and ja:dyba: are derived by adding -skia, -chonma
and -ba to the nominal stems.
|
|
pene + skia peneskia
|
‘money’ ‘to lend’ ‘to lend money’ |
pa:n + chonma pa:nchonma |
‘pan’ ‘to sell’ ‘to sell pan’ |
ja:du + ba ja:duba |
‘magic’ ‘to do’
‘to do magic’ |
|
rcespa + ba rcespaba
|
‘love’ ‘to love’
|
|
The verb rgya:laba ‘to be good’ is derived from the adjective rgya:la ‘good’.
|
|
rgya:la + ba rgyalaba
|
‘good’ ‘to be’ ‘to be good’
|
|
|
Notes
|
1.
The forms which end in -o have alternant forms
like bomon and bax8mon.
|
2. Purki resembles
Hindi in this respect. The ergative suffix-is can be compared
with -na in Hindi.
|
3. This statement does
not hold good in all cases. There are sentences in which the
object under question takes -a. Examine the following
sentences in Purki.
|
|
|
There are only two
verbs rdu s
‘beat’ and rgat ‘love’ in the whole of
data which are found to be exceptions to our generalization.
-a is an allomorph of the morpheme -la, which is homophonous
expressing both dative and locative cases. One possible
solution would be to treat khoa and kheraa
as dative or locative. I am not clear what case should be
assigned to these nouns.
|
4. There are two ways
of analysing the verbal roots which end in -s. One way is to
treat that the past tense is -s. occurs whenever the
verbal roots ends i -s (there are other environments also).
The alternative would be that these varbal roots also take -s
as the past tense maker. This treatment requires a
morphophonemic rule which would be as follows :
|
|
s -> Ę /s-
|
5. The past tense and
future tense suffixes are added to the verbal roots. But the
present tense suffix is added to the verbal stems. The tense
suffixes are not added either to the verbal roots or verbal
stems uniformly.
|
6. This statement is
valid only when the tense is either past or present. Only duk
occurs as the durative aspect when the tense is future. yot
is not found occurring if the tense is future. In my data, we
do not have sentence like,
|
uk khoa r
u
