Motivation and language behavior: a content analysis of suicide notes.

WHENEVER a person produces a message, whether it be conversation, an ordinary letter to a relative, or a suicide note, he employs a complex set of encoding habits. It seems reasonable to assume that these language habits are organized in much the same way as the habits underlying nonlanguage behavior and that the general principles of learning and performance therefore apply equivalently in both cases. This paper is concerned with the effects of motivation upon language behavior. It is assumed that the author of a suicide note—presumably written shortly before he takes his own life—is functioning under heightened motivation. Therefore, the structure and content of suicide notes should differ from both ordinary letters and from simulated suicide notes in certain ways predictable from a general theory of behavior. Following a brief theoretical discussion, we describe the application of a number of relevant content measures to a comparison, first, of suicide notes with ordinary letters to relatives and, second, of suicide notes with faked notes. Many of these measures differentiate in predicted ways suicide notes from normal control notes; a smaller number differentiate suicide from simulated suicide notes, suggesting that nonsuicidal individuals are able to adopt the state of the suicidal person in some respects but not in others. Language habits, like habits in general, appear to be organized into hierarchies of alternatives. We shall assume that increased drive has two distinct effects upon selection within such hierarchies: generalized energizing effects and specific cue effects (cf., Osgood, 1957, for a more complete analysis). The generalized energizing effects of drives