Now that we have seen the problem of Material Constitution and how it is, in effect, a problem of abstraction, we shall turn to the Problem of Many. Coincidentally, the problem of many is another one with which Peter Unger is closely involved — you may recall that I borrowed parts of his eliminativist solution to the problem of material constitution in my previous post.
In the problem of many we are asked to consider a cloud. From the ground, a cloud may have clean, sharply-delineated borders, but of course this is an illusion. When we look more closely, we realize that the cloud is made of many water droplets, and that what appeared at first to be a sharp border is in fact a fuzzy “trailing off” as the water droplets become gradually less dense.
The question then becomes, what is a cloud? If we simply define it as a collection of water droplets in the air then we have two problems. First, our definition is clearly too broad as effectively all air in our atmosphere contains some moisture. Second, and more troubling, is that we also seem to have an enormous number of clouds where there appears to be only one. After all, if I take only the left half of our cloud, that is itself a collection of water droplets, and thus a cloud in its own right. But this trick can be used to create a “cloud” for every possible subset of water droplets, which defies our understanding of clouds. Should we say that the main cloud is composed of millions of overlapping little clouds of every possible shape and size? That seems absurd.
As with material constitution, Unger’s solution is one from which I draw inspiration but do eventually deviate. Unger’s move was to claim that for certain conceptual reasons involving inconsistent definitions, there are no clouds. This, as the Stanford Encyclopedia article notes, is counter-intuitive, and I find Unger’s reasons for his conclusion rather confusing. At this point, my approach in the next few paragraphs should be obvious if you’ve read my previous posts (especially the most recent one on material constitution).
Once again, we are faced with a disagreement not about the nature of the underlying reality (which, for our purposes, can be talked about in terms of water droplets) but about how to define, delineate, conceptualize and refer to that reality. Unger is, in a sense, correct: there are no clouds in the fundamental underlying reality (just as there were no ships in our discussion of the Ship of Theseus). However, when I point in the sky and say “Look, a cloud” I am not talking nonsense; I am referring to the shared abstraction of a cloud, constructed via sociolinguistic negotiation.
In this view, there is one cloud and not millions simply because we all agree that there is only one cloud. If there were two cloud-shaped groups of water droplets very close to one another, it is entirely possible that one person would call them “two clouds” and another person would call them “one cloud with a very thin spot”. It would be wrong to assume that in this situation one of those people would be wrong and one would be right; wrong and right in that sense are a property only of facts about reality, not about our abstractions on top.
As long as you are consistent in your abstractions, you can have as many clouds as you want.