From the introduction: "The traditional distinction between inflectional and derivational morphology is a useful one, but it is quite complex and difficult to apply to actual forms. With this, as with other linguistic distinctions, there is a tendency to assume that we have an absolute binary distinction, a dichotomy, a hard-and-fast line, with everything on one side purely inflectional and everything on the other side purely derivational. I do not believe the facts fit such a model.
"I would claim that the distinction is better viewed as gradual or scalar along several parameters. In other words, rather than the dichotomy presented in 1.a, [below] the scale of 1.b is a more accurate representation, with its possibility of a morpheme falling in between the two poles on the scale, of one morpheme being more inflectional or more derivational than another, without it being possible to unequivocally call some morphemes either inflectional or derivational. This notion of a scale or cline has affinities to Pike's wave model, which in 1.c. is applied to this distinction. [figures]
"Such scalar distinctions are expected in what has been called the "prototype" model of categorization, which many cognitive psychologists believe more accurately represents normal human categorization than does the simple dichotomous model. Typically there will be several parameters along which contrasting categories will differ;"
Tuggy, David H.
"The inflectional/derivational distinction,"
Work Papers of the Summer Institute of Linguistics, University of North Dakota Session: Vol. 29, Article 8.
Available at: https://commons.und.edu/sil-work-papers/vol29/iss1/8