Towards a typology of coordinating compounds in Turkish

Although there is a rich literature on compounding in Turkish, coordinating compounds have not received enough attention from researchers in the field of morphology. However, the Turkish lexicon is characterised by a large number of coordinating compounds, belonging to different subclasses; this paper is a first attempt to sketch a typology of these word forms. We will first provide an overview of the different types of coordinating compounds in the language, exploring the limits of this phenomenon, and we will then analyse them in terms of five main parameters: a. semantic relation between the constituents and the referent of the whole compound, b. input and output word classes, c. morphological cohesiveness, d. prosodic cohesiveness, e. degree of conventionalisation/lexicalisation. We will argue that there is a fundamental divide between endocentric and exocentric coordinating compounds; the latter appear as more lexicalised and entrenched in the Turkish lexicon than the former, and also seem to be closer to prototypical words in the morphology-syntax continuum