I’ve posted previously on the notion of falsifiability in science, as it related to a discussion of the existence of a supreme being. More generally, though, the concept of whether a statement can be falsified is central to making a distinction between science, and, well, not science (sometimes this is called pseudoscience).
Nassim Taleb discusses this idea in reference to his now famous “black swan” example: even if someone observes 1000 swans, all of which are white, this someone cannot feasibly claim that “No swan is black” since he has not seen all of them, nor could he ever truly know if he has seen every last one. The existence of a black swan could, at any moment, render the statement false. Therefore, while it is possible to support the theory “All swans are white,” it is precisely in the nature of a theory that it cannot ever be verified: we will never be able to observe every single swan to be 100% sure they are all white, so we must rely on the fact that we observe some number less than all of them, and have not yet found one which is black.
Taleb explains in the spirit of Popper:
There are only two types of theories:
- Theories that are known to be wrong, as they were tested and adequately rejected ([Popper] calls them falsified).
- Theories that have not yet been known to be wrong, not falsified yet, but are exposed to be proved wrong.
Why is a theory never right? Because we will never know if all the swans are white. […] The testing mechanism may be faulty. However, the statement that there is a black swan is possible to make [since it requires merely one positive observation of a black swan]. A theory cannot be verified. […] It can only be provisionally accepted. A theory that falls outside of these two categories is not a theory. A theory that does not present a set of conditions under which it would be considered wrong would be termed charlatanism – it would be impossible to reject otherwise. Why? Because the astrologist can always find a reason to fit the past event, by saying that Mars was probably in line but not too much so. […] Indeed the difference between Newtonian physics, which was falsified by Einstein’s relativity, and astrology lies in the following irony. Newtonian physics is scientific because it allowed us to falsify it, as we know that it is wrong, while astrology is not because it does not offer conditions under which we could reject it. Astrology cannot be disproved, owing to the auxiliary hypotheses that come into play. Such point lies at the basis of the demarcation between science and nonsense […]
Nassim Taleb, Fooled by Randomness, p. 126-7
So Newtonian physics is falsifiable (turned out to be false), while astrology is not falsifiable, since it is not possible to prove it false. In the context of a supreme being, this is the distinction between the statement “God exists” and the statement “God does not exist”. The former is not falsifiable, since it is not possible to demonstrate the lack of existence of God (or a black swan, for that matter); the latter is falsifiable, for if God exists and can be observed, this would precisely disprove the theory. This is not to say the statement “God exists” cannot have personal, philosophical, or spiritual meaning – but, rather, that it cannot have a significance scientifically equivalent to the statement “God does not exist”. From a scientific perspective, the two statements are not on the same level.
EDIT (1/12/13): An extension on the notion comes from Tania Lombrozo‘s discussion of defeasibility:
On its face, defeasibility is a modest concept with roots in logic and epistemology. An inference is defeasible if it can potentially be “defeated” in light of additional information. Unlike deductively sound conclusions, the products of defeasible reasoning remain subject to revision, held tentatively no matter how firmly.
All scientific claims — whether textbook pronouncements or haphazard speculations — are held defeasibly. It is a hallmark of the scientific process that claims are forever vulnerable to refinement and rejection, hostage to what the future could bring. Far from being a weakness, this is a source of science’s greatness. Because scientific inferences are defeasible, they remain responsive to a world that can reveal itself gradually, change over time, and deviate from our dearest assumptions.
The concept of defeasibilility has proven valuable in characterizing artificial and natural intelligence. Everyday inferences, no less than scientific inferences, are vetted by the harsh judge of novel data: additional information that can potentially defeat current beliefs. On further inspection, the antique may turn out to be a fake and the alleged culprit an innocent victim. Dealing with an uncertain world forces cognitive systems to abandon the comforts of deduction and engage in defeasible reasoning.
Defeasibility is a powerful concept when we recognize it not as a modest term of art, but as the proper attitude towards all belief. Between blind faith and radical skepticism is a vast but sparsely populated space where defeasibility finds its home. Irreversible commitments would be foolish; boundless doubt paralyzing. Defeasible beliefs provide the provisional certainty necessary to navigate an uncertain world.
Recognizing the potential revisability of our beliefs is a prerequisite to rational discourse and progress, be it in science, politics, religion, or the mundane negotiations of daily life. Consider the world we could live in if all of our local and global leaders, if all of our personal and professional friends and foes, recognized the defeasibility of their beliefs and acted accordingly. That sure sounds like progress to me. But of course, I could be wrong.
[EDIT (4/15/13): Here is Margaret Heffernan on the importance of disagreement. Her anecdote in the first 5 minutes of the talk speaks to the central role of welcoming criticism and the search for disproof as a sign of science.]