Jackson's concept of "problem frame" has influenced requirements engineering (RE) by emphasizing the constituent problems that system requirements address. But RE is itself a problem, one that may be supported by implemented systems for which, presumably, RE is the real-world domain in which the system operates. Even when the intention is not to specify RE support technology, the concept of the problem frame illuminates the assumptions that inform much RE research and practitioner literature. This paper analyzes the contribution to RE of the major problem frames identified by Jackson. Notable RE papers are analyzed through the lens of problem frames, and it is shown that the research and professional community frequently talks at cross purposes because the problems being addressed are literally framed differently and the success criteria for RE are therefore defined in ways that do not permit comparison.