M.

Butt: Tense and Aspect in Urdu 1

Tense and Aspect in Urdu

Miriam Butt
Universität Konstanz
[email protected]
Journée d’études sur le temps et l’aspect
Paris, October 18, 2003
(Slightly) Revised Version for ParGram Meeting, Dublin, September 2004

1 Central Problem
• Semantic analyses of tense/aspect are meant to be crosslinguistically valid.

• However, the available tools would appear to be inadequate for crosslinguistic analysis.

• This talk: case study of a non-European language.

– How is the system organized?

– What concepts are basic?

2 The Language
Urdu is a South Asian language spoken in:

• Pakistan (national language)

• India (one of the 18 official languages)

• World-Wide due to the South Asian Diaspora (big populations in the U.K., U.S.A.,
Canada, etc.)

Urdu is closely related to Hindi.

Taken together these represent the second most spoken language in the world: 400 000 000.

Properties: SOV, subject and object agreement, non-nominative case including an ergative.
M. Butt: Tense and Aspect in Urdu 2

3 Tense
Methodology: Begin by trying to sort through the language roughly via traditional (Latin)
conjugation classes.

Reason:

1. This world view still influences much of modern linguistics, despite long-standing work
on Native American languages (for example).

2. Unlike the Native American languages (or Chinese, cf. Smith 1991), Urdu is an Indo-
European language like Latin.

Semantics: Ehrich’s (1992) DRT version of the basic Reichenbachian oncepts E,R and S.

Contextually Determined Relations: S (Speech Time) and R (Reference Time)

Intrinsic Relations: E (Event Time) and R

Roughly:
Contextually Determined
S,R R<S
Intrinsic E,R Present Past
Relations E < R Perfect Past Perfect
E>R — —
The notation “S,R” signifies that these stand in a relation to one another. This relation could
be either temporal overlap (S R) or temporal precedence (determined by the context).

3.1 Present
First Problem: Urdu only has a present tense for one verb: ho ‘be’.

(1) Present of Urdu be

Singular Plural Respect (ap) Familiar (tum)
1st hũ hẼ
2nd hE hẼ ho
3rd hE hẼ
ho- ‘be’

Present Tense: E R&R S

3.1.1 Subjunctive/Questions
The same morphology on other verbs signals either a subjunctive or a question.
M. Butt: Tense and Aspect in Urdu 3

(2) Urdu Subjunctive Paradigm

(Question)
Singular Plural Respect (ap) Familiar (tum)
1st mar-ũ mar-ẽ
2nd mar-e mar-ẽ mar-o
3rd mar-e mar-ẽ
mar- ‘hit’

Subjunctive:

(3) AgAr mẼ a-ũ, to Acch a ho-g-a

if I.Nom come-1.Sg so good be-Fut-M.Sg
‘If I were to come, that would be good.’

Question:

(4) mẼ Abh i a-ũ?

I.Nom now come-1.Sg
‘Should I come (over) now?’

3.1.2 Imperative
The same morphology is used for imperatives with the second persons and the first plural.

(5) kh ana kh a-o

food.Nom eat-2.Fam
‘Eat (your) food.’

(6) aye kh ana kh a-ẽ

come food.Nom eat-1/2.Resp
‘Come, let’s eat.’ (Schmidt 1999:98)

There is also some special morphology for extra-polite imperatives.

This morphology is confined to the handful of stems ending in vowels (and r.).

le ‘take’ de ‘give’ kAr ‘do’ pi ‘drink’ si ‘sew’

ap (2.respect) lijiye dijiye kijiye pijiye sijiye
‘please do do X’
M. Butt: Tense and Aspect in Urdu 4

3.1.3 Historical Development

Present Morphology

• The present morphology is a remnant of the old Sanskrit/Prakrit present and impera-
tive morphology (the two collapsed in Middle Indo-Aryan).

• It is the only piece of tense morphology that has survived into the modern ages.

Special Imperative

• The special imperative is generally taken to derived from an Old Indo-Aryan (OIA)
optative in -yā, which became -eyya in early Middle Indo-Aryan (MIA) and then
is supposed to have “hardened” in later MIA to -ejja-, -ijja- (Kellog 1893:229–230,
Beames 1872:108, Chatterji 1926:900).

• Alternative Analysis (Butt and Lahiri):

The source of the modern -j- is an allomorphic variant of the original Sanskrit verb gā
‘go’ (cf. modern Urdu ja-/ga- ‘go’). The person endings of the indicative are simply
remnants of the original inflections this verb carried. The -j- caused vowel raising on
the stem vowel.

3.1.4 Summarizing Questions

• Is it crosslinguistically odd not to have a “real” present tense? Or is it normal?

• There is a clear historical affinity between present, subjunctive (modality) and fu-
ture. Can formal semantic analyses reflect this affinity and even predict the historical
changes?
N.B. The notion of semantic maps (e.g., van der Auwera and Plungian 1998) is an
attempt in this direction, but not an ultimately satisfactory one as yet.

3.2 Future
The future is the only tense in Urdu which is formed purely inflectionally and which applies
regularly across the verbal inventory.

(7) Urdu Future Paradigm

Singular Plural Respect (ap) Familiar (tum)
M/F M/F M/F M/F
1st mar-ũ-g-a/i mar-ẽ-g-e/i
2nd mar-e-g-a/i mar-ẽ-g-e/i mar-o-g-e/i
3rd mar-e-g-a/i mar-ẽ-g-e/i
mar- ‘hit’
M. Butt: Tense and Aspect in Urdu 5

• The first piece of morphology after the stem is identical to the present/subjunctive
morphology.

• The -g- is derived from a Sanskrit participle of the verb gā ‘go’ (Kellog 1893:231, Beg
1988:191, McGregor 1972).

• The gender and number agreement morphology (a/i/e) is regular synchronically in

that it is also found on the perfect (section 3.3.2), imperfect (section 4.2) and and
progressive (section 5.1) forms, all descended from participles.
The morphology is unproblematic if the -g- is indeed associated with an old participle.

Evidence for On-Going Historical Change

• The -g- was separable until relatively recently, providing evidence for a change which
pressed a version of ‘go’ into service as a tense auxiliary.

• This auxiliary is turning into a tense affix via the status of a clitic.

• The -g- is still somewhat separable (and is written as a separate word in the Urdu
orthography, but not the Hindi one).

(8) vo [lıkh -e or pAr.h -e]-gi

Pron.3 write-3.Sg and read-3.Sg-Fut.Fem.Sg
‘She will read and write.’

Semantics, Absolute Future: E R&R>S

3.2.1 With Perfect Morphology

Perfect/past morphology (section 3.3.2) can also serve to express the future.

(9) mẼ Abh i a-yi

I.Nom now came-Perf.F.Sg
‘I’ll be right there (come right away).’

This is perhaps analogous to the German situation, where the present morphology is sys-
tematically ambiguous between present and future readings.

Immediate Future: E R&R≥S

M. Butt: Tense and Aspect in Urdu 6

3.2.2 Imminent Future

There is yet another way to express imminently impending actions.
• -vala ‘one’ (as in ‘the egg-eating-one’) in combination with a verbal noun.
• The dative/accusative ko in combination with a verbal noun.
(10) kıSti d.ub-ne vali hE
boat.F.Sg.Nom drown-Inf.Obl one.F.Sg be.Pres.3.Sg
‘The boat is about to sink.’ (Schmidt 1999:139)
(11) a. ıs lAr.ki ki Sadi ho-ne vali hE
this girl.F.Sg Gen.F.Sg marriage.Nom be-Inf.Obl one.F.Sg be.Pres.3.Sg
’This girl’s wedding is taking place soon.’ (Schmidt 1999:140)

a. ıs lAr.ki ki Sadi ho-ne ko hE

this girl.F.Sg Gen.F.Sg marriage.Nom be-Inf.Obl Acc be.Pres.3.Sg
’This girl’s wedding is taking place soon.’ (Schmidt 1999:140)

Imminent Future: E R & R > S & the event will take place with great certainty (i.e.,
minus the modal readings the future usually carries with it).

3.2.3 Summary and Questions

• The future is the only regular “finite” form for all verbs.
• Is this natural or exceptional? Cf. English and German which don’t have morphological
futures at all, but rely on futures culled from modals.
• A general theory of when verbal nouns plus case marking can be used in “tensed”
contexts is desirable. Is there any connection to the cases used for tense in Australian
languages (e.g., Evans 1995, Nordlinger and Sadler 2004)?

3.3 Past
There is no specialized past tense morphology. A form of the “perfect” morphology is used
to express past tense.

The verb ‘be’ forms a suppletive paradigm with th - (from ‘stand’) in that this form expresses
the past tense of ‘be’.
(12) Past of Urdu be
Singular Plural Respect (ap) Familiar (tum)
M/F M/F M/F M/F
1st th -a/i t -e/ı̃
h

2nd th -a/i th -e/ı̃ th -e/ı̃

3rd th -a/i t -e/ı̃
h

th - ‘be’
M. Butt: Tense and Aspect in Urdu 7

The verb is marked for gender and number because it is derived from an old participial form.

3.3.1 Imminent Action Again

Just as with the present tense, the past tense ‘be’ also allows a combination with verbal
nouns to express an imminent action in the immediate past.

(13) a. jAb d.akt.Ar sahıb bol-ne=ko th -e

when doctor sahib.M.Nom speak-Inf.Obl=Acc be.Past-M.Pl
to sAb log cUp ho gE-ye
though all people.Nom quiet become go.Perf.M.Pl
‘When the doctor was about to speak, everybody fell quiet.’ (Glassman 1986:233)

b. mem-sahıba cai bAna-ne=ko th -ı̃

Madam.F.Nom tea.F.Nom make-Inf.Obl=Acc be.Past-F.Pl
‘Madam was just about to make tea.’ (Glassman 1986:233)

3.3.2 With Perfect Morphology

The “perfect” is formed with the verb stem and the general gender/number morphemes a/i/e.

There used to be a -i- or -y- which explicitly marked this tense/aspect, but it disappeared
over the last few hundred years.

(14) Urdu Perfect/Past Paradigm

Singular Plural Respect (ap) Familiar (tum)
M/F M/F M/F M/F
1st mar-a/i mar-e/i
2nd mar-a/i mar-e/i mar-e/i
3rd mar-a/i mar-e/i
mar- ‘hit’

In isolation, this marking results in a past tense reading.

(15) Anjum aj sUbAh skul cAl-i

Anjum.F.Sg.Nom today morning school.F.Sg.Loc walk-Perf.F.Sg
‘Anjum walked/went to school this morning.’

Past: E R&R<S

Historical Origin:
This morphological marking has been termed “perfective” or “perfect”, presumably because
the (now null) morpheme is descended from an OIA past participial form in -tā (e.g., the
eaten apple).
M. Butt: Tense and Aspect in Urdu 8

4 Aspect
4.1 Perfect
In combination with the ‘be’ auxiliary, the perfect/past morphology yields standard present
and past perfect readings.
(16) Anjum=ne adnan=ko dekh -a hE
Anjum.F.Sg=Erg Adnan.M.Sg=Acc see-Perf.M.Sg be.Pres.3.Sg
‘Anjum has seen Adnan.’

(17) Anjum=ne adnan=ko dekh -a th -a

Anjum.F.Sg=Erg Adnan.M.Sg=Acc see-Perf.M.Sg be.Past-M.Sg
‘Anjum had seen Adnan.’
Present Perfect: E<R & R S
Past Perfect: E<R & R < S

4.2 Imperfect
The “imperfect” is formed with the verb stem, a -t- morpheme and the general gender/number
morphemes.
(18) Urdu Imperfect Paradigm
Singular Plural Respect (ap) Familiar (tum)
M/F M/F M/F M/F
1st mar-t-a/i mar-t-e/i
2nd mar-t-a/i mar-t-e/i mar-t-e/i
3rd mar-t-a/i mar-t-e/i
mar- ‘hit’
In combination with the ‘be’ auxiliaries it yields a habitual present/past reading.
(19) Anjum adnan=ko mar-t-i hE
Anjum.F.Sg.Nom Adnan.M.Sg=Acc hit-Perf.f.Sg be.Pres.3.Sg
‘Anjum hits Adnan.’

(20) Anjum adnan=ko mar-t-i th -i

Anjum.F.Sg.Nom Adnan.M.Sg=Acc hit-Perf.F.Sg be.Past-F.Sg
‘Anjum used to hit Adnan.’
Without the auxiliaries, this marking expresses unfulfilled wishes (Schmidt 1999:118).
(21) mẼ tala lAga-t-i ...
I.Nom lock put-on-Impf-F.Sg
‘Had I put on a lock, . . .’
It is also used in a continued narrative (Schmidt 1999:118). and is often referred to as the
“literary imperfect” (Montaut, p.c., October 2003)
M. Butt: Tense and Aspect in Urdu 9

5 Continuation
While Urdu has no prototypical present or past, it makes intense use of various ways of
expressing a continuing action (beyond the habitual and narrative detailed above).

5.1 Progressive
A general progressive meaning is achieved by combining the verb stem with a progressive
auxiliary rAh ‘stay’ (in the present or past). The progressive auxiliary is morphologically
perfect.

(22) a. Anjum adnan=ko mar rAh-i hE

Anjum.F.Sg.Nom Adnan.M.Sg=Acc hit stay-Perf.f.Sg be.Pres.3.Sg
‘Anjum is hitting Adnan.’

b. Anjum adnan=ko mar rAh-i th -i

Anjum.F.Sg.Nom Adnan.M.Sg=Acc hit stay-Perf.F.Sg be.Past-F.Sg
‘Anjum was hitting Adnan.’

Present Progressive: E R&R S & the event is a continuing one.

5.2 Iteration and Longer Continuation with the Progressive

The progressive with an imperfect main verb signals iteration.

(23) a. kUtta bh õk-t-a rAh-a

dog.M.Sg.Nom bark-Impf-M.Sg stay-Perf.M.Sg
‘The dog kept on barking.’
E R&R S & the event iterates over a give time span.

b. kUtta bh õk-t-a rAh-t-a

dog.M.Sg.Nom bark-Impf-M.Sg stay-Impf-M.Sg
‘The dog keeps on barking.’

E R&R S & the event is a continuing one over a long time span that has not ended.

5.3 Iteration and Longer Continuation with ‘go’

(24) kUtta bh õk-t-a ja-t-a hE
dog.M.Sg.Nom bark-Impf-M.Sg go-Impf.M.Sg be.Pres.3.Sg
‘The dog keeps on barking (willfully, over a long time).’

Features: long time span that has not ended yet, willfulness of actor, iteration, event itself
is of a longer duration
M. Butt: Tense and Aspect in Urdu 10

One can add yet another auxiliary: cAl ‘walk’ (data courtesy of the MT group in Lahore).

(25) kUtta bh õk-t-a cAl-a ja-t-a hE

dog.M.Sg.Nom bark-Impf-M.Sg walk-Perf.M.Sg go-Impf.M.Sg be.Pres.3.Sg
‘The dog keeps on barking (willfully, over a long time).’

And yet another iteration.

(26) kUtta bh õk-t-a cAl-a ja rAh-a hE

dog.M.Sg.Nom bark-Impf-M.Sg walk-Perf.M.Sg go stay-Perf.M.Sg be.Pres.3.Sg
‘The dog keeps on barking (willfully, over a long time, continuously).’

These indicate a yet greater degree of iteration and continuition: longer duration of the
iterated events.

5.4 Imperatives as to Habitual Actions

A special morphosyntactic construction is used to express imperatives which command/forbid
the general/habitual practice of an action.

(27) a. mAt bol-a kAr-o

not speak-Perf.M.Sg do-2.Fam
‘You shouldn’t keep speaking (saying things when you shouldn’t)!’

b. mıl-n-e a-ya kAr-o

meet-Inf-Obl come-Perf.M.Sg do-2.Fam
‘You should keep coming to visit.’

Davison analyzes the verbs (‘speak’, ‘come’) as nouns in this instance, so the doing of a certain
thing, namely “speaking” or “coming” is commanded/forbidden. The nominalization gives
the sense of continuation/habituality.

5.5 Summary
• Beyond these examples here, a long list of further combinations with various forms
of the future and participial forms of ‘be’ and some of the other auxiliaries are also
possible.

• The tense/aspect system is therefore quite complex and no good analysis exists yet to
date (though several dissertations have made an attempt).

• Basic features: beyond the standard notions of E,R,S and modality, one needs to
express different lengths (duration) of continuation and iteration of the event.
M. Butt: Tense and Aspect in Urdu 11

6 Complex Predicates
In addition to these elements of the verbal complex, light verbs also play a role.

The semantic contribution of the light verb is difficult to characterize. It indicates a range
of meanings from completion, inception, benefaction, force, suddenness, etc. (see Hook 1974
for a detailed study). The light verb does not predicate its own event — rather: it serves to
modify the event semantics of the main verb (Butt and Geuder 2001).

(28) a. nadya=ne xAt lıkh li-ya

Nadya.F=Erg letter.M.Nom write take-Perf.M.Sg
‘Nadya wrote a letter (completely).’

b. nadya=ne mAkan bAna di-ya

Nadya.F=Erg house.M.Nom make give-Perf.M.Sg
‘Nadya built a house (completely, for somebody else).’

c. nadya mAkan bAna pAr.-i

Nadya.F.Nom house.M.Nom make fall-Perf.F.Sg
‘Nadya fell to building a house.’

d. nadya gır pAr.-i

Nadya.F.Nom fall fall-Perf.F.Sg
‘Nadya fell (down, suddenly).’

e. nadya gır gE-yi

Nadya.F.Nom fall go-Perf.F.Sg
‘Nadya fell (down).’

f. nadya bol Ut.h -i

Nadya.F.Nom speak rise-Perf.F.Sg
‘Nadya spoke up/broke into speech (suddenly, forcefully).’

This construction provides evidence for the result projection (Butt and Ramchand 2003).

Basic Feature: Telicity (completion/inception).

7 Aspectual Verbs
In addition to the light verbs, there are also some aspectual verbs: cUk ‘pick up’, lAg ‘begin’.

These may look like light verbs, but they do not have the full meaning spectrum of light
verbs and are compatible with every main verb (light verbs generally impose selectional
restrictions).
M. Butt: Tense and Aspect in Urdu 12

(29) a. nadya mAkan bAna cUk-i hE

Nadya.F.Sg.Nom house.M.Nom make pick-up-Perf.F.Sg be.Pres.3.Sg
‘Nadya has built a house (finished it completely, already).’

a. nadya ga cUk-i hE
Nadya.F.Sg.Nom sing pick-up-Perf.F.Sg be.Pres.3.Sg
‘Nadya has sung (completely, already).’

(30) nadya mAkan bAna-n-e lAg-i hE

Nadya.F.Sg.Nom house.M.Nom make-Inf-Obl begin-Perf.F.Sg be.Pres.3.Sg
‘Nadya began building a house.’

Basic Features:

• lAg is a control verb that embeds another verb. Its basic meaning is ‘begin’.

• cUk is more difficult to analyze. It seems to be more like the light verbs in that the
syntax is similar and telicity is involved. However, it is too general to be a light verb
(though it is better with intransitives).

8 Summary and Outlook

• Much more work needs to be done on Urdu (and similar languages).

• There is no “ready made” semantic analysis that can do justice to the differing shades
of iteration, continuation and duration.

• There is also no overall theory that can do justice to the interaction between differ-
ent types of tense/aspect auxiliaries, morphology and complex predication (or to the
historical aspects).

Outlook:

• Currently under development: computational LFG grammar for Urdu within the Par-
Gram project (Butt, King, Niño and Segond 1999).

• ParGram stands for Parallel Grammars and the idea is to gear analyses so that they
are as crosslinguistically valid as possible.

• One part of this effort is the determination of a crosslinguistically valid feature space
(Butt et al. 2003).

• Glasbey (2001) concludes that temporal (discourse) information is well placed at the
level of f(unctional)-structure within LFG.

• So: encode the relevant information within a feature space? (And then let semantics
deal with it properly.)
M. Butt: Tense and Aspect in Urdu 13

(31)
  " #  
r-s precede/follow/overlap
tense
standard/far/close
   



 distance 



precede/follow/overlap
     



 e-r 




 tns-asp



 distance standard/far/close 
 



+/−
 



  telic  



aspect  
+/−
   





 dur 
 



+/−
 



  cont  



iter +/−

Notes:

• Tense/Aspect is made up of the Reichenbachian E,R,S relations.

• Notions like progressive are composed out of the duration, continuation and iteration
features.

• Aktionsart and modality still need to be encoded under separate feature bundles.

Assume a Complex Verbal Complex:

(32) V

V cont prog tense

V LV cAl ja rAh ‘be’

‘walk’ ‘go’ ‘stay’

References
van der Auwera, Johan and Vladmimir A. Plungian. . 1998. Modality’s Semantic Map.
Linguistic Typology 2(1): 79–124.

Beames, John. 1872–79. A Comparative Grammar of the Modern Aryan Languages of India.
Delhi: Munshiram Manoharlal. Republished 1966.

Beg, Mirza Khalil A. 1988.Urdu Grammar: History and Structure. New Delhi: Bahri Pub-
lications.

Butt, Miriam. 1995. The Structure of Complex Predicates in Urdu. Stanford, California:
CSLI Publications.
M. Butt: Tense and Aspect in Urdu 14

Butt, Miriam, Tracy Holloway King, Maria-Eugenia Niño, and Frederique Segond. 1999. A
Grammar Writer’s Cookbook. Stanford: CSLI Publications.

Butt, Miriam, Martin Forst, Tracy Holloway King, and Jonas Kuhn. 2003. The Feature Space
in Parallel Grammar Writing. In ESSLLI 2003 Workshop on Ideas and Strategies for
Multilingual Grammar Development.

Butt, Miriam and Wilhelm Geuder. 2001. On the (Semi)Lexical Status of Light Verbs. In
N. Corver and H. van Riemsdijk (eds.) Semi-lexical Categories: On the content of
function words and the function of content words. Berlin: Mouton de Gruyter.
Butt, Miriam, and Gillian Ramchand. 2003. Complex Aspectual Structure in Hindi/Urdu.
In The Syntax of Aspect, ed. Nomi Ertishik-Shir and Tova Rappaport. Oxford: Oxford
University Press. To appear.

Chatterji, Suniti Kumar. 1926. The Origin and Development of the Bengali Literature, Vol-
ume II. Calcutta: D. Mehra, Rupa & Co (1975 edition).
Ehrich, Veronika. 1992. Hier und jetzt: Studien zur lokalen und temporalen Deixis im
Deutschen. Tübingen: Max Niemeyer Verlag.

Evans, Nicholas. 1985. A Grammar of Kayardild: With Historical-Comparative Notes on

Tangkic. Berlin: Mouton de Gruyter.

Glasbey, Sheila. 2001. Tense, Aspect and the Temporal Structure of Discourse: Towards an
LFG Account. In M. Butt and T.H. King (eds.) Proceedings of the LFG01 Conference,
Hong Kong. http://csli-publications.stanford.edu/LFG.

Glassman, Eugene. 1976. Spoken Urdu. Lahore: Nirali Kitaben.

Hook, Peter E. 1974. The Compound Verb in Hindi. Center for South and Southeast Asian
Studies: The University of Michigan.
Kellogg, S.H. 1893. Grammar of the Hindi Language. Delhi: Munshiram Manoharlal Pub-
lishers Pvt. Ltd. Second Edition, reprinted 1990.
McGregor, R.S. 1968. The Language of Indrajit of Orchā. Cambridge: Cambridge Univer-
sity Press.

Nordlinger, Rachel and Louisa Sadler. 2004. Tense beyond the Verb: Encoding clausal tense/aspect/-
mood on nominal dependents. Natural Language and Linguistic Theory 22:597–641.

Schmidt, Ruth Laila. 1999. Urdu: An Essential Grammar. London: Routledge.

Smith, Carlotta. 1991. The Parameter of Aspect. Dordrecht: Kluwer Academic Publishers.