Many of us know that the term "ontology" applies to the a priori philosophical study of the nature of existence. Ontology is a branch of metaphysics (the attempt to coherently characterize reality a priori in the most general terms possible). So, for example, a lot of rationalists have only one set of things in their ontological inventory (inventory of the kinds of things that exist), namely those states of affairs which can be modeled with the scientific method. Traditionally, such thinkers are known as metaphysical materialists.

Apparently,"ontology" seems to have a different meaning in computer science, namely a shared specification of the types of concepts that can be represented, e.g. what kinds of classes, states of affairs, relationships, and I-don't-know-what-else. (See http://www-ksl.stanford.edu/kst/what-is-an-ontology.html.)

What I'd like to know is, do some linguists attempt to specify ontologies (more or less as computer scientists use that term) for natural languages, either specific natural languages or natural languages in general?


Yes, but the term "ontology" is not always used; it's pretty general, after all, just the Greek for "words about things". In the case of philosophy or science, it means "questions about 'things' -- i.e what are "things"? and what properties do "things" have?

Epistemology is usually studied (and cited) together with ontology, because epistemology tries to answer how we can know about reality, including "things", i.e, how do we know what a "thing" is? and how do we know what properties "things" have? Etc.

Fillmore's linguistic work on Deixis, for instance, is clearly ontology, as is his concept of Frame, programmed as Framenet.

And so is Berlin and Kay's work on color terms, which is squarely based on the structural differences between how humans perceive colors physiologically and how we talk about them.

All of physical ontology -- real "things" -- is based on the human body and its sensory systems, since we need to perceive things in order to be sure they exist, and since the sensations of a human body -- like our experience of position, gravity, movement, and color -- are the only things humans have in common that everyone can depend on others' understanding, too.

And that's why we take "things", and their properties, and their consequences, for granted in natural language; and why all this has to be spelled out in painstaking and pathetically detailed form for computers, which don't have bodies or experiences to refer to.

And why we use body-part metaphors so often. Protagoras got it right:

Παντων μετρον ανθρωπος -- Man is the measure of all things.

Or, as I like to put it, Ontology Recapitulates Physiology.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.