Falsification and theory-ladenness

Word count:~1400

Question:

Critically assess the claim that observation statements are theory-laden. Assuming the claim to be true, examine its implications for Popper's falsificationist account of science.

Introduction

This essay will firstly explain what is meant by the term 'theory-laden', before providing arguments for and against the claim that observation statements are theory-laden. A brief look at Popper's falsificationist account of science will be given. Then the implications for the falsificationist account of science will be examined assuming the claim that observation statements are theory-laden. The essay will conclude that observation statements are theory-laden, and that this does not cause problems for sophisticated variants of falsification (though they fall to other problems not outlined here).

Theory-laden

The term "theory-laden" refers to how statements or terms have meaning only in relation to a theory or a paradigm; in a different context the statements or terms might have either a different (possibly contradictory) meaning or no meaning at all.¹ "Mass" is one term that has different meaning in two different systems (Newtonian physics and Einsteinian physics). "Planet" is another example, this time from astronomy. The change from the Ptolemaic, geocentric, system to the Copernican, heliocentric, system meant a change in the definition of "planet".

Some philosophers have put forward the idea that all observations are "deeply permeated by theories", and thus we can not have direct access to the world. If all observation is theory-laden as they suggest, then the objectivity of science is undermined, as we can not know if our perceptions accurately capture information about the world.² This is contested; observations and experiences may or may not be theory laden, other animals have observations and experiences, and most of them would not have the cognitive ability to formulate theories. This claim might be narrowed to simply include 'scientific observations', which is more probable. When a scientist makes a measurement, they often have a particular result in mind, and thus might not pay attention to other results or anomalies. Whether observations are theory-laden does not affect whether observation statements, descriptions of experiences are theory-laden or not.

Observation Statements are Theory-laden

Observation statements are statements about facts etc. that are observed/perceived. An example might be "this rock has a mass of 1kg". Similarly, "that tree is green and brown". The claim has been made by a number of different philosophers (including Karl Popper, Thomas Kuhn and Paul Feyerabend) that such observation statements are theory-laden. Implicit assumptions about mass are included in the statement above about "this rock". The observation about the tree, includes assumptions about wavelengths and so on.

To explain something or describe it, people use a common language with certain implicit and built in concepts. This is especially the case when talking using scientific terms on or about scientific data. Technical language is based around a paradigm.³ This can be seen when paradigms change (the examples given above, in physics and astronomy, apply here as well). People from the two paradigms had different theories implicit in what they talked about.

Because of this difference in terminology, scientists often talk at cross purposes, neither understanding the other. However, it could be argued that the scientific terminology is simply a short-hand, and if the scientists used lower-level language, they could communicate.⁴ In the case of astronomy and the change from the Ptolemaic to the Copernican, rather then using "planet", people from both paradigms could communicate by using more descriptive terminology. Astronomical bodies for example to describe all of the Moon, Sun, Mercury, Venus, Mars etc. Copernicans to describe what are now called planets could have call them non-Moon, non-Sun astronomical bodies, which Ptolemaics would agree with as well.

Other arguments for and against the claim that observation statements are theory-laden have been put forward. However, the literature on the subject is quite complicated.⁵

Even if observation statements are theory-laden, it is still possible to compare two theories objectively. If an outcome of an experiment is predicted differently by two different theories, then if, within the terminology of each theory, one theory's prediction fails and the other succeeds, then we can say that the second is more successful.⁶

Falsification and science

Karl Popper's philosophy of science was partly a reaction to Hume's problem of induction.⁷ He also had other reasons, the main reason was a solution to the "problem of demarcation". The problem of demarcation is: what distinguishes "science" from "non-science" and "pseudo-science"? Popper proposed falsification as the solution to the problem of demarcation.⁸

According to Popper, science is made up of conjectures and hypotheses, facts can never tell us if these are true, but they can tell us if they are false.⁹ A scientific theory is justified if it is falsifiable, but un-falsified.¹⁰ Falsifiable means that there has to be the possibility that there exists a set of circumstances in which the theory is simply incorrect, "it can be falsified by possible observation events".¹¹ If a theory predicts X and X happens but it is conceivable that Y could have happened, then the theory is falsifiable. If the theory predicts X and Y happens, then the theory is falsified, and thus stops being justifiable. The falsifying event has to be reproducible as well.¹²Good theories are highly falsifiable. They rule out a large number of possible outcomes, yet they remain un-falsified because predictions do not turn out incorrect.¹³

Evidence supporting theories is not rejected by falsificationists. Theories can be corroborated by experimental results and observations that are as predicted by the theory. However, theories are not confirmed as such by supporting evidence, even if you have seen 1000s of white swans and no black swans, there still remains the possibility that black swans exist.

There have been a variety of criticisms of falsification since it was first proposed by Karl Popper. Some of these centre around the conception of theory-ladenness and how observation statements are theory-laden (which was one reason given for Popper to oppose induction). This is what shall be looked at next.

Implications for falsification, assuming observation statements are theory-laden

The claim that observation statements are theory-laden, has implications for Popper's falsification theory of science (assuming that it is true). Naïve falsification falls to attacks that many people, including Kuhn, have launched against it. However, Popper and others have defended a more sophisticated variant of falsification, one that does not fall foul of these attacks.

The theory that "all swans are white" is falsifiable (it is possible that it is incorrect). It can be falsified by, for example, pointing out the existence of a black swan. However, the statement that "there is a black swan" is theory-laden. So a problem is, which theory is incorrect, the theory that all swans are white, or the theory or theories that are implicit in the statement that there is a black swan? Observations might not invalidate a theory (show that it is false) because they are wrapped in a theory, and who is to say which theory is incorrect? One is testing theories against theories, not against facts.¹⁴This point has been made by Hempel,¹⁵ Kuhn and others.

So to get around this, if an experiment is performed, it depends on a methodological decision which theory to regard as the being beyond doubt, and which one is under test.¹⁶ Popper argued that another way to get around this problem, is to agree before an experiment or observation, "which observable situations, if actually observed, mean that the theory is refuted".¹⁷

Lakatos distinguished between what he called Popper₁ and Popper₂. Popper₁ had a base of naïve falsification. This is the position that was criticised by Kuhn. However, Lakatos claimed that Popper₂ was more sophisticated, and did not fall foul of this problem.¹⁸ Popper₂ focused on growth, not on refutation; it also had theoretical programmes.¹⁹ These theoretical complexes are refuted when replaced by another that explains everything the first did, explains facts that were not explained before (or not adequately explained) and makes new predictions or 'produces' novel facts that could not be made in the old system.²⁰

Conclusion

While some people claim that observation statements are not theory-laden, their arguments fall short of being convincing. Assuming that observation statements are theory-laden has implications for the falsificationist account of science. Naïve falsification fails to be convincing taking into account the theory-ladenness of observation statements. However, more sophisticated variants do not fall foul of the same problems.

Bibliography

Bhaskar, Roy. A Realist Theory of Science. Leeds, Great Britain, Leeds Books Ltd., 1975.

Chalmers, A.F. What is this thing called Science? 3^rd Ed. St. Lucua, Queensland, University of Queensland Press, 1999.

Charlesworth, Max. Science, Non-Science & Pseudo-Science. Victoria, Deakin University Press, 1982.

Chase, James. HPA213/313 Philosophy of Science Study Guide. University of Tasmania, 2006.

Chase, James (editor). HPA213/313 Philosophy of Science Unit Reader. University of Tasmania, 2006.

Churchland, Paul M. A Neurocomputational Perspective - The Nature of Mind and the Structure of Science. Cambridge, Massachusetts, A Bradford Book - The MIT Press, 1989.

Couvalis, George. The Philosophy of Science - Science and Objectivity. London, Sage Publications, 1997

Kuhn, Thomas S. The Structure of Scientific Revolutions 3^rd Ed. Chicago, The University of Chicago Press, 1996.

Pollock, John L. Knowledge and Justification. Princeton, New Jersey, Princeton University Press, 1974.

Popper, Karl. The Logic of Scientific Discovery. London, Hutchinson & Co, 1972

Popper, Karl. Objective Knowledge - An Evolutionary Approach. London, Oxford University Press, 1972.

Thornton, Stephen. "Karl Popper", http://plato.stanford.edu/entries/popper/. Accessed 22/09/2006.

Footnotes