A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The de re / de dicto distinction concerns the scope of descriptions relative to intensional contexts (belief reports, conditionals, modal statements). De re readings quantify over actual individuals (the description refers to a specific thing); de dicto readings quantify over properties (the description's content matters, not the actual referent). 'John believes the mayor is corrupt' can report John's belief about a specific person (de re) or about whoever happens to be mayor (de dicto).
Work through belief reports with definite descriptions, distinguishing readings by whether the speaker commits to the existence and identity of the described individual. Test predictions by considering what happens if the description's actual referent changes.
From Montague semantics and formal quantification, you know that sentences are interpreted by assigning them truth conditions relative to models — sets of individuals, a world, and an interpretation function. You also know that quantifiers like *every*, *some*, and definite descriptions like *the mayor* interact with other operators through scope: *Every student passed some exam* differs in meaning depending on whether *every student* or *some exam* takes wide scope. The de re / de dicto distinction is what happens when this scope interaction occurs specifically inside intensional contexts — constructions like belief reports, desire attributions, and modal statements that are evaluated not at the actual world but across possible worlds.
Start with the contrast. Consider: *John believes the mayor is corrupt*. The phrase *the mayor* could be picking out a specific individual — say, Jane Smith — in which case the sentence attributes to John a belief about Jane Smith herself, regardless of whether John knows she is the mayor. This is the de re reading: the Latin means "of the thing," referring directly to an object in the world. Alternatively, *the mayor* might be functioning as a role description — John's belief is that whoever holds the office of mayor is corrupt, and John may have no particular individual in mind. This is the de dicto reading: "of the saying" or "of the word," where the content of the description is what matters, not the actual referent.
In the formal semantics you know from Montague grammar, this distinction is captured through scope. A de re reading arises when the definite description takes wide scope over the intensional operator (in this case, the belief verb): for some individual x, x is actually the mayor, and John believes x is corrupt. The quantifier ranges over actual individuals before the intensional context is entered. A de dicto reading arises when the description takes narrow scope, falling inside the intensional operator: John believes [the mayor is corrupt], where the description is evaluated at the belief world, not the actual world. The two logical forms yield genuinely different truth conditions. The de re reading is true only if John has a specific (actual) person as the target of his belief. The de dicto reading is true whenever John's belief world contains a corrupt mayor, even if no such person exists in the actual world.
The philosophical significance extends beyond the formalism. De re and de dicto attitudes raise the question of intentionality — the "aboutness" of mental states. If John believes de dicto that the mayor is corrupt, his belief is about a description, a role. If he believes de re that a specific person is corrupt, his belief is about an individual in the world, even if he has the description wrong. This matters for cases of mistaken identity: suppose the mayor is actually Jane, but John thinks the mayor is Bob. In the de dicto reading, John's belief concerns whoever is mayor; in the de re reading, John's belief is about Jane specifically, whether or not he knows she's mayor. The same sentence can attribute radically different mental states depending on which reading is intended.
One practical test: change the actual world (or the facts about who fills a role) and see what happens to truth. If the sentence *John believes the mayor is corrupt* is true de re because it's about Jane, and Jane resigns from office tomorrow, the de re belief attribution remains true — John still has that belief about Jane. But the de dicto attribution tracks the role: if a new, honest mayor takes office and John updates his belief accordingly, his de dicto belief might become false. The distinction between tracking individuals and tracking descriptions runs through philosophy of language, epistemology, and the semantics of propositional attitude reports — and the formal tools of scope and possible worlds make it precise.