[Read Part I Here and Part II Here.]
As I’ve been at pains to show in recent essays, not only is “the science” on any given issue never “settled,” as partisans of a particular political persuasion are forever assuring us. It remains an open question as to what science itself even is.
From the first quarter to approximately the middle of the 20th century, an assortment of scientist-philosophers and scientifically-trained philosophers and mathematicians known as the “logical positivists” or “logical empiricists,” inspired as they were by David Hume’s contention that there were two and only two types of statements that were meaningful—those that were tautologically true and those that could be confirmed through observation—insisted that, essentially, the only genuinely meaningful propositions other than those that are true by definition are scientific in nature.
A genuinely scientific proposition, then, is one whose truth could be verified through experience.
Karl Popper, a contemporary of the positivists who would become one of the century’s most renown philosophers of science, was having none of this. For as influenced as they were by Hume, the positivists shared none of the Scottish empiricist’s skepticism with respect to the principle of induction, the principle that the future will resemble the past. It’s on the basis of this principle that scientists make the predictions that they make. Popper, however, agreed with Hume that this principle cannot be logically justified. Rather, it is due to habit that we assume that, say, because the sun has risen every day in the past that it will do so tomorrow. But we can’t ever be certain about this. We can’t prove or justify it.
Against this dominant empiricist-inductivist conception of science, Popper argued for a philosophy of science that was deductive and to which he gave the name of “critical rationalism.” He selected the latter term in order to contrast it with the classical empiricism of his rivals. Whereas the classical empiricists maintained that “basic statements” or “observation statements” were infallible, Popper contended that there were fundamentally three problems with this claim.
First, all observations are invariably theory-dependent. All observations presuppose a conceptual framework of one sort or other. As Michael Oakeshott once put it, what we see depends upon how we look.
Second, scientific theories are abstract. The terms of a theory and the relations expressed between them in the propositions of that theory are abstractions that, as such, are not found in experience. They can be put to the test only obliquely, by observing their consequences.
Third, the hypothetical character of all knowledge—including, clearly, scientific knowledge—is ineliminable. Science is as much the fruit of creative imagination as is any other human artifact, and science is born of the desire to discover solutions to problems that are themselves invariably the product of historically and culturally-specific circumstances.
For Popper, what demarcates a scientific theory from a non-scientific one is not that it can be established as true. No theory can ever be shown to be true because the principle of induction, being a presupposition born of habit, can never be justified. Even if one observed countless numbers of swans, and every swan ever observed was white, this still could not prove that all swans are white. Rather, what makes a theory a genuinely scientific one is that it can be falsified. The theory that all swans are white, then, is scientific inasmuch as it can be falsified by the observation of at least one swan that is not white.
To be certain, Popper is not arguing that a theory is scientific only if it is proven to be false. What he’s arguing is that a theory is scientific as long as we can conceive of hypothetical conditions under which it could be falsified. So, for example, the theory that the sun will rise tomorrow is obviously falsifiable in that we can conceive of the possibility that tomorrow it will not rise. Or, as Alan Chalmers noted, the fact that we can imagine a brick falling upward when released from a person’s hand distinguishes Newton’s theory of gravitation as a scientific theory, for the mere conceivability of this scenario decisively demonstrates its falsifiability.
Yet until and unless these conditions materialize in reality, we have good reason to accept that the sun will rise tomorrow and that Newton’s theory of gravitation are both true (even though we cannot prove their truth).
Einstein’s theory of relativity Popper thought emblematized a truly scientific account of reality. While Popper accepted it as true, this is not why he thought it was scientific. Einstein’s theory was “risky,” Popper said, in that it had implications that contradicted the claims of the dominant Newtonian model. For instance, gravity, Einstein held, deflects light. The truth of this proposition was confirmed in 1919 by Arthur Stanley Edington and Frank Watson Dyson via “the Edington experiment.” What made Einstein’s theory scientific for Popper, however, is that if the consequences that it forecasted were false, this could have been demonstrated.
Popper’s philosophy of science, we can see, is deductive in that a theory can be logically, decisively refuted through a single counter-example to its predictions.
In glaring contrast, and contrary to the pretensions of their proponents, Freudian psychoanalysis and Marxism are emphatically not scientific. And they’re not scientific because they are not falsifiable: They’re elasticity is such as to allow them to accommodate all possible states of affairs—however mutually irreconcilable these states may be.
Scientific progress occurs along lines similar to the evolutionary process, as Charles Darwin conceived it (the problems with Darwin’s theory we can leave aside for our purposes here). According to Darwinian theory, “natural selection” favors those genetic variations that are “fit.” It simultaneously eliminates those that are not. Analogously, Popper argued that scientific progress is a matter of eliminating those “tentative theories” that fail to withstand the rigorous testing to which they are subjected while favoring those that do and that, as such, reveal themselves to be, not truer, or even necessarily true at all, but, rather, more fit:
There’s a “problem-situation.” In response to it, rival “tentative theories” are formulated. Each theory is besieged by as many attempts to falsify it as are necessary to either achieve this task or prove its fitness, its pertinence to the problem in question. This is the process known as “error elimination.” Deficient theories are exposed for what they are as the theories that survive the selection process make room for the emergence of ever-more complex problems. And so on, and so on.
Critically, the fitness of any theory with respect to a particular problem-situation most certainly does not mean that it won’t be shown to be false in the future. The point is that a certain type of scientific progress can be said to have been made by way of the process of sorting out fit theories from unfit ones.
Popper was not without his critics. In a future essay, we will see that another giant in the philosophy of science, the physicist-turned-historian and philosopher of science, Thomas Kuhn, had a take on all of this at odds with that of Popper.
Leave a Reply