledge is derived solely from the experience has not been established up till now to be fully consistent. The empiricist denouncement of realism has been unsuccessful in clarifying many of the philosophical problems. This forced inquiry along the lines of weakening the empiricist position. Immanuel Kant did not accept the Locke's idea that mind was originally like a blank sheet of paper. According to him, space and time are preconditions for the perception of something as an object and, therefore, human beings are 'programmed' to think in categories of quantity, quality, casualty, possibility, necessity, existence etc. A human being's picture of the actual world reflects this a priori organisation of his sense perceptions and his actual observations. Kant, is thus, a rationalist asserting that the empirical knowledge is organised according to a priori principles but, like empiricist, he does not accept the possibility of knowledge of things-in-themselves. Contemporary philosophers and psychologists have also given up the idea that the mind was originally a 'blank sheet of paper'. They also deny the existence of anything like pure observation. In medical examples, theory-dependence of observations can quite obviously be demonstrated. However, the role of empiricist philosophy in the development of science, including the scientific medicine, should be viewed in historical perspective. The metaphysical theories of pre-empiricist era were far too extravagantly rationalist and realistic. Therefore, it was only natural that empiricists felt that science must start afresh with systematic observations and establishment of the laws of nature. Popper has asserted that we can never be quite sure about our theories being true. However, the efficacy of modern technology, which is based to some extent on scientific theories, proves beyond doubt that in some areas Popper's 'approximation to truth' has been achieved. It is quite true that the development of modern science is inextricably bound up with empiricist philosophy, but philosophical reflections and the results of modern science suggest that the position of classical empiricism is quite untenable. It is more appropriate to accept the: ¨ realist point of view on the ontological level that the purpose of science is to explore what really goes on in the world and on the epistemological level, accept existence of pure observations but deny that the observations are the only source of knowledge, ¨ existence of causal relationships and deny the empiricist succession theory of causation, ¨ view that if one event generates another event through some or the other mechanism (as is quite obvious in some bio-medical situations) then a causal relationship between the two events exists and then these events may be said to follow the laws of nature and ¨ the view that causal relationships and laws of nature are objective in the sense that they exist independently of observation. Though the realist position does not solve the problem of induction, the knowledge of underlying mechanism sometimes makes the problem less troublesome. For example, the general statement that human granulocytes contain glycogen synthetase is not only the result of inductive inference, it is also based on existing theoretical knowledge. The philosophy of science has been dominated by empiricist thinking for the last few centuries, but during the last few decades, many philosophers (Smart, J.J.C., 1963; Harre., R., 1970; Hacking, I., 1983; Bhaskar,R., 1975) have favoured a realist theory of science which points out that scientific knowledge has both a transitive and an intransitive aspect. Knowledge in the form of a scientific theory should be regarded as a changeable social product, and as such, it is transitive. However, the object of that knowledge that does not depend upon the existence of observer, is intransitive. The importance of this observation is that it reveals the deficiency of different views of science including the medical science. Extreme realists disregard the transitive aspect of scientific knowledge while extreme empiricists disregard the intransitive aspect. It may be proposed that a balanced philosophy of natural science must take into account both the aspects and the relationship between them.

ONTOLOGICAL AND EPISTEMOLOGICAL POSITIONS OF MODERN SCIENTIFIC MEDICINE EMPIRICISM AND REALISM IN MODERN MEDICAL THOUGHT Philosophers of ancient Greece and medieval Europe were largely concerned with speculations about the nature of world, purpose of life etc. The Greek and medieval European philosophers were largely realists. In keeping with such philosophical atmosphere, medical theory in the western world up to two or three centuries ago, was to a large extent based on mere speculation. The philosophical position of Hippocratic humoural pathological theory describing four 'humours' (blood, phlegm, yellow bile and black bile) with different qualities and other similar theories may be described as speculative realism at ontological level and extreme rationalism at epistemological level. From realist position, these systems believed in the existence of an underlying real disease mechanism but being rationalists to an extravagant degree, they also believed that the nature of underlying disease mechanism can be ascertained by pure reasoning alone. It can be pointed out here that the philosophical position of various traditional/folk e.g. Ayurved, Unani) and newer unorthodox (Homeopathy) therapeutic systems can also be shown to be speculative realism and rationalism in varying degrees. Under the influence of empiricism the philosophers and scientists developed a sceptical attitude towards ontological inquiries and began to assert that one must rely entirely on one's senses. Empiricist demand for careful observation had some effect on medical thinking as early as 17th century. Thomas Sydenham stressed the importance of clinical observations at the bedside and described a number of disease entities based on their clinical manifestations. In the 18th century, following Linnaeus, the botanists who were also physicians, attempted to classify diseases like plants. These disease classifications were inspired by empiricist thinking as no attention was paid to the mechanism of disease. Being no more than divisions and subdivisions of ill-defined symptoms, such classifications of diseases had no lasting effect on the development of modern scientific medicine A few decades later, medical scientists developed an interest in human physiology. This made them see physiological disturbances where before this, they had only seen anatomical disturbances. This new mode of observation and consequent thought process had profound effect on the disease classification. The interest in physiology lead to establishment of new disease entities also. During the second half of last century, foundations of modern microbiology were also laid down. Identification of infective agent of disease had very significant effect on the philosophy of disease classification for two reasons. Firstly, it became possible for the first time to define a large number of disease entities aetiologically and, secondly, the dogma of universality i.e. one cause results in one disease was further established. However, this dogma can be shown fallacious in case of most of the infectious diseases, as the bacterium or the virus must usually be regarded as one of the many non-redundant causal factors in the individual patient. The disease classification is still largely a mixture of disease entities defined in anatomical, physiological and microbiological terms. In addition, a number of immunologically defined disease entities are also being established as a result of increasing interest in immunology. The anatomists, physiologists, microbiologists and immunologists work under the influence of empiricist philosophy as far as they pay great attention to their empirical observations in the laboratory. However, they are unknowingly following the realist philosophical position as they try to discover the underlying mechanism of disease forgetting or ignoring that true empiricist position denies this possibility. This is the underlying but ignored or not clearly understood schism between a realist and empiricist trends in modern scientific medical thinking still persisting today which had historically started in 1830s and 1840s. A highly significant analysis of the medical problems from the empiricist point of view was given by Gavaret (Gavaret, 1840). Gavaret's ideas were premature for his time and were not given much attention. However, a century later, in 1940s, time was ripe for the revival of his ideas in the form of new critical clinical school. At that time, the combination of new scientific knowledge and new technology resulted in explosive growth of new therapeutic and diagnostic methods which had to be tested critically in accordance with Gavaret's tenets which may be summarised as following: 1. The Laws of Logic are insufficient for judgement of the effect of a given treatment for a given disease. These are also insufficient for the ranking of those treatments that are recommended for the same disease according to the degree of their effect. 2. Only the Law of Large Numbers, which is applicable to therapeutic investigations, can solve these two important problems. 3. The death rate obtained by statistical calculations is never a precise and correct expression of the effect of the treatment in question, but larger the number of observations, closer the approximation. 4. A Therapeutic Law, obtained by comparing small number of facts may be so far from truth that it can never be trusted. 5. A Therapeutic Law is never absolute, but must always be expressed as a range. The range decreases as the number of observations increases and its width can be determined by those numbers on which the statistical calculations are based. 6. For determining the preference between two therapeutic methods, the results of preferred method must not only be better but the difference must also exceed a certain limit which depends on the number of cases. 7. When the number of investigations is large and the difference is low which does not reach this limit, it can be ignored as non-existent. 8. The same rules and same conclusions are applicable to the solution of problems connected with the theory of epidemic constitutions. 9. The same rules must be followed to ascertain whether the death rate in a disease varies according to age, sex and location. 10. In investigations of aetiological problems, the Law of Large Numbers only serves to prove the presence or absence of a presumed cause, regardless of its nature. The cause should be determined itself by means of considerations of a different order. This last question is outside the domain of statistics. The first nine tenets show radical empiricism while in the tenth tenet Gavaret suddenly retreats from this position. Empiricist theme is introduced in first three tenets i.e. doctors should study as many patients as possible and must not base their therapeutic decisions on speculative theories and logical deduction. In other words, they must discard theory and rely on positive facts. Tenets 4 and 5 clearly indicate the empiricist view of law of nature i.e. the demonstration of a law of nature is no more than the demonstration of a regular association between observed phenomena. Tenets 1 to 9 are almost similar to the recommendations of present day statisticians indicating that Gavaret was more than a century ahead of his times. It may also be pointed out that in 10 tenet, Gavaret points out the limitation of statistical approach and points out that statistics only serve to establish the presence or absence of a causal factor, regardless of its nature. He also seems to disapprove of the empiricist view of causality and assumes that a causal relationship is more than a regular succession of events and there is room for 'considerations of a different order'. Vast majority of Gavaret's contemporary medical practitioners, who were interested in science, ignored his views. They continued to explore the mechanisms of disease in the hope that the accumulation of enough such knowledge would enable them to deduce the correct treatment in the individual patient by means of the 'Laws of Logic'. It is thus clear that the vast majority of doctors at that time were realists but they accepted and applied the empiricist tenets of observation and experimentation.

CURRENT MEDICAL THINKING The contemporary medical thinking is also pervaded by both realist and empiricist approaches. For example, a modern lecture on a particular disease describes the observed symptoms or signs of the disease and gives a theoretical account of the underlying disease mechanism that fits a multitude of carefully executed experiments. It then concludes the logical treatment of that particular disease. It may be pointed out that for a disease, results of individual investigation are usually numerous figures that have been subjected to statistical analysis following the empiricist approach. However, these statistical truths are not regarded as an end in themselves. Rather, these are thought of as only a means to the establishment of a coherent theory, which is believed to reflect the reality behind the observations. This view of science has been labelled realism under empirical control (Wulff, Pedersen & Rosenberg, 1986). This current view differs markedly from that of radical empiricists like Gavaret and his modern followers whose sole aim is to formulate the statistical 'laws of nature' that may serve to predict future occurrences. Thus, the current medical thinking agrees fully with realist approach that medical knowledge largely depends on acquisition of more knowledge about the functions of human body in health and disease. However, it also believes in the great need of empiricist statistical approach in such matter. In the science in general, this mixture of realism and empiricism has lead to the recognition of distinction between pure science, technology and technique. In the modern scientific medicine also similar distinction between biological medicine (the scientific level), clinical medicine (the technological level) and clinical practice (the technical level) has been recognised. Members of the medical profession engaged in the biological medicine (e.g. basic research in the laboratory) are medical biologists. Those engaged in clinical medicine (e.g. testing of new drugs) are clinical researchers while those engaged in clinical practice (e.g. examination and treatment of individual patients) are medical practitioners. Medical biologists, just like physicists, seek true knowledge and their philosophical position is that of realism under empirical control. Though realism underlies the biological medicine, it should be pointed out that the biological variation exists at every level and this makes it impossible to make exact predictions in the individual case. The prognosis of the individual patient must always remain a statistical truth. Further, the course of a disease must usually be regarded as the result of a persistent interaction between the inner mechanics of body and the environment. Therefore, an exact prognosis requires extensive biological knowledge as well as full control of the stimuli from outside. There are numerous examples in the past of the introduction of wrong or worthless diagnostic and therapeutic methods due to unwarranted self-confidence on the part of biological scientists. These have clearly demonstrated that deductions from biological theory are often unreliable and that the techniques used in clinical practice must be tested empirically. The controlled therapeutic trial for the comparison of two treatments is the prototype of clinical research. The statistical result, which is a typical example of inductive reasoning, is what Gavaret calls a 'therapeutic law' and it serves a predictive purpose. Clinical researchers also test diagnostic methods according to similar principles. In this way, the philosophical position of clinical researchers may be described as empiricism on the methodological level. They do not follow empiricism in the original philosophical meaning as they do not accept the ontological agnosticism of Hume and the logical positivists. They fully accept that we know quite much about disease mechanisms but they emphasise the incompleteness of biological knowledge and impossibility of precise clinical predictions due to biological variation. The clinicians holding these views represent the critical clinical school, which revived the ideas of Gavaret in 1950s and 1960s. Further, it has been pointed out by many workers that clinical practice must not be regarded as applied biological medicine and it is necessary to adopt empirical approach for the solution of clinical problems. An important difference between realism under empirical control in biological medicine and methodological empiricism in clinical research should be noted. The statistical methods are also used in biological medicine but the end-product is always a theory considered to be a picture of reality, which may be true or false. On the other hand, the end-product of statistical methods in clinical medicine is always a statistical 'law of nature' which serves to predict future events. Thus, it may be pointed out that the empiricism as a philosophical position is untenable. Clearly, it is possible to reach some insight into the working of nature. However, the realist belief that medical theories will ever form a completely true and exhaustive picture of reality is also quite naive.