Logophoricity is the binding relation that holds between a special class of pronouns and their antecedents. Logophoric elements, which occur in embedded clauses introduced by verbs of saying, thinking or feeling, must be bound by the antecedent whose speech, thoughts, or feelings are being reported (Clements 1975, p. 241). The phenomenon was first observed in African languages that have a distinct set of logophoric pronouns that are morphologically differentiated from regular pronouns. Logophoricity is also attested with logophoric reflexives, which are non-clause-bounded reflexive pronouns such as is found in Japanese (Kuno 1972, 1987) and Icelandic (Maling 1984).
The conditions under which logophoricity may occur vary across languages. Both syntactic and semantic approaches have been used to account for logophoricity.
The term logophor was introduced by Claude Hagège (1974) to distinguish logophoric pronouns from indirect reflexive pronouns. In particular, Hagège argues that logophors are a distinct class of pronouns which refer to the source of indirect discourse: the original speaker or individual whose perspective is being communicated, rather than the speaker currently relaying this information. George N. Clements (1975) expanded upon this analysis, arguing that indirect reflexives serve the same function as logophoric pronouns. However, indirect reflexives do not differentiate between logophoric and non-logophoric forms, in contrast with Hagège's logophoric pronouns which fall into a morphologically separate set (Reuland 2006, p. 3). For example, the Latin indirect reflexive pronoun sibi may be said to have two grammatical functions (logophoric and reflexive), but just one form (Clements 1975), as opposed to a language that designates a separate pronoun for each function. More recent analyses of logophoricity are in line with Hagège's original account, under which indirect reflexives are considered to be logophors, in addition to those pronouns with a special logophoric form (Clements 1972, p. 3).