Professor Aditi Lahiri’s latest project started on October 1st, 2022 and investigates the principle of Pertinacity - or persistence. This project was selected by the European Research Council (ERC), and is funded by the Engineering and Physical Sciences Research Council (EPSRC, UKRI).
Centuries of linguistic research has stated that phonological change is inescapable. Some have claimed syntax to be inert, changing only due to phonological or semantic change. This project takes an unusual and challenging view that phonology is pertinacious and changes in existing phonological systems are governed by two principles of Pertinacity:
Either a particular phonological pattern persists but is extended to apply to new forms and different outputs emerge: [A] same pattern, different outputs
Or output forms look alike, but the underlying phonological system alters due to changes elsewhere in the grammar: [B] different pattern, same outputs
The Pertinacity project will allow us to set out our expectations, and better understand the reasons behind the whys and why nots of phonological change. Classical historical research will be combined with psycho- and neurolinguistic experimentation and computational speech recognition to explore the central issues of linguistic change and stability, diversity and uniformity.
We plan to investigate similar internal phonological processes and external loans via contact, across 3 language families (West & North Germanic, Indo-Aryan), and exploring 3 main themes:
A. THEORETICAL: DIACHRONIC & SYNCHRONIC ANALYSES
Internal: Determine the extent to which comparable phonological processes lead to different phonological contrasts in related languages
External: Determine the effect of adaptation of similar loans on grammars of related languages
B. PSYCHOLINGUISTIC & NEUROLINGUISTIC EXPERIMENTATION
Language processing: Investigate experimentally to what extent similar underlying patterns with differential outputs govern phonological processing in related languages
C. COMPUTER SPEECH RECOGNITION
Simulate the effect of different representations with our Flexible Speech Recognition system (FlexSR)
