428

falling-together of two paradigms, one with the partial
/-ya-/ in initial (occasionally medial) position in the
suffix, and one without. This is not a case of phono-
logical alternation of /ya/ ~ /:/, as in almost all
cases the /-ya-/ is followed by /:/ mark of Remote
tenses. There are indications that forms with /-ya-/
gloss as more remote (farther in the past) than forms
without it, but at present the contrast is not clear
enough to postulate two separate tenses. In any case,
complete paradigms of forms with and without initial
/-ya-/ were obtained for only one dialect, Jopoqueri,
while most dialects had a mixture of the two forms,

with one favored over the other: Juli has mostly /-ya-/
forms, Socca and Morocomarca have one /-ya-/ form each,
and La Paz has six. One dialect, Sitajara, apparently
has no /-ya-/ forms, as sources interpreted stems with
/-ya-/ as containing /-ya-/ causative. In dialects

that do have /-ya-/ Remote forms, the causative may pre-

cede the Remote.

Chura.ya.ya:t.wa. ‘I made X give to Y.' (La Paz)

Socca is the only dialect for which all forms
without /-ya-/ were obtained without preceding suffixes
in the verb stem. In other dialects, except for 373

/-:na/, RDK person/tense suffixes without /-ya-/ almost