A dramatic title for a rather banal point.
I was reading about modal realism recently, and it seemed pretty dumb to me, so here is a quick sketch of a counterargument:
A key concept in studies of modality is that of possible worlds. We say that an event or a state is possible when there is a possible world corresponding to that state. So that if we say something like - “The eiffel tower could have been 2000 metres tall”,whether this is a true statement or not depends on whether there exists a possible world in which the Eiffel Tower (or some object in that world somehow corresponding to the Eiffel Tower in our world - transworld identity is a tricky question!) is 2000 metres tall.
The important thing to note is that we never need to know anything about a possible world to make a modality decision except its theoretical existence. That means that to know facts about modality is to simply have a decision procedure corresponding to whether such a possible world exists.
However, this decision procedure cannot be empirical.To make a decision about a modal statement such as the Eiffel Tower example, we do not first have to jet around in some sort of trans-world spaceship and verify its existence. Indeed, even theoretically, there is no way to interact with any possible world other than our own. They are entirely walled off in separate realities.
Because of this, we can only make the decision using information availiable locally to us, in our reality. This means that our decision procedure, which acn be thought of as a functoin which takes in a state or event and returns whether a possible world corresponding to succh a state exists or not, and from there we decide whether a modal statement is true or false, can be rewritten entirely so that the function just takes in the state of a modal sentence and returns directly whether it is true or not, thus rendering the possible worlds entirely superfluous.
The two such functoins are entirely isomorphic from our perspective, as we can only have information local to our reality, and thus the decision as to the existence of the posible world, and the truth or falsity of the modal statement are the same.
As such, Occam’s razor would command that we should get rid of the concept of a possible world entirely as any sort of causal or metaphysical entity. When we ask questions about modality, we are not questioning the existence of some kind of possible world, we are instead asking about our “modality function”
As the key argument of Lewis in proposing modal realism is the simplicity argument - that is, why do we need to postulate such worls if they do not exist. Needing to postulate them is evidence towards their existence, I reply: I deny the first premise. We do not need to postulate them at all. Thus, his argument falls.
“The modality function” which I refer to is indeed mysterious. However it is far less mysterious than an infinity of existent and metaphysically relevant “possible worlds” which we must add to our ontology. A functoin that uses only local informatoin to compute modality about a state is both metaphysically simpler, and much more in line with reasonabel intuitions about reality, in which the existence of worlds which are entirelly causally distinct from our own can affect the truth or falsity of statements here.