One planet, two names, and a discovery that could have been a tautology.
Reference
The relation between an expression and the thing it picks out.
- Theme
- Logic, language and interpretation
- Tradition
- Analytic
- Term fixed
- 1905
In plain terms
A name picks something out. How it manages that is not obvious.
The natural answer is that a name is shorthand for a description - Aristotle means the man who taught Alexander and wrote the Ethics. Then the name refers to whoever fits.
It cannot be right. If it were, then Aristotle taught Alexander would be a trivial truth rather than a historical fact, and if it turned out he taught nobody the name would refer to someone else entirely, which is absurd. The alternative is that names work by a chain: someone was named, and the name has been passed on, and you refer to that person by standing at the end of the chain - whether or not you know anything true about them.
As a philosopher would put it
Frege introduces Bedeutung alongside sense; Russell's theory of descriptions analyses definite descriptions away, so that the present King of France is bald is false rather than meaningless, and reference failure stops being catastrophic.
The descriptivist theory that dominated until 1970 makes a name equivalent to a cluster of descriptions the speaker associates with it. Kripke's Naming and Necessity dismantles it with three arguments: the modal objection, that the person could have failed to satisfy the descriptions; the epistemic objection, that identifications which should then be trivial are not; and the semantic objection, that a speaker with entirely false beliefs about someone still refers to them.
His alternative is the causal-historical chain: a name is fixed in an initial baptism and passed on with the intention of referring to the same thing. Putnam extends this to natural kinds - water refers to H₂O in virtue of the stuff, not of anyone's beliefs - which yields semantic externalism and the conclusion that meanings are not in the head.
Where you meet it
Babylonian astronomers recorded the appearances of a bright object before sunrise and a bright object after sunset. The Venus tablet of Ammisaduqa, from the seventeenth century BC, records the risings and settings across twenty-one years, and the two were treated as the same body.
The Greeks initially did not. Phosphoros, the light-bringer, appeared in the morning; Hesperos, the evening one, appeared at dusk. The identification is credited variously to Pythagoras and to Parmenides, and it is not a matter of definition - nothing about either name entails the other. It was found out by watching.
That is why Frege used the case. Two names, one object, and a statement joining them that is necessarily true if true at all and yet had to be discovered. If a name were simply a label for its bearer, Hesperus is Phosphorus would say no more than Hesperus is Hesperus, and an astronomical discovery would have been an exercise in restating a tautology.
What is actually disputed
How reference is secured.
The causal-historical account is now standard and has recognised gaps. It says little about how the initial fixing works, it struggles with names whose bearers never existed, and it has no clear story about reference change - names do shift what they pick out over time, and the chain account has trouble saying when.
Descriptivism has not disappeared either. Two-dimensional semantics revives a descriptive component while accepting Kripke's modal points, and defenders argue that the causal theory cannot explain how a chain is anchored without something descriptive at the start.
The deeper worry is Quine's inscrutability of reference. His argument is that no behavioural evidence distinguishes a term referring to rabbits from one referring to undetached rabbit parts, so reference may be indeterminate in a way that no theory of how it is secured can repair.
Commonly confused with
- Meaning 1892 Two expressions can refer to the same thing and mean different things. That is the observation the whole subject starts from.
- Intentionality 1901 Intentionality is the mind's directedness. Reference is a relation of language, and whether one reduces to the other is an open question.
Defined against
- Sense Frege's distinction is precisely that these come apart, and every subsequent theory has to say how
Philosophers
- Bertrand Russell Descriptions, and how to speak of what is not there
- Ludwig Wittgenstein The name-object picture, proposed and later dismantled
- Willard Van Orman Quine The inscrutability of reference
- Rudolf Carnap Reference within a linguistic framework
- John Stuart Mill The early distinction between denoting and connoting
Works
- Principia Mathematica The logical treatment of denoting
- Tractatus Logico-Philosophicus Names as simple signs standing for objects
- Philosophical Investigations The dismantling of that picture
- Naming and Necessity The causal-historical theory
- The Foundations of Arithmetic Frege's setting for the sense-reference distinction
Schools
- Logical Atomism Where names and objects are paired
- Vienna Circle Where reference is framework-relative
- Ordinary Language Philosophy Where the name-object model is questioned
- Cambridge Analysis Where the formal machinery is built
- Pittsburgh School (Sellarsian Realism) Where reference is tied to inferential role
Related concepts
Sense 1892 · Meaning 1892 · Proposition 1903 · Sign and symbol 1868 · Intentionality 1901
Frege's Bedeutung is the origin; Russell's 1905 theory of descriptions is what made reference a technical problem with proposed solutions, and Kripke overturned the resulting consensus in 1970.