Falsification falsified. Basic statements are those statements, which assert,

and the Duhem’s Thesis



We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

In this paper, I will argue that the Duhem
Thesis does present a problem for Karl Popper’s falsification method.  This paper will proceed as follows. First, I
will outline Popper’s critic of verification and his proposed solution to the
demarcation problem, falsification. After this, I will consider the Duhem’s
Thesis, which is seen to be damaging to the logic of falsification. Due to the
limited length of this paper, unfortunately be unable to consider Quine’s
Thesis, which combined with Duhem’s, forms what is known as the Duhem-Quine



In rejecting the methodology of the logical positivists,
verification, Karl Popper proposed that falsification was a more appropriate
scientific methodology and demarcation criterion.  Falsification involves the attempt in
scientific research to falsify or refute scientific statements and hypotheses. The
notion of falsifiability can be illustrated in the form of modus tollens in the following way (H represents the hypothesis; O
represents the empirical observation):

1.     If
H, then O

2.     Not-O

Therefore, not-H

For Popper, whether a theory could be
falsified, and thus deemed scientific, depends on whether its basic statements
can be tested and falsified. Basic statements are those statements, which
assert, “that an observational event is occurring in a certain individual
region of space and time”.1 For example, the statement “There
is a raven in the space-time region k”
would be one of the basic kind.2
These play a vital role in Popper’s work, as he notes, because they are needed
“in order to decide whether a theory is to be called falsifiable…And we also
need them for the corroboration of falsifying hypotheses, and thus for the
falsification of theories”.3
According to Popper, a theory can be considered scientific if and only if it
can categorise its basic statements into one of the following: the potential
falsifiers (the basic statements with which it is inconsistent), and the non-contradictory
basic statements (the basic statements which do not contradict it.4
The scientific theories will be the ones in which the former category is not

At first light, the notion
of falsification seems logically appealing. However, as will be shown in the
next section, Duhem’s Thesis posed a significant problem for Popper’s
demarcation criterion.


3.     The Duhem-Quine Thesis

In his most influential work, The Aim and Structure of Physical Theory, Duhem asserted that it
was not possible for scientists to isolate a single hypothesis for
experimentation, and as a result, crucial experiments are impossible.6
The presence of implicit auxiliary hypotheses surrounding the central
hypothesis, “there is no principled way to distribute blame among the
conjuncts, and thus that it is impossible to say to what degree the refutation
disconfirms the central hypothesis”.7
Either the central hypothesis is false, or one of the many implicit auxiliary
hypotheses, or they are all false, but there is no way of knowing.

Duhem uses the example of the test O. Weiner
carried out to test F. E. Neumann’s assumption that the vibration in a ray of
polarised light is parallel to the plane of polarization. Once the experiment
had been performed, however, the predicted phenomenon was not produced, and
Weiner concluded on this basis that Neumann’s proposition concerning the
parallelism of the polarised ray of light and the plane of polarisation was
According to Popper, this was how science should be practiced. However, as
Duhem asserts, the experimenter can never be sure that the result of the
experiment was caused by the proposition, which was being studied, for a whole
number of other factors were implicitly involved in the experiment which may
have had an effect on the result.9
Thus, Duhem wrote:

In sum, the physicist can never subject an isolated
hypothesis to experimental test, but only a whole group of hypotheses; when the
experiment is in disagreement with his predictions, what he learns is that at
least one of the hypotheses constituting this group is unacceptable and ought
to be modified; but the experiment does not designate which one should be


The problems identified by Duhem can be represented by
the logic of modus tollens (the
variable (A1 ) represents the all of the possible
auxiliary hypotheses present):

1.     If
H, and (A1 )
then O

2.     Not-O

Therefore, not
H, or not (A1 )

As the above shows, it is no longer the case

In light of this, Popper’s notion of falsification seems
to have lost all the appeal it once had, and now seems to offer nothing over
the logical positivists’ verification. Rather than testing a single hypothesis
in an experimental vacuum, Duhem suggests that science should aim at testing
and falsifying whole groups of hypotheses. For example, it is not possible to
test Newton’s first law as an isolated hypothesis, since in order to test and
observe the consequences of Newton’s first law, theoretical assumptions (such
as the laws of gravity) and auxiliary hypotheses (such as the effect of the
gravitational forces on extra-terrestrial objects) must be included.11

            When a
falsifying result is produced in an experiment, what options does the scientist
have, if not falsifying the central hypothesis? The scientist can either:
attempt to protect the hypothesis by changing one of the auxiliary hypothesis, introduce
an ad hoc hypothesis, or discard the
hypothesis entirely. Popper argued that since the introduction of an ad hoc hypothesis reduces the
falsifiability of the theory, the theory becomes less scientific.12
In order to maintain the scientific status of the theory and save it from being
discarded, the scientist must change one of the auxiliary hypotheses. Both
Duhem and Quine suggested this as a solution. The next section will look into
these ideas further and question whether adjusting the auxiliary hypotheses is
an appropriate method in scientific research.


of Auxiliary Hypotheses

The Duhem-Quine Thesis is that it is not logically
possible to isolate and test a single hypothesis without the presence of an
unknown number of implicit auxiliary hypotheses. The challenge this presents
for Popper’s falsification is that it is not possible to falsify specific
hypotheses, since the presence of the auxiliary hypotheses may have had an
effect on the observation. Popper, in fact, recognises this issue when he writes:

The attribution of the falsity to some
particular hypothesis that belongs to this set of premises is therefore risky,
especially if we consider the great number of assumptions which enter into
every experiment.13


In light of this, Popper suggests that the whole system should
be falsified (including the auxiliary hypotheses, or in Popper’s terms, the
“initial conditions”), rather than individual, isolated hypothesis.14
However, this is acknowledgment by Popper is clearly not enough to defend
falsificationism from the objection. A more adequate defence is required if
falsification is to be maintained as a methodology within scientific research,
and as the answer to the demarcation problem.

Popper believed that all empirical observations are dependent on and laden with
theory; he denied the empiricist view that it is possible to observe the world
in an objective way, and believed that knowledge and facts are not simply
presented to us through experience in the world.15 In
both everyday life and scientific practice, observations are theory-laden, and
in fact, our observations are dependent upon theory. Background assumptions,
theoretical and metaphysical presuppositions (auxiliary hypotheses) are vital
for the observations we make to be meaningful; without them, our observations
would not make sense. Popper claims that experimental results or observation
reports “are always interpretations
of the facts observed; that they are interpretations
in the light of theories (original emphasis)”.16

            The critique posed by the
Duhem-Quine thesis is that it is impossible to test and falsify single
hypotheses, since the presence of implicit auxiliary hypotheses may have had an
effect on the result. In essence, whilst testing a single hypothesis, there may
be present certain other factors, auxiliary hypotheses, which affect the
observation. This critique rests on the basis that observations are free of
background assumptions. Duhem seems to suggest that auxiliary hypotheses exist
separately from observations, and therefore can be adjusted in order to
maintain the central hypothesis. This gap which has been created between the
observations and the auxiliary hypotheses is exploited by Popper’s critics, as
it allows them to question experimental observations on the basis of auxiliary
hypotheses. Through this
separation, the Duhem Thesis makes it possible to illustrate that the
observation may have arisen from the auxiliary hypotheses rather than the
hypothesis under examination. However, as the previous paragraph illustrates,
Popper was in fact aware of
the presence of auxiliary hypotheses in this work of the theory-dependent
nature of observation. For Popper believed that observations were inseparable,
since observations are “interpretations
in light of theories”.17
Therefore, for Popper, observations have value in virtue of the presence of auxiliary






One possible objection against this idea is that




In summary, the Duhem-Quine Thesis asserts that it is not
possible to isolate, test and falsify a single hypothesis due to the presence
of auxiliary hypotheses.


Popper accepts this









David. Knowledge and Social Imagery.
London: Routledge and Kegan Paul, 1976.


Duhem, Pierre. The Aim and Structure of Physical Theory.
New York, NY: Atheneum, 1962.


Karl Raimund. Conjectures and
Refutations: The Growth of Scientific Knowledge. New York: Basic Books,


Karl Raimund. The Logic of
Scientific Discovery. London: Hutchinson, 1972.


Karl Raimund. Realism and the Aim of
Science. London: Hutchinson, 1983.


Strevens, M. “The Bayesian
Treatment Of Auxiliary Hypotheses”. The British Journal For The Philosophy Of Science 52, no. 3
(2001): 515-537. doi:10.1093/bjps/52.3.515.

1 Karl Raimund Popper, The Logic of
Scientific Discovery (London: Hutchinson, 1972), 103

2 Ibid., 101

3 Ibid., 100

4 Ibid., 86

5 Ibid.

6 Pierre
Duhem, The Aim And Structure Of Physical Theory (New York, NY:
Atheneum, 1962), 183-190

7 M. Strevens,
“The Bayesian Treatment Of Auxiliary Hypotheses”, The British
Journal For The Philosophy Of Science 52, no. 3 (2001): 515-537,
doi:10.1093/bjps/52.3.515, 516

8 Ibid., 184

9 Ibid., 184-185

10 Pierre Maurice
Marie Duhem, The Aim And Structure Of Physical Theory (Princeton:
Princeton University Press, 1991), 187

11 Donald
Gillies, Philosophy Of Science In
The Twentieth Century (Oxford: B. Blackwell, 1993), 212

12 Karl Raimund
Popper, Conjectures And Refutations (London: Routledge, 2002), p37-38

13 Karl Raimund
Popper, Realism and The Aim Of Science (London: Hutchinson,
1983), 187

14 Ibid.; Popper, The Logic of
Scientific Discovery, 41-42; 76

15 David
Bloor, Knowledge And Social Imagery (London: Routledge and
Kegan Paul, 1976), 53

16 Popper, The Logic of Scientific
Discovery, 107 (footnote* 3)

17 Ibid., 107 (footnote* 3)