Blurb:

As the first case-study I take a domain that has attracted considerable interest in linguistic typology, and which epitomises the inadequacies of the “atheoretical” or “theory neutral” approaches typical of present day typology. This is the domain of grammatical roles and clause transitivity. I outline the received approach to these domains in linguistic typology, which assumes three universal primitives, S, A, and O, and attempts to account for grammatical relations cross-linguistically in terms of so-called alignment theory, which is concerned with how S, A, and O are grouped together in particular languages. This approach is subjected to critique (much of which has been common knowledge for a very long time, though conveniently swept under the carpet), and an alternative is suggested that begins with language-particular categories. I demonstrate the advantages of this approach, and discuss how typology can be done from this perspective.