Left-headed compounds in a right-headed language: Evidence from Modern Greek dialects

Angela Ralli¹ & George Chairetakis²

University of Patras^{1, 2}

In recent decades dialectal studies have contributed to theoretical linguistics, allowing scholars not only to test hypotheses on language structures and language properties on a much larger empirical basis, but also to refine them in a non-trivial fashion. Dialects offer a valuable test-bed for the identification of the grammatical principles and have been used to pin down differences and variation patterns that usually escape in theoretical studies which are based on standard languages.

In this presentation, we intend to discuss the issue of headedness in Greek compounding. We will show that although Standard Modern Greek is a clearly right-headed language and its compounds are also right-headed (Ralli 2013), certain Modern Greek dialects show a number of left-headed formations. This is not the first time that these structures attract the attention of linguists. Andreou (2014) has already observed the phenomenon in the Greek-based Cypriot and Grekanico and has claimed that it goes back to Ancient Greek, although it was never productively built.

Our work draws evidence from the Data Base of Dialectal Compounds (DICOMP), which has been developed and stored at the Laboratory of Modern Greek Dialects of the University of Patras (https://www.lmgd.philology.upatras.gr), and contains 13.780 dialectal compounds with their properties, taken from the entire range of Modern Greek Dialects. We will discuss that among the endocentric compounds there is a total of about 120 left-headed structures, and that, interestingly, almost all of them are located in the Aegean and Ionian islands, Crete, Cyprus, and South Italy. See (1) for indicative examples:

(1) Cretan malorupo < mal(i) rup(os)

wool of bad quality wool dirt

Heptanesian afedabelo < afed(is) abel(i)

master of vineyard master vineyard

Cycladic skololambro < skol(i) lambr(i)

feast of Easter feast Easter

Although we agree with Andreou that these structures were not unknown in Ancient Greek, we argue that the language-contact factor has played a significant role in their pattern revival, since the areas where these compounds appear are exactly those which have been under Italo-Romance domination (Ralli 2019), and, as is well known, Italo-Romance compounding is principally left-headed (Scalise 1992). Moreover, we claim that innovative left-headed compounds have also been admitted as possible structural patterns, on the basis that a language can accept innovations due to contact if they correspond to its own tendencies, in accordance with Jakobson (1938) and Ralli (to appear).

We will accompany our argumentation by a detailed information regarding the data of left-headed compounds, as provided by DICOMP.

Selected references

Andreou, M. 2014. Headedness in word formation and lexical semantics: evidence from Italiot and Cypriot. Ph.D. Diss, University of Patras.

Jakobson, R. 1938. Sur la théorie des affinités phonologiques entre les langues. In Actes du Quatrième Congrès International de Linguistes, 48–59. Copenhagen: Munksgaard.

Ralli, A. 2013. Compounding in Modern Greek. Dordrecht: Springer.

Ralli, A. 2019. Greek in contact with Romance. In the Oxford research encyclopedia of linguistics. Oxford: Oxford University Press. https://oxfordre.com/linguistics/browse?page=14&pageSize=20&sort=authorsort&subSite=linguistics.

Ralli, A. To appear. Matter versus pattern replication in compounding: Evidence from the Asia Minor Greek dialectal variety. Morphology (special issue edited by F. Gardani et al.).

Scalise, S. 1992. Compounding in Italian. Rivista di Linguistica 4. 1: 175–200.