World Sanskrit Conference (WSC) (17th : 2018)

How to Translate a Gerund? : The “Gerund/absolutive” in Epic-Puranic and Classical Sanskrit Literature Söhnen-Thieme, Renate 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


70440-Sohnen-Thieme_Renate_How_to_Translate_Gerund_2019.pdf [ 570.52kB ]
JSON: 70440-1.0379842.json
JSON-LD: 70440-1.0379842-ld.json
RDF/XML (Pretty): 70440-1.0379842-rdf.xml
RDF/JSON: 70440-1.0379842-rdf.json
Turtle: 70440-1.0379842-turtle.txt
N-Triples: 70440-1.0379842-rdf-ntriples.txt
Original Record: 70440-1.0379842-source.json
Full Text

Full Text

 How to Translate a Gerund? The “Gerund/absolutive” in Epic-Puranic and Classical Sanskrit Literature Renate Söhnen-Thieme Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13, 2018, Section 2: Linguistics.  Section Convenors: Madhav Deshpande and Jan Houben
General Editor: Adheesh Sathaye Published by the Department of Asian Studies, University of British Columbia, for the International Association for Sanskrit Studies. DOI: 10.14288/1.0379842.
URI: Suggested Citation Format: MLA:
Söhnen-Thieme, Renate. “How to Translate a Gerund? The “Gerund/absolutive” in Epic-Puranic and Classical Sanskrit Literature.” Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13, 2018, Section 2: Linguistics. Edited by Madhav Deshpande and Jan Houben, 2019. DOI: 10.14288/1.0379842. APA:
Söhnen-Thieme, R. (2019). How to translate a gerund? The “gerund/absolutive” in epic-puranic and classical Sanskrit literature. In M. Deshpande and J. Houben, (eds.) Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13, 2018, Section 2: Linguistics. DOI: 10.14288/1.0379842. Chicago:
Söhnen-Thieme, Renate. 2019. “How to Translate a Gerund? The “Gerund/absolutive” in Epic-Puranic and Classical Sanskrit Literature.” In Proceedings of the 17th World Sanskrit 
Conference, Vancouver, Canada, July 9-13, 2018, Section 2: Linguistics, edited 
by Madhav Deshpande and Jan Houben. DOI: 10.14288/1.0379842. Proceedings of the 17th World Sanskrit Conference, July 9-13, 2018 University of British Columbia, Vancouver, CanadaTHE   17TH    WORLD   SANSKRIT  CONFERENCEVANCOUVER, CANADA • JULY 9-13, 2018Copyright © 2019 by the author. Content is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license (CC BY-NC-ND 4.0).वैधुसव ्मकबंटुकुअ ारा यसं ृ ता यनसमवायःINTERNATIONAL ASSOCIATION OF SANSKRIT STUDIES THE 17TH WORLD SANSKRIT CONFERENCE, VANCOUVER, CANADA, JULY 9-13, 2018 How to Translate a Gerund?  The “Gerund/absolutive” in Epic-Puranic and Classical Sanskrit Literature Renate Söhnen-Thieme South Asia Department, School of Oriental and African Studies,  London, U.K. Abstract The article is based on the observation and collection, over several decades, of examples of the so-called “gerund,” “indeclinable participle,” or “absolutive” (the last of these mainly in German). It starts with the discussion of the terms and definitions, in a number of grammars, of a linguistic category which has become especially prevalent in Indian languages and literatures and may be regarded as a genuinely Indian development, since there is only scarce evidence of a similar form being used in other Indo-European languages. In its second part the paper proceeds to present a commented collection of selected examples that demon-strate its various types and the functions in which this particular category of a verbal noun is used. It concludes, with the attempt to give a satisfactory defini-tion plus description of this category which would be applicable at least to the vast majority of the examples, establishing the types of relationship between the gerund/absolutive phrase and the main sentence. This may also be helpful for achieving an idiomatic translation that corresponds as closely as possible to the Sanskrit original. The examples are selected mainly from classical Sanskrit nar-rative literature (epics, kāvya), but they include also specimens from the Brāh-maṇas and examples taken from the Indian grammatical literature. Keywords: Sanskrit grammar, syntax, grammatical categories, translation problems. 1. Introduction: Terms and definitions The oldest Sanskrit grammar known to us, the Aṣṭādhyāyī of Pāṇini, gives sparse, but more or less exhaustive information about this linguistic item, identifying it by the artificial term ktvā, based on the suffix -tvā with which it is formed (and Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13 2018, Section 2: Linguistics, edited by Madhav Deshpande and Jan Houben, 2019. DOI: 10.14288/1.0379842.  SÖHNEN-THIEME 2which can be replaced by -ya in the case of verbs with a preverb).  Paṇini explains 1its formation (i.e. the way in which it is added to a certain form of the root),  its 2status as an “indeclinable,”  and its main, mutually exclusive, uses in a sentence 3(which will be discussed later in detail).  He does not explain the formation of 4the suffix itself (i.e. that it is actually a petrified instrumental case of a verbal noun formed by the suffix -tu-), nor is he interested in identifying it by a term denoting a word category more specific than “indeclinable;” the term ktvā is un-mistakable and therefore sufficient. It is, however, a monolingual term (perhaps similar to the term “wh-forms,” used in modern English grammar).  Therefore Western grammarians 5had to find expressions, not so much for the formation, but for the function of this grammatical phenomenon within a sentence. In French and German grammars the term “absolutif/Absolutiv” is used,  which emphasizes its inflec6 -tional disconnection from any member of the sentence defined by inflection (no “agreement” with any noun, pronoun, or adjective), i.e. its status as an “indeclin-able” (in agreement with Pāṇini's first rule where ktvā is mentioned). In English, unlike German, a functionally perhaps comparable category can be found, the “gerund,” which appears to be somewhat similar to the Latin usage,  e.g., docendo discimus – “we learn through teaching.” The only problem in 7 Cf. P. 7.1.37ff. 1 Cf. P. 6.4.18,31ff., 7.4.43ff.2 Cf. P. Cf. P. 3.4.18ff.4 Cf. Quirk and Greenbaum, passim.5 Renou, Stenzler, Mayrhofer, Morgenroth, Thumb-Hauschild (so also even Gonda in 6English medium). Cf. J. B. Hofmann, Lateinische Syntax und Stilistik, p. 369: “Das Gerundium erfüllt die 7Funktion eines Verbalsubstantivs, das den Infinitiv in den obliquen Kasus ergänzt und ähnlich wie dieser die verbale Rektion aufweist: legendi librum wie legere librum.” The gerund in English, however, can be used in all cases; but both Latin and English ge-runds are quite different from the Sanskrit “gerund,” which is indeclinable.  How to Translate a Gerund?  3English is that the suffix -ing is very ambiguous, being used for three different categories:  8(a) for a noun (similar to the German suffix -ung, e.g., in “Wohnung,” English “dwelling;” “Zeichnung,” English “drawing,” etc.); (b) for a present participle (e.g., “somebody dwelling or drawing a por-trait”), which is in English, unlike in French, German or Sanskrit, as in-declinable as any other adjective or even adverb;  and then  9(c) for the gerund, which is something in between (e.g. “because of killing Vṛtra”, “after drawing a portrait”), since it resembles Sanskrit in taking an accusative object (without being a participle), and this makes “gerund” a suitable term for this category.  One has to bear in mind, however, that the English -ing form (c) is not a participle  – even if participle and gerund share the verbal faculty of taking an 10accusative object – because a participle would automatically agree with the grammatical subject of the clause; the English gerund, on the other hand, can be used adverbially with a preposition, like a noun (often a good solution for a rela-tively literal translation), and without any obvious agent to agree with, as often found in Sanskrit passive sentences and in impersonal clauses, as will be shown. The other, perhaps minor, problem is that the English gerund can be used in any case function, whereas the Sanskrit ktvā can be only used in the instrumental or as an (indeclinable) adverb. In the following, I am going to present a collection of examples for as many possibilities of using the Sanskrit “gerund” in a simple sentence as I could find, followed by examples with gerunds in more complex syntactical constructions.  Moreover there are fluent transitions between these categories; cf. Quirk and Green8 -baum, p. 391, note. Therefore it does not make much sense to call the Sanskrit gerund an “indeclinable par9 -ticiple” (cf. Thumb-Hauschild §635, Whitney §989, Speyer §380), because in Sanskrit participles are not only declinable (defined by case, number and gender), but also dis-tinguished by tense (past or present), and voice (active or passive). Thus it is rather an “indeclinable gerund” (see above, fn. 7). It is often translated with a perfect active participle (“having killed Vṛtra,” “having 10drawn a portrait”).  SÖHNEN-THIEME 42. Discussion of Sanskrit Examples 2.1. Let us start with Pāṇini's first rule on the usages of “ktvā-words” (3.4.18), con-cerning its usage with alam: alaṃ-khalvoḥ pratiṣedhayoḥ prācāṃ ktvā – “(it is used) after alam and khalu in the sense of a prohibition, according to the Eastern tradi-tions.”  Examples: 11Rām. 2.25.14: tad alaṃ te vanaṃ gatvā – “So enough of your going into the wilderness!” (Rāma to Sītā, after explaining the dangers and discomforts to be met with in the wilderness). Here we have a clear indication of the original in-strumental function of a verbal noun, as can be seen from a parallel alam con-struction with a noun, cf. MBh. 3.262.24: alaṃ te śaṅkayā – “Stop worrying!”  I have not met with any example for khalu with a gerund, but the gerund is also used in the instrumental function with kim, in the sense of “what is the use of?” e.g. Dkc. ucchv. 4: kiṃ tava gopayitvā – “What use is hiding it from you?” (= “Why should I not tell you?”), somewhat parallel to instrumenal constructions like BhG 10.42: kiṃ jñātena tavārjuna – “Why should you need to know this, Arjuna?” 2.2.  The most frequent use of the Sanskrit gerund is described in Pāṇini's last rule concerning the use of gerunds, 3.4.21: samānakartṛkayoḥ pūrvakāle – “If two ac-tions have the same agent, it is used for the earlier action.” Here it tends to be often equated with the active perfect participle (not with a gerund) in English; but this is mostly not really idiomatic and sometimes misses out on the under-tones of an original instrumental function. Examples:  (1) índro vṛtráṃ hatvā́ devátābhiś cendriyéṇa ca vyār̀dhyata (TS
“Because of killing Vṛtra, Indra was separated from the gods as well as deprived of his manhood.”  12 This represents its use in the function as an instrumental.11 The same causal relationship is also to be assumed for the example adduced by Hock 12(1991: 63): (19) (a) (h)atvā vṛtraṃ vijitya yuṣmābhir me 'yaṃ saha somapītha(ḥ) – “Because of (my) being victorious by killing Vṛtra this Soma-drinking together with you is mine” (= “I am entitled to drink Soma with you”). Hock’s translation, “Having … having …” is not really idiomatic English; it is considered to be “literal,” but the conception of it being “literal” is based on the idea of it being a perfect participle/active past participle. How to Translate a Gerund?  5(2) kroṣṭor vaṃśaṃ hi śrutvaiva sarvapāpaiḥ pramucyate (BrP, ASS, 13.213).
“For merely by listening to the genealogy of Kroṣṭṛ one is released from all sins.” Thus the two verbal actions are not only connected in a temporal sequence, but also in causal dependance, and the translation with a gerund (with a preposition) seems to be the better solution. Sometimes it is, however, better to translate a temporally related gerund construction simply with a subordinate clause using “when,” e.g.: (3) taṃ deśam āgamya … apaśyam iṣuṇā … tāpasaṃ hatam (Rām. 2.57.27).
“When I went to that place, I saw the ascetic struck by (my arrow).” (Daśaratha recalling a former misdeed of his before his death.) If it is only a sequence of actions, it is expressed in English best by parat-actically connected past tense verb forms (“he took his hat and went away,” rather than “having taken” or “after taking his hat he went away”). This is espe-cially suitable for sentences with chains of such gerunds ending with a finite verb (which are also often found in Pāli narratives). Examples: (4) atha siṃhaṃ … ṛkṣarājo … nihatya maṇiratnaṃ tad ādāya prāviśad guhām (BrP 16.27).
“Then the king of bears killed the lion, took that precious jewel, and en-tered the cave.” (5) āyurvedaṃ bharadvājāt prāpya … tam aṣṭadhā punar vyasya śiṣyebhyaḥ pratyapādayat (BrP 11.38).
“After obtaining the Āyurveda from Bharadvāja, he (Dhanvantari) subdi-vided it eightfold and conveyed it to his disciples.” (6) śucau deśe pratiṣṭhāpya sthiram āsanam ātmanaḥ, tatraikāgraṃ manaḥ kṛtvā … upaviśyāsane yuñjyād yogam ātmaviśuddhaye (MBh 6.28.11-12 [BhG 6.11-12]).
“In a pure place he should establish a firm seat for himself, concentrate his mind on one point and enter yoga for the purification of his self.” What is important in this case is that the two (or more) actions ought to be in the same mood, here the optative, according to the main verb. That is mostly valid also for examples with the main verb in the imperative mood or future   SÖHNEN-THIEME 6tense.  An alternative would be here to translate the gerund with a present par13 -ticiple. Examples: (7) mayi sarvāṇi karmāṇi saṃnyasya … nirāśīr nirmamo bhūtvā yudhyasva (MBh 6.25.30).
“Entrust all your actions to me and fight without any expectation or self-interest” (or: “Entrusting…fight…”) (8) sarvadharmān parityajya mām ekaṃ śaraṇaṃ vraja (MBh 6.40.66 [BhG 18.66]).
“Give up all traditional rules and take refuge only with me!” (or: “giving up…take refuge…”) The pūrvakāle of Pāṇini's rule is not always strictly observed; there are cas-es where the actions are more or less simultaneous (e.g. Renou's example from RaghuV 2.62) : māyāṃ mayodbhāvya parīkṣito 'si – “I have put you to the test by 14conjuring up an illusion.”  152.3. As for -kartṛkayoḥ in Pāṇini's rule one has to remember that kartṛ- denotes the function of the “agent,” not the syntactical role of the “subject,” thus it can be identified, in a sentence, in any syntactical role that would be defined by the case ending. In an active sentence the agent of a gerund is usually (but not necessari-ly) the subject in the nominative, whereas in passive sentences the agent appears usually (but not necessarily)  in the instrumental case. (Actually it may appear in 16any case, as required in the sometimes more sophisticated construction of a sen-tence, or it may be left out altogether).   17 One has to remember that the gerund/absolutive is not defined by tense, mood, or 13voice, apart from being indeclinable; this would depend on the action it preconditions.  Cf. Renou, vol.1, p.129; here the gerund seems to have the function of an instrumental.14 The last three examples quoted before would also confirm Renou’s observation, that the 15absolutive often expresses a simultaneous action, not only a preceding one (p.129). E.g., in the first two examples in § 2.2 above, the kartṛ of the gerund is identical with 16the subject of the passive main verb, whereas the instrumental appears in the function of a negated sociative (rather an apadāna than a kartṛ). For the nom. case with a passive main verb, cf. the first two examples above, for others 17see below. How to Translate a Gerund?  7Even if the gerund was originally mainly used in active sentences, the use in passive constructions became more popular with the development of the past participle constructions in colloquial language, which can be observed in direct speeches in the epics. Examples (direct speech of Daśaratha): (9) dvipo 'yam iti matvā hi bāṇenābhihato mayā (Rām. 2.58.13).
“Thinking it was an elephant I struck him with an arrow,” or:  (10) … śabdam ālakṣya mayā gajajighaṃsunā visṛṣṭo 'mbhasi nārācaḥ … (Rām. 2.58.15).
“On hearing this sound, I released an arrow towards the water, wishing to shoot an elephant” (11) mayāpi tat pitṛvyasanam ākarṇya … so 'bhihitaḥ … (Dkc. 4, told by Arthapāla).
“When I had heard about this disaster concerning my father, I said to him (i.e. Pūrṇabhadra)…” From the Purāṇas: (12) ity uktvā tāṃ pariṣvajya kṛtas tena manorathaḥ (PdP 5,51.24).
“After these words he (Indra) made his wish come true by embracing her.”  18The passive construction is also very popular in sentences with a so-called “polite imperative,” expressed either with a third person imperative passive or a gerundive, where the agent may be left out. Examples: (13) sītā madvacanād vācyā samāśvāsya prasādya ca bhartā te kuśalī rāmo … (MBh 3.264.56).
“On my behest Sītā ought to be told, with consolations and propitia-tions, ‘Your husband Rāma is in good health …’”  (14) kathyatāṃ ca drutaṃ gatvā paulomyā vacanaṃ mama (ViP 5,30.49).
“Go quickly and tell Paulomī what I have said!” (lit.: after going quickly, let my words be told to Paulomī). Another interesting example with a sophisticated imperfect passive of the desiderative: (15) śmaśānam upanīya kṛpāṇena prājihīrṣye (Dkc. 4).
“I was intended to be taken to a cemetery and slain with a knife.”  Here the two gerunds have quite clearly distinctive functions: the first is pūrvakāle, the 18second is used as an instrumental.  SÖHNEN-THIEME 8In all these instances the kartṛ- is not mentioned, but somehow obvious from the context. In a further example even two agents that are not mentioned (and two cases for the object) seem to be involved (the teacher for the gerund, the students for the passive phrase): (16) na hy anupadiśya varṇān iṣṭā varṇāḥ śakyā vijñātum (Patañjali, on vārtt. 16 in the introduction).
“For without instruction concerning the (classes of) phonetic sounds, the correct sounds cannot be understood.” 2.4. As for the two actions, which have to be inferred from the dual samānakartṛkayoḥ in Pāṇini's rule, the second one need not be a genuine action, it can also be a state or condition, e.g., (17) na mādṛśīm … dūṣayitvā pītāmṛtasyāpi tavāsti mokṣaḥ (Rām 3,48.24).
“After violating (a woman) like me there is no escape for you, even if you had drunk the nectar of immortality.” (18) tat sarvaṃ kathayām āsa gautamāya tapasvine | 
tasthau prahasya sa munir mahendraṃ ca vinindya ca || (BrVaiP 61.43).
“All that she (= Ahalyā) told Gautama, a practiser of special austerities;
and the ascetic stood there laughing and mocking great Indra.” Here the verb sthā is used rather as an auxiliary, which can be observed also in other attestations, e.g.: (19) saṃveṣṭya kṣāram udadhiṃ plakṣadvīpas tathā sthitaḥ (ViP 2,4.1).
“Likewise (the continent) Plakṣadvīpa surrounds (lit.: “stands surround-ing”) the salt ocean.”  2.5. For similar constructions with sthā, where the agent is not relevant at all, Pāṇini provided the preceding rule (3.4.20) parāvarayoge ca – “and in connection with ‘beyond’ and ‘on this side,’” which is obviously relevant for geographic descrip-tions like the one given above, cf. the examples given by the Kāśikā: atikramya tu parvataṃ nadī sthitā 
“Beyond (lit.: ‘after crossing’) the mountain there is a river.” aprāpya nadīṃ parvataḥ sthitaḥ
“There is a mountain on this side of (lit.: ‘without reaching’) the river.”  How to Translate a Gerund?  9Here no action is involved in the main phrase of the sentence (for the agent one may think of somebody exploring the geographical situation). 2.6. At this point, before I consider more complex examples, I would like to introduce diagrams for the patterns dealt with so far. First there are three diagrams show-ing basic structures, representing three forms of statements: active, passive, and statement of a situation or condition, with the two layers of subordinate gerund and main verb. These are found in the middle of the boxes, with the agent on the left, and the two objects of the two verbal actions on the right hand side. Diagram 1. Active. Diagram 2. Passive. Diagram 3. Descriptive Statement.      The following diagrams are the models of possible “Modifications,” with corresponding examples, e.g., when there is only one object (e.g., tad) for both actions, then when the object is the same, but in a different case in both actions (which works well with a neuter); then we have the same object, but in different cases (acc. with the gerund, nominative with the passive main verb), with two different agents to be assumed (but not expressed) for the two different actions. Layer of gerund actionAginstrgerund ±objaccLayer of main action verbact ±objaccLayer of gerund actionAginstrgerund objaccLayer of main verb action verbpass objnomLayer of modifying clause ± agent gerund ±objaccLayer of main clause = subjnom verbal action ———  SÖHNEN-THIEME 10Diagram 4. Modifications: Active.   Diagram 5. Modifications: Passive.    Then there are two versions of the polite imperative, with the agent to be imagined; in the second case the object is expressed in the nom. in the main action, but imagined in the acc. in the gerund. Diagram 6. Passive (imperative).   The last three diagrams exemplify patterns which have no object in the main action, since the verbs, or verbal expressions, to which the gerunds are connected, express not an action but a condition (twice a form of sthā-, and once as a verbal noun + as-). Aginstrgerund obj =verbact objacc(tvam)kṛtvātadmā vināśaya(Ag) ≠ gerund objacc  =(Aginst) verbpass objnom(ācāryeṇa) anupadiṣya varṇān(śiṣyaiḥ) na śakyā vijñātum varṇāḥindreṇayajñair avāpyadevatvambhujyate(Ag = ) gerund objacc  =Aginst verbpass objnom(Ag = ) gerundAginst verbpass objnom(yuṣmābhiḥ)gatvākathyatām vacanam(Ag = ) gerund (objacc  = )Aginst verbpass objnom(tvayā)samāśvāsyavācyā Sītā How to Translate a Gerund?  11Diagram 7. Descriptive statement.   2.7. Before concluding, I would like to present some more complex examples with gerunds not connected to the main action. Here it is of course important to es-tablish to which action the gerund is related (the verbal expressions connected with an agent, mostly participles, are underlined): (20) sa tad rājā vacaḥ śrutvā … duḥkhārto na kiṃcid vyājahāra (MBh 3.261.2).
“Tormented with grief by hearing this speech, the king did not say any-thing.” (nom.) (21) tato rudraprayuktena siṃhaveṣeṇa līlayā |
devyā manyukṛtaṃ buddhvā hato dakṣasya sa kratuḥ || (BrP 39.50).
“Dakṣa’s sacrifice was playfully destroyed by (Vīrabhadra) in a lion’s at-tire, employed by Rudra because of becoming aware of Devī’s anger.” (instr.) (22) uttiṣṭhatas tasya … mahāvarāhasya mahīṃ vigṛhya … śarīram … munayaḥ stuvanti (ViP 1.4.29).
“The ascetics praise the body of the huge boar when he rises holding up the earth.” (gen.) (23) mālāṃ pragṛhya devyāṃ tu sthitāyāṃ devasaṃsadi … (BrP 36.27).
“When the Goddess was standing there in the assembly of the gods, holding a garland in her hand …” (loc. abs.) agentnom gerund objacc= Aginst verbintr ———sa muniḥprahasya mahendramtasthau ———(agent?) atikramya parvatamnadī sthitā ———(agent?) gerund objaccsubjnom verbintr ———(agent) gerund objacc= indir. objgenverbintr ———tavadūṣayitvā mādṛśīṃna … asti mokṣaḥ ———  SÖHNEN-THIEME 12(24) tayā … puruṣair abhigṛhya tarjitayā daṇḍapāruṣyabhītayā nirbhin-naprāyaṃ rahasyam (Dkc. 4).
“By her, (having been) arrested and threatened by the guards, afraid of harsh punishment (lit.: ‘the harshness of punishment’), the secret was divulged in its essential part.” (instr.) Similarly, from the Raghuvaṃśa, two examples spoken by Rāma, address-ing Sītā, on their flight back to Ayodhyā:(25) puraṃ niṣādādhipater idaṃ yasmin mayā maulimaṇiṃ vihāya | 
baddhāsu jaṭāsu arudat sumantraḥ kaikeyi kāmāḥ phalitās taveti || (Ragh.13.59). 
“Here is the city of the Niṣāḍha king, where Sumantra lamented, when I had braided my hair, after putting away my royal diadem ‘O Kaikeyī, your wishes have come to fruition!’” (Here the loc. abs. has to be converted into active, in order to arrive at a digestible translation.) (26) addhā śriyaṃ pālitasaṃgarāya pratyarpayiṣyaty anaghāṃ so sādhuḥ |
hatvā nivṛttāya mṛdhe kharādīn saṃrakṣitāṃ tvām iva lakṣmaṇo me || (Ragh. 13.65).
“Now this noble (brother of mine) will hand back to me, who kept (my father's) promise, the royal glory unpolluted, in the same way as Lakṣ-maṇa handed you back to me well-protected, when I had returned from slaying Khara and the other rākṣasas in battle.” (dat.) Referring to a past participle as first member of a bahuvrīhi compound: (27) tayā dattāni bhojyāni … bhuktvā labdhabalāḥ santaḥ … (MBh 3.266.41).
“Having regained our strength by eating the food she gave us…” (implied instr.) Without any identifiable agent: (28) kiṃ hi kṛtvā viṣaṇṇas tvaṃ (Rām 2.27.5).
“Why (lit. ‘because of which action’) are you desponding?” And recently I came across an especially tricky complex construction in the MBh: (29) draupady uvāca: droṇaputrasya sahajo maṇiḥ śirasi me śrutaḥ |
nihatya saṃkhye taṃ pāpaṃ paśyeyaṃ maṇim āhṛtam |
rājan, śirasi taṃ kṛtvā jīveyam iti me matiḥ || (MBh 10. 11.20).
[After she was asked by Yudhiṣṭhira to give up her plans of suicide (i.e.- How to Translate a Gerund?  13fasting to death)], Draupadī said: “I have heard about a jewel on the head of Droṇa's son, born together with him. If I were to/could see the jewel taken away [by x] after killing that evil person in battle, I could live on, placing it on your head, o king – This is my decision.” This contains, of course, the hidden adhortation that someone should kill Droṇa’s son Aśvatthāman, take off the jewel and give it to her, but since she does not know what will happen, she cannot yet name the agent (it might be Bhīma, Arjuna, or even Kṛṣṇa). It is mainly the reluctance to express the agent, which makes such exam-ples difficult to translate. Conclusion Looking at the list of examples, one may notice that the most frequent construc-tion of the gerund (the first basic model in the diagrams), is underrepresented (only ex. 4-5 and BhG teaching in 6-8), although many examples could have been adduced from the epics, Purāṇas, and classical court literature, so far as they are narratives. As for passive constructions, it is striking how many more are found in the quotation of direct speeches than in narrative sections. There is the past partici-ple construction in direct speech (e.g., examples 3,9-10) and later in the Purāṇas (ex. 12 and 21) and in the Dkc. (11), then the different possibilities of an indirect, “polite” imperative (ex. 13-14). In all these examples the agent is thought of as an instrumental, even if it is omitted in the text; sometimes it is left open to whom it refers, as in ex. 16 (Dkc.) and with the first gerund in ex. 29 (apart from the po-lite imperatives).  Since past tenses are in Prakrit generally expressed with past participles and the agent of a transitive verb in the instrumental, one may ask whether that has influenced also spoken Sanskrit (e.g., in the epics), and may later also have been taken over in narrative texts (like the Purāṇas, Pañcatantra literature, etc.). But this would have to be studied further. Finally, Western attempts to classify or categorize the Sanskrit gerund are to be revisited. Many grammars classify the Sanskrit ktvā simply as an “indeclin-able participle,” which is contradictory in itself, as could be shown, and are con-sequently mostly worried about the problem of the agent. It should have become clear that it may also express a causal relationship; thus its relationship to the main action can be preceding, modifying or accompanying the main action. It may also modify subordinate actions expressed in verbal adjectives (participles),   SÖHNEN-THIEME 14including verbal adjectives in loc. abs. constructions or in bahuvrīhi compounds. Since it is an “indeclinable,” it is connected with a verb (or a verbal expression) which it modifies, and since it is derived from a verbal root, it seems justified to call it a “verbal adverb,” as I ventured to define it in my translation of the Stenzler primer.  (Pāṇini’s descriptive approach seems, after all, to be the most appro19 -priate.) In English the most literal translation would often be to use an English gerund, not a past active participle. How to transform it into idiomatic good English, is another question, which depends on the individual case.  20Bibliography Dasgupta, Probal. 1986. “Gerunds, Participles, and the Theory of Categories.” Bulletin of the Deccan College Postgraduate & Research Institute 46: 7-18. Delbrück, Berthold. Altindische Syntax. Deshpande, Madhav M. 1981. “Sanskrit Gerund Constructions: Syntactic Dispu-tations.” Indo-Iranian Journal 23: 167-185. Gonda, Jan. 1967. “On the Use of the Absolutive in Sanskrit.” In Kavirāj Abinanda-na Grantha, edited by Baburam Saksena. Lucknow: Akhil Bharatiya Sans-krit Parishad, 262-265. [= Selected Studies, vol. III, Leiden, 1975, 91-94.] Haspelmath, Martin, E. König, ed. 1995. Converbs in Cross-Linguistic Perspective. Structure and Meaning of Adverbial Verb Forms (Adverbial Participles, Gerunds). Berlin: Mouton de Gruyter. Hock, Hans Henrich. 1987. "Reduced-clause and Clause-union Absolutives and Participles in Vedic Prose.” Select Papers from SALA 7 (VIIth South Asian Lan-guages Analysis Round Table), edited by Elena Bashir, Madhav M. Deshpande, and Peter Hook. Bloomington: Indiana Univ. Linguistics Club, 182-198. ———. 1991.”Possessive Agents in Sanskrit?” Studies in Sanskrit: A Volume in Honor of the Centennial of Speijer’s Sanskrit Syntax, edited by H.H. Hock. Delhi: Mo-tilal Banasidass.  It seems that its non-definedness on either nominal agreement or verbal categories 19made it an especially popular linguistic implement in the further development of Sanskrit and other Indo-Aryan languages.  This is convincingly exemplified in L. Rocher’s article (1980), “A Note on the Sanskrit 20Gerund.” How to Translate a Gerund?  15Hofmann, J. B. 1965. Lateinische Syntax und Stilistik, neubearbeitet von A. Szantyr. (Lateinische Grammatik, 2. Band.) München: Beck. Verbesserter Nachdruck, München 1972. Jeffers, Robert and Robert Kantor. 1984. “A History of the Sanskrit Gerund.” Indo-germanische Forschungen 89: 99-103. Mayrhofer, Manfred. 1965. Sanskrit-Grammatik, mit sprachvergleichenden Erläute-rungen. Berlin: de Gruyter. Morgenroth, Wolfgang. 1977. Lehrbuch des Sanskrit. 2. erw. u. überarb. Auflage. München: Hueber. Neisser, Walter. 1928. Review of: Hanns Oertel, The syntax of Cases in the narrative and descriptive prose of the Brāhmaṇas. I. The disjunct of cases (Heidelberg 1926). Zeitschrift für Indologie und Iranistik 5: 319-326. (= Kleine Schriften, ed. by Peter Das, Wiesbaden: Franz Steiner 1980.) Pāṇini. 1887. Aṣṭādhyāyī. Ed. and transl. by Otto Boehtlingk. Leipzig. Reprinted Hildesheim: Georg Olms, 1964. Quirk, Randolph and Sidney Greenbaum. 1973. A University Grammar of English. Harlow, Essex: Longman Group. Renou, Louis. 1930. Grammaire sanscrite. 2 Vols. Paris: Adrien-Maisonneuve.  Rocher, Ludo. 1980. “A Note on the Sanskrit Gerund.” Recherches de Linguistique: Hommage à Maurice Leroy, edited by J. Bingen, A. Coupez, and F. Mawet. Université libre de Bruxelles, 181-188. Söhnen[-Thieme], Renate. 1985. “Die Konstruktion des Absolutivs im Sanskrit.” Grammatische Kategorien: Funktion und Geschichte. (Akten der VII. Fachtagung der Indogermanischen Gesellschaft, Berlin, 20.-25. Februar 1983), edited by Bernfried Schlerath and Veronica Rittner. Wiesbaden: Dr.Ludwig Reichert Verlag, 478-489. Speijer, J. S. 1886. Sanskrit Syntax. Leiden. Reprinted in India, Delhi: Motilal Bar-sidass, 1973. Stenzler, A. F. 1915. Elementarbuch des Sanskrit. (Grammatik – Texte – Wörter-buch). 9. Auflage, fortgeführt von Richard Pischel, umgearbeitet von Karl F. Geldner. Gießen: Alfred Töpelmann. [17. Auflage, nachgedruckt Berlin: de Gruyter, 1980.]   SÖHNEN-THIEME 16Stenzler, A. F. 1992. Primer of the Sanskit Language, translated by Renate Söhnen[-Thieme]. School of Oriental and African Studies: London. [2nd revised edition 1997.] Thumb, A. and R. Hauschild. 1958-59. Handbuch des Sanskrit. 2 vols. Heidelberg: Carl Winter. Whitney, Dwight D. 1879. Indische Grammatik, umfassend die klassische Sprache und die älteren Dialekte. (Bibliothek Indogermanische Grammatiken, Band II.) Leipzig: Breitkopf und Härtel.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items