Oxford University logo
Phonetics Laboratory
Faculty of Linguistics, Philology, and Phonetics

Mapping prosodic convergence in Cyprus:

a geo-historical acoustic investigation of the effects of insularity at a linguistic crossroads

Research funded by John Fell Fund award 0011309

PI: Elinor Payne

CoI: Mary Baltazani

Cypriot Greek collaborator: Spyros Armostis

Data Science collaborator: Rainer Simon


Some background...

Throughout history, languages have frequently come into sustained contact with each other as a result of socio-political changes, migration and demographic shift, often creating multilingual societies. While the effects of contact have been studied for a range of linguistic features, much less is known about its effects on prosody, even though prosody may be particularly susceptible to contact influence, given that it is rarely represented in writing and arguably passes more easily under the radar of prescriptivism.

What we are doing...

In this project, we investigate how geographical and temporal factors influence the sharing of prosody across typologically diverse languages spoken within a specific area, namely Cyprus. Analysing audio recordings of Cypriot varieties of Greek, Turkish and Arabic, we test the hypothesis that prolonged, close contact leads to some degree of prosodic convergence. We ask: being an island, and therefore topographically limited, does Cyprus represent a kind of linguistic ‘melting pot’, with its own distinct prosodic signature? We compare patterns with non-Cypriot varieties of these languages, i.e. varieties that are phylogenetically close but geographically distant, as a benchmark.

To read more about our project goals and the historic backdrop to the complex linguistic diversity found in Cyprus, please check out this recent presentation, where we also explain our methodology and report some initial findings:

  • Using quantitative modelling of intonational contours, we compare tunes for polar questions in Cypriot Greek with those of Athenian Greek, confirming previously reported phonological similarity between the two, with some phonetic differences in tonal alignment (Grice, Ladd & Arvaniti 2000). We also find these patterns in Cypriot Greek speakers’ L2 English. (English is widely spoken in Cyprus as an L2).
  • In a follow-up analysis (under review), we examine polar question tunes in Cypriot Turkish, and find both Greek-like variants (presumably due to contact with Cypriot Greek) and variants that are more like Standard Turkish, with distribution likely conditioned by sociolinguistic factors.

Next up...

COMPARING these tunes with polar questions in Cypriot Arabic, a highly endangered language spoken by fewer than a thousand people.

CONDUCTING a more detailed prosodic investigation of geographical and cross-generational variants of Cypriot Greek and Cypriot Turkish, and how these may have been shaped by more recent displacements on the island, as a result of the events of 1974.

CONSTRUCTING a digital historical atlas, incorporating audio files and linguistic analyses, to dynamically depict the geo-temporal backdrop to the prosodic variation we find.

On the horizon...

In many respects, the linguistic and geo-historical complexity of Cyprus can be seen as a microcosm of the larger Eastern Mediterranean area, over which centuries, if not millennia, of geo-political change and human displacement have brought about multiple pockets and layers of linguistic contact, a dynamic which continues to the present day. Research indicates that residues of such contact may be traceable in prosodic patterns, punctuating the otherwise (presumed) ‘smooth’ gradient of geographical variation for any given language (Tavakoli et al. 2019) with discontinuities and convergence clustering effects.

Our investigation of prosodic convergence in Cyprus is therefore also a pilot for a more general investigation of these dynamics, over the wider Eastern Mediterranean. This will involve further languages (e.g. Italo-Romance; Slavonic), topographical spaces (e.g. coastlines, archipelagos), and historical and social factors, and will extend the geographical scope of our digital atlas.

The wider project also involves the development of a convergence index, based on a fusion of acoustic and psychoacoustic measurements, and using a range of methodologies (e.g. perceptual experiments, statistical analysis and machine learning). This will allow us to explore possible correlations with other aspects of human behaviour affected by migration, developed through interdisciplinary engagement with the Pelagios network.


Grice, M., Ladd, R., Arvaniti, A. 2000. On the place of phrase accents in intonational phonology. Phonology 17(2), 143–185.

Tavakoli, S., D. Pigoli, J. A. D. Aston and J. S. Coleman. 2019. A Spatial Modeling Approach for Linguistic Object Data: Analysing dialect sound variations across Great Britain. Journal of the American Statistical Association Volume 114, Issue 527, 1081‒1096. https://doi.org/10.1080/01621459.2019.1607357; Preprint at https://arxiv.org/pdf/1610.10040


We presented our project and our first findings on the comparison of the polar question tune in Athenian Greek and Cypriot Greek at the Multidisciplinary Approaches to Migration and Multilingualism (MAMM2022) conference, held in Nicosia in October 2022. Here's the presentation.