1

UNIT 3 HISTORICAL REALISM

Contents

3.0 Objectives
3.1 Introduction
3.2 Lakatos: Enriching Popper and Kuhn
3.3 Shapere: Transcending Classical Empiricism and Rationalism
3.4 Larry Laudan: Science - A Problem-Solving Enterprise
3.5 Let Us Sum Up
3.6 Key Words
3.7 Further Reading and References

3.0 OBJECTIVES

Some philosophers of science in the later decades of the 20th century made attempts to envisage a
picture of science and philosophy of science, which would be more comprehensive, by avoiding
the extremes of both Logical Positivism and Historicism. They pay due importance to the
historical, social and non-rational aspects of science, while maintaining the rational character and
a sort of realistic picture of science intact. These postpositivist thinkers can be categorized as
Historical Realist Philosophers of Science. This lesson seeks to familiarize the students with
some of the salient contributions of some of the philosophers of science belonging to the school
of Historical Realism.

3.1 INTRODUCTION

Some historians like, R. G. Collingwood, for whom “perception and history are identical” (1924,
204), argue for a realistic view of history. They tend to take ‘history’ as the exact description of
what happened in the past. But there are strong reactions against this view; history necessarily
involves interpretation of what, how and why something happened. It is never a clear description
of the past as it all depends on who writes history. Critiques like Goldstein argue that historians
cannot be taken to be as ‘constituting history’ (1976). In philosophy of science also the role of
history cannot be exaggerated, as history of science is not a real description of what happened in
the past. Historical Realism, the contemporary dominant school of philosophy of science, is said
to be a mid-way position between Logical Positivism (LP) and Historicism, in viewing the role
of history of science in envisaging science. The historical realists argue that both made the
mistake of going to the extremes: LP absolutized science, while Historicism went to the other
extreme of relativizing science and rendering it almost irrational. So the historical realists as
opposed to Logical Positivism say everything in science is changeable, as nothing is final and
absolute; however, against historicists, they show that these changes take place in a rational and
responsible manner, and therefore ‘anything’ cannot go in science. They understand rationality in
 2

science in a non-traditional way and new criteria for rationality emerge from the scientific
community. All these go to show the compulsory need and relevance of history of science to do
philosophy of science. We discuss some of the contributions of three philosophers of science of
this school: Lakatos, Laudan and Shapere.

3.2 IMRE LAKATOS (1924-74): ENRICHING POPPER AND KUHN

Lakatos, born in Hungary in 1922, lived in a very disturbed times of the World War II. Though
his mother and grandmother were murdered, he could escape. Pursuing a political career, by
1947, he became a powerful figure with the Hungarian Ministry of Education. He was
nevertheless imprisoned in 1950; later he escaped to Vienna in 1956; thence to Cambridge,
where he worked for his second doctorate under R.B. Braithwaite. Attracted by Popper’s ideas
he regularly attended his lectures. Lakatos’ famous work, Proofs and Refutations, a collection of
his articles in philosophy of mathematics, posthumously published in 1976. He challenges LP’s
claim of sharp distinction between the context of discovery and the context of justification. He
started teaching in England. The government did not grant him citizenship, and so he remained
technically ‘stateless’. In 1974 he had a sudden death of heart attack.

Scientific Research Program (SRP)

Science is made up of many Research Programs, involving a series of theories. Scientific

revolution consists in one research program superseding, not necessarily rejecting, the previous
one. There are series of theories which are related to one other like T1, T2, T3, T4.... Tn, where
each being an improvement upon the preceding one. Example: Research Program of Copernican
system. In 1543, Copernicus gave us the theory of heliocentrism. Though it was rather crude, the
basic idea of having the stationary sun at the center was correct. But other aspects like epicycles,
circular orbits of the planets were not correct; let this be COPl; Then came Kepler, who removed
the epicycles and replaced the circular orbits with the elliptical ones; but Kepler could not say
why planets move in ellipses; this we call COP2. Then Hookes explained the elliptical path with
centrifugal force and this can be COP3. Newton, with his mathematical maneuver, made it still
better and let this be COP4. Laplace made the theory even better, by explaining scientifically the
stability of the solar system, which Newton could not do; in fact, he postulated a plumbing God.
Laplace's theory can be COP5. Thus the Copernican system reached its betterment with the
contributions of many, each one improving the previous one. Science must be assessed in terms
of such SRP, not of individual theories (Popper and logical positivists hold that a theory or a law
is the basic unit of inquiry in science). SRP is much more comprehensive than a theory. (e.g.
Instead of looking at individual kings, like Karikalan, Raja Raja Cholan, etc. of the Chola period
of the ancient South India, we must take the whole of Chola period as one unit).

Two Parts of a SRP: i) Negative Heuristic or Hardcore - Every SRP has a core part, giving the
very identity of the programme. It is the metaphysical presuppositions of the programme.
Individual theories may be rejected but as long as the hard core is kept intact and the programme
keeps going. It is the non-negotiable part that cannot be compromised or changed. It is forbidden
to modify or reject it; since it forbids any change it is called negative heuristic. (Examples for the
hardcore: Sun-centredness in the Copernican system; the mechanical view of the world in the
Newtonian system). ii) Positive Heuristic or Protective Belt - Positive heuristic is the changeable
 3

part of the SRP that protects the hardcore from being challenged. It allows changes to keep the
hardcore intact. It consists of partially articulated sets of suggestions. It specifies the directions in
which the research must proceed. It directs us how to take care of the anomalies and modify the
research program. It also contains metaphysical principles. (E.g. Orbits of the planets, either
elliptical or circular in the Copernican System).

Evaluation of a SRP: It is evaluated in terms of the nature of the problemshift. It is progressive

if it is able to explain all the problems of the facts of the previous programme and to predict new
facts. It is perfectly rational to pursue the new programme even if any one of the conditions is
met with. A progressive research program is theoretically progressive if every successive theory
has some excess empirical content and more predictive power than its predecessor and it is
empirically progressive if some of its excessive empirical content is corroborated. Thus SRP
seems to give a rational picture of science because at least some of the predictions must be
established. A progressive research program is one, which is both theoretically and empirically
progressive. A problemshift is scientific if it is at least theoretically progressive, and if not, it will
be pseudoscientific. (Thus he does away with the falsification criterion to demarcate science
from non-science). A Degenerating programme would be not only not giving new facts but also
not explaining those of the previous programme. So it is not progressive both theoretically or
empirically.

Rejection of a SRP: SRP is never rejected in haste even if it is degenerating and it is only
shelved aside. According to LP, if a theory cannot be verified (for Popper, if not falsified!), it has
to be thrown away. But Lakatos argues that we can know whether a research program is
degenerating only by hindsight. For example, Newton initiated the particle theory of light. As a
genius, he also knew the wave theory but he realized that the wave theory was not tenable for
light. But by the crucial experiment conducted by Fizzau and Foucoult the particle theory was
rejected. Einstein resurrected it. Hence it is rational to work on degenerative problems instead of
rejecting it outright.

Proliferation of SRP: There is no possibility of inductively confirming a theory. There are lots
of problems in induction philosophically, though it works in actual practice. The hardcore of a
research program is may be false. We can know whether it is true or false only by inductively
conforming. But since induction has problems, no amount of testing theories of research program
can guarantee the validity of theory. It is irrational to assume that any particular research
program is true for ever. So the truth of a research program cannot be absolutely guaranteed. If
truth can be guaranteed, then it is enough to have only one research program. If not, then it is
good to have many research programs and by comparing them accept what is better. We can only
talk of comparative superiority not of absolute superiority of SRP.

Lakatos, Kuhn and Popper

Lakatos aims at a sort of a synthesis of Popper and Kuhn, as though grafting the revolutionary
ideas of Kuhn on the Popperian tree; Popper gives him, so to say, the frame work and Kuhn
gives him the contents. Going along with Kuhn, he saw that the basic unit of analysis in science
must be much broader than a single theory. Lakatos’ SRP is very similar to Kuhn's paradigm but
Lakatos demands many SRPs at a given time. Paradigm, for Kuhn, gives the worldview but for
 4

Lakatos, SRP does not give any worldview. Kuhn’s normal science period has a dogmatic nature;
the scientists in it cannot easily challenge the paradigm and try to absorb the anomalies into the
existing system. But SRP rules out such dogmatism, as Lakatos combines it with the Popperian
attitude of dropping a degenerating SRP, ‘under certain objectively defined conditions’. Thus,
the continuity in science in Kuhn’s scheme is ‘socio-psychological’, while in Lakatos’
framework it is ‘normative’ (Lakatos, 1978, 90). Further, he finds Kuhn’s account of theory
change (adoption of a new paradigm) to be not rational; further, it fails to distinguish actual
science from pseudo science. But Lakatos, with the notions of ‘progressive’ or ‘degenerating’
problemshift, tries to make the Kuhnian account more rational, which would also demarcate
science from non-science.

Lakatos called himself a disciple of Popper. In many ways he followed Popper; for instance, like
Popper, he also argued for proliferation of theories and comparative superiority of theories.
However, though Lakatos began as a popperian, slowly he moved on to critique Popper’s views.
His paper “Changes in the Problem of Inductive Logic’ (1968) was actually a defence of
Popper’s view of inductive logic, while “Popper on Demarcation and Induction” (1974) was a
serious critique of Popper’s solution of Hume’s problem. Popper and Logical Positivists saw an
unbridgeable gap between the ‘context of justification’ and the ‘context of discovery’. They all
insisted that science and logic must be concerned only with the context of justification not with
the context of discovery. Disagreeing with this, Lakatos proposes a new concept of philosophy of
science, which dilutes any such distinction.

The unit of inquiry for Lakatos is not a theory but a SRP. Popper rejected a theory if it was not
falsifiable, but Lakatos is not for any hasty rejection of SRP. Popper’s falsification was proved to
be untenable, as Duhem-Quine showed that any theory could survive falsification with ad-hoc
modifications. Lakatos proposes a modified falsification: A theory can be considered falsified
only when an alternative theory is proposed with the following features: the new theory has more
empirical content than the falsified one; it is able to predict more novel facts, which were
improbable or even forbidden by the first theory; it solves all the problems solved by the previous
theory; and it has more content that is corroborated (1970, 116). Therefore, unlike Popper,
Lakatos argues that a theory cannot be dropped even if it is falsified, unless there is an alternative
theory is available. A theory can be dropped, as pointed out above, only when the whole of SRP,
of which it is a component, becomes degenerative and gets rejected as a whole SRP.

Remarks:

i) Lakatos succeeds to a certain extent in integrating Kuhn’s ideas of Science as a social

enterprise with the Popper’s normative methodology. Like Kuhn, Lakatos is convinced that
science is always guided by a theoretical framework. But he makes his SRP normative, as Popper
makes falsification.

ii) In the framework of SRP, a theory becomes a better theory by superseding the previous one.
With this move towards betterment, Lakatos carefully avoids the issue of truth and verisimilitude
of Popper’s ideas.
 5

iii) By making problemshift as the criterion for the demarcation of science and pseudoscience,
Lakatos is not drawing a strict line between scientific and non-scientific theories, but he
recommends a certain methodology, by which, one has to construct theoretically progressive
problemshifts, and avoid going in for degenerating ones.

iv) Lakatos gives due importance to historical dimensions of philosophy of science. A

methodology must study both successful and not-so successful episodes of history of science and
look for influence of the ‘external factors’ (e.g. political or religious interference) in science.
History of science is very essential to do a proper philosophy of science because, as he rightly
declares: “Philosophy of science without history of science is empty; history of science without
philosophy of science is blind” (Lakatos, 1971, 91).

v) SRP is better described as a theory of historiography, rather than a ‘methodology of research

programme’. For, it does not give us strict rules of practice; it is not very clear as to when to
abandon a SRP. Critics have pointed out that if a methodology has not given instructions about
the choice from competing research programmes, it is not really a methodology. Lakatos seems
to give only a ‘framework’ or ‘theory of historical appraisal’ and not any ‘scientific and practical
methodology’.

v) Musgrave argues that ‘hardcore’ of SRP is not historically correct and methodologically
effective; for example, the scientists before 1850 did not treat Newtonian gravitation law as a
part of hardcore. The elements in the hard-core are also not refutable, but treated as the
fundamental theoretical postulates or axioms of a theory. It is a sort of a convention amidst the
scientists that makes them hold onto hard-core in spite of anomalies (Musgrave, 1978, 110). The
role of the individual scientists is minimized, while the significance of convention is stressed
more. Further, Lakatos does not elaborate how certain hypotheses are to be chosen for ‘hardcore’
category.

vi) Edwin Hung (1997, 403) and other critics point out: Lakatos insists that in a progressive SRP
the successive theories must have more ‘empirical content’, but this is a vague concept, as in
many of the cases, it is not possible to ascertain this. Also, he claims that the latter theory must
have more explanatory power than the previous one, but he has not given a clear notion of
explanation, say, whether it is a contextual or causal or edificatory theories of explanation.
Finally, if Lakatos distances himself from the question of truth or nearness to truth (Popper’s
verisimilitude), why are the scientists expected to choose the progressive programmes and what
would be the criterion to choose one programme from the other?

Check Your Progress I

Note: Use the space provided for your answers.

1) What is a SRP in Lakatos’ scheme? What is its significance in understanding science? How is
it evaluated?

………………………………………………………………………………………………………
………………………………………………………………………………………………………
 6

………………………………………………………………………………………………………
…………………………………………………………………………………

2) Compare and Contrast Lakatos’ SRP with Kuhn’s Paradigm.

………………………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………
…………………………………………………………………………………

3.3 DUDLEY SHAPERE: TRANSCENDING CLASSICAL EMPIRICISM AND

RATIONALISM

Shapere, one of the leading philosophers of science of our times, had his doctoral studies at
Harvard and worked at several prestigious institutes. His famous works include Galileo: A
Philosophical Study (1974) and Reason and the Search for Knowledge – Investigations in the
Philosophy of Science (1984). He critiques both logical positivist and the post-positivist
philosophies of science to arrive at a balanced understanding of science. He has taken it up as his
life-vocation to delve into the philosophical foundations of science. He is known for his
innumerable, even complicated, case studies from history of science to substantiate his position.
He has taken philosophy of science as his life-mission. About the project of his recent works,
The Rational Dynamics of Inquiry and The Values of Knowledge, he mentioned to me in a
personal interview on 6th October, 2006, at his residence, in North Carolina, USA, “I mainly do it
[writing the books] for myself; it clears up what I am”.
Shapere’s important ideas like methodology, observation, meaning and incommensurability in
science, the notions of domain and goal success, the absence of specific doubt in science, the role
of ‘the given’ in experiments, the need to learn how to learn etc. – all seem to be a well-
connected web, as though they are different nodes in a fishing net. However this section
discusses only two main contributions: the notion of observation and contingent interactional
empiricism.
Observation as a “Concept Schema”
To understand the contemporary sophisticated science, Shapere argues, it is better that one treats
the important concepts like observation, meaning, reality, truth, knowledge, and so on as
‘concept schema’, rather than individual concepts. For, the meanings of these concepts are not
given once and for all as one individual concept; rather they develop over the years, with all
related aspects. A concept schema implies that: i) it is formulated in terms of a ‘framework’ that
remains reasonably stable over a significant period of time, or over contemporaneous areas of
inquiry; ii) It is treated as part of ‘an approach to inquiry’ and the interpretation of its results.
Classical empiricism wrongly equates observation with sense perception. With the insistence of
sense perception, classical empiricism is not only insufficient to understand science and its
problems, but also it is largely irrelevant and even positively obstructing the process. For, even if
all our knowledge were shown to be ultimately based on sense perception we would still not
know or understand knowledge fully, as our sense perception covers only a limited section in the
 7

whole range of Electro Magnetic (EM) spectrum. The normal light, which is visible to our
human eyes, is just one portion of the vast spectrum (0.4 – 0.7µm). Gamma rays, with extremely
short wavelength as short as a billionth of wavelength of the visible light, are at the one end and
at the other end are the radio waves, with trillion times higher wavelength of the visible light.
“The eye” therefore, “comes to be regarded as a particular sort of electromagnetic receptor,
capable of ‘detecting’ electromagnetic waves of the ‘blue’ to ‘red’ wavelengths, there being
other sorts of receptors capable of detecting other ranges of that spectrum. This generalized
notion of a receptor or detector thus includes the eye as one type” (Shapere, 1982, 505).

There is no ‘pure’ observation and all observations are theory-laden. Modern science does not
hold “evidential as perceptual”, because sense perception is often unreliable (seeing the half-
immersed stick as bent) and incapable (in the areas of too small or too big in size or with the
objects too near or too far) (Shapere, 2000). The conditions for something to be observed
(observable) are: i) information is received by appropriate receptor; ii) that information is
transmitted directly, without any interference, to the receptor from the entity” (Shapere, 1988b,
308); and iii) “The information is transformed by appropriate devices into humanly-accessible
information which is (eventually) perceived by a human being” (Shapere, 1982, 517). This
interpretation makes observation intelligible to humans.

What is observable / unobservable is, thus, determined by a number of factors, like: the
instruments used; the theoretical knowledge which tells us the nature of interactions; the
theoretical possibilities of detection of particular interactions, and how the particular interactions
give information about its sources. Thus the concept of observation is not a single notion to be
captured by a logical or a priroi analysis. It is a concept schema evolving over the years,
intertwined with the methodology and background knowledge.
The Scientific Status of Unobservable in Modern Science: There are certain things that can never
be perceived directly like the particles in the cloud chamber, as one can observe only the track of
the particle. Modern cosmology teaches that the part of the universe will not be observable by us
unless it enters our horizon, and “if the universe is infinite, that, at any given time, there will
always be regions which are unobservable” (Shapere, 2000, 159). Particle Physics and modern
cosmology encounters problems and theories for which observational or experimental tests
appear impossible, even in principle. For instance dark matter and dark energy cannot be directly
observed but can be known only from the effects they create. Classical empiricism might ignore
all such entities and theories as unscientific as they are unobservable and untestable, but Science
considers all these as legitimate objects of scientific study, though they are unobservable in
principle.

Of course, Shapere does not permit any bizarre entity / theory to be treated as scientific. He gives
guidelines to distinguish between legitimate and wild / loose speculation in accepting something
observable in science: i) If that entity is logically and mathematically implied by something that
is already observable or has observable consequences; ii) If it is needed for consistency
considerations, even though it is not implied by the observable parts of the theory; and iii) If it
provides answers to problems concerning the observable parts of the theory with which no other
solution deals successfully. These guidelines of course are to be taken in spirit, not in letters.
However, despite the liberality, it is by no means the case that ‘anything goes’ (Shapere, 2000,
159).
 8

Rational Descendent of Classical Empiricism: Transcending Traditional Rationalism and

Empiricism
Traditionally two major epistemological groups, Empiricism and Rationalism, have tried to
clarify the issues related with knowledge and knowing. While empiricism maintains that all
pieces of knowledge (concepts & beliefs) are rooted in experience, for their source and
justification, rationalism argues that at least some knowledge is attainable by reason, independent
of experience, without any interaction with the reality outside. Problems with empiricism: i) it
fails to clearly clarify what is meant by pure observation, which does not need any interpretation,
and to show how all the other non-observational beliefs are rooted in observation; ii) it fails to
realize that sense experience does involve inference, as there is no pure ‘given’ available. If the
importance of the ‘given’ in experience is stretched too far, one would land in ‘solipsism’, as one
cannot justify believing in anything beyond one’s own experiences; one can’t speak of past or
future, which are basically inferential in nature, going beyond the ‘given’ of here and now
(Shapere, 1988b, 301); iii) Empiricism’s claim, “all our knowledge is based on sense experience”
– is it analytic (a matter of definition)? or empirical (a matter of fact)? In both cases it would be
highly problematic: “if analytic, it can, on its own principles, tell us nothing about the matters of
fact with which it seems to be concerned; but if it is empirical, it cannot be established with
absolute certainty” (Shapere, 2000, 161). Therefore empiricism itself (like its conclusion) is an
empirical doctrine, and so it can bring itself to a stage at which it outgrows itself. Problem with
rationalism: i) It could not establish any substantive truths about reality, totally independent of
any connection with the senses; ii) Even the compromise proposed by Kant to synthesize both
empiricism and rationalism is also problematic; for, Kant’s a priori intuitive forms of Space-time
category and the Principle of Causality have been seriously questioned by the contemporary
physics [See: Shapere, 1988].
Shapere views the knowledge-seeking process (science) as Rational Descendant of Classical
Empiricism: it is empiricism, because we have to interact with the world to learn about it; and it
is rational descendant as well, because we have to learn what it is to interact; with the help of the
background information we have to learn to how to learn. It is neither an aprioristic rationalism
nor a sensory empiricism, though it takes certain important aspects from both of them into
consideration: namely, we bring some background information to interpret the data available,
however that background information is changeable for specific reasons; it is neither relativist
nor foundationalist, due to the fact that the observation–situation is infused with background
information and there is ‘the given’ in the whole of inquiry; it is both historical… and rational,
because even what counts as ‘reason’ too emerges over the periods of history, and finally it is far
more concerned with content than with logical form, though logic is not excluded (Shapere 1995,
25-26).
Remarks
i) Shapere’s balanced approach: Though Shapere does not take extreme positions his views
deserve serious consideration; he evokes a lot of reactions from philosophers and philosophers of
science, probably because he attempts at a balanced view between the positivists and the
historicists. Not just a compromise, but he seeks to transcend both by inculcating the strengths,
and avoids the weaknesses, of each other, offering “a coherent and often attractive vision of
scientific inquiry” (Nickles, 1985, 310).
ii) The Role and the Need for Background Information: Shapere has shown, rightly so, as to how
science bases itself on background information and at the same time preserves its rational
 9

character in terms of the ‘given’. Modern science has taught us that we need ‘to learn how to
learn’ about reality, how to think and speak about it, how to refer to what we study and how to
judge the results of our investigation (Shapere, 2001, 200). The demarcation between the
observable and the unobservable, the scientific and the non-scientific, scientific possibilities and
impossibilities, scientific problems and pseudo-problems – all these “are not something given
once and for all, but rather shift as our knowledge and understanding accumulates” (Shapere,
1978, 1000). Therefore, science is an enterprise that has evolved over generations, building on
background information. This reveals the central importance of historical dimension in science.

iii) Shapere on Observation: As observation is not sense perception, Shapere argues that humans
are needed only at the end of the process of observation, where it has to be interpreted. Jan Faye,
however, argues that an observation as such needs some human observers. Observation involves
some sort of beliefs and therefore what is done by an instrument is in fact only a measurement. If
no such beliefs are involved in observation there is hardly any difference between a camera
‘seeing’, a scanner ‘reading a document’ and human observing. Since beliefs are involved in
human observation, we are not merely seeing something, but are seeing that ‘something is the
case’ and “this intentional component is that which elevates perception to observation” (Faye,
2000, 173).
In analysing the notion of observation, we suggest, he includes consensus in the scientific
community as fourth condition. Perhaps, he thinks that the notion of background information
takes care of this aspect. But I think it is so important in modern science that it needs a specific
mention. This consensus is also one of the tools to overcome the experimenters’ regress in
modern science, especially when an experiment is done to decide whether an entity exists or not.
The issue is: a correct result of an experiment is one that occurs when the experimental apparatus
is functioning properly (and it uses proper method), but we check proper function of the
apparatus by whether or not the experiment returns the correct result (http://www.galilean-
library.org/blog/?p=105). If the results are positive one can conclude that entity/field/force in
question exists; if negative, it does not exist. But there can be methodological or mechanical or
mathematical errors and because of which the said entity is detected, or not detected. There
seems to be no other way of checking it out independent of all these procedures. Such areas of
investigations largely rely upon the consensus in the scientific community.
iv) Shapere, a Seeker of Knowledge with Openness: Shapere invites us to learn how to learn
from nature. He comes across a genuine seeker of knowledge with much openness: “We must be
prepared for the possibility that there are indeed more things in heaven and earth that are dreamt
of in our present picture of the universe. Even other universes” (Shapere, 1987, 331).
Check Your Progress II
Note: Use the space provided for your answers.
1) How does Shapere enrich the notion of observation? What are its implications in
understanding modern sophisticated science?
………………………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………
2) Shapere claims to have transcended classical empiricism and rationalism to have a holistic
picture of science. Is his claim justified? Explain.
 10

………………………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………
…………………………………………...……………………………………..

3.4 LARRY LAUDAN: SCIENCE – A PROBLEM-SOLVING ENTERPRISE

As Laudan he acknowledges in his prominent work, Progress and Its Problems (1977), he is
always happy that he was taught by and / or to work with many stalwarts of philosophy of
science, as Hempel, Kuhn, Feyerabend, Popper, Lakatos, Adolf Grunbaum and so on. This in a
way makes it clear that all these people have influenced his thinking. He attempts to take the
works of Kuhn and Lakatos further to have a better understanding of science and its
methodology. In his Science and Hypothesis (1981) he traces the method of hypothesis from
Galileo and Descartes to Peirce, to show the important role of hypotheses in the natural sciences
and to study the philosophical efforts to analyze the structure of hypothetico-deductive
explanation. His another important work Science and Values (1984) is not about ethics of science
and scientists; it is about the cognitive values involved in science and its methodologies. He
focuses on rationality in science, rather than morality in science, though the latter is also of great
importance.

Empirical Problems in Science

All the questions about the functions of nature - why things fall to the earth, why offspring
manifests the characteristics of the parents - and many such queries are empirical problems. He
differentiates between problems and facts, between solving a problem and explaining a fact.
Problems arise in a given context; the theoretical background information informs us what, and
what not, to expect, and this in turn decides what is normal and peculiar. What is problematic in
one domain of inquiry may not be problematic in another context. A scientific theory is a
‘solution to an empirical problem’. Here one need not worry about the truth or falsity of the
theory, because its ability to solve a problem at a given time is that which matters and it may
even turn out to be ineffective to solve later. Science is nothing but a ‘problem-solving system’.
Science, therefore, does not aim at truth, but only on problem-solving ability.

Science solves empirical problems, roughly in terms of explaining the phenomena. Science has
to explain not only the usual phenomena but also anomalies. If an empirical problem p has been
solved by a theory, then for other theories in the relevant domain, which cannot solve the
problem, p becomes an anomaly. So unless we have a theory to explain a phenomenon, it will
not be considered an anomaly. [According to Newton’s theory, there was some discrepancy in
determining the orbit of Mercury; they postulated the presence of another unknown planet to
explain the discrepancy of Mercury, but unfortunately this planet was never discovered. The
discrepancy was not considered as an anomaly, until Einstein’s general theory of relativity
explained the discrepancy in 1916]. Solving an anomaly, however, is only a secondary aim of
science. In fact, anomalies are not in nature, but they reveal the lacuna in theories; “Anomalies
are anomalies for theories. They are symptoms of diseases – not disease of nature, but diseases of
 11

theories” (Edwin Hung, 408). Thus empirical problems can be either problems of explanation or
anomalous problems. The progress of science can be assessed in terms of the problem-solving
effectiveness of theories (Laudan, 1977, 68).

Conceptual Problems

In science conceptual problems are more significant than empirical problems. Empirical
problems are related to experiments, observation etc., but conceptual problems are richer. More
than empirical problems, conceptual problems enable science to grow. When a theory is
proposed to solve some empirical problems, very often they end up creating some conceptual
problems, which can be either internal or external: “Conceptual problems arise for a theory, T, in
one of the two ways:
(i) When T exhibits certain internal inconsistencies, or when its basic categories of analysis are
vague and unclear; these are internal conceptual problems. [E.g. Bohr’s model of atomic theory
can be an example for the internal conceptual problem; his model involved the picture of
classical electrons, yet he spoke about their radiating energy in discrete quanta, an idea given by
new quantum physics much later].
(ii) When T is in conflict with another theory or doctrine, T’, which proponents of T believe to
be rationally well founded; these are external conceptual problems (Laudan, 1977, 49). [E.g.
When Copernicus proposed the heliocentric theory, it had external conceptual problem, as it was
inconsistent with the prevailing Aristotelian theory of geocentrism. Conceptual problems can
also arise from differing worldviews of the scientists. [E.g. When Newton proposed the
gravitational theory between the sun and the planets, his critics like Leibniz and Huygens were
not ready to accept it because they could not understand the notion of ‘action-at-a-distance’, as
they were engrossed with the Cartesian worldview of action-through-contact]. Further,
conceptual problems can also arise due to the difference in the methodologies. [E.g. problems
due to LP’s inductive method or Popper’s Hypothetico-deductive method].

Research Tradition (RT)

In order to improve upon the notions of Kuhn’s paradigm and Lakatos’ SRP, Laduan proposes
‘Research Tradition’ (RT), which provides science with ontology and methodology.
Like a paradigm, RT specifies the kinds of fundamental entities to be used in theories, and it
defines its own empirical problems. However the difference between paradigms and RT is: the
replacement of one paradigm by another is like a gestalt switch, sudden and abrupt, and also like
a psychological conversion, whereas the RT changes step by step, as they have hard-core, which
does not easily give in to changes. Similarly, unlike Kuhn’s paradigm, there can be more than
one competing RT can exist. RT can evolve over time, and sometimes two traditions can get
amalgamated (Laudan, 1977, 104).
The efficiency and adequacy of a RT can be assessed at a given time or can be done over a period
of its lifetime. The RT with the most problem-solving efficiency is accepted; however other
traditions are not rejected immediately; they also can be taken for further pursuit, because they
may be able to solve it in future. Scientists may decide to pursue a RT though they may not
accept it and there can be several RTs at a time: “while it is only reasonable to accept one
research tradition at any one time, it is not unreasonable to pursue several research traditions
simultaneously” (Edwin Hung, 412). For Kuhn there can be only one paradigm; Feyerabend
 12

insists that scientists work in different paradigms for science to be progressive; and for Lakatos it
is better to be with the progressive SRP, though it is not irrational to stay with the degenerating
SRP; but Laudan by cleverly distinguishing between accepted tradition and pursuing tradition,
has achieved a sort of compromise between all the three methodologies.

Remarks

i) To say that science aims only at solving the problems may not be a right description.
Scientists do expect and believe that theories give some sort of truth or insights about the
nature and structure of the world. Only for outsiders science may appear just to be a problem-
solving activity and one will not know what scientists do and how they look at science.
Problem-solving may be one of the important aspects of science but this fails to capture the
richness of what science can and does accomplish. Moreover, in science many problems are
not yet solved (may be, never) like the origin of the universe.
ii) One may question that by solving problems science does not actually make progress though
one may get better insights. By solving a problem 4-3=1, what has one accomplished?
iii) Laudan does not give any criterion to show what is, and what is not, a satisfactory solution.
Also the problem-solving effectiveness is not a quantitative measurement; Further, solving an
empirical problem is in fact to explain the problem. But he does not explain what a scientific
explanation or a theory is, nor the relationship between them. Without making all these
elements clear his methodology is rather empty (Hung, 1997, 413).
iv) Laudan agrees that there is no theory-neutral language or theory-free observation. But the
theory-ladenness does not lead to the famous problem of incommensurability. For he
explains that the assumptions to understand a problem and those to solve it are different.
Different people can propose different theories to solve a problem, but the understanding of
the problem may be common. For instance, about the nature of light, particle theory
(Newton), longitudinal wave theory (Huygens) and transverse wave theory (Young and
Fresnel) were proposed, but since all shared the common problem about light, these theories
were comparable. But since Laudan has not analysed what an explanation is, and theory is,
we cannot understand cross-theoretical explanation, and thereby the problem of
incommensurability is left unaddressed.

Check Your Progress III

Note: Use the space provided for your answers.
1) How does Laudan distinguish between empirical and conceptual problems in science? What
are the advantages of the latter ones?
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
....................................................................................................................................
2) Compare and Contrast Laudan’s RT and Kuhn’s Paradigm
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
....................................................................................................................................3) Is Laudan’s
RT more adequate than Lakatos’ SRP in understanding science? If so, how? And if not, why not?
 13

............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
....................................................................................................................................

3.5 LET US SUM UP

The approaches of historical realist philosophy of science have been explained with the help of
the salient features of Lakatos, Shapere and Laudan. Each one has tried to enrich the picture of
science and the role of history of science in doing philosophy of science. Lakatos’ SRP, Sphere’s
notion of observation as concept schema and Laudan’s RT are sincere attempts to show how
science can be rational, while still being faithful to the socio-historical and non-rational factors.
This lesson has discussed some of their merits and demerits.

3.6 KEY WORDS

Unobservables in Science: It refers to the entities, fields, events or phenomena which cannot
directly be observed (e.g. electron, magnetic field, radio waves etc). Scientific realists claim that
they exist in reality, independent of us, while antirealists deny it, claiming that they are only
convenient tools to deal with nature.
Scientific Research Programme: It is the basic unit of science according to Lakatos. Each
programme has unchangeable hardcore and changeable protective belt.
Research Tradition: According to Laudan science is made up of research traditions. Though it is
similar to Kuhn’s paradigm, it does a better job to maintain the rational character of science.

3.7 FURTHER READING AND REFERENCES

Collingwood, R. G. Speculum Mentis. Oxford: Clarendon Press, 1924.

Faye, Jan. “Observing the Unobservable.” in The Reality of the Unobservable. Ed. E. Agazzi,
and M. Pauri. Netherlands: Kluwer Academic Publishers, 2000. 165-175.

Goldstein. Historical Knowing. Austin: University of Texas Press, 1976.

Laudan, Larry. Progress and Its Problems – Towards a Theory of Scientific Growth. California:
University of California Press, 1977.

----------- Science and Hypothesis. Dordrecht: Holland, 1981.

----------. Science and Values – The Aims of Science and Their Role in Scientific Debate.
California: University of California Press, 1984.
 14

Lakatos, I. “Falsification and the Methodology of Scientific Research Programmes.” in Criticism

and the Growth of Knowledge. Ed. Lakatos and I. Musgrave. Cambridge: Cambridge University
Press, 1970.

----------- The Methodology of Scientific Research Programmes: Philosophical Papers. Two

Vols. J. Worral and G. Currie. Ed. Cambridge: Cambridge University Press, 1978.

Palter, Robert. “Philosophic Principles and Scientific Theory.” Philosophy of Science. 23 (1956).
111-35.

Shapere, Dudley. “The Concept of Observation in Science and Philosophy.” Philosophy of

Science. 49(1982). 485-525.

----------- “Unity and Method in Contemporary Science.” in Scientific Knowledge. ed. by Janet
Kourany. California: Wadsworth Publishing Company, 1987.

----------- “Rationalism and Empiricism: A New Perspective.” Argumentation 2 (1988). 299-312.

----------- “On the Methods of Science.” in The Creation of Ideas in Physics. Ed. Leplin.
Netherlands: Kluwer, 1995. 13 – 27.

---------- “Reasons, Radical Change and Incommensurability.” in Incommensurability and

Related Matters. Ed. Hoyningen-Huene and Sankey. Netherlands: Kluwer Academics
Publishers, 2001. 181-206.