UC Santa Barbara
Himalayan Linguistics
Title
Making and agreeing to requests in Old Tibetan
Permalink
https://escholarship.org/uc/item/6bq059cw
Journal
Himalayan Linguistics, 21(1)
Author
Hill, Nathan
Publication Date
2022
DOI
10.5070/H921156970
Copyright Information
Copyright 2022 by the author(s).This work is made available under the terms of a
Creative Commons Attribution-NonCommercial-NoDerivatives License, available at
https://creativecommons.org/licenses/by-nc-nd/4.0/
Peer reviewed
eScholarship.org
Powered by the California Digital Library
University of California
Himalayan Linguistics
Making and agreeing to requests in Old Tibetan
Nathan W. Hill
Trinity College Dublin
ABSTRACT
The verbs གསོལ་ gsol ‘request’ and གནང་ gnaṅ ‘agree, grant', because of their complementary semantics and
parallel syntax, provide a convenient window through which to caste light on the two forms of subordinate
clauses that they both govern, namely infinitives and terminative verbal nouns.
KEYWORDS
Old Tibetan, switch reference, subordination
This is a contribution from Himalayan Linguistics, Vol. 21(1): 29-39
ISSN 1544-7502
© 2022. All rights reserved.
This Portable Document Format (PDF) file may not be altered in any way.
Tables of contents, abstracts, and submission guidelines are available at
escholarship.org/uc/himalayanlinguistics
Himalayan Linguistics, Vol. 21(1). © Himalayan Linguistics 2022
ISSN 1544-7502
Making and agreeing to requests in Old
Tibetan
Nathan W. Hill
Trinity College Dublin
1 Introduction
The verbs གསོལ་ gsol ‘request’ and གནང་ gnaṅ ‘agree, grant’, have clearly complementary
meanings; those addressed by requests either grant them or not. In addition, གནང་ gnaṅ is an honorific
verb and གསོལ་ gsol ‘request’ is a humilitic verb (Kitamura 1975). The verbs གསོལ་ gsol ‘request’ and གནང་
gnaṅ ‘agree, grant’ also share a parallel syntactic ability to govern subordinate clauses. 1 These
subordinate clauses take two forms, which I call ‘terminative verbal nouns’ and ‘infinitives’.2 The
verbal nouns are those subordinated verb forms nominalized with the suffix པ་ -pa or བ་ -ba, to which
the terminative case marker ར་ -r is added. The infinitives are subordindated verb forms followed
directly by དུ་ -du and its allomorphs. 3 Two verb stems make themselves available to function as
infinitives, namely the present and the future. Terminologically we can thus further distinguish ‘future
infinitives’ and ‘present infinitives’. The verb གསོལ་ gsol ‘request’ governs the future infinitive and གནང་
gnaṅ ‘agree, grant’ governs the present infinitive (Garrett et al. 2013). To my knowledge the grounds
on which verbs select either infinitives or terminative verbal nouns, and why some verbs govern
present infinitives and others future infinitives, has not yet been discussed in print.
This essay attempts to elucidate these questions from the vantage point of the behavior or
གསོལ་ gsol ‘request’ and གནང་ gnaṅ ‘agree, grant’. In order to ensure that we investigate a single linguistic
system, the evidence examined here comes exclusively from Version I (mss. D + A) of the Old Tibetan
Rāmāyaṇa ( Jong 1989), a paraphrase in Tibetan of a well known Indic epic.4
1
This research builds directly on unpublished work of Abel Zadoks, in particular the sixth chapter of Zadoks (2017).
The reader, like one anonymous referee, may prefer a different terminology than that used here, but since the English
terminology for Tibetan verbal forms remains far from conventionalized and since a subordinate verbal form by any
other name smells just as sweet, I implore the reader to bear with these terminological choices.
2
3
Taking inspiration from the French distinction between a “complément d’objet direct” that directly follows a verb
and a “complément d’objet indirect” where an à or de interposes itself between the verb and its object, one can draw a
distinction between a Tibetan ‘direct infinitive’ and an ‘indirect infinitive’. Where the subordinate verb directly
precedes the matrix verb, this is the direct infinitive. In Version 1 of the Old Tibetan Rāmāyaṇam, I identify only the
three matrix verbs ནུས་ nus ‘be able’ དགོས་ dgos ‘need’ ཕོད་ phod ‘dare’ as governing a direct infinitive. Where the terminative
converb interposes itself between the subordinate verb and the matrix verb, this is the indirect infinitive. Nonetheless,
because both གསོལ་ gsol ‘request’ and གནང་ gnaṅ ‘agree, grant’ govern the indirect infinitive, it suffices for the purposes
of this essay to understand ‘infinitive’ as always meaning ‘indirect infinitive’.
4
Manuscript A has the shelf mark IOL Tib J 0737-1 and manuscript D the shelf mark IOL Tib J 0737-3.
Himalayan Linguistics, Vol 21(1)
2 The verb གསོལ་ gsol ‘request’
Subordinate clauses governed by the verb གསོལ་ gsol ‘request’ show a striking distribution,
whereby infinitives occur in direct speech (§2.1) and terminative verbal nouns in the narrative frame
(§2.2).
2.1 The verb གསོལ་ gsol ‘request’ in direct speech
All examples of གསོལ་ gsol ‘request’ governing infinitives in Version I of the Old Tibetan
Rāmāyaṇa occur in direct speech. In most examples, the speaker requests of the addressee that the
addressee undertake some action.
(1)
པྲིན་ཡིག་བརྫང་དུ་གསོལ་ཞེས་གསོལ་པ་དང་༎
« prin-yig
« letter
brdzaṅ-du
send\FUT-CVB.TRM
gsol
» źes gsol-pa-daṅ
//
request » QUOT request-NMLZ-ASS //
‘I request that [you] send a letter,’ he said. (Rama A 273)
(2)
བདགི་བུ་མོ་མེ་ག་སྀ་ན་འབུལ་ན་བཞེས་སུ་གསོལ་ཞེས། བུ་མོ་ཁྲིད་དེ་འོངས་ནས་བུལ་བ་དང་༎
« bdagi
« me.GEN
bu-mo Me-ga-sĭ-na
girl
Meghasenā
źes
/ bu-mo
QUOT / girl
khrid-de
lead-CVB.SF
ḫbul-na
bźes-su
gsol
»
offer-CVB.LOC accept.HON-CVB.TRM request »
ḫoṅs-nas
come\PST-CVB.ELA
bul-ba-daṅ
//
offer-NMLZ-ASS //
‘I offer my daughter Meghasenā, I request that [you] accept her,’ thus [he] brought his
daughter and offered her. (Rama D 39-40)
(3)
དེ་རྣམས་ལ་རྨར་གསོལ། ཞེས་མཆི་ནས༎
de
that
rnams-la
PL-ALL
gsol
/ źes
mčhi-nas
request / QUOT say-CVB.ELA
rmar
ask.HON.CVB.TRM
//
//
‘I request that [you] ask them,’ he said. (Rama A 181-182)
In example (4) the agent of the subordinate verb is not the addressee, but generic.
(4)
མྱྀ་སྡུག་་གཟུགས་ཆན་འདྀ་ལས་པྲོག་དུ་གསོལ་ཞེས་མཆིསྟེ།
myĭ
NEG
sdug gzugs
pretty form
čhan ḫdĭ-las
possess this-ABL
prog-du
rob\FUT-CVB.TRM
30
gsol
request\HON
źes
QUOT
Hill: Making and agreeing to requests in Old Tibetan
mčhiste /
say\CVB.SF /
‘I beg to be taken away from this ugly creature!’ (Rama A 186)
Example (4) may lead us to wonder whether examples (1), (2), and (3) are also amendable to
inter-pretation as passives, viz. ‘request that a letter be sent’, ‘request that she be accepted’, and ‘request
that they be asked’. If so, the implication in these sentences that the addressee is the one meant to
undertake the action of the subordinated verb is a pragmatic result of the discourse situation rather
than a syntactic consequence of the construction. Two further considerations weigh in favor of this
passive interpretation. First, passives are cross-linguistically typical of the indirectness appropriate to
polite requests; to say ‘please be seated’ lacks the impatient air of ‘please sit down’. Second, a patient
focused meaning characterises the future stem in general (Tillemans 1988, Tillemans and Herforth
1989, Tillemans 1991a, Tillemans 1991b). There are also considerations that speak against an
analysis of examples (1)-(4) as passives; for instance, in example (2) the speaker is of course only
offering his daughter to the Ṛṣi and not simply asking that she be taken of his hands. Nonetheless,
if we permit outselves to entertain the hypothesis that that the future infinitive, when governed by
གསོལ་ gsol ‘request’, is syntactically passive and pragmatically a polite request, this hypothesis dovetails
nicely with the wider tendency for future infinitives of transitive verbs to put the patient in focus.5
Example (5), with the unambiguous transitive subordinated verb ‘hunt’, makes this patient
prominence clear.
(5)
འུ་ནྀ་ བཟླུ་བའྀ་རི་དགས་ཡིན་བས། བསྙག་དུ་མྱི་རུང་སྟེ།
ḫu
this
nĭ
NF
bzlu-baḫĭ
deceive-NMLZ.GEN
ri-dags yin-bas/
deer
to be-NMLZ.ERG/
bsñag-du
hunt-CVB.TRM
myi ruṅ-ste
/
NEG suitable-CVB.SF /
This is a deceptive deer and is not suitable to be hunted (Rama A 145-146)
The clause བསྙག་དུ་མྱི་རུང་ bsñag-du myi ruṅ ‘is not suitable to be hunted’, in which མྱི་རུང་ myi ruṅ ‘is not
suitable...’ governs the future infinitive, includes no overt noun phrase, neither the hunter nor the
quarry, but the unsuitableness obtains only to the quarry.
2.2 The verb གསོལ་ gsol ‘request’ in the narrative frame
In the narrative frame the verb གསོལ་ gsol ‘request’ governs the terminative verbal noun and
not the infinitive. Naturally, those making the requests, those to whom these requests are made, and
Obviously, when the subordinated verb is intransitive it is the sole argument that is in focus, e.g. མཚོ་ལ་བརྟེན་པའྀ་ངང་དང་ངུར༎
གུད་དུ་འཔུར་དུ་གནས་མ་མཆིས༎ mtsho-la brten-paḫĭ ṅaṅ-daṅ ṅur // gud-du ḫpur-du gnas ma mčhis // Ducks and geese which live on
the lake have nowhere to fly for shelter (Rama A 348).
5
31
Himalayan Linguistics, Vol 21(1)
those who would engage in the requested activities are in the narrative frame all third person. Still, a
variety of co-reference relationships are available among these parties; let us distinguish the three
parties as ‘speaker’, ‘addressee’, ‘undertaker’.
In example (6) the addressee and the one performing the requested activity are both the king
(speaker ≠ addressee = undertaker).
(6)
རྒྱལ་པོ་གཤེགས་པར་གསོལ་ནས།
rgyal-po
king
gśegs-par
go\HON-NMLZ.TRM
gsol-nas
/
request-CVB.ELA //
They asked the king to come. (Rama A 115)
In example (7) the one requesting and the one performing the requested activity are both
Lakṣana (speaker = undertaker ≠ addressee).
(7)
གཅུང་ལག་ཤ་ནས་སྔར་བརྒལ་བར་གསོལ་ནས༎
gčuṅ
Lag-śa-nas
younger.brother Lakṣana.ERG
sṅar
first-TRM
brgal-bar
cross-NMLZ-TRM
gsol-nas
//
request\HON-CVB.ELA //
The younger brother Lakṣana ་asked to cross first.
(Rama A 314-315)
In example (8) the Devaputras are requesting a boon from Mahadeva, when the goddess of
speech interferes. The request is for supernatural abilities; since no action is requested it is not
meaningful for either the speaker or addresee to undertake the requested action (speaker ≠ addressee,
no undertaker).
(8)
གང་ལ་མདའ་འཕངས་ཕོག་ཆིང་འགུམ་བར་གསོལ་བར་བསམས་པ་དང་། མདའ་དང་པོ་འཕང་སྟེ་འགུམ་བར། ལྷ་མོས་བསྒྱུར་ཏོ༎
gaṅ-la
mdaḫ ḫphaṅs
who-ALL arrow shoot.PST
phog-čhiṅ
hit.PST-CVB.CONT
bsams-pa-daṅ /
mdaḫ
think.PST-NMLZ-ASS / arrow
daṅ-po ḫphaṅ-ste
first
shoot.PST-CVB.SF
lha-mos
goddess.ERG
ḫgum-bar
gsol-bar
die-NMLZ.TRM request-NMLZ.TRM
ḫgum-bar
/
die-NMLZ.TRM /
bsgyur-to
//
change.PST-FIN //
‘They intended to ask that whoever they shot would be hit lethally, but the goddess changed
it so the first arrow they shot would be lethal.’ (Rama A 52-54)
32
Hill: Making and agreeing to requests in Old Tibetan
In sum, the co-reference relations among the speaker, addressee and undertaker are in no way
encoded by this construction.
3 The verb གནང་ gnaṅ ‘agree, grant’
The verb གནང་ gnaṅ ‘agree, grant’ also governs both infinitives and terminative verbal nouns,
but the obvious division of labor that worked in the case of གསོལ་ gsol ‘request’ with infinitives in direct
speech and terminative verbals nouns in the narrative frame, does not here obtain. Instead, the present
infinitive construction is used when the subordinate verb is intransitive and its sole argument is the
same as the agent of གནང་ gnaṅ (§3.1), whereas if either of these criteria is not met, we instead find
the terminative verbal noun construction (§3.2).
3.1 Present inf initives with the matrix verb གནང་ gnaṅ
All examples of the verb གནང་ gnaṅ ‘agree, grant’ governing infinitives in Version I of the Old
Tibetan Rāmāyaṇa happen to take the motion verb གཤེགས་ gśegs ‘go, come’ as their subordinate verb.
Nonetheless, as we will see when we look at the distribution of terminative verbal nouns governed
by གནང་ gnaṅ ‘agree, grant’, it seems likely that the plot of the story and not some grammatical
constraint is the reason that we see infinitives only with this verb. The intransitivity of the subordinate
verb is probably the salient factor.
(9)
མྱི་འགྲོ་འོ་ཞེས་བྱུང་ནས༎ གཤེགསུ་མ་གནང་ངོ་༎
«myi
«NEG
ḫgro-ḫo »
go\PRS-FIN »
źes
QUOT
byuṅ-nas
//
arise\PST-CVB.ELA //
gśegsu
go\HON.CVB.TRM
ma
NEG
gnaṅ-ṅo //
grant\HON-FIN //
He said: ‘I shall not go’ and did not agree to go. (Rama A 35)
(10)
མ་ཧ་དེ་བ་ནྀ་གཤེགསུ་གནང་བ་ཡང་མྱི་འདྲའ་ན༎
Ma-ha-de-ba
Mahadeva
nĭ
NF
gśegsu
go\HON.CVB.TRM
gnaṅ-ba
grant\HON-NMLZ
ḫdraḫ-na
//
seem-CVB.LOC //
‘It doesn't seem like Mahadeva will agree to come.’ (Rama A 37)
33
yaṅ
myi
WF
NEG
Himalayan Linguistics, Vol 21(1)
(11)
ཚ་བོ་རྣམས་ཚེ་གཆིག་ལང་ཀ་པུ་རར་གཤེགས་སུ་ཇི་གནང་ཞེས་
tsha-bo rnams tshe
nephew PL
time
gčhig
ART
Laṅ-ka-pu-rar gśegs-su
Laṅkāpūra.TRM go\HON-CVB.TRM
ǰi
what
gnaṅ »
grant\HON»
źes
QUOT
‘Would you nephews agree to go to Laṅ kāpūra sometime?’ (Rama A 8)
(12)
ལྷའི་བུ་རྣམས་གཤེགསུ་གནང་སྟེ༎
lhaḫi
gods.GEN
bu
son
rnams gśegsu
PL
go\HON.CVB.TRM
gnaṅ-ste
//
grant\HON-CVB.SF //
The Devaputras agreed to go. (Rama A 9)
Looking for other examples of the present infinitive construction in the text, we find that
among the verbs that Garrett et al. (2013) give as governing this construction our text offers only བྱེད་
byed ‘do’ and this only in one instance.
(13)
རྟ་མགོ་འུད་ཆེས་བཆད་ནས། རྫུ་འཕྲུལ་སྟོར་ཏེ། ཡོ་ཡོ་ནས་། མྱི་དང་སྤྲེ་འུ་དམག་གྀ་སྟེང་དུ་འགྱེལ་དུ་བྱེད་པ་དང་
rta-mgo
horse-head
yo
teeter
ḫud
ḫud
čhes
bčhad-nas
QUOT cut-CVB.ELA
yo-nas /
teeter-CVB.ELA /
ḫgyel-du
fall-CV-TRM
myi-daṅ
men-ASS
/ rdzu-ḫphrul
/ magic
stor-te /
lose-CVB.SF /
spre-ḫu
monkey
dmag-gĭ
army-GEN
steṅ-du
top-TRM
byed-pa-daṅ
do-NMLZ-ASS
The horse-head ‘swoosh’ was cut off. [The demon] lost his magical power. He swayed to and
fro and made (as if ) to fall on the army of the men and monkeys. (Rama A 319-20)
So, it is fair to say that co-reference between the subject of the super-ordinate and subordinate
verb is part of the meaning of the present infinitive construction.
3.2 Terminative verbal nouns with the matrix verb གནང་ gnaṅ
The terminative verbal noun is used whenever one of the two conditions (viz. coreference
and intransitivity, §3.1) calling for the present infinitive construction do not obtain. Thus, we have
34
Hill: Making and agreeing to requests in Old Tibetan
examples of co-reference but with transitive subordinate verbs (§3.2.1) and examples with
intransitives but no co-reference (example 20).6
3.2.1 Examples of co-reference, but with transitive subordinate clauses
In example (14) the speaker, a Ṛṣi, agrees to himself accept in marriage Meghasenā, the
daughter of Man-lya-pan-ta.
(14)
ཁྱོད་ཀྱི་བུ་མོ་ཡང་༎ ཁབ་དུ་བཞེས་པར་གནང་ངོ་ཞེས་བྱུང་ནས༎
khyod-kyi
you-GEN
bu-mo yaṅ //
girl
WF //
khab-du
wife-TRM
bźes-par
take-NMLZ.TRM
gnaṅ-ṅo
źes
grant\HON-FIN QUOT
byuṅ-nas
//
occur-CVB.ELA //
‘I consent to take your daughter as consort,’ he replied. (Rama A 1)
In example (15) the addressees, the Devaputra, are asked to agree to themselves take revenge
against the gods.
(15)
ཚ་བོ་ཁྱེད་ཀྱིས་ལན་གླན་ཞིང་ལྷ་རྣམས་ཁ་གདག་པར་ཇི་གནང་ཞེས་གསོལ་པ་ལས༎
tsha-bo
nephew
khyed-kyis
lan
glan-źiṅ
you.HON-ERG answer answer-CVB.CONT
kha gdag-par
kha gdag-NMLZ.TRM
ǰi
what
gnaṅ
grant\HON
źes
QUOT
lha
god
rnams //
PL
//
gsol-pa-las
request-NMLZ-ABL //
‘Would you agree to take revenge and vanquish the gods?’ he asked. (Rama A 20-21)
In (16) it is both Rama who does not agree and Rama who would rule (if he agreed to).
(16)
རྒྱལ་སྲིད་མཛད་པར་མྱྀ་གནང་ན་ཡང་༎ ...
rgyal-srid
reign
mdzad-par
myĭ
do-NMLZ.TRM NEG
gnaṅ-na
grant\HON-CVB.LOC
yaṅ // ⋯
WF // ⋯
‘Even if you don't agree to reign […] (Rama A 88-89)
6
An anonymous referee proposes that examples (14)-(17) do not necessarily involve co-reference, and thus one can
thus draw the stronger conclusion that the infinitive is used for coreference and the terminative verbal noun is used
for switch-reference. I remain open minded about the possibility of this analysis, but believe it is appropriate for me
to stay with the weaker analysis and hope that the referee will pursue this matter elsewhere.
35
Himalayan Linguistics, Vol 21(1)
In a letter to king Rama, Hanumān requests that Rama does not rebuke him.
(17)
བཀའ་མྱི་འབུབ་པར་ཇྀ་གནང་ཞེས་གསོལ་ནས༎
bkaḫ
word
myi
NEG
ḫbub-par
jĭ gnaṅ
send down.FUT-NMLZ.TRM what grant\HON
źes
QUOT
gsol-nas
request-CVB.ELA
‘Would you agree not to rebuke me?’ he asked. (Rama A 351-352)
3.2.2 Examples where there is no co-reference between the one agreeing and the one acting
In examples (18) and (19) the addressee is asked to agree to let the speaker(s) do something.
In both examples the subordinate verb is transitive.
(18)
བདག་ནྀ་ཆགས་འོག་གི་བློན་པོ་བགྱིད་པར་ཇི་གནང་ཞེས་
bdag
me
nĭ
NF
čhags ḫog-gi
shoe below-GEN
blon-po
minister
bgyid-par
do.PRS-NMLZ.TRM
ǰi
what
gnaṅ
grant
źes
QUOT
‘... would you allow me to act as minister under your shoe?’ (Rama A 88-89)
(19)
ལྷ་མོ་ལ་པྱག་འཚལ་བར་ཇི་གནང་ཞེས་གསོལ་པ་དང་༎
lha-mo-la
queen-ALL
pyag
salute
ḫtshal-bar
ǰi
do-NMLZ.TRM what
gnaṅ
grant\HON
źes
QUOT
gsol-pa-daṅ
//
request-NMLZ-ASS //
‘Would you allow us to salute the queen?’ they asked. (Rama A 414-415)
In example (20) a demon accidentally asks for the boon of sleep. The one granting the boon
and the one sleeping are not the same. This example is intransitive.
(20)
རེ་ཤིག་ན་སྲིན་པོ་བུམ་རྣ་ཞེས་བྱ་བ༎ སྤུན་སེམས་ཤན་ཐམས་ཆད་ལ་ཟ་བར་དངོས་གྲུབ་བུབས་པ་ལས༎ ལྷ་འྀ་དབང་པོས༎
ཚྀག་ལ་དབང་བ་འི་ལྷ་མོ་གཆིག་ལྕེའི་ཐོར་ཏོ་ལ་སྤྲུལ་ཏོ། གཉིད་ལོག་པར་ཇྀ་གནང་ཞེས་བསྒྱུར་ནས། དུས་ཐམས་ཤད་དུ་གཉིད་ལོག་པ།
re
while
śig-na
ART-LOC
srin-po Bum-rna
demon Many-Eared
thams-čhad-la za-bar
dṅos grub
all-ALL
eat-NMLZ.TRM siddhi
źes
QUOT
bya-ba //
do-NMLZ //
bsgrubs-pa-las //
achieve-NMLZ-ABL //
36
spun sems śan
fellow creatures
lhaḫĭ
god.GEN
Hill: Making and agreeing to requests in Old Tibetan
dbaṅ pos //
power.ERG //
tshĭg-la
speech-ALL
thor to-la
tip-ALL
sprul-to /
gñid log-par
transform-FIN / sleep-NMLZ.TRM
bsgyur-nas
/
change-CVB.ELA /
dus
time
dbaṅ-baḫi
lha mo
have power.NMLZ.GEN goddess
thams-śad-du
all-TRM
jĭ
what
gčhig
ART
gnaṅ
grant\HON
lčeḫi
tongue.GEN
źes
QUOT
gñid log-pa /
sleep-NMLZ /
‘Once there was a demon named ‘Many-Eared’ who practised to acquire the power to eat all
fellow creatures but, by the power of the gods, a goddess of speech transformed onto his
tongue tip and changed [his request] into ‘would you allow me to sleep,’ whence he would
sleep all the time.’ (Rama A 301-303)
The goddess of speech also interfered with the wording of a request for a boon earlier in the
narrative (example (21)). This example includes both གནང་ gnaṅ and གསོལ་ gsol as matrix verbs. The
verb བགྱིད་ bgyid ‘do’, which is subordinate to གནང་ gnaṅ, is transitive.
(21)
སྲྀད་གསུམ་ལ་དབང་བྱེད་པར་གསོལ་བར་བསམས་པ་དང་། ལྷ་ལ་དབང་བགྱྀད་པར་ཇྀ་གནང་ཞེས། ལྷ་མོས་བསྒྱུར་ཏོ༎
srĭd
gsum-la
world three-ALL
dbaṅ byed-par
power do\PRS-NMLZ.TRM
bsams-pa-daṅ /
lha-la
think.PST-NMLZ-ASS / god-ALL
lha-mos
goddess.ERG
gsol-bar
request-NMLZ.TRM
dbaṅ bgyĭd-par
power do.PRS-NMLZ.TRM
jĭ
what
gnaṅ
grant\HON
bsgyur-to
//
change.PST-FIN //
‘They intended to ask for power over the three worlds, but the goddess changed [this request]
into ‘would you grant us power over the gods?’’ (Rama A 50-51)
Example (22) is interesting because the second person addressee is potentially co-referenced
as the subject of མཇལ་པར་ mǰal-par and it is not entirely obvious whether མཇལ་པར་ mǰal-par should be
considered a transitive or intransitive verb. Nonetheless, since the terminative verbal noun is not used
when both the subordinate verb is intransitive and there is coreference between the subjects of the
matrix and subordinate verbs, we can conclude from the use of the terminative verbal noun that either
the verb མཇལ་ mǰal is transitive (‘you consent to meet me’) or at least that there is no coreference (‘you
consent to us meeting’).
(22)
མྱི་པྲད་དུ་མྱྀ་རུང་ན༎ ལན་ཆིག་མཇལ་པར་ཇི་གནང་ཞེས་
myi
NEG
prad-du
myĭ
meet-CVB.TRM NEG
ruṅ-na
be suitable-CVB.LOC
37
//
//
lan
time
čhig
ART
Himalayan Linguistics, Vol 21(1)
mǰal-par
ǰi
meet.HON-NMLZ.TRM what
gnaṅ
grant\HON
źes
QUOT
‘If it were not an option not to meet, would you consent to meeting once?’ (Rama A 380-381)
4 Conclusions
In conclusion, future infinitives constitute a sort of passive that can be used as a polite
imperative and present infinitives require an intransitive subordinate verb and coreference between
the agent of the matrix verb and the sole argument of the subordinate verb. When neither the future
or present infinitive are appropriate, the terminative verbal noun is used.
A B B R E VI A T I O N S
ABL
ALL
ART
ASS
CONT
CVB
ELA
ERG
FIN
FUT
GEN
HON
IMP
LOC
ablative
allative
article
associative
continuative
converb
elative
ergative
finite suffix
future
genitive
honorific
imperative
locative
NEG
NF
NMLZ
PFV
PL
PROX
PRS
PRT
PST
QUOT
SF
TRM
WF
negative
narrow focus
nominalizer
perfective
plural
proximate
present
particle
past
quotative
semifinal
terminative
wide focus
REFERENCES
Garrett, E.; N. W. Hill; and A. Zadoks. 2013. “Disambiguating Tibetan verb stems with matrix verbs
in the indirect infinitive construction”. Bulletin of Tibetology 49.2: 35–44.
Jong, J. W. de. 1989. The story of Rāma in Tibet: text and translation of the Tun-huang manuscripts.
Wiesbaden: Franz Steiner.
Kitamura, H. 1975. “The honorifics in Tibetan”. Acta Asiatica 28: 56–74.
Tillemans, T.; and D. Herforth. 1989. Agents and actions in classical Tibetan: the indigenous
grammarians on Bdag and Gzhan and Bya byed las gsum. Wien: Universität Wien, Arbeitskreis für
Tibetische und Buddhistische Studien [Wiener Studien zur Tibetologie und Buddhismuskunde
21].
38
Hill: Making and agreeing to requests in Old Tibetan
Tillemans, T. J. F. 1988. “On bdag and gzhan and related notions of Tibetan grammar”. Tibetan
Studies, 491–502. Munich: Kommission für Zentralasiatische Studien, Bayerische Akademie der
Wissenschaften.
Tillemans, T. J. F. 1991a. “Note on bdag don phal ba in Tibetan grammar”. Études Asiatiques 45.2:
311–323. https://doi.org/10.5169/seals-146922
Tillemans, T. J. F. 1991b. “gSer tog blo bzan tshul khrims rgya mtsho on Tibetan verb tenses.” In: E.
Steinkellner (ed.), Tibetan History and Language. Studies dedicated to Uray Géza on his seventieth
birthday, 487–496. Vienna: Arbeitskreis für Tibetische und Buddhistische Studien [Wiener
Studien zur Tibetologie und Buddhismuskunde 26].
Zadoks, A. 2017. “Syntax and semantics of the Tibetan verb”. Unpublished manuscript.
Nathan W. Hill
[email protected]
39