The article begins by exploring briefly the role of the elderly in sociolinguistic theory and research. After an outline of the parameters of speech accommodation theory together with a new schematic model, it is argued that speech accommodation theory is a profitable framework for elucidating the sociolinguistic mechanics of, and the social psychological processes underlying, intergenerational encounters. A recent conceptual foray in this direction, which highlights young-to-elderly language strategies, is then overviewed with some illustrations. Contrastive data from a case study are then introduced, a discourse analysis of which allows us to conceptualize various elderly-to-young language strategies. This interpretive analysis suggests important avenues for extending speech accommodation theory itself. A revised, more sociolinguistically elaborated version of this framework is then presented which highlights strategies beyond those of convergence, maintenance, and divergence and leads to the conceptualization of over- and underaccommodation. Finally, and on the basis of the foregoing, a new model of intergenerational communication is proposed and Ryan et al.'s (1986) “communicative predicament” framework duly revised. (Accommodation theory, elderly, overaccommodation, case studies, discourse management, stereotypes, underaccommodation, interdisciplinary)

This article critically examines contemporary interactional studies of the cultural specifity of human language, conducted mostly in modern, multiethnic, industrialized societies (e.g., Clyne 1979; Gumperz 1982a, 1982b; Valdés & Pino 1981). What is often presented as the “linguistic evidence” for miscommunication in such contexts is in fact, we argue, the locus of the violations of the cooperative principles of discourse and human interaction, such as the Principle of Charity (Davidson 1974) and the Principle of Humanity (Grandy 1973). The conclusions these studies arrive at are vitiated by the fact, for which considerable empirical evidence exists, that the native speaker's repairability threshold depends crucially on nonlinguistic variables (Hackman 1977). Only a cross-cultural analysis of how or whether these misconstruals entail analogous consequences, regardless of who is being misunderstood by whom, can, we argue, produce the sort of evidence these studies claim to unearth.

Literature on language death offers abundant information on the grammatical, phonological, lexical, and sociolinguistic processes that a dying speech form can undergo. However, work remains to be done in the area of narrative skills and performance. This article examines the creative manipulation of certain narrative devices, including bilingual lexical resources from modern Greek and Tosk Albanian in stories offered by a residual group of fluent Albanian speakers in Greece. In a community, which is highly variable from the point of view of the allocation of its Arvanítika (Albanian) linguistic and sociolinguistic skills to various population segments, fluent speakers manage to achieve a significant performance breakthrough by foregrounding and evaluating information important for their attitude building and vital to their social existence. Narrative performances become ways of relating historical events and past experiences in present-day life. The article makes the point that in studies of language death a more intensive use of the ethnography of speaking paradigm can be of great value for the detection of sensitive areas of speech behavior and change. (Narrative performance, language death, ethnography of speaking, Balkan sociolinguistics, Tosk Albanian, Greek)