A common way of discussing generalisation is to search for one conception--a monist view. Another approach is to create a dichotomy between quantitative and qualitative research, each having a single definition--a dualist perspective. A pluralist view is argued for here, i.e. the existence of several lines of reasoning, each of which can be evaluated in specific cases in terms of usefulness, strengths and weaknesses. Five different lines of reasoning about generalization, which could possibly be useful in qualitative research, are presented. They are: "studies that undermine established universal 'truths'", "the ideographic study", "enhancing the generalisation potential by maximizing variation", "generalisation through context similarity" and "generalisation through recognition of patterns". Each is critically commented on.