Generics are generalizing statements of the form ‘Fs are G’, like ‘Birds fly’ for example. They express something about the characteristic properties of a kind and its members, without the use of an explicit quantifier. That is to say, generics do not use a quantifier to generalize a property, like one does in ‘Most birds fly’ or ‘Birds mostly fly’. Instead, we interpret generic statements as generalizations in the absence of any explicit quantifier. Despite the ubiquity of generics in our language use, it has proven very difficult for theorists to spell out their truth-conditions. To see the complexity of the problem, consider the following examples:
- Tigers are striped.
- Birds lay eggs.
- Sharks attack bathers.
- Books are paperbacks.
One reason for the difficulty of spelling out the truth-conditions of these generics is immediately clear; they allow for many exceptions. We intuitively judge that statement (1) is true, but some tigers are albino’s. Furthermore, we also think that (2) is true, yet only the female and fertile birds lay eggs. Similarly, it seems true that ‘Sharks attack bathers’, although in fact few of them ever do. This does not mean that generics are just loose generalizations, however. Even though most books are in fact paperbacks, (4) is false.
According to some proposals, generics state that normal members of the kind have the property. Others believe that generics express something about the probability of a member of the kind having a property. The details of the most promising semantic theories are intricate and complicated, but none of them are without counter-examples. One goal of this project is to defend a new, epistemically motivated semantic account of generics. For a draft of my answer to this problem, see here.
Finding a correct semantic theory about generics is interesting in its own right, at least for philosophers of language and for linguists. But this debate also has wider implications. Some of the most objectionable stereotypes have a generic form. Consider statements like:
- Muslims are terrorists.
- Women are submissive.
It is clear that these statements are objectionable, but why is that exactly? Perhaps one would want to say that these are false overgeneralizations. But given that generics can often be true despite the fact that there are many counterinstances, it is not clear whether we should really think of them as simply overgeneralizing a property. Furthermore, if generics can be true even when most members of the kind do not instantiate the property, it’s doubtful that one could falsify one of these stereotypes simply by providing evidence of counterexamples.
A second goal of this project is to spell out what exactly someone claims by uttering generic stereotypes, and how one can best falsify their statements if that happens. To get a sense of my answer to these questions, I hold that generics say that a property is characteristic of a kind, but I also believe that there are different ways for a property to be characteristic of a kind. So one strategy in aiming to falsify a generic is to provide evidence that a property is not characteristic of a kind in any of these ways. But there are also some conditions that no generic can satisfy while still being true. Hence a second and often more efficient strategy to falsify a generic stereotype is to show that one of these conditions holds.