Concepts are not static
Appendix D endnote 20, from Lisa Feldman Barrett.
Some context is:
Even the highest-level, multisensory summary of Happiness,” represented by sets of neurons in the default mode network, can be different each time. [...] Within a concept, there can be several different goals.
The same concept can be constructed in the service of several different goals, none of which are core (chapter 5). The concept "Bird" can involve a goal to fly, to eat, or to have a pet. The goal of "Awe" can be to inspire, to challenge your worldview, to gain compliance, or to put things in perspective. None of these goals is more “core” than any other. And for a given concept, certain instances will meet the requirements of a certain goal better than others. A concept has no core features, just features that are relevant to particular goal in a particular context.
While it is easy to mistake a goal for the core of a concept, we should try not to. A concept is not a “thing” that exists in the brain, any more than a “species” is a static dividing line, or that “space” is a physical thing that the universe expands into. “Concept,” “species,” and “space” are theoretical ideas. Following Darwin, a species was a theoretical, goal-based concept, consisting of a population of uniquely different individuals. Any summary of this species — the average of the population, for example — is an abstraction over all the variation. A species is goal-based, because it is defined by a function — which individuals breed successfully with each other — rather than by similarities in their perceptual features like their body structures.
It is a verbal convenience to talk about “a concept" as if it is a thing, an object, rather than as a remembered construction, or an event. The tyranny of English is that it invites essentialism.
In reality, a human brain comes equipped with the ability to learn patterns. With the help of words, intentionally spoken by other people in rich social environment, that brain can find commonalities — can create patterns that would not otherwise be there — and then transmit those patterns to others. The word “dog” helps babies see the commonalities (a summary) across a tiny Chihuahua and a huge Saint Bernard. The made-up word “toma” allows three-month-old babies to summarize across different dinosaurs that vary widely in color and shape. And the word “anger” can presumably allow children to summarize goals across instances as well. But these summaries are not "the" concepts. When I write “you have a concept for anger” this translates as “you have many instances that you have categorized, or that have been categorized for you, as anger, and each is captured as a pattern in your brain, and you can reconstruct these, or combine them in novel ways.” The "concept” refers to all the knowledge you have about anger in your conceptual system that you have acquired over your lifetime. When you “use a concept,” you are really constructing an ad-hoc instance of a concept on the spot.
Notes on the Notes
- E.g., for a discussion, see Mayr, Ernst. 2004. What Makes Biology Unique? Considerations on the Autonomy of a Scientific Discipline. New York: Cambridge University Press.