Lenition is common in both phonological and sound change processes. In opening processes, which involve increasing aperture, labial and dorsal consonants tend to retain some labial or dorsal constriction even if they fully vocalise. It is less straightforward what happens with coronal consonants that vocalise, since vowel production rarely involves the tongue tip or blade. Vocalization of coronal consonants often involves a reconfiguration of, or loss of, the coronal gesture, as in approximants like [j ɹ]. Danish has historically seen extreme lenition in the unaspirated stop series /b d g/ in prosodically weak positions, a process which has had fascinating phonetic and phonological consequences, not least because /d/-vocalization has not obviously led to a reconfiguration of the coronal gesture.

In Modern Standard Danish, the reflexes of /b d g/ in prosodically strong positions are voiceless unaspirated [p t k], whereas in weak positions, the reflexes are typically semivocalic. /b/ somewhat inconsistently surfaces as [ʊ̯] or [p] in weak positions. /d/ surfaces as the ‘soft d’, a sound which is often transcribed as [ð], but is well-known to be more open than a fricative. /g/ either surfaces as [ʊ̯ ɪ̯] or elides in weak positions, determined mostly by the identity of preceding vowels.

This presentation will present and discuss the outcome of stop vocalization in Modern Standard Danish, with particular focus on /d/-vocalization. Several open questions remain about the stop vocalisation process, including: (1) What exactly is the acoustic and articulatory outcome of /d/-vocalization? I will present analyses of spontaneously produced corpus data and recently collected articulatory data using electromagnetic articulography probing this question. (2) How does stop vocalization unfold in the long and short term? This will be discussed with reference to historical written sources, and I will outline the goals of a recently started project aimed at tracking and analyzing the relatively recent past of /d/-vocalization in a large longitudinal speech corpus. (3) Does phonetic and phonological evidence support an analysis where [p t k] are linked to semi-vowels in the synchronic phonology of Danish? This question will be discussed in light of the new data.