A study examined the ways that people talk to policy and decision makers about the future and constructed a model of future talk. Specifically, speeches given to government policy makers about science and technology were analyzed in order to discover how people communicate information about the future. Each incident was reviewed and assigned a descriptive label or categories. The category system was developed from the data to reflect the nature of future talk. Categories included non prescriptive predictions, indirect prescriptions (prediction), prescriptions, contrasting time references, comparisons, personal/common vision, problem identification, synergistic references, emphasis on a macro view, future as an entity, emphasis on rate of change, and miscellaneous. These categories were in turn divided into three groups: causal cluster, temporal cluster, and communication cluster. Three predominant themes emerge as characteristics of how people talk about the future. First, future talk relates past and present events in order to foster better understanding and acceptance of the changes to come. Second, cause and effect relationships provide a fuller context for imagining and understanding possible futures in science and technology. Third, talk about the future tends to deal with competitive issues and comparisons. (A figure illustrating a model of future talk is included.) (SRT)