James Garvie on Popper

James Garvey, The Twenty Greatest Philosophy Books  Continuum, London 2006

The authors of the 20 greatest books are Aquinas, Aristotle, Ayer, Berkeley, de Beauvoir, Descarates, Hegel, Hobbes, Hume, Kant, Locke, Marx, Mill, Neitzsche, Plato, Popper, Rousseau, Sartre, Schopenhauer and Wittgenstein.

“Popper is certainly among the twentieth century’s most influential philosophers of science. It is possible to think of him as the most influential one…Others, though, have less time for him. He fought his corner vigorously, they say, but not very well.” 

One would like to know who “they” are, and their arguments. According to Garvey there are five “troubles” with Popper’s view and he listed these at the end of this short chapter, leaving the impression that whatever influence Popper might have exerted, his ideas can be safely put aside from serious consideration.

Garvey approached “The Logic of Scientific Discovery” (LSD) as a response to Hume’s critique of induction, so Popper’s two problems concerned the role of induction in science and the difference between science and psuedo-science. “His solution to both involves the idea of falsification”. It would help to point to the obsessions of the logical positivists with meaning and nonsense, and the way that induction was supposed to be the hallmark of science.

“The Logic of Scientific Discovery makes two very large claims about scientific inquiry. First, Popper argues for a version of falsificationism, the view that science proceeds not by verifying the truth of hypotheses, but by putting aside false ones. Falsificationism, Popper contends, solves, or at least saves us from, the dreaded problem of induction. Second, Popper argues that scieintic propositions are different from pseudo- or non-scientific propositions because they are empirically falsificable”.

It would help to make a distinction between Popper’s views on the LOGIC of scientific investigation and what counts as PROGRESS in science.

As to the logic, the issue is “What can you do say about a general statement when you have in your hands a true (and relevant) statement of fact?”

Answer: there is an asymmetry between falsification and verification: the true statement of fact can identify an error but it cannot confirm a truth. You might add that the location of the error is clear in the case of statements like “all swans are white” but in the world of research there are problems of data quality and also the Duhem problem, so the precise location of the error may be difficult to find.

On progress, science advances when people invent better theories. It is not just a matter of refuting theries and putting them aside, a la Garvey’s statement, because better theories are required to take the place of the ones that are put aside. In Chapter 10 (section 5) in Conjectures Popper specified three requirements for a good new theory: first to extend from a single big insight that unifies disparate fields of study, second to be independently testable and third to pass some new and severe tests. I have an idea that David Deutsch is one of the few people who has pointed to this part of Popper’s work.

Garvey has conflated issues from logic, from history, from research practice and from practical application of theories. People who read the two pages that he wrote on induction and demarcation will get a glimmering of the alternative to positivism but that is undone by the three pages that follow. 

Troubles with Popper’s view

1. “Are we to think of scientists as genuinely in the business of trying to falisfy and not prove their theories?”

2. “Popper’s view, anyway, seems to ignore the social dimension of science as it is practised, a fact exploited by later philosophers of science such as T S Kuhn and Paul Feyerabend”.

3. “If Popper is right and we can never know whether or not a theory is true, then, rather bizarely, science tells us nothing about the world except what is not true of it. If all we can know is that a theory is false, we end up with no positive scientific beliefs about the world”.

4. “Further, if Popper is right, in what sense can science improve life or help us to make practical choices?”

5. The Duhem problem ( the role of  other theories and statements of initial conditions in testing theories).

On 1. Of course scientists do all kinds of things but it helps to use data in the most effective (and cost effective manner), which is to test theories by means of “critical experiments” which can help to revise the status of competing theories.

On 2. Jarvie pointed out that far from ignoring the social dimension of science, Popper led the way with a “social turn” that can be identified in The Logic of Scientific Discovery and also in The Open Society, especially Chapter 23 on the social nature of science.

Chapter 23

On 3. Popper is not concerned with beliefs (with reasons), he has focussed on well tested theories as a practical substitute for beliefs.

On 4. See 3.

On 5. The Duhem problem is just the way things are. Simply pointing to it does not discredit Popper’s thoeries unless you can produce some other approach that does better to handle the theory-dependence of observations.

The treatment of Popper contrasts with the strange deference that Garvey extends to Ayer’s “Language, Truth and Logic” and his breathless and admiring treatment of Wittgenstein.

.