Describe the distribution of the following pairs of sounds: [p] and [b]; [t] and [d]; [k] and [ɡ]; [kw] and [ɡw]; and [s] and [z

Question

Passamaquoddy (Native American language of the Algonquian family, spoken in Maine and New Brunswick) The following data from Passamaquoddy show a number of different voiced and voiceless consonants. Note that [kw] and [ɡw] are labiovelar stops. (O'Grady et al., 2017).


mɛtka              'he stops dancing'

keɡuw             'all day'

pak                  'fib'

altɛstaɡən        'wooden dish used in dice game'

tʃibilkʷe           'he has epilepsy'

padazikhal       'he makes him choke on smoke'

kʷədaɡən         'his throat'

kadʒiptun        'he takes it by hiding it' satkʷ           '300'

ptəɡʷap            'bag'

 

wabɛjidʒik       'white men' hustiwin 'host'

mitsut              'fork'

hezis                'older brother'

tabaɡən            'sled'

əbəs                 'tree'

alkʷɛbu            'he sits around'

əptan               'woman's coat' litpəzuwin   'tribal council member' tʃiksədəmən  'he listens to it'

Describe the distribution of the following pairs of sounds: [p] and [b]; [t] and [d]; [k] and [ɡ]; [kw] and [ɡw]; and [s] and [z]. Do you expect to have a similar conclusion for all of these pairs? If yes, describe in prose and in feature notation a SINGLE phonological rule that occurs between these pairs. If no, explain why


Details
Purchase An Answer Below

Have a similar question?