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

Sūtra Paraphrases in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti Ben-Dor, Sharon 2019

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
70440-Ben-Dor_Sharon_Sutra_paraphrases_2019.pdf [ 1.02MB ]
Metadata
JSON: 70440-1.0380177.json
JSON-LD: 70440-1.0380177-ld.json
RDF/XML (Pretty): 70440-1.0380177-rdf.xml
RDF/JSON: 70440-1.0380177-rdf.json
Turtle: 70440-1.0380177-turtle.txt
N-Triples: 70440-1.0380177-rdf-ntriples.txt
Original Record: 70440-1.0380177-source.json
Full Text
70440-1.0380177-fulltext.txt
Citation
70440-1.0380177.ris

Full Text

 Sūtra Paraphrases in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti: The Interpretation of Ablative Case in the Ātmanepada and Parasmaipada Sections Sharon Ben-Dor Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13, 2018, Section 3: Vyākaraṇa.  Section Convenors: Malhar Kulkarni and Peter Scharf
General Editor: Adheesh Sathaye Published by the Department of Asian Studies, University of British Columbia, on behalf of the International Association for Sanskrit Studies. DOI: 10.14288/1.0380177.
URI: http://hdl.handle.net/2429/71102. Suggested Citation Format: MLA:
Ben-Dor, Sharon. “Sūtra Paraphrases in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti: The Interpretation of Ablative Case in the Ātmanepada and Parasmaipada Sections.” Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13, 2018, Section 3: Vyākaraṇa. Edited by Malhar Kulkarni and Peter Scharf, 2019. DOI: 10.14288/1.0380177. APA:
Ben-Dor, S. (2019). Sūtra Paraphrases in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti: The Interpretation of Ablative Case in the Ātmanepada and Parasmaipada Sections. In M. Kulkarni and P. Scharf (eds.) Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13, 2018, Section 3: Vyākaraṇa. DOI: 10.14288/1.0380177. Chicago:
Ben-Dor, Sharon. 2019. “Sūtra Paraphrases in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti: The Interpretation of Ablative Case in the Ātmanepada and Parasmaipada Sections.” In Proceedings of the 17th World Sanskrit 
Conference, Vancouver, Canada, July 9-13, 2018, Section 3: Vyākaraṇa, edited 
by Malhar Kulkarni and Peter Scharf. DOI: 10.14288/1.0380177. 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). http://creativecommons.org/licenses/by-nc-nd/4.0/वैधुसव ्मकबुंटुकअ ारा यसं तृा यनसमवायःINTERNATIONAL ASSOCIATION OF SANSKRIT STUDIES THE 17TH WORLD SANSKRIT CONFERENCE, VANCOUVER, CANADA, JULY 9-13, 2018 Sūtra Paraphrases in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti: The Interpretation of Ablative Case in the Ātmanepada and Parasmaipada Sections Sharon Ben-Dor Abstract Various scholars have noted the similarities among the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti. The examination of the paraphrases of the authors of these texts on the sūtras occurring in the section which concerns the addition of the parasmaipada and ātmanepada endings confirms that these paraphrases share a common pattern. This pattern might have belonged to an earlier source which the three authors were using. Although the Mahāvṛtti is consistent and keeps the same pattern in its paraphrases of the sūtras, the Cāndravyākaraṇavṛtti and Kāśikāvṛtti have inconsistencies and variations in respect to the way the sū-tras are paraphrased. There are clear reasons for the variations found in the Cān-dravyākaraṇavṛtti, and they are the result of the author’s own intention. In con-trast, the variations found in the Kāśikāvṛtti in paraphrases of the parallel sūtras are probably the result of later modifications to the original text, which may have had consistency in the way the sūtras were paraphrased. The variations found in the current text originate from two opposite approaches in the way additional specifications are mentioned in the commentary. The version of the paraphrases which appears in the current Kāśikāvṛtti is a mixture of the paraphrases of these two approaches.  Keywords: Vyākaraṇa; Cāndravyākaraṇavṛtti; Kāśikāvṛtti; Mahāvṛtti; Jainendra; Cāndra. Proceedings of the 17th World Sanskrit Conference, Vancouver, Canada, July 9-13 2018, Section 3: Vyākaraṇa, edited by Malhar Kulkarni and Peter Scharf, 2019. DOI: 10.14288/1.0380177. BEN-DOR 2I. The similarity between the Cāndravyākaraṇavṛtti and the Kāśikāvṛtti has been not-ed previously.  In addition, scholars have pointed out that the Kāśikāvṛtti shares 1resemblance with Abhayanandi’s Mahāvṛtti.   2In this paper, I compare these commentaries on the sūtras which occur in the section on the additions of parasmaipada and ātmanepada endings.  In both 3the grammars of Candra and Jainendra, most of the sūtras in this section are quite identical to Pāṇini’s sūtras and they are given in almost the same order. Therefore, this section is beneficial for comparing these commentaries regarding various aspects. Here, I examine the way the authors paraphrase the sūtras of this section and I focus particularly on the way the commentators interpret the  See, for example, Aussant 2009. 1 It is accepted that the Cāndravyākaraṇavṛtti is earlier than the Kāśikāvṛtti and the 2Mahāvṛtti. The latter two are probably from the same period (7th CE) and it is uncertain which of the two texts is earlier. In addition, it is claimed (Bronkhorst 2004: 52) that the Mahāvṛtti has been inspired by an earlier lost commentary, namely Devanandin’s commentary on the Jainendra-vyākaraṇa. There is disagreement among scholars re-garding the source of the Kāśikāvṛtti and whether it was influenced by Devanandin’s commentary or by another Pāṇinian source. See Aussant (2009: 192) for a brief sum-mary on the different opinions. In this paper, I do not aim to settle this matter. Yet, I will argue that the Mahāvṛtti strictly follows and has preserved the style of an earlier version of the paraphrases. This version is found also in Cāndravyākaraṇavṛtti and the Kāśikāvṛtti in respect to some sūtras. Therefore, their source should be earlier than the Cāndravyākaraṇavṛtti. The Kāśikāvṛtti and Mahāvṛtti have directly or indirectly used the same source. In the Aṣṭādhyāyī, this section starts with A 1.3.12 (अनदुा&िङत आ+नपेदम  ्) up to A 1.3.93 (िट3च 3ृपः), in Cāndravyākaraṇa, C 1.4.046 (यथापाठम)् - C 1.4.145(िट कऌपः) and in Jainendra-vyākaraṇa, J 1.2.6 (ङनदुा&देतो दः) - J 1.2.89 (िट च 3ृपः). The grammars use different termi-nology. For convenience, in this paper I use Pāṇini’s terminology when I refer to the grammars in general. In cases where I refer to a particular grammar, I mention its own terminology. Sūtra Paraphrases 3ablative ending occurring after the upasargas which are mentioned in the ma-jority of the sūtras occurring in this section.   4As in other vṛtti-type commentaries, the comments of the Cān-dravyākaraṇavṛtti, the Kāśikāvṛtti and the Mahāvṛtti include a paraphrase for each sūtra. From a brief reading of the paraphrases found in these commentaries, one can observe that in many cases the authors of these texts paraphrase the same corresponding sūtra differently. In addition, particularly in the Kāśikāvṛtti, there are variations in the style of the paraphrases and it seems that the authors of the Kāśikāvṛtti did not aim for consistency in respect to the manner they paraphrase Pāṇini’s sūtra. One may ask whether a commentary should have consistency in respect to the manner of how the sūtras are paraphrased. Puruṣottamadeva in his Bhāṣāvṛtti, for example, is quite consistent in respect to the style of his para-phrases to Pāṇini’s sūtras.  In addition, as we shall see, also the Cān5 -dravyākaraṇavṛtti and particularly the Mahāvṛtti, keep a particular pattern in re-spect to many sūtras, and in cases where there are variations, one can find rea-sons why the authors choose to paraphrase the sūtra in question in a different way than their typical way. The authors of the Kāśikāvṛtti, on the other hand, do not seem to be consis-tent in respect to the way they paraphrase Pāṇini’s sūtras. A reason for inconsis-tencies in the Kāśikāvṛtti may be that the authors of the Kāśikāvṛtti did not aim for consistency, and they phrased each sūtra in the way which they found it to be the most suitable for the sūtra, whether it was their own originality or they simply picked one of the paraphrases from the variety of paraphrases which they found  A comparison of the paraphrases found in the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti and 4the Mahāvṛtti on the section on the addition of parasmaipada and ātmanepada end-ings raises many valuable issues. In this paper, I discuss one of them which I think highlights the relation between the texts and the character of each of them and pro-vides also information about their common source. In notes 94 and 97, I refer to other features in the Kāśikāvṛtti’s paraphrases which raise similar issues. See BhV (24-34).5 BEN-DOR 4in the various sources that they were using.  Yet, in this paper, I show that by 6comparing the paraphrases of the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti of the sūtras in the section of parasmaipada and ātmanepada endings, one can find that the three texts share a common pattern. This pattern may have originated from a source earlier than the Cāndravyākaraṇavṛtti, and the authors of the Kāśikāvṛtti and Mahāvṛtti were influenced by that source, directly or indi-rectly. In addition, as we shall see, one can find some reasons for the inconsis-tency in the Kāśikāvṛtti on this matter, and it appears that the original text of the Kāśikāvṛtti has had consistency. The reason why it is absent from the current text is an integration of different approaches regarding the way of specifying the scope of Pāṇini’s sūtras and addressing difficulties which the commentators found in Pāṇini’s Aṣṭādhyāyī.  II.1. In general, both parasmaipada or ātmanepada endings can be added after verbal bases.  In the section on the addition of parasmaipada and ātmanepada endings 7(A 1.3.12: अनदुा&िङत आ+नपेदम  ्– A 1.3.93: िट च 3ृपः), Pāṇini presents restrictive rules which account for instances where only parasmaipada or ātmanepada end-ings are added after a base. In this section, there are two major rules; A 1.3.12 (अनदुा&िङत आ+नपेदम )् and A 1.3.78 (शषेा>त ?िर परAपैदम )्. According to A 1.3.12, after ver-bal bases which are marked with anudatta or with the anubandha ṅ, only ātma-nepada endings are added. According to A 1.3.78, when an agent is to be signi-fied (kartari), only parasmaipada endings are added to the remainder (śesa), that is when the conditions for the addition of only ātmanepada after the bases men-tioned in A 1.3.12-1.3.77 are not met. Most of the sūtras in the section on the parasmaipada and ātmanepada endings may be taken as extensions or excep- As they say themselves in their opening verses. KV I.1-4: व&ृौ भाFे तथा धातनुामपारायणािदष।ु6िवJकीण ?L तML िNयते सारसङPहः॥ इTपुसङUानवती शVुगणा िववतृगढूसZूाथा ?। [\ु]^पिसिVवृ ?ि&िरयंकािशका नाम॥ [ाकरणL शरीरं पिरिनि`तशाaकाय ?मतेावत।् िशbः पिरकरबdः िNयते ऽL Pfकारणे॥ In their commentary on the Kāśikā’s introductory verse, the Nyāsa and the Padamañjarī refer to particular authors (Culli, Bhaṭṭi, Nallūra and Kuṇi) of vṛttis. Nyāsa I.2 (on the KV’s in-troductory verse): तZ च विृ&ः पिणिनJणीतानां सZूाणां िववरणं चिूgभिhनgूरािदिवरिचतम॥् PM I.2 (on the KV’s introductory verse): तZ सZूाथ ?Jधा नो Pfो वृि&ः। सा चहे पािणिनJिण ता नां सZूाणांकुिणJभिृतिभराचायijवरिचतं िववरणम॥् Cf. A 3.4.77 (लL), A 3.4.78 (ितlिmझ िसoथpिमqArाताझंथासाथाsंिमtिहमिहङ)् A 3.2.124 (लटः7शतशृानचावJथमासमानािधकरण)े and A 3.2.125 (सuोधन े च). Sūtra Paraphrases 5tions to A 1.3.12 and A 1.3.78. These sūtras mainly refer to particular bases with certain conditions: when the base is preceded by one or more particular upasar-gas, is used either transitively or intransitively, or is used in a specific meaning. In such specified conditions, only parasmaipada or ātmanepada endings are added after the bases mentioned in these sūtras. The majority of the sūtras in the section on the addition of parasmaipada and ātmanepada endings refer to specific bases which are preceded by a specific upasarga. In such sūtras the bases and upasargas are mentioned with a “techni-cal ablative” which Pāṇini introduces by A 1.1.67 (तAािदv&ुरL). According to this sūtra, in cases of ambiguity with the ordinary meaning of the ablative, a stipu-lated grammatical operation applies to the item following the item which is re-ferred to by a term mentioned in the ablative. In other words, the ablative end-ing has the meaning of “after.”  For example, in A 1.3.17 (नjेवशः) both the terms ni 8and viś are mentioned in the ablative. Accordingly, the meaning of this sūtra is “after (the verbal base) viś which is preceded by (the upasarga) ni, only āt-manepada endings are added.”  This sūtra accounts for forms such as niviśate.  9 10Otherwise, when viś is not preceded by ni, parasmaipada endings are added to this base according to A 1.3.78 which accounts for forms such as praviśati.  11Some of the sūtras in the section on parasmaipada and ātmanepada end-ings mention one specific upasarga with an ablative ending. Other sūtras men-tion a list of upasargas which are compounded and the compound ends with the same ending. As we shall see, in their paraphrases of such sūtras, the Cāndra-vyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti share a common pattern for dis-playing the upasargas and the ending. Yet there are many irregularities, particu-larly in the Cāndravyākaraṇavṛtti and the Kāśikāvṛtti. These irregularities are a result of two major difficulties which the commentators raise, and as we shall see, they have an effect on the way the authors display the upasargas and clarify the technical ablative ending in their paraphrases. Therefore, it is necessary to clarify these difficulties before presenting the commentators’ paraphrases.   In his vārttika on A 1.1.66/1.1.67 (Bh I.172.15), Kātyāyana argues that this type of ablative 8ending expresses also an immediate sequence. Thus, according to him, the meaning of this ending is “immediately after.”       The term ātmanepada continues by anuvṛtti from A 1.3.12.9 Present, 3rd person, sg., ātmanepada of ni viś.10 Present, 3rd person, sg.,  parasmaipada of pra viś.11 BEN-DOR 6II.2.  With respect to the sūtras that refer to upasargas, the commentators raise the question of why such sūtras do not refer to any other item which has an identical sound sequence. In other words, how does one know that these sūtras refer only to an upasarga and not to any other identical units? For example, A 1.3.17 (नjेवशः) refers to the verbal base viś when it is preceded by ni. One may argue that this sūtra may undesirably be applicable also to the case of madhuni viśanti bhramarāḥ ‘bees enter the honey’ on the basis that the base viś is preceded by ni. In this case, the form ni of madhuni  is not an upasarga, but the final n of the nominal stem 12madhun and the locative ending ṅi. In order to remove this difficulty, The Bhāṣya suggests adding the term “upasarga” to the sūtras. This addition would provide the knowledge that the sūtra refers only to an upasarga and not to any other identical item.   13 madhuni (←madhun + i ← madhunum + ṅi [A 7.1.73: इकोऽिच िवभwौ] ← madhu + ṅi ).12 The suggestion is mentioned in Bh I.279.9 (on A 1.3.19): उपसग ?Pहणं कत ?[म।् परा जयित सनेिेत॥13And in Bh I.280.2 (on A 1.3.21): उपसग ?Pहणं कत ?[म।् इह मा भतू ।् अन ु Nीडित माणवकम॥् In addi-tion, it is discussed in the Bhāṣya on A 1.3.60. See notes 44 and 45. In this passage, Kā-tyāyana says that the term upasarga is needed for removing another difficulty which I discuss in section II.3. It is noteworthy that the statements upasargagrahaṇaṃ kar-tavyam in Patañjali’s comments on A 1.3.19 and A 1.3.21 appear as paraphrases to vārt-tikas, but the vārttikas are not mentioned. It may be that the view that the sūtras are applicable to such cases is not Kātyāyana’s view, but a later one. In addition, in cases such as madhuni viśanti bhramarāḥ, the derivation has reached its final stage and the parasmaipada ending was already added after the base. Therefore, the application of A 1.3.17 in respect to madhuni viśanti bhramarāḥ, for example, involves rejecting the deci-sion that was already made regarding which type of ending should be added, and thus removing the parasmaipada ending and starting the derivation again. Such a proce-dure is questionable and such difficulties may be irrelevant to Pāṇini’s system. Sūtra Paraphrases 7Later commentators invoke some additional paribhāṣās for explaining why the sūtras do not refer to other identical units which are not upasargas.  14However, it is generally accepted that the desired knowledge is obtained by the way of association (sāhacarya).  Accordingly, one knows that the term in ques15 -tion is an upasarga because the other term or terms which are mentioned in the sūtra are clearly upasargas. Patañjali mentions this reasoning while explaining how one knows that the term parā in A 1.3.19 (िवपरा{ां जःे) does not refer to parā in parā jayati senā, “the superior army wins.”  Thus, Patañjali rejects the suggestion 16of adding the term upasarga to the sūtra. II.3. In addition to the difficulty mentioned above, Kātyāyana raised another difficul-ty in respect to sūtras such as A 1.3.17. It is traditionally accepted that the techni-cal ablative which is provided for by A 1.1.67 requires an immediate sequence  The Kāśikāvṛtti, for example, invokes the paribhāṣā arthavadgrahaṇe nānarthakasya in 14respect to the case of madhuni viśanti bhramarāḥ. KV p.55 (on A 1.3.17): न|ेपसग ?LPहणमथ ?व}हणे नानथ ?कL Pहणिमित तAािदह न भवित मधिुन िवशि~ मराः ॥ According to this paribhāṣā, a meaningful grammatical form, mentioned in a sūtra, does not refer to meaningless sound sequences. Arguing that the form ni in madhuni is meaningless is based on the Pāṇinīyas’ view that the augment num is considered the final part of the form to which it is attached (C.f. A 1.1.47: िमदचोऽ€ा\रः). Therefore, the form madhuni is divided into the base madhun and the case ending i. The Nyāsa refers to this issue. Nyāsa II.28 (under KV on A 1.3.17): मधशुL सl‚केवचन इको ऽिच िवभwािवित निुम कृत े मधनुीित ^पंभवित तL चावयवो िनशोऽन थ ?को ॥ It is expressed by the paribhāṣā sahacaritāsahacaritayoḥ sahacaritasyaiva grahaṇam – 15paribhāṣā 111 in Nāgeśa’s Paribhāṣendubhāskara. Bh I.279.9-13 (On A 1.3.19): उपसग ?Pहणं कत ?[म ् (vt. 1)। परा जयित सनेा इित । त&jह वw[म।् न16वw[म।् यƒिप तावदयम प्राशो bापचार उपसग ?…ानपुसग ?…ायं त ु ख िवशो ऽbापचार उपसग ? एव॥ A 1.3.17 (नjेवशः) does not include another term that can indicate that the term ni is an upasarga. However, the context in which this sūtra is given, that is in a section that concerns a base preceded by upasargas, provides the knowledge that A 1.3.17 refers to an upasarga and not to any other identical unit. Sāyaṇa uses this reasoning and argues that the term ni in A 1.3.17 is an upasarga by sāhacarya with the term pari, etc., mentioned in the following sūtras that provide ātmanepada endings. िनिवशते । नjेवश इित त&रसू Zोपा&ःैपया ?िदिभः साहचया ?ि]िरहोपसग‰ गŠृत े न त ु त‹ित^पक इvŒे िनिवशत भयािदvZ तङभावः Jित^पकापŽेो b[ः ॥MDhV p. 489 (on root VI.131). The example quoted here by Sāyaṇa is taken from Raghu-vaṃśa 12.38. BEN-DOR 8between the item mentioned in the ablative and the item on which the gram-matical operation occurs.  Accordingly, A 1.3.17 is applicable only to cases where 17the base viś is immidiately preceded by ni. However, it is desired that this sūtra is applicable also to cases such as nyaviśata ( ← ni aviśa + ta ← ni aviś + śap + ta ← ni aviś + ta ← ni aṭviś + laṅ ← ni viśa  + laṅ ).  In this case, viś does not follow immediately 18 19ni because the augment aṭ intervenes. Therefore, one may argue that A 1.3.17 is not applicable to this case. Later commentators usually use a paribhāṣā to re-move this difficulty. For example, in respect to nyaviśata, the Kāśikāvṛtti says that on the basis of the paribhāṣā yadāgamās tadguṇībhūtās tadgrahaṇena gṛhyante, the augment aṭ does not intervene between the ni and viś.  20I shall now discuss how the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and Mahāvṛtti paraphrase the sūtras which include upasargas with a technical abla-tive ending. As we shall see, the difficulties mentioned in this section affected the way the Cāndravyākaraṇavṛtti and the Kāśikāvṛtti paraphrase some of the sū-tras in the section on the addition of parasmaipada and ātmanepada endings. III.1. In the paraphrases of sūtras which include one specific upasarga, the common way of the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti is to mention the upasarga in a compound with the direction word “pūrva.” The compound ends with an ablative ending and the number of this ending is in agreement to  Cf. Bh I.172.15 (A 1.1.66 vt.1)17 DhP VI.130.18 A 3.2.111 (अनƒतन े लङ)्, A 6.4.71 (ललृ‘डुदा&ः), A 3.4.77 (लL) A 3.4.78 19(ितlिmझिसoथpिमqArाताझंथासाथाsंिमtिहमिहङ)्, A 3.1.68 (कत ?िर शप )्, A 6.1.87 (आ’णुः), A 6.1.77 (इको यणिच). Pāṇinīyas accept that the augment aṭ is added in the stage of the l affixes (lāvasthāyām) that is, before their replacement by the ending by A 3.4.78. Kātyāyana and Patañjali (Bh I.286.1-287.2) reach this conclusion after discussing the order of ap-plication of A 3.4.77/A 3.4.78 and A 6.4.71. Their arguments on this matter involve many details. Therefore, I do not mention them here. Obviously, if A 3.4.77/A 3.4.78 apply before A 6.4.71, the difficulty does not occur. KV p.55 (on A 1.3.17): यदागमाr}हणेन गŠृ~े तनेाटा नािr [वधानम।् “िवशत॥ On the basis of A 201.1.46 (आƒ~ौ टिकतौ) Pāṇinīyas consider the augment aṭ as an integral initial part of the base to which it is added. According to this paribhāṣā, a speech form referring to an item also refers to the item with an augment added. Thus, the term viś in A 1.3.17 refers also to the form aviś ( aṭ + viś ) which immediately follows the upasarga ni. Sūtra Paraphrases 9the number of the bases which are qualified by the compound. For example, the paraphrases of the three texts for A 1.3.24 (उदो ऽनsू ?कम ?िण) and its corresponding sūtras C 1.4.66 (उदो ऽनsू”हायाम)् and J 1.2.19 (उद ईह)े are:  KV on A 1.3.24: utpūrvāt tiṣṭhater anūrdhvakarmaṇi vartamānād ātmane-padaṃ bhavati.  21CV on C 1.4.66: utpūrvāt tiṣṭhater anūrdhvaceṣṭāyāṃ vartamānāt taṅānā  22bhavanti.  23MV on J 1.2.19: utpūrvāt tiṣṭhater īhārthe vartamānāt do  bhavatī.   24 25In addition, for A 1.3.33 (अधःे Jसहन)े and its corresponding sūtras C 1.4.79 (अधःे शwौ) and J 1.2.28 (Jसहन े ऽधःे):  KV on A1.3.33: adhipūrvāt karoteḥ  prasahane vartamānād ātmanepadaṃ 26bhavati.   27CV on C 1.4.79: adhipūrvāt kṛñaḥ śaktau vartamānāt taṅānā bhavanti.   28MV on J 1.2.28: adhipūrvāt kṛñaḥ prasahane 'rthe do bhavati.  29Clarifying the significance of this technical ablative with the direction word pūrva seems to be a standard way of that time and also of an earlier time. There are several instances in the Mahābhāṣya where a preceding item is men- KV p.57 (on A 1.3.24). 21 Candra refers to ātmanepada endings collectively by the compound taṅāna.22 CV I.142 (on C 1.4.66).23 Jainendra uses the name da for Pāṇini’s name ātmanepada.24 MV p.26 (on J 1.2.19).25 Notice the variation in the way the authors mention the verbal base. Here, the 26Kāśikāvṛtti mentions the base with the affix śtip. Whereas, the Cāndravyākaraṇavṛtti and the Mahāvṛtti mention the same base in its basic form with the anubandha. In some cases, the Kāśikāvṛtti includes the term dhātu (verbal base) in its paraphrases (See, for example, KV on A 1.3.30 below). In other instances, the Kāśikāvṛtti quotes the Dhātupāṭha.   KV p.60 (on A 1.3.33).27 CV I.145 (on C 1.4.79).28 MV p.27 (on J 1.2.28).29 BEN-DOR 10tioned with this direction word. For example, in the Bhāṣya on A 3.1.71 Patañjali clarifies A 3.1.72 (संयसश्च), which concerns the verbal base yas when preceded by the upasarga sam, by saying “sampūrvāt yasaḥ”.  30III.2. In the paraphrases of the sūtras which list some upasargas, the common way of the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti is to list the upasargas and to end the list by iti together with the word evam followed by pūrva that ends in ablative. The number is decided according to the number of bases in which it is in apposition. For example, for A 1.3.22 (समवJिव{ः pः) and its corresponding sūtras C 1.4.65 (सिंवJावात)् and J 1.217 (pो ऽविवJा–): KV on A 1.3.22: samavapravītyevaṃpūrvāt tiṣṭhater ātmanepadaṃ bhavati.   31CV on C 1.4.65: saṃviprāvetyevaṃpūrvāt tiṣṭhates taṅānā bhavanti.   32MV on J 1.2.17: avavipretyevaṃpūrvāt saṃpūrvac  ca tiṣṭhater do bhavanti.  33 34And for A 1.3.30 (िनसमपुिव{ो —ः), C 1.4.76 (िनस[ंपु{ो —ः) J 1.2.25 (िनस[ंपुाद ्—ः): KV on A 1.3.30: nisamupavītyevaṃpūrvād dhvayater dhātor  ātmanepadaṃ 35bhavati.   36 Bh II.60.13-15 (on A 3.1.71): अनपुसगा ?िदित िकमथ ?म।् आयLित JयLित। अनपुसगा ?िदित श˜मकत ु?म।् कथम ्30आयLित JयLित। सयंस…vेतेत ्िनयमाथ ?म ्भिवFित। सपंवूा ?ƒसो ना“पवूा ?िदित ॥ The same way is found also in Kātyāyana’s vārttikas. For example: Bh II.88.8-9 (On 3.1.124): समवपवूा ?– (Vt.2)। समवपवूा ?–िेत वw[म।् समवस™य ?ः॥ In addition, although Pāṇini uses the technical ablative for specifying the location of the grammatical operation, he sometimes mentions the pre-ceding item with the direction word pūrva. For example, in A 5.1.92 (सšिरपू वा ?› च), which refers to the stem vatsara when preceded by sam or pari. KV p.57 (on A 1.3.22).31 CV I.141 (on C 1.4.65).32 The term sam continues to J 1.2.17 by anuvṛtti from J 1.2.16 (समो ऽकूजे).33 MV p.16 (On A1.2.17).34 Notice that here the authors include the term dhātu.35 KV p.59 (on A 1.3.30).36 Sūtra Paraphrases 11CV on C 1.4.76: nisaṃvyupetyevaṃpūrvād dhvayates taṅānā bhavanti.  37MV on J 1.2.25: nisaṃvyupetyevaṃpūrvād dhvayater do bhavanti.  38The authors of the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti use this pattern in many of the paraphrases for the sūtras which refer to in-stances where a base is preceded by an upasarga. However, there are exceptions. As we shall see in respect to this matter, the Mahāvṛtti shows the most consisten-cy. In addition, I will show that for most of the variations found in the Cāndra-vyākaraṇavṛtti there is a clear reason, and the variations are justified by the au-thor himself. I discuss each of these texts separately. IV.1.  In the section of Jainendravyākaraṇa on the addition of parasmaipada  and āt39 -manepada  endings, forty-one sūtras refer to specific upasargas which precede 40verbal bases. The Mahāvṛtti shows consistency in its paraphrases of these sūtras. For clarifying the technical ablative ending which is used in the sūtras, its para-phrases display an identical pattern; a single upasarga is followed by pūrva and a list of upasargas is followed by ityevaṃpūrva. As shown in section 3, this is the pattern common to all the three commentaries, the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and the Mahāvṛtti.  In the Mahāvṛtti, only one sūtra is paraphrased differently. In the vṛtti on J 1.2.42 (सJंतरेAतृौ), the Mahāvṛtti paraphrases the sūtra: saṃpratipūrvāj jānāter asmṛtyarthe do bhavati.  41As this sūtra concerns two upasargas, one would expect to have a para-phrase similar to the paraphrase presented in the Kāśikāvṛtti on the correspond-ing sūtra A 1.3.46 (सœित{ामनाान)े. This paraphrase includes ityevam after the list of the upasargas sam and prati:  CV I.144 (on C 1.4.76).37 MV p.27 (On J 1.2.25).38 Jainendra uses the name ma for Pāṇini’s name parasmaipada. 39 da in Jainendravyākaraṇa.40 MV p.29 (on J 1.2.42).41 BEN-DOR 12saṃpratītyevaṃpūrvāt jānāter anādhyāne vartamānād ātmanepadaṃ bha-vati.  42IV.2.  In the section of the Cāndravyākaraṇa on the addition of parasmaipada and āt-manepada  endings, thirty-nine sūtras refer to specific upasargas that precede 43verbal bases. For clarifying the technical ablative ending of the upasargas men-tioned in twenty-six sūtras, the Cāndravyākaraṇavṛtti uses the pattern which is shown in section III and is common to all three commentaries. For the other thirteen sūtras, the Cāndravyākaraṇavṛtti uses a different pattern when clarifying the ablative ending after the upasargas mentioned in the sūtras. In most of these sūtras, the author himself justifies the different interpretation of this ending, and it becomes clear that it is in order to remove the difficulties which are al-ready raised in the Mahābhāṣya and discussed in sections II.2 and II.3.  IV.3. In respect to the difficulties mentioned in sections II.2 and II.3, the Cāndra-vyākaraṇavṛtti does not invoke any paribhāṣā. The author uses an alternative so-lution proposed by Patañjali in the Bhāṣya on A 1.3.60. In A 1.3.60, vt. 3, Kā-tyāyana says that it should be added that the restrictive rules, such as A 1.3.17 (नjेवशः), which concern verbal bases that are preceded by a specific upasarga, re-fer also to cases where the augment aṭ intervenes between the base and the up-asarga.  Patañjali gives the counter-examples of nyaviśata and vyakrīṇīta.  In 44 45respect to these cases, Kātyāyana and Patañjali do not invoke the paribhāṣā yadāgamās tadguṇībhūtās tadgrahaṇena gṛhyante.  In his following vārttika, Kā46 - KV p.62 (on A 1.3.46). 42 taṅāna in Cāndravyākaraṇa.43 A 1.3.60 vt.3 (Bh I.285.22): उपसग ?पवू ?िनयमे ऽžवाय उपसŸानम ॥्44 Bh (I.285.23-24) on A 1.3.60 vt.3: उपसग ?पवू ?L िनयमे ऽžवाय उपसŸानम ्कत ?[म।् “िवशत [Nीणीत।45 क पनुः कारणं न िसित। अटा [विहत¡ात॥् Patañjali mentions here also the case of vyakrīṇīta ( ← vi aṭḍukṛñ + laṅ ) which is given in respect to A 1.3.18 (पिर[व{ेः िNयः). This case raises the same issue as the case of nyaviśata. Recall that the Kāśikāvṛtti invokes this paribhāṣā for removing this difficulty. See note 20. 46 Sūtra Paraphrases 13tyāyana justifies the addition of this statement.  Patañjali, on the other hand, 47rejects this addition and proposes some other solutions, among them, taking the ending of neḥ as genitive and not as the expected ablative. Patañjali argues that this type of a genitive case ending expresses qualified-qualifier relation (viśeṣaṇa ṣaṣṭhī).  This solution removes both of the difficulties mentioned in sections 2.2 48and 2.3. By considering the ending of the term ni in A 1.3.17 as a viśeṣaṇa ṣaṣṭhī, this sūtra refers only to upasargas and does not refer to any other identical items because this relation can occur only between the base viś and the upasarga ni. That is, ni qualifies the meaning of viś. Consequently, A 1.3.17 is not applicable to the case of madhuni viśanti bhramarāḥ,  because here ni does not qualify viś. In 49addition, the same interpretation removes also the difficulty of applying the sū-tra to cases where the upasarga and the base do not occur in an immediate se-quence. A viśeṣaṇa ṣaṣṭhī does not imply an immediate sequence because a qual-ified-qualifier relation between the upasarga and the base occurs also in cases where an item intervenes. Thus, by considering the ending of the term neḥ as a  A 1.3.17 vt.5 (Bh I.287.14): तAापसŸानम॥्47 Bh. I.287.2-4 (on A 1.3.60 vt. 5): अथवा न ेिरित नषैा प¢मी । का तjह । िवशषेणष`ी । नये‰ िविशः । क…48नjेविशः । िवशFेः । [विहत…ािप श˜ते िवशषेियतमु ् ॥ In the Bhāṣya which precedes this argument, Patañjali rejects the use of the paribhāṣā yadāgamās tadguṇībhūtās tadgra-haṇena gṛhyante in respect to this matter on the basis that the augment aṭ is added to the aṅga viś + śap and not to the root viś. This involves the question about the order of the application of the rules and the following discussion in Bhāṣya revolves around this question. At the end of the Bhāṣya on this sūtra, Patañjali presents the option that the term ni in A 1.3.17 ends with the genitive case ending, which expresses a relation of qualified and qualifier (viśeṣaṇa ṣaṣṭhī). See section II.2. 49 BEN-DOR 14viśeṣaṇa ṣaṣṭhī, A 1.3.17 is applicable also to cases where the augment aṭ inter-venes.   50The Cāndravyākaraṇavṛtti adopts the alternative solution proposed by the bhāṣya on A 1.3.60 vt. 5. In most of the sūtras where the commentators find such difficulties, the Cāndravyākaraṇavṛtti interprets the ablative ending of the up-asarga as a genitive case ending expressing a qualified-qualifier relation. For example, it paraphrases C 1.4.51 (नjेवशः):  ner viśeṣyād viśas taṅānā bhavanti.   51Accordingly, this sūtra specifies that after the base viś when it is qualified by ni, only ātmanepada endings are added. The Cāndravyākaraṇavṛtti on this sū-tra simply mentions the form nyaviśata  as an example were this sūtra applies. 52Moreover, it argues that the use of the genitive case in neḥ is done so that this sūtra would not be applicable in the case of madhuni viśanti bhramarāḥ.   53 The Kāśikāvṛtti on A 1.3.55 (समः Jित£ान)े uses this reasoning for explaining how this sūtra 50is applicable to cases such as saṃprayacchate ( ← saṃpra dā + laṭ ). This sūtra concerns the base dāṇ (DhP I.97) when preceded by the upasarga sam. In saṃprayacchate, the upa-sarga pra intervenes between the dā ( → yacch ) and sam. KV p.55 (on A 1.3.55): समः Jशने[वधान े कथमा+नपेदं भवित। समो इित िवशषेणे ष`ी न प¢मी॥ Yet, in its paraphrases of the sūtra, the ending is displayed as a technical ablative: dāṇ dāne parasmaipadī. tataḥ saṃpūrvāt tṛtīyāyuktāt ātmanepadaṃ bhavati sā cet tṛtīyā caturthyarthe bhavati. Whereas, the Cāndra-vyākaraṇavṛtti paraphrases the corresponding sūtra C 1.4.88 (दाणः सा च–ेत¤ु ?थ”): samo viśeṣyād dāṇas tṛtīyāntena yuktāt taṅānā bhavanti sā cet tṛtīyā caturthyarthe bhavati. In the Mahāvṛtti, a corresponding sūtra to A 1.3.55 is missing. However, it is given as a vārtti-ka in its comments on J 1.2.50. The Mahāvṛtti paraphrases in a similar way as the Kāśikāvṛtti: sampūrvād dāṇo bhāyoge do bhavati sā ced abarthe bhā. In its comments on the vārttika, the Mahāvṛtti does not argue in the same way as the Cāndravyākaraṇavṛtti and the Kāśikāvṛtti. It simply says that in saṃprayacchate the upasarga sam is in connection (saṃbandha) with the base. Therefore pra does not intervene and the vārttika concerns also such cases. MV p.19 (1.2.50): दाण… सा चदेबथ” ऽिशb[वहारे इित वw[म।् सšवूा ?¥ाणो भायोग े दोभवित सा चदेबथ” भा।… दाLा सJंय¦त।े वषृ§ा सJंय¦ते कामकुः। सम इित सबंdे ता। तने Jशने [वधान ं नभवित॥The vārttika follows the arguments of the Bhāṣya (Bh I.284 2-8) on A 1.1.55 and a similar vārttika is also found in the Kāśikāvṛtti on this sūtra (p.54). CV I.139 (on C 1.4.51).51 See Section II.3.52 CV I.139 (on C 1.4.51): नjेवशFेाि¨शrङाना भवि~। िनिवशते “िवशत। ष©ा िनद”शः िकम।् मधिुन िवशि~53मराः॥ Sūtra Paraphrases 15IV.4. In its paraphrases of thirteen sūtras which include a technical ablative, the Cāndra-vyākaraṇavṛtti uses a pattern different from the pattern shown in section III for clarifying this ending. In eight of these paraphrases, the Cāndravyākaraṇavṛtti interprets the ending of the upasarga as conveying a qualified-qualifier relation as it does in respect to A 1.4.51 (नjेवशः).  Yet, in respect to C 1.4.58 (Nीडो ऽनपुिर{ां54च), the Cāndravyākaraṇavṛtti uses a different interpretation for removing the same difficulties. In this sūtra, the upasargas anu and prati are mentioned and the upasarga āṅ continues from the previous sūtra C 1.4.54 (आङो दः). C 1.4.58 concerns the verbal base krīḍ when preceded by one of these three upasargas. The Bhāṣya on Pāṇini’s sūtra A 1.3.21 (Nीडो ऽनसुपंिर{…), which is parallel to C 1.4.54, mentions the counter-example anu krīḍati māṇavakam (“he plays with the boy”) where this sūtra may undesirably be applicable.  55In the cases of madhuni viśanti bhramarāḥ and in parā jayati senā respective-ly, ni and parā do not qualify the meaning of the base viś and ji, whereas, in the counter-example anu krīḍati māṇavakam, anu may be considered as a qualifier of the meaning of the base krīḍ. Therefore, the Cāndravyākaraṇavṛtti provides an-other solution to this difficulty. This is clear from the way it paraphrases the sūtra: krīḍater anuparibhyām āṅā ca yuktāt taṅānā bhavanti. Here, the Cāndra-vyākaraṇavṛtti does not interpret the ablative ending of the upasargas as under-lining a sequence of a preceding and following item or a qualified-qualifier rela-tion. According to this paraphrase, only ātmanepada endings are added after the base krīḍ when it is in connection with (yukta) the mentioned upasargas.  This is 56based on a Pāṇinian method which is not introduced by Candragomin’s sūtras.  In should be noted that the upasargas mentioned in C 1.4.53 (िवपरा{ां जःे ) clearly end 54with ablative. Yet, Cāndravyākaraṇavṛtti paraphrases the sūtra: etābhyāṃ viśeṣyāj jaya-tes taṅānā bhavanti. Here, the Cāndravyākaraṇavṛtti uses a qualified-qualifier relation as one of the explanations for why this sūtra is not applicable to cases such as parā jayati senā (“the superior army wins.”) The Cāndravyākaraṇavṛtti is silent about the type of ending the term viparābhyāṃ ends with. CV I.139 (on C 1.4.53): एता{ां िवशFेाªयतrेङानाभवि~। िवजयते पराजयत।े कथं ब«िव जयित वनं परा जयित सनेिेत। नाZ िवपराशा{ां जयितjवशFेत।ेअ“ो“साहचया ?¨ ा Jािदपिरपिठतयोिरह Pहणम॥् Considering one nominal ending as having a meaning of another nominal ending (vibhaktivipariṇāma) is common in the Bhāṣya.  Bh I.280.2 (on A1.3.21): उपसग ?Pहणं कत ?[म।् इह मा भतू।् अन ुNीडित माणवकम॥्55 CV I.140 (on C 1.4.58): Nीडतरेनपुिर{ामाङा च यwुा&ङाना भवि~। अन ुNी डत े पिरNीडत आNीडत।े इह56कAा] भवित माणवकमन ुNीडतीित । नाZ Nीडितरननुा यwु इित न भिवFित॥ BEN-DOR 16By A 1.4.59 (उपसगा ?ः िNयायोग)े,  Pāṇini provides the name upasarga to the pra, etc. 57when they are used in connection with an action.  In respect to Cāndravyākaraṇavṛtti on C 1.4.58, the Mahāvṛtti on the parallel sūtra J 1.2.15 (Nीडो ऽनपुया ?ङः) should be mentioned. The Mahāvṛtti on J 1.2.15 para-phrases this sūtra: anuparyāṅ ityevaṃpūrvāt krīḍo do bhavati.  58The Mahāvṛtti comments that on the basis of sāhacarya, the term anu in this sūtra refers only to an upasarga  and mentions the counter-example 59māṇavakam anu krīḍati.  The same argument is given also in the Kāśikāvṛtti on 60the parallel sūtra A 1.3.21.  In its following comments, the Mahāvṛtti adds the 61explanation found in the Cāndravyākaraṇavṛtti and says that the sūtra is not ap-plicable to such cases because anu in māṇavakam anu krīḍati is in connection with the nominal base māṇavaka and it is not in connection with the verbal base krīḍ. In such cases, anu is not an upasarga.  It appears that Mahāvṛtti in respect to 62this matter shares the information found in the Cāndravyākaraṇavṛtti.   63 In general, Candra avoids using technical names. Therefore, he does not provide any 57name to refer to upasargas. He refers to such units collectively by the Gaṇa title prādi (“pra, etc.”). They are listed in Cāndravyākaraṇavṛtti (I.41) on C 1.1.109 (गदमदयमो ऽJादःे).  MV p.25 (on J 1.2.15). The Kāśikāvṛtti (p. 56) paraphrases the parallel sūtra A 1.3.21: krīḍṛ 58vihāre etasmād anusaṃparītyevaṃpūrvād āṅpūrvāc cātmanepadaṃ bhavati.  gi in Jainendra’s terminology. 59 In the Bhāṣya, the counter-example is given in this order: anu krīḍati māṇavakam. But in 60the Cāndravyākaraṇavṛtti, Kāśikāvṛtti, and Mahāvṛtti, it is given in this order: māṇavakam anu krīḍati. KV p. 56 (A 1.3.21): समा साहचया ?द¬ािद|पसग‰ गŠृत े तनेहे कम ?JवचनीयJयोग े न भवित। माणवकमन ुNीडित॥61 MV p.25 (on J 1.2.15): िगसाहचया ?दनोग”रवे Pहणािदह न भवित। माणवकमन ु Nीडित। माणवकेन सहvेथ ?ः। भाथ ?62इvननुा योग इिoगितस£ंाJितषधे… ॥ The Mahāvṛtti mentions here J 1.4.14. In addition to providing the second case ending for a nominal base which is in connection with anu that conveys the meaning saha (with), this sūtra negates the term gi (upasarga) for anu in such cases. Cf. Bh I.346. (on A 1.4.84) and KV on A 1.4.84, A 1.4.85 and A 1.4.94.  The Mahāvṛtti seems to mention that anu is not an upasarga (gi) by J 1.4.14 as the reason 63why the term anu which is known as an upasarga by sāhacarya does not refer to anu in māṇavakam anu krīḍati. The Cāndravyākaraṇavṛtti cannot argue in the same way as the Mahāvṛtti because Candra avoids using technical names and he does not give any names to the group of items named upasarga, karmapravacanīya, or upapada by Pāṇini.  Sūtra Paraphrases 17IV.5. In the paraphrases of C 1.4.74 ([दुrपः) and C 1.4.92 (सJंतरेAतृौ), the Cāndra-vyākaraṇavṛtti displays the upasargas and the ablative ending in a similar man-ner. Each of these sūtras includes two upasargas: vi and ud in C 1.4.74 and sam and prati in C 1.4.92. The Cāndravyākaraṇavṛtti paraphrases C 1.4.74:  ver udaś ca parāt tapater avyāpyāt svāṅgāpyāc ca taṅānā bhavanti.   64And C 1.4.92:  samaḥ prateś ca parāj jānāter asmaraṇe vartamānāt taṅānā bhavanti.   65Such a formation is found quite a few times in the Cāndravyākaraṇavṛtti’s paraphrases of other sūtras occurring in other sections.  The reason why these 66two sūtras, particularly C 1.4.74, are paraphrased differently by the Cāndra-vyākaraṇavṛtti may be that the author follows the Bhāṣya’s suggestions mentioned in respect to the parallel sūtras A 1.3.27 (उि¨{ा~पः) and A 1.3.46 (सJंित{ामनाान)े. In his vārttika on A 1.3.27, Kātyāyana suggests that this sūtra applies also to cases where the object of the action is one’s own limb.  This suggestion is not imple67 -mented in C 1.4.74 itself, but the Cāndravyākaraṇavṛtti includes it while para-phrasing the sūtra. The author might have composed his own paraphrase that is different from the paraphrase in its source which he was using. This may be the reason why he follows a different pattern when displaying the upasargas and the ablative ending. The variation in respect to the paraphrase of A 1.4.46 may be due to a similar reason. Here the Cāndravyākaraṇavṛtti replaces Pāṇini’s term anā-dhyāna by asmṛti.   68In the paraphrases of three other sūtras, the Cāndravyākaraṇavṛtti displays the ablative ending of the upasargas in a different way than shown in section III. I discuss them while discussing the irregularities found in the Kāśikāvṛtti.   CV I.143 (on C 1.4.74).64 CV I.148 (on C 1.4.92).65 Compare, for example, CV I. 52 C 1.1.148 (आसमः ­ोः): āṅaḥ samaś ca parāt sravater ṇo bha66 -vati and CV I.95 C 1.3.20 (“दुो Pः): ner udaś ca parād girater ghañ bhavati.  Bh I.281.21-22 (on A 1.3.27). ®ा¯कम?का– (vt.1)। ®ा¯कम?का–िेत वw[म।् उ&पते पाणी। िवतपते पाणी।67उ&पते पृ` म।् िवतपते पृ` म॥् Compare with KV p.62 (on A 1.3.46).68 BEN-DOR 18IV.6. In the Aṣṭādhyāyī, in the section on the addition of parasmaipada and ātmane-pada endings, forty-two sūtras refer to specific upasargas that precede verbal bases. In the paraphrases of thirty-seven sūtras, the authors of the Kāśikāvṛtti present the ablative ending of the upasargas in the same pattern as shown in section III. In the paraphrases of five sūtras, the ablative ending of the upasargas is displayed differently. A 1.3.19 (िवपरा{ां जःे) concerns the base ji when preceded by the upasargas vi and parā. In addition A 1.3.39 (उपपरा{ाम  ्) concerns the base kram when preceded by the upasargas upa and parā. The Kāśikāvṛtti paraphrases these sūtras as fol-lows: KV on A 1.3.19: viparāpūrvāj jayater dhātor  ātmanepadaṃ bhavati.   69 70KV on A 1.3.39: upaparāpūrvāt kramater vṛttyādiṣv artheṣu vartamānād 
ātmanepadaṃ bhavati.   71The Mahāvṛtti paraphrases the corresponding sūtras J 1.2.13 (िवपराजःे) and J 1.2.35 (परोपात)् as follows: MV on J 1.2.13: viparetyevaṃpūrvāj jayater do bhavati.   72MV on J 1.2.35: propetyevaṃpūrvāt kramer do bhavati.   73As both A 1.3.19 and A 1.3.39 concern two upasargas, one would expect that the list of the mentioned upasargas would be followed by ityevaṃpūrvat as it is done in respect to other sūtras and as it appears in the Mahāvṛtti on the corre-sponding sūtras. In fact, the Osmania edition of the Kāśikāvṛtti notes that there are manuscripts that read these paraphrases in the same manner as in the Mahāvṛtti:  Notice that here the authors include the term dhātu.69 KV p. 55 (on A 1.3.19).70 KV p. 61 (on A 1.3.39).71 Mahāvṛtti p.25 (on J 1.2.13).72 Mahāvṛtti p.28 (on J 1.2.13).73 Sūtra Paraphrases 19For A 1.3.19: viparetyevaṃpūrvāj jayater ātmanepadaṃ bhavati.  74For A 1.3.39: upaparetyevampūrvāt kramater vṛttyādiṣu artheṣu vartamānād ātmanepadaṃ bhavati.  75These manuscripts may have preserved an authentic version of the Kāśikāvṛtti’s paraphrases of these sūtras. Alternatively, it may be that the writers of the manuscripts were under the influence of the frequent pattern occurring in other similar sūtras. In addition, whether the versions of the two paraphrases in the published editions are originals or later revisions, the reason why the techni-cal ablative is clarified in a different way may be the call for specifying directly that these sūtras refer only to the upasargas by including the term upasarga in the paraphrases. This may be seen in respect to the paraphrases of the Kāśikāvṛtti and Cāndravyākaraṇavṛtti which I discuss in the next sections.  IV.7. The Kāśikāvṛtti and the Cāndravyākaraṇavṛtti display the ablative ending of the upasargas in a different pattern than shown in section 3 in respect to some other sūtras which are not discussed in previous sections. It is useful to consider the case of these sūtras together with their parallel sūtras. A 1.3.18 (पिर[व{ेः िNयः) and C 1.4.52 (पिर[वाि°यः) refer to the upasargas pari, vi and ava. A 1.3.39 (उपपरा{ाम ्) and C 1.4.85 (परोपात ्) refer to the upasargas parā and upa. A 1.3.42 (Jोपा{ां समथा ?{ाम ्) and C 1.4.88 (Jोपादार±)े refer to the upasargas pra and upa. The Cāndravyākaraṇavṛtti and the Kāśikāvṛtti paraphrase these sūtras as follows:  1. CV on C 1.4.52: paryādipūrvāt krīṇātes taṅānā bhavanti. KV on A 1.3.18: parivyavebhyo uttarasmāt krīṇāter ātmanepadaṃ bhavati. 2. CV on C 1.4.85: paropābhyām eva prādibhyāṃ parāt kramater vṛttyādiṣv artheṣu taṅānā. bhavanti KV on A 1.3.39: upaparāpūrvāt kramater vṛttyādiṣv artheṣu vartamānād 
ātmanepadaṃ bhavati. 3. CV on C 1.4.88: propābhyāṃ parāt kramater ārambhe taṅānā bhavanti.   KV (p. 66) note 12 (ms. pha). Notice that in this version the term dhātu is missing. 74 KV (p. 61) note 15 (ms. ḍha - a printed book of Liebich which includes notes on variant 75readings). BEN-DOR 20KV on A 1.3.42: propetyetābhyāṃ parasmāt kramater ātmanepadaṃ bhavati tau cet propau samarthau tulyārthau bhavataḥ. As we can see, although these sūtras include a few upasargas, both the Cāndravyākaraṇavṛtti and the Kāśikāvṛtti do not end the list of upasargas with the expected ityeveṃpūrvāt.  In addition, instead of the direction word pūrva, the 76Cāndravyākaraṇavṛtti on C 1.4.85 and C 1.4.88 uses the direction word para. The Kāśikāvṛtti on A 1.3.18 and A 1.3.42 uses respectively the direction words uttara and para with their pronominal declension.   77In respect to quite a few sūtras which include a technical ablative, the Cāndravyākaraṇavṛtti clarifies this ending with the direction word para  In addi78 -tion to the Kāśikāvṛtti’s paraphrase of A 1.4.17 which I will discuss later, the Kāśikāvṛtti while paraphrasing A 8.3.79 (िवभाषटेः) uses the direction words para and uttara for clarifying the technical ablative.  In addition, the Nyāsa under the 79Kāśikāvṛtti on A 5.4.107 (अ[यीभावे शर‹भिृत{ः), uses para with a pronominal de-clension while paraphrasing the gaṇa-sūtra 158 (Jितपरसमन{ुो ऽ²ः),  and a much 80 Compare with the Mahāvṛtti on the corresponding sūtras: 
76MV p.25 (on J 1.2.12: पिर[वNीयः): parivyavetyevaṃpūrvāt krīṇāter do bhavati | 
MV p.28 (on J 1.2.35: परोपात)्: paropetyevaṃpūrvāt kramater do bhavati |
MV p.28 (on J 1.2.38: Jादार±)े*: prapūrvāt krama ārambhe do bhavati | Notice that in its paraphrase of J 1.2.35, the Mahāvṛtti does not include the semantic con-dition vṛttyādiṣv artheṣu which is mentioned in the paraphrases of the Cāndra-vyākaraṇavṛtti on C 1.4.85 and the Kāśikāvṛtti on A 1.3.39. The Mahāvṛtti on J 1.2.35 (see note 90) mentions this requirement as preliminary information, before presenting the paraphrase of the sūtra and also as an additional specification after presenting the sūtra examples.  * A 1.3.42 and C 1.4.88 include the upasargas pra and upa. On the other hand, J 1.2.38 mention directly only the upasarga pra. Yet, the Mahāvṛtti comments that upa occurs in the sūtra by anuvṛtti from J 1.2.35. MV p. 28 (on J 1.2.38): परोपािदvत उपािदित वत ?त॥े Cf. A 7.1.16 (पवूा ?िद{ो नव{ो वा). 77 For example see note 79.78 KV p.157 (on A 8.3.79): iṇaḥ parasmāt iṭaḥ uttareṣām ṣīdhvaṃluṅliṭām yaḥ dhakāraḥ tasya 79mūrdhanyādeśo bhavati vibhāṣā. Compare with Cāndravyākaraṇavṛtti on the correspond-ing sūtra C 6.4.l00 (वटेः): iṇaḥ parād iṭaḥ pareṣāṃ sīluṅliṭāṃ dhakārasya ḍhakāro bhavati vā. Nyāsa VI.316 (on A 5.4.107): अिŽशा‹ितपरसमि¬vते{ेः परAाh³भवित॥80 Sūtra Paraphrases 21later text, the Bālamanoramā, clarifies the technical ablative in this manner in its paraphrases of the several sūtras.   81The reason why the Kāśikāvṛtti on A 1.3.18, A 1.3.39 and A 1.3.42 and the Cāndravyākaraṇavṛtti on C 1.4.52, C 1.4.85 and C 1.4.88 present a different pattern for clarifying the technical ablative may be found by considering the comments of the commentators on all of these sūtras. Although the Cāndravyākaraṇavṛtti and Kāśikāvṛtti do not mention in their comments on each of these sūtras that these sūtras refer only to upasargas, it appears that there was a call for specify-ing this and it affected the way they paraphrase the sūtras. Moreover, it was al-ready mentioned in section 4.6 that one manuscript of the Kāśikāvṛtti on A 1.3.39 reads the paraphrase with ityevaṃpūrvāt after the list of the upasargas and it may be that the published editions of Kāśikāvṛtti present a revised version of this paraphrase. Nevertheless, the following discussion may also explain why most of the manuscripts used by the editors of these editions read the paraphrase of Kāśikāvṛtti to A 1.3.39 without “ityevam.”  IV.8. As it was mentioned, the Cāndravyākaraṇavṛtti and Kāśikāvṛtti paraphrase C 1.4.52 (पिर[वाि°यः) A 1.3.18 (पिर[व{ेः िNयः) as follows: CV on C 1.4.52: paryādipūrvāt krīṇātes taṅānā bhavanti KV on A 1.3.18: parivyavebhyo uttarasmāt krīṇāter ātmanepadaṃ bhavati On C 1.4.52, The Cāndravyākaraṇavṛtti does not say that this sūtra refers to upasargas. However, on the parallel Pāṇinian sūtra A 1.3.18, the Kāśikāvṛtti com-ments that this sūtra applies only to upasargas and presents the counter-exam-ple of bahuvi krīṇāti vanam.  In addition, the Osmania edition of the Kāśikāvṛtti 82notes that one manuscript reads the paraphrase of A 1.3.18 with iti after the list of the upasargas and without the direction word uttara:  parivyavetyebhyaḥ krīṇāter ātmanepadaṃ bhavati.   83 The BM (III.550) paraphrases A 1.3.18: parivyavebhyaḥ parasmāt krīñdhātor ātmane-81padam ity arthaḥ, A 1.3.19: viparābhyāṃ parasmāj jidhātor ātmanepadam ity arthaḥ and A 1.3.21 (BM III.551): tathā ca anusaṃparyāṅ ebhyaḥ parasmāt krīḍadhātor ātmanepadam ity arthaḥ.  KV p.55 (on A 1.3.18): पया ?दयो उपसगा ? गŠृ~।े तनेहे न भवित ब«िव Nीणाित वनम॥्82 KV p. 55 note 9.83 BEN-DOR 22In both versions (parivyavebhyo uttarasmāt… or parivyavetyebhyaḥ…), it seems that a term such as upasargebhyaḥ, which would qualify parivyavetyebhyaḥ or parivyavebhyaḥ is missing. This becomes evident when comparing this para-phrase with the Kāśikāvṛtti’s paraphrase of A 1.3.42:  propetyetābhyāṃ parasmāt kramater ātmanepadaṃ bhavati tau cet propau samarthau tulyārthau bhavataḥ. The Kāśikāvṛtti on A 1.4.42 does not comment that this sūtra refers only to upasargas. However, the Tara and the Ramlal Kapur Trust editions of the Kāśikāvṛtti read the paraphrase of this sūtra with the term upasarga:  propety etābhyām upasargābhyām parasmāt kramater ātmanepadaṃ bha-vati…   84This was probably an earlier version of the Kāśikāvṛtti’s paraphrase. The Padamañjarī evidently had this version of the Kāśikāvṛtti as it quotes “propety etā-bhyām upasargābhyām.” The Padamañjarī comments that it is necessary to specify that this sūtra concerns only upasargas so that it would not be applicable to any other identical items which are not upasargas.  The necessity to specify that A 851.3.18 applies only to upasargas so that it would not be applicable to bahuvi krīṇāti vanam, as the Kāśikāvṛtti argues, and the similarity between the paraphrases of A 1.3.18 (parivyavebhyo uttarasmāt or parivyavetyebhyaḥ) and of A 1.3.42 (propety etāb-hyām [upasargābhyām] parasmāt) indicate that this paraphrase included the term upasarga which was omitted afterward.  This is also indicated from the 86Kāśikāvṛtti on A 1.3.17 which I discuss later. In fact, similar versions of para-phrases for the corresponding sūtras are found in the Hṛdayahāriṇī of Nārāyaṇadaṇḍanātha on the Sarasvatīkaṇṭhābharaṇa of Bhojadeva. Nārāyaṇa-daṇḍanātha’s paraphrases include the term upasarga. For example, KV on A 1.3.18: parivyavebhyo uttarasmāt krīṇāter ātmanepadaṃ bhavati.  KV (Tara ed.) II.51 and KV (Ramlal Kapur Trust ed.) p. 55. The latter notes (note 3) that 84this word is not found in all manuscripts. The Osmania edition (p. 62, note 1) notes that two manuscripts read the paraphrase with the term upasarga. Padamañjarī II.51 (on KV to A 1.3.42): अ“ो ऽ“साहचया ?‹ोपयो|पसग ? योP ?हणं न Jाितपिदका~-85कम?Jवचनीययोना ?िप िNया~रसuिdनोना ?´ नथ ?कयोिरvाह Jोपvेतेा{ामपुसगा ?{ािमित॥ The Padamañjarī does not give any counter example.  This does not necessarily mean that these versions are the authentic versions of the 86paraphrases of the original Kāśikāvṛtti. In my opinion, they are later adaptations of the original text. This I think is indicated by the Kāśikāvṛtti on A 1.3.17 which I discuss later.  Sūtra Paraphrases 23Hṛdayahāriṇī on S 3.1.50: (paryādibhya) upasargebhyaḥ (parasya) krīṇāteḥ kartaryātmanepadāni bhavanti.  87In addition, in a much later text, the Kāśikāvṛttisāra (1717 CE), the term upa-sarga is included in the paraphrase of this sūtra:  KVS on A 1.3.18: parivyavopasargebhyo uttarasmāt krīṇāter ātmanepa-dam.  88IV.9. A similar situation occurs in respect to the Cāndravyākaraṇavṛtti on the C 1.4.88 that corresponds to A 1.4.42. In respect to A 1.4.88, the Cāndravyākaraṇavṛtti does not specify that it concerns only upasargas.  In its paraphrase of this sūtra, the 89Cāndravyākaraṇavṛtti uses a similar pattern (i.e. propābhyām parāt) as in its para-phrase of C 1.4.85 for clarifying the technical ablative:  CV on C 1.4.88: propābhyāṃ parāt kramater ārambhe taṅānā bhavanti.  CV on C 1.4.85: paropābhyām eva prādibhyāṃ parāt kramater vṛttyādiṣv artheṣu taṅānā. As we can see, in respect to C 1.4.85, the Cāndravyākaraṇavṛtti specifies that this sūtra refers only to upasargas by including the term prādi while paraphras-ing this sūtra. Thus, it may be that also in its paraphrase of C 1.4.88 this term  Hṛdayahariṇī III.7 (on S 3.1.50). The parentheses are originally in the edition. 87Nārāyaṇadaṇḍanātha includes the term upasarga in many of his paraphrases to the sūtras in this sections. See also note 91.  KVS I.52 (on A 1.3.18).88 prādi in Cāndravyākaraṇa. 89 BEN-DOR 24was meant to be included because according to the Kāśikāvṛtti on the parallel sū-tra A 1.4.42, it is necessary to specify that the sūtra refers only to upasargas.   90IV.10. As we can see, the call for specifying that the sūtras mentioned above refer only to upasargas had an effect on the way the Cāndravyākaraṇavṛtti paraphrases the sūtras. The Cāndravyākaraṇavṛtti clearly paraphrased the sūtras in a way that would specify that the sūtra refers only to upasargas. In respect to the Cān-dravyākaraṇavṛtti, one can see a consistency in respect to this matter. On the oth-er hand, the Kāśikāvṛtti has variations for which the authors do not present any justification. The Kāśikāvṛtti on A 1.3.17 may give us a clue about the inconsisten-cy in respect to this matter.  The Kāśikāvṛtti paraphrases A 1.3.17 (नjेवशः):  neḥ parasmāt viśaḥ ātmanepadam bhavati.  91This pattern is comparable to the Kāśikāvṛtti’s paraphrases of A 1.3.18: (parivyavebhyo uttarasmāt…) and of A 1.3.42: (propety etābhyāṃ [upasargābhyām]  92parasmāt…). As it was mentioned before, in respect to all these three sūtras, the  Both the Kāśikāvṛtti and the Mahāvṛtti on the parallel sūtras A 1.3.39 and J 1.2.35 (Jपात)् 90argue that this specification is necessary so that the sūtra would not be without a pur-pose. The addition of only ātmanepada endings after the verbal base kram which is used in the meanings of uninterrupted undertaking (vṛtti), enthusiasm (sarga) and accomplishment (tāyana) is already provided by the previous sūtra, A 1.3.38 (विृ&सग ?तायनषे ुNमः) or J 1.2.34 (विृ&सग ?तायन े Nमः). According to the Kāśikāvṛtti and the Mahāvṛtti, A 1.3.39 or J 1.2.35 are restrictive sūtras (niyama sūtra). In the meanings of vṛtti etc., they re-strict the addition of only ātmanepada endings after the verbal base kram when it is preceded only by the upasargas upa and pra and not any other upasargas. Without A 1.3.39 and J 1.2.35, in the specified condition only ātmanepada endings would be pro-vided for after the base kram also when it is preceded by any upasarga. KV p.61 (on A 1.1.39): विृ&सग ?तायनिेµित वत ?त।े उपपरापवूा ?°मतवेृ ?¶ािदµथ”ष ु वत ?मानादा +नपेदं भवित। िकमथ· तह¸दम¹ुत।ेउपसग ?िनयमाथ ?म।् सोपसगा ?पपरापवूा ?दवे ना“पवूा ?िदित। उपNमते पराNमत।े उपपरा{ािमित िकम।् सºामित।व¶ृािदिµvवे। उपNामित पराNामित॥ MV p. 28 (on J 1.2.35): विृ&सग ?तायन इित वत ?त।े परोपvेवेपंवूा ?°मदे‰भवित। पराNमत।े उपNमत।े िसVे »ार±‰ िनयमाय परोपा{ामवे ना“Aा’ःे। अनNुामित।व¶ृािदिµvवे।पराNामित। उपNामित॥ KV p.55 (on A 1.3.17). Compare with Hṛdayahāriṇī (III.7) on S 3.1.49 (नेवर्शः): 1.49: ner upa-91sargāt parasmāt viśéḥ kartaryātmanepadāni bhavanti. See note 84. 92 Sūtra Paraphrases 25commentators argue that it is necessary to specify that they concern only up-asargas. Therefore, as in the case of the Kāśikāvṛtti’s paraphrase of A 1.3.42, also in respect to the paraphrase of A 1.3.17 it seems that the term upasarga is missing (i.e. ner *upasargāt* parasmād viśaḥ) . In addition, the Kāśikāvṛtti on A 1.3.17 starts 93with the preliminary statement:  śeṣāt kartari parasmaipadam iti parasmaipade prāpte nipūrvād viśa āt-manepadaṃ vidhīyate.  Here, for clarifying the technical ablative in the sūtra, the Kāśikāvṛtti uses the pattern common to the Cāndravyākaraṇavṛtti, the Kāśikāvṛtti, and Mahāvṛtti. In fact, in respect to quite a few sūtras in the section of addition of the para-smaipada and ātmanepada endings, the Kāśikāvṛtti has paraphrases which are similar to the preliminary statement given in the Kāśikāvṛtti on A 1.3.17.  There94 -fore, also this statement may be taken as a paraphrase to this sūtra. Thus, it seems that we have here two paraphrases.  One, which is not affected by the call 95for specifying that the sūtra refers only to the upasarga ni, and therefore the technical ablative used in the sūtra is clarified in the common manner.  On the 96 See note 91 for the striking similarity to Hṛdyahāriṇī’s paraphrase of the corresponding 93sūtra S 3.1.49.  For example: KV p.58 (on A 1.3.29): śeṣāt kartari parasmaipadam iti prāpte sampūrve-94bhyaḥ gamyṛcchi-pracchi-svaraty-arti-śru-vidīti etebhyo akarmakebhyo dhātubhyo ātma-nepadaṃ bhavati. KV p.61 (On A 1.3.38): śeṣāt kartari parasmaipade prāpte vṛttyādiṣv artheṣu kramer dhātor ātmanepadaṃ bhavati. But in KV p.55 (On 1.3.19), preliminary in-formation is given before the paraphrase: śeṣāt kartari parasmaipadam ity asyāpavā-daḥ. viparāpūrvāj jayater dhātor ātmanepadaṃ bhavati. The expression śeṣāt kartari para-smaipade prāpte or its variants are not included in the Kāśikāvṛtti’s paraphrases for all the sūtras which can be taken as exceptions to A 1.3.78. The insertion of this phrase maybe a result of a view that the relation of the commented sūtra with A 1.3.78 should be specified within the paraphrase. Whereas, the paraphrases in which this expression is missing may have originated from an approach that holds that this can be specified as an additional comment. The Kāśikāvṛtti on 1.3.19, where this matter is mentioned as preliminary information, may indicate that earlier, this statement was given in such a way and that it was included as part of the paraphrase later.  As it was mentioned before (see section IV.7 and note 81), instead of the direction word 95pūrva, the paraphrases in the Bālamanoramā hava para with a pronominal declension (i.e parasmāt). Interestingly, its paraphrase for A 1.3.17 is: nipūrvād viśa ātmanepadaṃ syāt.  Compare with Mahāvṛtti (p. 25) on J 1.2.11: nipūrvād viśo do bhavati.96 BEN-DOR 26other hand, the version of the other paraphrase (neḥ parasmāt viśaḥ…) is a result of this call. In this version of the paraphrase the term upasarga was probably omitted in the intervening time because, as Patañjali, the Kāśikāvṛtti and other commentators argue, it is unnecessary to mention it. Alternatively, it might be omitted because of a merge of some version of the paraphrase into one para-phrase.  V. The Cāndravyākaraṇavṛtti, Kāśikāvṛtti, and Mahāvṛtti have probably shared infor-mation, whether directly or by another source which they have used. As we saw, one can find common features in their paraphrases of the sūtras occurring in the section on the addition of the parasmaipada and the ātmanepada endings. The particular common way in which the three commentaries display the upasargas and ablative endings may originate from that common source in which the up-asargas and the ablative were displayed in such a manner. The Mahāvṛtti is con-sistent in respect to the way it displays the upasargas and ablative endings and probably preserves the version of the paraphrases which appeared in that earlier source, particularly in respect to the presentation of the upasargas and the abla-tive ending. Because the same pattern is found also in the Cāndravyākaraṇavṛtti, it is likely that these versions of the paraphrases are earlier than the Cāndra-vyākaraṇavṛtti. The Kāśikāvṛtti was more popular than the Cāndravyākaraṇavṛtti and the Mahāvṛtti, and therefore it was more vulnerable to modification.  In addition, as we saw, the paraphrases of the Cāndravyākaraṇavṛtti in which the ablative and the upasargas are displayed in a different way can be ex-plained, and they are a result of an attempt to specify the desired scope of the sūtras, as the author himself clarifies. Thus, the Cāndravyākaraṇavṛtti chose to remove difficulties by providing a paraphrase which would specify the desired scope of the sūtra. The Mahāvṛtti, on the other hand, prefers to remain close to the literal meaning of the sūtras and prefers to resolve difficulties in its addi-tional comments to the sūtra. This may be the approach also of the earlier source. The Kāśikāvṛtti seems to have both ways, it sometimes presents a brief paraphrase clarifying only the meaning of the sūtra, but in respect to other sū-tras it has additional specification(s) which are meant to remove difficulties. It appears that the Kāśikāvṛtti that we have today is a result of these two approaches.  97 The variations in the presentation of the verbal bases in the Kāśikāvṛtti’s paraphrases 97and particularly the insertion of the term dhātu (See notes 26, 35 and 69, 74) may have originated for the same reason as the reason for the insertion of the term upasarga. Sūtra Paraphrases 27One is to remove difficulties by providing additional specifications within the paraphrases, and the other is to remain close to what the sūtra says. Over the years, the paraphrases of these two approaches have been merged and this may be one of the causes of the inconsistency in respect to the style of the paraphras-es in the Kāśikāvṛtti that we have today. A new critical edition of the Kāśikāvṛtti consulting additional manuscripts might shed light on the matter.  Acknowledgments I am grateful to Malhar Kulkarni and Peter Scharf for their valuable comments on earlier versions of this paper. This paper is based on a research funded by the Kone Foundation (Grant Number 088333).  Abbreviations A Aṣṭādhyāyī. Bh Mahābhāṣya of Patañjali.  BhV Bhāṣāvṛtti of Puruṣottamadeva. BM Bālamanoramā of Vāsudeva Dīkṣita. CV Cāndravyākaraṇavṛtti (See Candragomin). DhP Dhātupāṭha. (Reference by gaṇa and serial number as in Katre Sumitra M). J Jainendravyākaraṇa’s sūtras taken from the Mahāvṛtti. KV Kāśikāvṛtti of Vāmana and Jayāditya.  KVS Kāśikāvṛttisāra of Balabhadra Prasāda Tripāthī. MDhV Mādhavīyā Dhātuvṛtti of Sāyaṇa. MV  Mahāvṛtti of Abhayanandi. Nyāsa  Kāśikāvivaraṇapañjikā of Jinendrabudhi. (See Vāmana and Jayāditya [Kāśikāvṛtti Tara edition, 1988]). PM Padamañjarī of Haradatta. See Vāmana and Jayāditya (Kāśikāvṛtti Tara edition, 1988). S Sarasvatīkaṇṭhābharaṇa of Bhojadeva.  BEN-DOR 28Bibliography Aussant, Emilie. 2009. “Kāśikāvṛtti and Cāndravyākaraṇa: A Comparison of the Pratyāhārasūtra Section.” In Studies in the Kāśikāvṛtti. The Section on Pra-tyāhāras, Critical Edition, Translation and Other Contributions. 191-214. Edited by Pascale Haag and Vincenzo Vergiani. Balabhadra Prasāda Tripāthī. Kāśikāvṛttisāra. Part I. Critically edited and embel-lished with a Sanskrit commentary called Sudhā by Chandrabhanu Tripa-thi. Allahabad: Ganganatha Jha Kendriya Sanskrit Vidyapita. 1992  Bhojadeva. Sarasvatīkaṇṭhābharaṇa of Bhojadeva with the commentary of Nārāyaṇa-daṇḍanātha. Edited by Sambhasiva K. Shastri. Trivandrum: Trivandrum Sanskrit Series No. CXL, 1938. Bronkhorst, Johannes. 2004. “More on the sources of the kāśikā.” In Problems in Vedic and Sanskrit Literature (Ganesh Umakant Thite Felicitation Volume). Edited by Maitreyee Deshpande, 47-54. Delhi: New Bharatiya Book Corporation. Candragomin. Cāndravyākaraṇa of Candragomin. Part I. Edited by Chatterji, Kshi-tish Chandra Poona: Deccan College Postgraduate & Research Institute, 1953. Candragomin. Cāndravyākaraṇa of Candragomin. Part II. Poona: Deccan College Postgraduate & Research Institute, 1961. Gaṇa-sūtra (Reference by serial number as in Vāmana and Jayāditya 2008). Hṛdayahāriṇī of Nārāyaṇadaṇḍanātha (See Bhojadeva). Katre, Sumitra M. 1989. Aṣṭādhyāyī of Pāṇini. Delhi: Motilal Banarsidass. Patañjali. The Vyākaraṇa-Mahābhāṣya of Patañjali. Edited by F. Kielhorn. Third Edition by K. V. Abhyankar, Revised and Furnished with Additional Rea-dings, References and Select Critical Notes. Three volumes. Poona: Bhan-darkar Oriental Research Institute, 1962-72. Puruṣottamadeva. Bhāshāvṛtti (a commentary of Pāṇini’s grammar) by Purushottama-deva. Edited by Swami Dwarikadas Shastri. Varanasi: Tara Publications, 1971. Sāyaṇa. Mādhavīyā Dhātuvṛtti by Sāyaṇācārya. Edited by Swami Dwarikadas Shas-tri. Varanasi: Tara Book Agency, 2000.  Sūtra Paraphrases 29Vāmana and Jayāditya. Kāśikā of Vāmana and Jayāditya with Nyāsa or Kāśikā-Viva-raṇapañjikā of Jinendrabuddhi and Padamañjarī of Haradatta Miśra with the Bhāvabodhinī by Jaya Shankar Lal Tripathi. Ten volumes. Varanasi: Tara Book Agency, 1988. Vāmana and Jayāditya. Pāṇinīyvyākaraṇasūtravṛttiḥ Kāśikā of Vāmana and Jayāditya. Edited by Shri Narayan Mishra. Varanasi: Chaukhambha Sanskrit Sans-than, 1996. Vāmana and Jayāditya. Śrīvāmajayādityavircitā pāṇinīyāṣṭādhyāyīsūtravṛttiḥ Kāśikā. Edited by Vijayapalo Vidyavaridhi Hariyana: Ramlal Kapur Trust, 1997. Vāmana and Jayāditya. Kāśikā by Vāmana & Jayāditya. Edited by Aryendra Sharma. Hyderabad: Sanskrit Academy, Osmania University, 2008. Vāsudeva Dīkṣita. Vaiyakāraṇa-Siddhāntakaumudī of Bhaṭṭojidīkṣita with Bālamano-ramā of Vāsudevadīkṣita and Tattvabodhinī of Jṇānedrasarsvatī. Four volumes. Edited by Caturveda Giridharaśarmā and Parameśvarānandaśarma Bhās-kara. Varanasi: Motilal Banarsidass, 1987-1991.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.70440.1-0380177/manifest

Comment

Related Items