Secondary articulations in Tashlhiyt Amazigh
Secondary articulations in Tashlhiyt Amazigh are the phenomena I am investigating in my PhD thesis. Tashlhiyt is an Amazigh language mainly spoken in Morocco. It has two distinctive secondary articulations in its phoneme inventory: Labialization in the subset of dorsal consonants and Pharyngealization in the entire set of coronals.
Another particularity of Tashlhiyt is the fact that it makes strong use of consonant clusters. This is of great importance, because secondary articulations are usually signaled on the vowels surrounding the modified consonant. This raises the question in which way these distinctions are phonetically implemented, if the preceding, the following, or both segment positions are occupied by consonants.
In my work I investigate the production of these secondary articulations in different contexts, these are: in intervocalic position [V_V], in post-consonantal and pre-vocalic position [VC_V], in post-vocalic and pre-consonantal position, and in inter-consontal position [VC_CV]. I rely on Electromagnetic Articulography (EMA) as well as acoustic measurements in my research. In the final step, I use these measurements and analyses to model the plain/labialized and plain/pharyngealized contrasts.
EMA data analysis
At the beginning of my PhD I was introduced into the realm of Electromagnetic Articulography, a technique that I am using in my phonetic investigations of secondary articulations. I am analyzing EMA data without the usage of external programs, but with own custom scripts for the data extraction and processing,e.g., landmark detection and measurements. One output is a tool that makes EMA data available to non-technical users, a converter called ema2wav that can be used to transform this data into wave files in order to use PRAAT for the display and analysis. See the GitHub repository for more information. I am also currently developing an open-source editor tool for the display, analysis and processing of EMA data.
When I started my work on secondary articulations, I was curious about the distribution of secondary articulations in the languages of the world. Although some information is available based on the phenomenal work of Maddieson (1984), nowadays larger datasets of the phoneme inventories of the languages of the world are available have not yet been analyzed in this regard. In my work on quantitative phonology, I rely on freely available datasets like PHOIBLE for phoneme inventories and GLOTTOLOG for geographical information/information on the association to language families. One result is an overview of secondary articulations in the languages of the world in this poster presentation. See in the map below an overview of secondary articulations in languages across the world.
I am involved in several other projects where I assist mainly with technology-related matters. My collaborations include:
- Investigation of palatalization in Italian by Maria Giavazzi, Bowei Shao and Sam Mitchell from ENS-PSL, Paris, France (forced-alignment, stimuli preparation, EMA data processing)
- Phonetics of stress in Kambaata by Yvonne Treis from INALCO, Paris, France (alignment, phonetic analysis)
- ema2wav, a conversion tool for EMA data, together with Simon Roessig from Cornell University, Ithaca, USA, and with Doris Mücke and Lena Pagel from the IfL Phonetics, Cologne University, Cologne, Germany and Anne Hermes.
- Various aids in relation to data processing (EMA, acoustics, text) and statistics within my lab