Referential distinctions

The other day I read Data and Reality, a book from the 1970s pretending to be about databases, but is actually about the philosophy of language, and ontology. Kent splits a lot of hairs around words used to refer to things, and I thought it’d be interesting to collect the kinds of dichotomies I’ve seen. Note that different people might use these dichotomies to refer to different things.

map/territory: The map is not the territory, in the same way that a painting of a pipe is not a pipe. Here, map refers to a mental model of something, and territory is the thing itself. If you point at something on the map, you could be referring to the map, or you could be referring to the place it represents.

signifier/signified: Our words for this dichotomy come from Saussure; in the original French, it’s signifiant and signifié. Signifier and signified both refer to the map part of map/territory. The signifier represents the things drawn on the map and the material it’s made of, and the signified refers to the concept the map conveys, the Platonic ideal form. Terminology-wise, map is to territory as signifier/signified is to referent.

use/mention: The use/mention distinction is the difference between blue being a color, and “blue” having four letters. The same word can refer to the object, or to the word itself. Consider how “orange” is named after the fruit. Here, orange does double duty, the signifier referring to what it signifies, but also to itself.

type/token: How many letters are in the word “letters”? There are seven letters, but only five distinct letters; this is the type/token distinction. The type is the category, the token is a member of the category; before I learned the name for this I called it the prototype/instance distinction. A subtler example is when a white horse is not a horse. Here, the ambiguity comes from types. A white horse, as a token, is of type horse. But the type of white horses isn’t the type of horses. The SEP entry gives many more examples.

intension/extension: Are six feet and seventy-two inches the same thing? They refer to the same length, but their meaning is different. These have the same extension, but a different intension. And note the spelling: they both end in sion, not tion. The set of unicorns and the set of people with Social Security number starting in 000 are both the empty set; does that mean they’re the same set? These words mostly appear when talking about logic or semantics.

sense/reference: This pair we owe to Frege; in the original German it’s sinn and bedeuteung. People disagree about what Frege meant with these words. One possible interpretation is intension/extension. Another is that intension/extension only refers to general properties, and sense/reference refers specifically to names that represent objects. Either way, I think people will agree that Clark Kent and Superman have different senses, even if they could have the same reference.

connotation/denotation: Means something different in linguistics and semantics. In general linguistics, denotation is a literal dictionary definition, while connotation includes other associations. A fire denotes flame, and connotes warmth or destruction. In formal semantics, it’s roughly intension/extension. When people do make a distinction, intension/extension refers to a specific instance, and connotation/denotation refer to the set of all possible intensions/extensions. What makes this even more confusing is that some people translate Frege’s sense/reference distinction as sense/denotation…

de dicto/de re: The sentence “I want to meet the tallest person in New York” has two interpretations: either I want to meet the tallest person in New York, whoever they are; or there’s a specific person I have in mind that I want to meet. This is the de dicto/de re distinction. Compare this with the sentence “The tallest person in New York wants to meet me” which doesn’t have this ambiguity. This distinction appears specifically in propositional attitude reports, sentences that report what people want, believe, or know. Compare Donnellan’s attributive/referential distinction.