The Gettier Problem in the Philosophy of Science.

Philosophy of science has a curious relationship to epistemology, or the theory of knowledge.

There is little doubt that in the 20th century the philosophy of science took on a life of its own largely independent of epistemology. Interestingly, given that we begin by making an historical observation, some of the most important philosophers of the 20th century, and some of the most important works of the 20th century, were by philosophers of science or of philosophy of science.

Philosophy of science has become an autonomous sub discipline of philosophy, and many undergraduate and graduate philosophy courses teach epistemology and philosophy of science autonomously.

This is curious for scientific knowledge is a species of knowledge. The problems of scientific knowledge pondered by philosophers of science have more than a whiff of traditional epistemology to them. For instance, perhaps the *key* problem of epistemology has been the problem of scepticism. Philosophers of science, to no small degree, are interested in providing some grounding to science given the problem of scepticism.

Scientists, to use the expression of Richard Feynman, stopped worrying about such issues long ago, and simply got on with doing science. However, at times, epistemological angst creeps into the sciences in a serious way. Some of the fiercest debates in theoretical physics today are concerned with the nature of the scientific enterprise itself.

I myself have the impression that philosophy of science, in part, became important because paradigm shifting advances in scientific knowledge, such as highly abstract pure mathematics, relativity, and quantum theory, provoked epistemological angst, as it were, which contributed to the rise of logical positivism and its associated concern with science.

I think sociological and historical reasons also contributed, such as the second industrial revolution and the much closer relationship that was forged between science and the state.

The fall of logical positivism has not improved matters, at least not in philosophy if not the sciences.

It would be interesting to do a historical study of epistemological angst with reference to the Kuhnian framework. Are paradigm shifts accompanied by an upsurge in epistemological work that seeks to put science on a firmer epistemological foundation? It is easy to imagine how it could do so.
Normal science is not a time, one feels, when many are worried by epistemological niceties.

But when the paradigm shifts all is torn asunder. One could, perhaps, fit the rise of logical positivism into such a historical and structural framework.

Anyway, that is not really my concern here.

I am more interested in the Gettier counterexamples. The analysis of the concept, knowledge, has been dominated since antiquity with a tripartite conception of knowledge. That is, to know p is to have a belief that p, that p be true, and the belief that p is justified or warranted.

Much of epistemology has been devoted to accounting for the justification criterion. Theories abound. However, Gettier, in a classic two page paper (which wouldn’t cut the mustard in the neoliberal university), up ended this account of knowledge when he showed that it is possible to have a justified true belief yet not to be in the possession of knowledge. Gettier counterexamples take the form;

Smith justifiably believes that P.
P is false.
Smith correctly infers that if P is true, then Q is true.
So, Smith believes Q, justifiably.
Q is true, but not because of P.
So, Smith has a justified true belief that Q.

When I read philosophy of science I can’t help but get this hunch, this feeling, this gut intuition, that a lot of it focuses on the justification criterion, like in traditional epistemology, but only in the context of scientific knowledge.

But if knowledge is *not* justified true belief, as Gettier held it not to be, then philosophy of science should be just as infected by the Gettier problem as epistemology is. But, because of its autonomous status, philosophy of science is not affected by the Gettier problem to the same degree.

One way of sidestepping the Gettier problem is to adopt one or another stance of epistemology naturalised, to borrow from Quine. Perhaps what is required is a philosophy of science naturalised.

Can you use naturalistic inquiry to defend naturalistic inquiry? Is there, perhaps, a paradox of self reference here? If there is, could not philosophy of science be incomplete in similar fashion to the way we say mathematics is incomplete?

We started with epistemology as a general case and then moved on to philosophy of science as a specific case.

Could we now seek to move, by analogy, from the specific to the general? Namely, if philosophy of science is incomplete because of paradoxes of self reference could not epistemology be shown to be incomplete for similar reasons?

Can you come to know what knowledge is without first knowing what knowledge is?

The study of the philosophy of science, from the early 20th century onward, with reference to the Gettier problem would, at a minimum, be a most fascinating study. At the outer edge of speculative fancy, a formal demonstration of the incompleteness of epistemology would be a significant and original contribution to human knowledge.

It could be that we are a species, through the clear light of reason, that can come to *know* that one cannot *know* knowledge.

Now such a theorem ought be called Socrates’ theorem if any should. One, the tripartite conception comes from him via Plato, and two Socrates knew that he knew nothing.


  1. Karen Barad’s account of Bohr’s philosophy-physics provides an interesting take on standard constructive perspectives. Knowledge in science only exists once the mechanisms are in place for observation and measurement. If this is extended to all knowledge – even the common sense kind – we might perhaps stop worrying so much about Gettier’s challenge.

  2. I am not familiar with Barad’s take on constructivist epistemology, but certainly mechanisms of observation and measurement are important even in the everyday domain. Although, of course, one needs to qualify this position given that observation and measurement in science are types of quantitative experimentation that abstracts from our everyday surroundings, which we don’t do typically as a part of our everyday lives; observation in the domain of the familiar is more akin to what is sometimes called naive empiricism.

  3. One possiblility opened by Gettier’s paradoxes (originating with Russell and assuming Chisholm’s definition of knowledge) is that epistemology (like the philosophy of science) has hit a dead end. Or, more acceptably, that rationality has inherent limits. One other deadend — not unavoidable either — is relativism. Relativism is a self-contradictory position, but one that we are left with if we wish to ground knowledge and science in historical reality (if not, then we are faced with irrelevancy, cf., Feynman’s remark). This means philosophy needs to be re-thought.

Comments are closed.