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


Greek in Contact

UPDATE: We have been awarded an ESRC grant (ES/R006148/1) for a 42-month project Intonation and diachrony: a phonetic investigation of the effects of language contact on intonational patterns


We are extending the scope of our work to include more contact varieties of Greek.


We presented the results of our work on Cappadocian Greek polar questions and continuation rises at the TIE 2018 conference in Gothenburg. Our results indicate that with respect to those two melodies, Cappadocian, a Greek dialect historically in long term contact with Turkish, resembles Turkish, not Athenian Greek. Our poster is available  here


Cretan Greek 


Our most recent work investigates the implications of historical interaction of Greek with Italian and Turkish speakers for intonation change in Cretan speech. We compare the declarative falling melody of Standard Modern Greek (SMG, as spoken in Athens) with that of Cretan Greek (CG) and Italian, the latter two chosen because their speakers historically interacted.


 Although it is impossible to demonstrate that Venetian as spoken 400 years ago had the same intonation as modern Italian, the fact that the declarative fall pattern is currently found in geographically separate contemporary varieties is likely to be due to contact factors. We argue that this similarity is not coincidental but stems from daily contact of Cretans with Italian speakers during the long period of Venetian rule of Crete.


Declarative falls (arrows show the accentual syllable) in Cretan and Italian show similarities: relative to the accentual vowel, they display an early peak, in contrast to SMG, which has a late peak.

Functional Data Analysis confirms the AM comparison. Left: The Cretan (**) and Italian (••) fall starts earlier and is much steeper than in SMG (°°). Right: The slope and curvature of CG pattern with Italian and have a higher quadratic curvature than SMG.

Have a look at our poster presentation for LabPhon16 and the recordings of the declarative fall presented in section 4 of the poster: Cretan Greek, Italian, Athenian Greek



Asia Minor Greek 



Anatolian Greek Dialects


Anatolian Greek dialects until 1923. Demotic in yellow. Pontic in orange. Cappadocian in green, with green dots indicating individual Cappadocian Greek villages in 1910. Source: https://en.wikipedia.org/wiki/Cappadocian_Greek#/media/File:Anatolian_Greek_dialects.png

This project forms part of a wider research agenda seeking to determine how historical change and language variation arise out of language contact situations. The goal of the present pilot is to build a corpus of archival recordings of Asia Minor Greek, a contact variety showing a mixture of Greek and Turkish features. The analysis of the data will be carried out in the later stages.  

We have built up an 85-hour audio corpus of 188 speakers of Asia Minor Greek varieties (Cappadocia, Pontus, Istanbul, Smyrna) going back to the late 19th century. For comparison, we also have audio of 66 hrs of Standard Modern Greek and 2½ hrs of Turkish speech. Our acquisitions come from institutional holdings: Bibliothèque Nationale de France, Humboldt Universität Lautarchiv, Academy of Athens and online platforms such as Epitropi Pontiakon Meleton and YouTube. A number of academics have also shared their field recordings. The resultant corpus constitutes a substantial resource for diachronic investigations of Greek speech, with the birthdates of informants spanning a century. The oldest material, recorded in 1917, features speakers born in the 1890s. The most recent recordings capture the speech of third generation bilingual Asia Minor Greek speakers born in the 1990s. The data comprises a range of speech styles, from word lists to dyadic conversations. 
Research questions:  
  • What is the role of language contact in intonation change? 
  • How long can contact influences last in the recipient variety? 
  • Which aspects of intonation are subject to variation? 
  • Does intonation show the same principles and processes as segmental change? 
To achieve this goal, we compare the intonation of the donor languages (Greek and Turkish) with Asia Minor Greek. We hope to contribute to a better understanding of contact Greek-Turkish speech and yield new data for modern historical studies of this region. Finally, the incorporation of languages from increasingly wider geographic areas also supports subsequent explorations deeper into the past, potentially allowing the reconstruction of proto-intonation for some branches of Indo-European. We will seek funding for a project to trace the historic development of intonation in Greece and its neighbours.
In the pilot we focus on continuation rises (comma intonation) with a view to extending our investigations to other functions of intonation. Below are some examples of continuation rise melodies from our speech corpus. Here, the Asia Minor Greek contour is a rise-fall-rise, as is the Turkish sample, but the Modern Standard Greek contour is a low level followed by a rise.

Continuation rises in Asia Minor Greek vs. Standard Greek and Turkish

Audio clips: Asia Minor Greek (magenta), Turkish (red), Modern Standard Greek (blue)
(These open in a new tab. If the audio does not play immediately, reload the page.)
Further examples:

Asia Minor Greek: AMGr1, AMGr2, AMGr3, AMGr4

Turkish: Turkish1, Turkish2, Turkish3, Turkish4

Standard Modern Greek: SMG1, SMG2, SMG3, SMG4


We presented a poster at Modern Greek Dialects and Linguistic Theory Conference, Rhethymno, October 2016. The full paper  will be available from MGDLT 7 online proceedings.


Investigators: Mary Baltazani, Joanna Przedlacka, John Coleman

Supported by an award from Oxford University's John Fell Fund 152/075