If you prefer a video version of this chapter, click here:
A theory is stronger when it is falsifiable.
Good scientific theories are testable, but no theory can be completely tested because scientists cannot observe the future or past. This is one reason why Karl Popper proposed that good scientific theories are those that can be falsified in principle. This means a good theory is one in which we can imagine what would make it false. If we can imagine what would make it false then we can, in principle, test it. We may not have the equipment yet, but we can, in principle, test it.
To clarify, consider the following two theories that are falsifiable in principle:
Theory 1: There is a planet between Mercury and Earth.
Theory 1 is falsifiable because we can imagine what would make it false. For example, imagine we looked through a telescope in the area between Mercury and Earth. Imagine we did this for years and even traveled there, but never saw a planet. If that happened, the theory would be false. Since we can imagine what would make it false, the theory is falsifiable.
Of course, in reality, Theory 1 is true. When we look through a telescope, we see Venus between Mercury and Earth. So, we can say the theory is falsifiable in principle, yet true in reality. Indeed, we are more justified in believing Theory 1 because we can imagine the conditions that would make it false if it were false, and test to see if those conditions are present.
Consider Theory 2: All swans are white.
This is a good theory because one can imagine what would make it false (e.g. observing a nonwhite swan). Unlike the first theory, this theory also turns out to be false since there are black swans.
Notice how these falsifiable theories take risks. We can imagine what would make them false and then test to see if they are false. They are bold theories that dare us to test them. When no tests can disconfirm or falsify them in reality, then we are more justified in believing them. We are more justified in believing them precisely because we know exactly what would make the theory false, but do not find those falsifying conditions in reality.
Now, some theories do not take risks; they are unfalsifiable. This is usually a weakness of these theories.
Theory 3: Nonspatial and nontemporal fairies live inside my nose.
Theory 3 is unfalsifiable because I cannot imagine anything that could make it false. Again, I cannot imagine any test (e.g.looking through a telescope) that would show this theory to be false. Since there is no way to disprove this theory, it is unfalsifiable. This is not a strength of the theory because we cannot really test it. So, most people will not take my fairy theory seriously precisely because it is unfalsifiable.
Theory 4: Everyone is always selfish.
Under some definitions of selfish, this theory is unfalsifiable. If you cannot imagine any conditions where someone could act unselfishly, then we cannot test this claim. If we cannot test it, then we should not be confident that it is true.
In short, the central idea behind falsifiability is that good theories (especially in science) are testable in some way.
However, there are exceptions. Some philosophers argue a few unfalsifiable claims are justified.
Theory 5: I am currently conscious.
There is nothing I can imagine that would make Theory 5 false, but I am more justified in believing that I am currently conscious than I am in believing anything else. I recommend thinking deeply about Theory 5 in order to identify why it is certainly true even though it is unfalsifiable.
There are other exceptions as well, which we will explore in the exercise (e.g. space and time exist, I am conscious, & every possible object either exists or does not exist).
For now, simply remember that a theory is usually stronger if it is falsifiable in principle.
Discuss whether the following claims are falsifiable in principle. If they are unfalsifiable, discuss whether one is justified in believing them.
- For every action, there is an equal and opposite reaction.
- Everyone is always selfish (i.e. every human motivation is exclusively selfish).
- There is something beyond the limits of space and time. Or, space and time exist.
- Invisible gremlins live in my freezer.
- There is something it is like to be me. That is, I am conscious; I am having subjective experiences.
- The car is in the garage.
- For every entity, it either exists or does not exist.
- God exists
- The Earth is 8,000 years old, and humans lived with dinosaurs.
- Yes, Newton’s laws are falsifiable in principle.
- No, it is not falsifiable in principle because no matter what happens, one could list a possible selfish motivation/interpretation. However, this theory is unfalsifiable only if we use a certain definition of selfish. See Psychological Egoism Video on YouTube for a discussion of this point.
- No, neither is falsifiable in principle. Something outside of space/time cannot be observed and would make no difference to space/time. So everything would be the same regardless of whether the first statement is true or false. The same holds for the second statement (arguably) because, even if there were no space/time, people can’t help but live as if there is space/time. Another way to get at this truth is to argue that observation presupposes space so one cannot prove space exists through observation.
- No, this is probably unfalsifiable in principle. If one looks in the fridge, I reply they cannot be seen because they are invisible. If one pokes around, I argue they are too fast to be felt/touched. If one cannot hear them, I argue one must believe they exist to hear them. No matter what test is devised, I have an ad hoc reply to save my theory.
- No, it is not falsifiable in principle. A scientist can infer that you are probably conscious based on brain behavior, but only you know with certainty that there is something it is like to be you. How can you doubt that there is something it is like to be you in the moment you are experiencing something it is like to be you? You cannot make me doubt that I am currently aware. See Chalmers or Searle on consciousness.
- It seems falsifiable in principle. It seems falsifiable because one could build a building with 2+2=5 math to prove that 2+2=5 is false (since the building would collapse). That is, 2+2=4 works in the real world, and 2+2=5 doesn’t. However, deeper issues about the nature of mathematics may arise in such discussions.
- It is falsifiable in principle. Go look in the garage. If you see nothing, it is probably false.
- It is not falsifiable in principle. Logic commands that things either exist or do not exist. What other option could there be? Notice that even though it is unfalsifiable, it is rational to believe things either exist or do not exist. This is one limitation to falsifiability. Of course, many of these examples are not "scientific" theories, which is what falsifiability was originally about. Still, many people attempt to apply falsifiability beyond its scope and so all these examples are relevant.
- God exists is a difficult one because it depends on how you define God. If God is a timeless, spaceless being that is also omnipresent, then there’s nothing empirical that could falsify a person’s belief in God. If God is essentially meaning or purpose, then there’s nothing that would falsify belief in God since there is no way to measure or test for meaning and purpose. However, if God is a physical being that, say, lives behind some planet, then the belief in God is falsifiable in principle. We can imagine visiting that area to see if God exists. Also, if you claim your conception of God is logically consistent and based on such consistency, then a logical inconsistency may disprove such a belief.
So, does the unfalsifiablity of belief in God make the belief weak? Not necessarily. After all, “I am currently conscious” is unfalsifiable, but I am more justified in believing that than I am anything else. The key is to think about how God would be known if God existed.Since people have different concepts of God, some would be testable and some would not be testable.
- Yes, we can imagine what would make these statements false. The first is probably false if we use techniques like carbon dating and find something older than 8,000 years old. Also, we can calculate the rate at which a river is carving through rock and use its depth to calculate its age. If it’s older than 8,000 years old, then it’s false. There are many imaginable conditions that would make this false. Tests verify that the earth is much older than 8,000 years old.As for humans and dinosaurs, we can look at the fossil record. If we never find humans living with dinosaurs, the theory is most likely false.
Application and Value
A theory that is falsifiable in principle seems stronger than one that is unfalsifiable in principle (assuming all other qualities of the theories are equal). This is especially true in science.
However, some foundational beliefs seem to be true even though they are unfalsifiable, but it is important to remember that they are not justified because they are unfalsifiable. In short, one can use the falsifiability test to evaluate many claims and theories. If it is unfalsifiable, we need to give very strong reasons for why the theory is justified.