This project aims to investigate the synchronic properties of five apparently related syntactic configurations in three genetically related Modern Greek dialects: (i) Cappadocian, (ii) Pharasiot, and (iii) Pontic. Until 1923 (i-iii) were spoken in Turkey. After 1923 the speakers were relocated to Greece due to the Greek-Turkish population exchange. With 2800 and 25 speakers respectively, (iii) are on the verge of extinction today. The configurations to be examined are (1) headed relative clauses (cf. English 'the book that I saw'), (2) free relative clauses (what I saw), (3) sentential subjects ([that I saw the book] was a lie), (4) complement clauses to verbs (he knows [that I saw the book]), and (5) complement clauses to prepositions (for (the fact) [that I saw the book]). Our knowledge about (1-5) and the relation among them across (i-iii) is rather fragmentary and is based on data from small size corpora dating before 1923. In this project, I first intend to offer a synchronic documentation of (1-5) across (i-iii) with data to be drawn from spoken corpora collected after 2000. Second, I intend to offer a comparative syntactic analysis of (1-5) in (i-iii) against the framework of generative grammar, with elicited data based on native speakers' intuitions. The documentation and analysis of (1-5) will provide us with a deeper understanding of the structural affinity among (1-5) and enable us to locate the notion of genetic relatedness among (iiii) on empirical grounds.