The ability to measure pain across diverse cultures is important for understanding the universal aspects of pain and expediting nursing intervention. The McGill Pain Questionnaire (MPQ) is the most valid and reliable single multidimensional pain instrument available for measuring pain. Although it has been translated in several languages, most efforts, including two Norwegian translations, have resulted in a variety of new versions, all lacking sufficient faithfulness to the original MPQ to allow qualitative or quantitative cross-cultural comparisons. There is a need for direct translations that maintain the original denotation, connotation and numerical value of the MPQ and thus provide a base for future cross-cultural studies of pain. This paper reports on the development and initial testing of a direct translation of the McGill Pain Questionnaire into Norwegian (NMPQ). The translation process involved three phases: translation, back-translation and consensual. In order to evaluate its performance and validity, the NMPQ was administered to a group of adult surgical patients at two different points in time during the postoperative period. The NMPQ was examined for its feasibility, its sensitivity in detecting decreases in intensity of pain postoperatively and for construct validity. A visual analogue scale was used to check for converging validity, and Spielberger's state anxiety scale was used to assess discriminate validity. The initial testing of the NMPQ with adult surgical patients suggests that the NMPQ is culturally acceptable, relevant, sensitive to fluctuations in pain and numerically consistent with the original MPQ. The moderate levels of validity attend lend considerable assurance to the instruments readiness for use in cross-cultural studies of pain.