Where do categories come from?

 

1. Similarity comes in degrees. 
2. But belonging to a kind or sharing a property do not.
3. Hence (from 1 and 2), similarity cannot be belonging to a common kind or sharing a property.
4. Unless we (i) amend our conception of belonging to allow for degrees, or 
5. (ii) we understand ‘being more or less similar’ as ‘having more or less properties or kinds in common.’
6. But (ii) requires kinds to be previous to similarity, (otherwise, the number of kinds in common would depend entirely on the number of objects involved in the comparison and not on their actual similarity). 
7. Thus we must have at least some categories given a priori before comparison. 
8. Psychological evidence backs this hypothesis up: Comparison is a combination of bottom-up and top-down cognitive and perceptual processes (Rehder and Hastie 2001)
9. This raises two fundamental questions: 

• The genealogical question: where do these a priori categories come from? Do they all come from the same place, so to speak?
• The normative question: what are the right categories to ground our comparisons? Is there only one right set of basic categories or is it possible to have many different ontological fundamental systems of categories.

10. The genealogical question is usually answered psychologistically, i.e., studying how we actually categorize, but our apriori categories could also be conventional (a la Carnap 1950) and historical (a la Friedman 1992).
11. If the former, then they may be natural and unchangeable, therefore of less philosophical interest; but if the later, there is room for philosophical critique!
12. The normative question is answered by doing metaphysics. 
13. Some people have proposed that it can be naturalized, looking at natural science.
14. But this seems to assume that what natural science does is not just what we do when we categorize.
15. Otherwise there would be a vicious circularity here.
16. Consequently, we ought to reject the naturalist standpoint in metaphysics.
17. Including the very notion of a natural kind and the natural/artificial distinction.
18. The question is no longer what kinds are natural and which ones are less fundamental/real, but what kinds are relevant for our present purposes.
19. Thus, Valore (2021, 2024) proposes that we must embrace categorization as “a deliberate, purpose-oriented choice” of categories.

My summary of some of the many interesting points in  Paolo Valore's “Ontology and Ordering Strategies: Kantian Legacy within the Quantificational Paradigm”

References:

• Carnap, Rudolf. "Empiricism, semantics, and ontology." Revue internationale de philosophie (1950): 20-40.
• Friedman, Michael. "Causal laws and the foundations of natural science." The Cambridge Companion to Kant 3 (1992): 161-199.
• Rehder, Bob, and Reid Hastie. "Causal knowledge and categories: the effects of causal beliefs on categorization, induction, and similarity." Journal of Experimental Psychology: General 130.3 (2001): 323.
• Valore, Paolo, Maria G. Dainotti, and Oskar Kopczyński. "Ontological Categorizations and Selection Biases in Cosmology: The Case of Extra Galactic Objects." Foundations of Science 26.3 (2021): 515-529.
• Valore, Paolo, and David Witzel. "Ontological Categorizations and Ontological Relativity in Biomedical Sciences. A Case Study." Sophia 16.2 (2024): 423-443.