In several writings I have claimed that the basis of knowledge organisation (KO) must be found in subject knowledge, and that researchers and practitioners in KO must achieve knowledge about the domains that they are organising. Domain knowledge is not neutral, but rather is based on competing epistemologies and worldviews, and the classifier is therefore participating in struggles related to worldviews. Different traditions, approaches and paradigms in knowledge organisation research (and practice) can best be understood as more or less associated with one of four epistemologies: empiricism, rationalism, historicism/hermeneutics, or pragmatism/critical theory (of which only the last position fully acknowledges the non-neutrality of knowledge organisation). Ranganathan – and the whole facet-analytic school – has formerly been exemplified as a rather clear example of rationalism. Some have objected to this claim, and Satija (1992), Tennis (2003), Mazzocchi (2013b), Herre (2013), and Dousa and Ibekwe-SanJuan (2014) have each provided important arguments that need to be considered. This paper therefore takes these authors’ studies as the point of departure and examines the arguments that have been raised in relation to my position.