In this paper, I defend Szűcs’s (2018) approach to Hungarian finite complement clauses. I argue that the grammatical function COMP is not necessary, and the criticism voiced by Laczkó (2021) can be satisfactorily addressed by considering general principles regarding coordination and the argument-structure of certain verbal and nominal predicates. Laczkó (2021) presents some evidence that seem to support COMP, but I rather propose treating them as (thematic) ADJUNCTs and in the case of simple event nouns, as instances of POSS. Supporting data chiefly cited from Hungarian but theoretical and cross-linguistic considerations are also added. I also discuss some related issues, such as some unresolved issues with regards to adverbials, and the nature of demonstrative pronouns involved in clausal complements. I conclude with some meta-theoretical remarks advocating a generally restrictive view of LFG’s inventory of grammatical functions.
Péter Szűcs (Tue,) studied this question.