Research on Japanese-English comparative syntax has argued that the Japanese-type languages lack functional categories. Nevetheless, certain syntactic phenomena in Japanese have often been argued to support intricate phrase architecture with multiple layers of functional projections as assumed in the English-type languages. Faced with this apparent contradiction, this paper rethinks the role of functional categories in the theory of parameters from the viewpoint of honorification in the nominal domain in Japanese. The result shows that parametric variation between the two types of languages should be attributed to a more fine-grained difference as to how categories with lexical content are integrated into their syntactic computation.