Scientific generalizations

While scientists often express their findings using quantified generalizations (e.g., “All electrons are negatively charged”), they also communicate using unquantified generic generalizations. Examples include:

  1. Americans overestimate social class mobility.
  2. Sound waves carry gravitational mass.
  3. Infants expect leaders to right wrongs.

In my research, I have focused on several topics related to this use of generic language in science. Given that natural languages, such as English, provide scientists with the tools to quantify, specify, and hedge their generalizations, why do scientists also talk in generic terms about their domain? What is the purpose of uttering generics such as (1-3) versus other – quantified, specified, or hedged – types of generalizations?

To answer these questions, it is important to understand what generic sentences mean. It is well-known that generics are not universal or majority generalizations. In my work, I have argued that they also do not express normality generalizations or probabilistic generalizations, as others claim. Instead, I have proposed that generics express kind-qualifying generalizations and that their content depends on the contextually operative criteria for qualifying as a kind.

For example, my account argues that the scientific generic ‘Americans overestimate social class mobility’ uttered in the context of social psychology says that overestimating social class mobility is one of the properties based upon which the category American qualifies as a kind for social psychology. What this sentence entails crucially depends on the (contextually determined) criteria for scientific kinds in social psychology.

According to this theory, generic language is used in science not to support probabilistic reasoning about the members of a category, nor to facilitate reasoning about the normal members of a scientific category, but to succinctly convey complex content that facilitates kind-based reasoning. Furthermore, this theory also explains the benefits and dangers of using generic language in science. Specifically, it highlights the dangers of miscommunication, especially when scientific generics are read by novices who interpret them according to a folk understanding of kinds.