Smith points at Jones at a party and says 'The man drinking champagne is a philosopher,' but Jones is actually drinking tonic water. Using Donnellan's distinction, what is the most accurate analysis?
ASmith's statement is false because Jones is not drinking champagne, so the description fails
BSmith used the description referentially — successfully referring to Jones despite the inaccurate description — and the statement is true if Jones is a philosopher
CSmith used the description attributively, meaning whoever is actually drinking champagne at the party is a philosopher
DThe statement lacks a truth value because the description fails to apply to anyone
In the referential use, the description serves as a conversational vehicle for identifying a specific individual the speaker has in mind. Smith intends to talk about Jones; the description 'man drinking champagne' is just the means of pointing him out. If Jones is a philosopher, the statement is communicatively successful — Smith said something true about Jones even though the description was inaccurate. In the attributive use (option C), the description must correctly apply for reference to occur.
Question 2 Multiple Choice
A philosophy student argues: 'Donnellan's distinction proves Russell's theory of descriptions is simply wrong.' A more careful response would be:
ADonnellan's distinction proves that both Russell and Frege were correct about descriptions
BRussell could accommodate the referential use via pragmatics: semantic content (what the sentence means) can diverge from speaker meaning (what the speaker intends), and the referential use is a pragmatic phenomenon, not a counterexample to the semantics
CDonnellan's distinction only applies to fictional contexts, not everyday speech
DRussell's theory was never meant to account for cases where descriptions happen to be inaccurate
The Gricean/Kripkean response is that the referential use exploits a semantic description for pragmatic purposes. The Russellian semantic analysis still applies to what the sentence *means*; what varies is speaker meaning — what the speaker *intends* to communicate. The distinction between semantic reference and speaker reference preserves Russell's semantics while acknowledging that use outstrips semantic content. Donnellan showed language use is richer than logical form, not that the logical form is wrong.
Question 3 True / False
In an attributive use of a definite description, if no one actually satisfies the description, the speaker's statement refers to no one and has no subject.
TTrue
FFalse
Answer: True
In the attributive use, the description's role is to pick out whoever has the relevant properties — the reference is description-governed. If the detective says 'The murderer is insane' and the death was accidental (no murderer exists), the statement has no subject: it applies to whoever is the murderer, and there is no such person. This contrasts with the referential use, where the speaker has a specific person in mind and successfully refers to them even if the description is inaccurate.
Question 4 True / False
Donnellan's referential-attributive distinction shows that the meaning of a sentence is fully determined by the speaker's intent at the moment of utterance.
TTrue
FFalse
Answer: False
Donnellan's distinction actually highlights the *gap* between semantic content (the sentence's meaning, determined by linguistic convention) and speaker meaning (what the speaker intends to communicate). The same sentence has the same semantic content whether used referentially or attributively — what changes is the speaker's purpose and the context. The point is not that intent determines meaning, but that intent creates a dimension of communicative use that semantic content alone cannot capture.
Question 5 Short Answer
Explain the difference between the attributive and referential use of a definite description, using an example.
Think about your answer, then reveal below.
Model answer: In the attributive use, 'the F' means whoever actually has property F — the description does its full descriptive work. Example: a detective who says 'The murderer is insane' without a suspect in mind means whoever turns out to have committed the crime. In the referential use, the speaker uses 'the F' to pick out a specific individual they have in mind, and successful reference doesn't depend on that person actually being F. Example: someone pointing at an innocent person says 'The murderer must be clever' — they mean that person, not whoever actually committed the crime.
Attributive uses are description-governed: truth conditions depend on who actually satisfies the description, and if no one does, reference fails. Referential uses are person-governed: truth conditions depend on the individual the speaker intends to pick out, and successful reference doesn't require the description to apply accurately. The same sentence can have different truth conditions depending on which use is at play — which is the key evidence that semantic content alone underdetermines what a speaker is talking about.