This study analyzes the editorials in Science and Nature published between 2000 and 2012 about careers in science. Of the total body of documents, 8.8 % dealt with science careers. The editorials were manually classified by topics and then mapped using the VOSviewer. This revealed six easily distinguishable clusters: career conditions in science, the attractiveness of science as a career, merit-based career policies, the effect of research funding on careers, specific groups underrepresented in science, and mobility of scientists. The paper summarizes the main thrust of the arguments in these editorials. There is strong agreement about the problems in scientific careers, but less consensus on the solutions to these problems. The paper also explores whether mapping on the basis of automatically identified terms could have provided adequate results, but concludes that manual classification is needed.
