# If a Theory Can Explain Anything, It Explains Nothing In the [Fabric of Reality](Fabric%20of%20Reality.md) Deutsch argues quite persuasively that if a theory can explain anything, it actually explains nothing. This is tightly linked to the idea of a good [Explanation](Explanations.md). A good explanation is *hard to vary*. By definition this means that it is tightly constrained. It says something *specific* about the world, and if we slightly altered what it said, that alteration would be a *worse* explanation. Consider the explanation of "it was meant to be". This can be used to explain *anything*. Lose your job? It was meant to be. Have an unplanned pregnancy? It was meant to be. Squirt ketchup on your shirt? It was meant to be. The explanation "the gods did it" is of the same form, and they are both exceedingly bad explanations because they can explain anything. But wait, what about [Computation](Computation.md)? Computation is used to explain a wide variety of phenomenon. Does its breadth mean that it actually explains nothing? The same could be said about any of the strands in The [Fabric of Reality](Fabric%20of%20Reality.md). As the scope of what they can explain expands does that mean they start to explain less? Most definitely not! There is a key distinction here between "explaining many things" and "explaining *any* things". If an explanation is hard to vary and explains a broad set of phenomena in a detailed and constrained manner, that is a good explanation. If it can explain anything you throw at it and it is easy to vary, it is an awful explanation. Again take "it was meant to be". It can be applied to anything one would like, and it could just as easily be changed to "the gods willed it to be so". A good explanation should have *consequences*. [Predictions are Consequences of Explanations](Predictions%20are%20Consequences%20of%20Explanations.md). A theory that can explain anything will not have any consequences. This is because *by construction* it was designed to provide no constraints and rule out no possibilities. Again consider the theory "The gods willed it to be so". Why did they will it to be this way? Why not another way? Why these gods and not some other gods? They simply do not have *any consequences* because they have *no constraints*. Now what about the explanation of the seasons that the Greeks came up with? In a sense their explanation was "the gods did it", but in another sense they *massively* improved upon that explanation. Their explanation went like this: Long ago, Hades, god of the underworld, kidnapped and raped Persephone, goddess of spring. Then Persephone’s mother, Demeter, goddess of the earth and agriculture, negotiated a contract for her daughter’s release, which specified that Persephone would marry Hades and eat a magic seed that would compel her to visit him once a year thereafter. Whenever Persephone was away fulfilling this obligation, Demeter became sad and would command the world to become cold and bleak so that nothing could grow. While this explanation is unmistakably false, it still is a massive improvement over "the gods did it". This explanation constrains how exactly the gods did it. It starts trying to provide the *how*. Of course it is incorrect, as the counter example of Australia would show. But it still provides some constraints. And it also can only be applied to seasons. We can't take this explanation and use it to explain free fall. Thus, a good explanation must be specific enough that it has consequences that can be tested or checked. --- Date: 20241118 Links to: [Fabric of Reality](Fabric%20of%20Reality.md) [Explanations](Explanations.md) Tags: References: * []()