Skip to main content
SpringerLink
Log in

“Memory” Revisited: What Sāmavedic Technical Literature Tells Us About Smṛti’s Early Meaning

  • Published:
Journal of Indian Philosophy Aims and scope Submit manuscript

Abstract

In this paper, I build on recent scholarship concerning the early semantic history of the word “smṛti,” which has been shown to denote “tradition” in the early dharmasūtra material. I seek to add nuance to this work by examining the meaning of smṛti in the early Sāmavedic technical literature. This corpus helps elucidate one of the processes whereby smṛti came to refer to something textual. This paper argues that smṛti’s earliest textualized referent may have been fixed or semi-fixed individual statements rather than definite texts.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. See Klaus (1992) for an in-depth consideration of the original signification of the root smṛ and its derivatives in the Vedic corpus.

  2. Scharfe (2002, p. 16) demonstrates that Taittirīyāraṇyaka 1.2.1—which is often cited as the earliest passage that uses smṛti in the sense of “tradition” or “traditional texts derived from the Veda”—does not use the word in either of these senses. See also Brick (2006, p. 288).

  3. I would like to thank the anonymous reviewer for an especially clear presentation of the importance of the chronological progression of the dharmasūtras for the evaluation of the changing meaning of smṛti in that corpus.

  4. GDhS 1.1 literally states that “the Veda is the root of dharma” (vedo dharmamūlam), but Wezler (2004, pp. 637–638) has convincingly argued that this metaphor should be understood to mean that the Veda is “the cause of the knowledge of dharma.”

  5. Olivelle (2018, p. 53) makes the interesting suggestion that Gautama uses smṛti to mean “orally articulated recollections.” This meaning is consonant with what we see in some of the early Sāmavedic material. Unfortunately, he does not discuss this instance or his interpretation in detail.

  6. The Baudhāyanadharmasūtra (BDhS) includes “smārta” in its list of dharmapramāṇas. Brick (2006, p. 295) argues that this word means “the dharma connected with smṛti” and should be taken as pointing to something textualized, though he (2006, p. 291) also notes that the BDhS (1.2.8) may elsewhere use smṛti in the sense of an untextualized tradition.

  7. Brick (2006, p. 289, n. 11) has identified most of the relevant instances of smṛti in the early sūtra literature. I would only add that it also appears in the Drāhyāyaṇaśrautasūtra, which is a more recent and expanded version of the Lāṭyāyanaśrautasūtra. Outside of the dharmasūtras and the Sāmavedic technical literature, smṛti occurs in HIraṇyakeśi-Satyāṣāḍhaśrautasūtra 3.1.2, p. 1:268; Ṛgvedaprātiśākhya (RPrŚ) 11.63; and Vārāhagṛhyasūtra 6.32. The first of these states that “domestic rituals are gotten through the memory of Vedic words” (vaidikānāṃ śabdānāṃ smṛtigrahaṇāni laukikāni) and provides one of our early attestations of what would become the classical theory of smṛti. The second appears in a chapter of the RPrŚ that, given its composition in the classical vaṃśastha meter and supplementary character, is probably a later addition to the text (Shastri, 1959, pp. 62, 76). The last example is discussed in some detail by Brick (2006, p. 296).

  8. The interpretation for smṛti I propose here was prompted by the insightful comments of the anonymous reviewer of this article, who pointed out the possibility that the Nidānasūtra (NidS) might use smṛti, when compounded with ācārya, to refer to statements attributed to venerable seers.

  9. Bühler (1898, p. 139), Olivelle (1999, p. 59), and Brick (2006, p. 292) all translate aparam as “the superior position,” which accounts for the distinction between eke and aparam. One might also consider translating aparam as “another position,” especially because the parallel sūtra in the Hiraṇyakeśidharmasūtra 27.3.57, 10:173 reads eke in the place of aparam.

  10. Bühler (1898, p. 139) understands adhikāra to signify the right to perform sacrificial rituals, whereas Olivelle (1999, p. 59) and Brick (2006, p. 292) take it to mean that the child comes under the purview of a certain set of rules.

  11. My translation largely follows the earlier translations of Bühler (1898, p. 139), Olivelle (1999, p. 59), and Brick (2006, p. 292). There are several salient points where Bühler’s translation disagrees with those of Olivelle and Brick.

  12. Here Bühler no doubt follows the commentator Haradatta: upanayanam api parāmṛśatas tacchabdasya niṣṭhāśabdasāmānādhikaraṇyāt strīliṅgatā | sā niṣṭhā tad upanayanam avasānam anadhikārasyeti. “Even though the word ‘that’ (i.e., sā) refers to [the neuter] word ‘initiation,’ it is qualified by the feminine gender because of its co-referentiality with the [feminine] word ‘limit.’ That is the limit [i.e.,] that initiation is the termination of the absence of qualification [for the performance of sacrificial action].”

  13. For the first, see ĀpDhS 2.23.10: traividyavṛddhānāṃ tu vedāḥ pramāṇam iti niṣṭhā tatra yāni śrūyante vrīhiyavapaśvājyapayaḥkapālapatnīsaṃbandhāny uccair nīcaiḥ kāryam iti tair viruddha ācāro ’pramāṇam iti manyante. “It is the firm view of the most eminent scholars of the triple Veda, however, that the Vedas are the ultimate authority. The rites using rice, barley, animals, ghee, milk, and potsherds and involving the participation of the wife that are prescribed in the Vedas must be performed with the loud and soft recitation of ritual formulas, they hold, and any practice opposed to those rites is devoid of authority.” This translation is Olivelle’s (1999, p. 67). For the second, see ĀpDhS 2.29.11: sā niṣṭhā yā vidyā strīṣu śūdreṣu ca | ātharvaṇasya vedasya śeṣa ity upadiśanti. “The knowledge found among women and Śūdras forms the conclusion, and they point out that it is a subsidiary component of the Atharva Veda.” This is also Olivelle’s translation (1999, p. 72).

  14. GDhS 11.25: vipratipattau traividyavṛddhebhyaḥ pratyavahṛtya niṣṭhāṃ gamayet. “Given conflicting [evidence, the king] should confer with those learned in the three Vedas and deliver his judgement (niṣṭhā).” My translation follows Olivelle’s (1999, p. 97) with some modifications to the wording.

  15. Wezler (2004, p. 631) suggests that samaya in the phrase dharmajñasamaya might mean something like “agreement that exists or appears in practice.” Perhaps smṛti points to something similar, at least insofar as this text is concerned.

  16. ĀpDhS 1.7.31: samāvṛttasyāpy etad eva sāmayācārikam eteṣu. “Even after he has returned home, the accepted practice is that he should behave towards these individuals exactly the same way as he behaved when he was a student.” This translation is Olivelle’s (1999, p. 16).

  17. This suggestion was made by the anonymous reviewer of this article. Brick (2006, pp. 299–300) discusses the connection between the increasing use of these verses and the textualization of smṛti in some detail, though he does not suggest that smṛti ever referred to them individually or collectively. See also Olivelle (1999, pp. xxviii–xxx) for a discussion of the citation of verses in the dharmasūtras.

  18. Bronkhorst (2007, p. 189) draws our attention to these particular examples, which he has taken from Bhatṛhari’s commentary on the Mahābhāṣya.

  19. For a more extensive overview of the Sāmavedic literature, see Howard (1977, pp. 8–9) and Parpola (1968b, pp. 1:67–76).

  20. There are, as we will see, several sāmans that are exceptionally composed over several ṛks in the Prakṛtigāna.

  21. The Uttarārcika contains most of the sets of two or three ṛks over which the extensions are supposed to be performed; see Howard (1977, p. 8–9).

  22. See Parpola (1968b, pp. 1:75–76), though note that he is in disagreement with Caland (1931, pp. xi–xiii) about the date of the two ūha songbooks.

  23. For the parallel sūtras, see Drāhyāyaṇaśrautasūtra (DŚS) 16.1.5-7. Note that Dhanvin, the commentator of the DŚS, reads DŚS 16.1.7 as: tisṛṣv ācāryāḥ smṛter yathānyāni sāmāni. This reflects a combination of LŚS 6.1.6 and LŚS 6.1.7, the latter of which likewise reads yathānyāni sāmāni “Just as in the case of other sāmans.” See also Ranade (1998).

  24. Caland (1931, p. 323) notes that the Calcutta edition of the Sāmaveda mistakenly sets this sāman over a single ṛk and that the correct reading can be found in the Grantha edition, where the Santani is chanted over three stotrīyās.

  25. Technically speaking, it is composed over three stotrīyās; the latter two being made by the division of a single verse from the Pūrvārcika (Caland 1931, p. 323).

  26. Agnisvāmin on LŚS 6.1.4: yathāmnāya ekarce dṛṣṭāni prayoge tṛce kriyante evam idam api santani tṛce dṛṣṭaṃ prayoge triṣu tṛceṣu kriyate. “Just as [sāmans] set over a single ṛk in the Prakṛtigāna (āmnāya, lit. traditional text) should be performed over three verses during their use [in the Soma sacrifices], so too the Santani, which is set over three verses [in the Prakṛtigāna], should be performed over three verse-triplets.” Similarly, see Dhanvin on DŚS 16.1.5: yathaikasyām ṛci dṛṣṭaṃ rathantaram uttarayor apy ṛcor eva gīyate | sādṛśyasya saṃpatyartham evaṃ tṛce dṛṣṭaṃ santany uttarayor api tṛcayor eva geyam. “Just as the Rathantara, which is set over a single verse [in the Prakṛtigāna], is sung over two further verses, the Santani, which is set over three verses, should be sung over two further verse-triplets in order to bring about a similarity [with the Rathantara in terms of the transformation it undergoes for use in the Soma sacrifices].”

  27. Agnisvāmin on LŚS 6.1.6: evaṃ hi ācāryaiḥ gītaṃ smṛtir ūhagītiḥ. “The ‘memory’ (smṛti) that ‘it was sung in this way by the teachers,’ i.e., the Ūhagīti. See also Dhanvin on DŚS 16.1.7: saiva kuta iti cet vasiṣṭhāder ūhakārasya smṛter iti brūmaḥ. “Why is [the sāman thought to be completed over each ṛk in the Prakṛtigāna]? We say it is on the basis of the memory of the ūha-makers such as Vasiṣṭha.”

  28. Dhanvin on DŚS 16.1.7: smṛter eva kiṃ mūlam iti cet pratyabhijñeti brūmaḥ | chāndasapāṭhe yathā prathamāyām ṛci santanino gānaṃ tathaivottarayor api pratyabhijānīmaḥ | prastāvas tu kevalaṃ na dṛśyate. “Now what is the root of that memory? We say it is recognition. Just as we recognize the tune of the Santani over the first ṛk in the Prakṛtigāna (chāndasapāṭha, lit. reading derived from the Pūrvārcika), we recognize it over the second two as well. Only the introductory portion (prastāva) is absent.” See also 16.1.14: chāndasapāṭhe prastāvādarśanam eva smṛter mūlam. “The basis of the memory is the absence of the introductory portions in the Prakṛtigāna.”

  29. Ūhagāna Daśarātraparva 5.10, p. 100–101; 7.1, p. 139; Saṃvatsaraparva 3.13, p. 250–251, Ahīnaparva 2.7, p. 420–421; 2.10, p. 422; Satraparva 1.6, p. 478; 4.4, p. 514–515; and 4.7, p. 515–516. Here I refer to the Ūhagāna of the Kauthuma śākhā because the LŚS is connected with this śākhā.

  30. Weber (1886, p. 27) provides a list of all the sūtras where the LŚS cites the Ārṣeyakalpa; some appear under the abbreviated name Kalpa.

  31. LŚS 6.1.10: ekarceṣu coddhṛtaṃ kalpe. “In the [Ārṣeya]kalpa, [the Santani] is extracted [for application] to sāmans composed over a single ṛk.” This argument refers to Ārṣeyakalpa (Caland, 1908, p. 142–143) 10.2, 5th day: bārhadgirasyarkṣu santani. “The Santani [is performed to the] verses of the Bārhadgiri sāman;” 7th day: kāṇvasyarkṣu santani. “The Santani [is performed to the] verses of the Kāṇva sāman.” Both the Bārhadgiri and Kāṇva sāmans are set to a single ṛk in the Prakṛtigāna. See Āraṇyakagāna 3.1.16 and Grāmegeyagāna 7.1.28.

  32. See also DŚS 16.1.13-14.

  33. See Dhanvin on DŚS 16.1.13: pūrṇaṃ hi sāma geyam iti nyāyāt. “Because of the principle that a sāman is to be sung in a complete form.”

  34. Dhanvin on DŚS 16.1.13: yadi pratyṛcaṃ sāmasamāptiḥ tarhi pratyṛcaṃ pāñcabhaktikam eva gāyen nottarayoḥ prastāvalopaḥ kāryaḥ. “If the sāman is completed with every ṛk-verse, then each of those verses should be sung with all five bhaktis. The prastāvas should not be elided for the latter two verses.” The fact that the prastāvas must be added comes out more clearly in Agnisvāmin’s commentary on LŚS 6.1.12: ya uttare stotrīye luptaprastāva āmnāyete tayor api prastāvaḥ kartavyaḥ. “One must perform a prastāva even for the two subsequent stotrīya-verses that are transmitted without them.”

  35. Agnisvāmin on LŚS 6.1.13: uttarayoḥ stotrīyayor ācāryāḥ prastāvaṃ necchanti | kasmāt | smṛteḥ | evaṃ hi smaryate santany ūhagītau. See also Dhanvin on DŚS 16.1.14: ūhakārasmṛter nottarayoḥ prastāvaḥ. “On the basis of the memory of the ūha-makers, there is no prastāva for the latter two verses.” According to Parpola (1968b, p. 1:66), the Anustotrasūtra, which examines the form sāmans actually take in ritual performance, also connects the word smṛti with ūha.

  36. Parpola (1968b, pp. 1:133–136) provides a detailed comparison of parallel passages from the two texts and argues that the LŚS must have reformulated part of the NidS.

  37. This interpretation was first pointed out to me by the anonymous reviewer of this article.

  38. NidS 2.1, p. 22: ṛṣikṛtāḥ svid ūhā3 anṛṣikṛtā iti.

  39. The pūrvapakṣa is cited in full in the appendix.

  40. This opening statement responds to the adversary’s initial position, which reads: anṛṣikṛtā iti vai khalv āhuḥ. “Indeed, they say that the modified sāmans (ūhāḥ) were not created by the seers.” The editor suggests emending ṛṣikṛta ity aparam to ṛṣikṛtā ity aparam. My translation follows his suggestion.

  41. I prefer to read eva along with the related manuscripts collectively labeled “C” by the editor of the critical edition. My translation follows this reading.

  42. NidS 2.1, p. 22–23.

  43. The phrase attributed to Vasiṣṭha appears to be a modified and abbreviated form of a common formulation found especially in the Jaiminīyabrāhmaṇa. See, for example, Jaiminīyabrāhmaṇa 1.217: śrutakakṣaḥ kākṣīvataḥ paśukāmaḥ tapo ’tapyata | sa etat sāmāpaśyat. “Desiring livestock, Śrutakakṣa, the son of Kakṣīvat, practiced austerities. He saw this sāman.” See Caland (1919, p. 84) for a German translation.

  44. See Aṣṭādhyāyī 4.2.7: dṛṣṭaṃ sāma. “[The affix ḌaṆ is applied after the name of an individual by whom a sāman has been seen in the sense of] a sāman seen [by him].” See also Jaiminīyabrāhmaṇa 3.26: yad u vasiṣṭho ’paśyat tasmād vāsiṣṭham ity ākhyāyate. “Vasiṣṭha saw it, therefore it is called the Vāsiṣṭha sāman.” The anonymous reviewer of this article pointed out that a number of Vedic texts connect Vasiṣṭha with the Rathantara sāman and Bharadvāja with the Bṛhat sāman. See, for example, Ṛgveda 10.181.1-2 and Aitareyabrāhmaṇa 1.21, p. 16.

  45. Ūharahasyagāna Kṣudraparvan 1.7, p. 703–704. See also Puṣpasūtra 2.29.4. But note that, according to Caland, the Bṛhadrathantara sāman is “Die rathantara-Singweise auf den stotrīyās des Bṛhat.” See Kṣudrasūtra 2.5 n. 6 (Caland 1908, p. 182).

  46. This reading was suggested by the anonymous reviewer of this article.

  47. Several manuscripts collectively labeled as “C” read vāsiṣṭhabharadvājam, and instead of the dual ete, they read anenaiva. If we were to accept this set of readings, perhaps the meaning would be something like “Because of this very [reference], they call it the sāman seen by Vasiṣṭha and Bharadvāja.” It is also worth noting that the reading ete is only found in one manuscript (B), albeit the editor states this is his most reliable one. The others read anye or the more problematic enye.

  48. Compare, for example, with Pañcaviṃśabrāhmaṇa (PB) 11.6.3, where abhivadati is used to mean that the priest “addresses” the second day of the bahiṣpavamāna. See also PB 12.1.8, where tad eva tad abhivadati seems to mean that the priest “addresses” or “refers to” the antarikṣa by chanting a sāman composed on a ṛk (RV 9.65.16) that includes the word antarikṣa. For a translation of these two passages, see Caland (1931, pp. 260, 272).

  49. This interpretation was suggested by the anonymous reviewer of this article.

  50. I am here adopting with some modification the insightful view of the anonymous reviewer of this article.

  51. Contrast with NidS 6.5, p. 138: eṣa ācāryasamayaḥ. “This is the convention of the teachers.” This phrase refers to conventional way of performing the Mādhyandinapavamāna during the Gargatrirātra ritual—the normal Yaudhājaya sāman is replaced the Yaudhājayekākṣarānta sāman. This reference does not seem to involve a textualized tradition of any kind. The Ārṣeyakalpa 6.3, p. 78 does not make note of this convention, though Caland (n. 3) mentions it in a footnote of his edition.

  52. NidS 2.1, p. 23.

  53. For the date of the MīSū, see, for example, McCrea (2010, p. 124 n. 2).

  54. Śābarabhāṣya on MīSū 9.2.1: kim ayam apy ārṣo nitya uta puruṣapraṇīta iti. Technically, this statement refers to a specific instance of Sāmavedic ūha cited in Śabara’s commentary, but the point is obviously to deal with Sāmavedic ūha more generally. I have translated the line with this larger frame of reference in mind.

  55. By this I mean MīSū 1.3, which the later tradition would come to call the “Smṛtipāda.” But, see also the use of smṛti in MīSū 7.1.10-11.

  56. Śābarabhāṣya on MīSū 9.2.1: evaṃ hi smaranti | mantrabhūtāny etāny ārṣāṇi nityānīti.

  57. This meaning of smṛti seems similar to one that Olivelle (2018, p. 59) suggests in a different context. See note five of my paper.

  58. There is a possible counterargument to this interpretation. The MīSū’s pūrvapakṣa might want to draw attention to the existence of the ūha songbooks, which are, according to Śālikanātha, studied in the same manner as Vedic texts. If smṛti were to refer to these songbooks, then the adversary’s point might be to highlight that these songbooks are studied and treated in a manner identical to other Vedic texts. See Śālikanātha’s commentary on Bṛhatī 9.2.1-2.

  59. For example, SVB 1.5.13 contains the term abhojya, which Olivelle (2002, p. 345) informs us appears nowhere else in the Vedic corpus apart from the late Gopathabrāhmaṇa (1.3.19). SVB 3.5.3 contains an etymological explanation for the term cakravartin. Kane (1946, p. 66) tells us that in the Vedic literature this word first appears in the late Maitrāyaṇīyopaniṣad. Similarly, SVB 3.9.1 contains the word maṭha in the sense of a “hovel,” a term that Scharfe (2002, p. 172) states first appears in the later sections of the Mahābhārata (12.139.29) and in the Baudhāyanadharmasūtra (3.1.14).

  60. The direction of borrowing appears clear given the GDhS makes additions to a liturgical utterance prescribed along with a water-libation. Compare SVB 1.2.7 with GDhS 26.12. See Bühler (1898, p. li) and Olivelle (1999, p. 74)

  61. Burnell’s edition reads kāmākṣarasaṃyuktaḥ and Konow’s (1893, p. 79) translation seems loosely to follow it (Wunschworten). Sharma’s apparatus reveals that this is the reading transmitted in Sāyaṇa’s commentary. Bharatasvāmin, who lived around 1300 CE, reads kāmāḥ kṣurasaṃyuktāḥ.

  62. Sharma seems to emend his edition of Sāyaṇa’s commentary to read kāmāḥ kṣurasaṃyuktāḥ, though he notes in his apparatus that variants appear in the manuscripts he labels A (kāmākṣarasaṃyuktāḥ) and B (kāmākṣarasaṃyukto) as well as in Burnell’s edition. According to Sharma’s introduction (p. xi), A and B are the only manuscripts he consulted that include Sāyaṇa’s commentary. It is possible that he is here following Satyavrata Samasramin’s 1895 edition. I unfortunately do not have access to this work. Insofar as the root text is concerned, it seems that Sharma’s better manuscripts read kāmāḥ kṣurasaṃyuktāḥ, but a few others provide a reading closer to Sāyaṇa’s as it appears in the manuscripts reported by Sharma.

  63. One might consider whether the reading akṣara transmitted in Sāyaṇa’s commentary is meant in some way to refer to nidhanas and not to observances. I am, however, unaware of any Sāmavedic technical literature that uses the word akṣara this way, though it is sometimes used to specify the number of akṣaras that appear in a nidhana.

  64. Is it possible that Sāyaṇa actually read something like kāmā akṣarasaṃyuktāḥ and the two words were mistakenly written together? This explanation would at least account for the fact that the gloss niyamasaṃyuktāḥ seems to ignore the word kāma. Or, if the singular reading of B is correct, perhaps part of the o-ligature was dropped by scribal error?

  65. SVB 1.5.15, 2.5.5 (Burnell 2.5.4), 3.6.11, and 3.6.13.

  66. The critical edition appears to have been misnumbered as 2.5.3 has been skipped. In Burnell’s edition and Konow’s (1893, p. 58) translation, this passage is numbered 2.5.5.

  67. SVB 2.5.5: kṛṣṇavrīhīṇāṃ nakhanirbhinnānāṃ piṣṭamayīṃ pratikṛtiṃ kṛtvā piṣṭasvedaṃ svedayitvā sarṣapatailenābhyajya tasyāḥ kṣureṇāṅgāny avadāyāgnau juhuyāt. “Making an effigy out of ground up black rice broken up with the fingernails, he should let the dough rise, rub it with mustard oil, and slice the limbs with a razor offering them into the fire.” I am here following Konow’s (1893, p. 58) German translation.

  68. My translation follows Bhat’s rendering (1987, p. 311), though I have made a number of modifications.

  69. Could the image point to something like Kaṭhopaniṣad 3.13-14: yacched vāṅmanasī prājñas tad yacchej jñāna ātmani | jñānam ātmani mahati niyacchet tad yacchec chānta ātmani || uttiṣṭhata jāgrata prāpya varān nibodhata | kṣurasya dhārā niśitā duratyayā durgaṃ pathas tat kavayo vadanti. “A wise man should curb his speech and mind, control them within the intelligent self; He should control intelligence within the immense self, and the latter, within the tranquil self. Arise! Awake! Pay attention, when you’ve attained your wishes! A razor’s sharp edge (kṣurasya dhārā) is hard to cross—that, poets say, is the difficulty of the path.” This translation is Olivelle’s (1998, p. 391) with my addition in parentheses.

  70. See Konow (1893, pp. 78–79) for a German translation that agrees in sense with mine.

  71. An analogous concern with the assignment of observances can be felt in 2.1.1-2: athātaḥ kāmyānām | anādeśe trirātram upavāsaḥ puṣyeṇārambhaḥ. “Now, on to [the ritual procedures performed] for particular objects. In the absence of a specific instruction, one should perform a fast for three days and begin in the month of Pauṣa.”

  72. For example, Konow’s (1893, p. 79) “den Andeutungen der Smṛti.” Liṅga does, albeit seldomly, appear in the meaning of “character” in some early texts, for example Bṛhadāraṇyakopaniṣad 4.4.6. But this meaning would here make liṅga redundant; the word smṛti would suffice on its own.

  73. For the phrase yathāliṅgam, which means “[apply the mantra] in accordance with its indicatory mark,” see, for example, ĀpŚS 1.5.5, 3.20.2, 4.1.8, and 4.10.1. The ĀpŚS uses liṅga without yathā in 19.18.4 and 19.18.8.

  74. The SVB sometimes prescribes particular endings (nidhanas) that relate to the function of the ritual. Were liṅga to be taken separately from smṛti as referring to indications found in the underlying verses of the sāmans or to their Prakṛtigāna forms, these nidhana would be of no relevance as they are specifically enjoined by the SVB and not part of the underlying sāman. There is some further discussion of the issue of the relationship between the words underlying particular sāmans and their function in the SVB in Konow (1893, pp. 22–23).

  75. NidS 4.2, p. 65: tāsāṃ dharmo niyujyate na vyāhared nāśnīyād abhinaddhākṣam āsīta | … | etam evāhaṃ brahmacaryaṃ dharmam u manya iti gautama. “A prescribed course of conduct (dharma) is enjoined for [the Purīṣa sections. The performer] must neither speak nor eat, and he must sit with his eyes covered … I think this very [course of conduct] is brahmacārya—so says Gautama.” It is not entirely clear where this quotation begins, but at least part of it is attributed to Gautama.

  76. This is especially clear in examples like NidS 3.3, p. 44: atirātrīṃ prathamāṃ manya iti dhānañjayyaḥ. These sorts of expressions were probably not fully fixed though they appear to have been transmitted verbally. Some seem to have developed as an oral commentary on older texts; for example, LŚS 10.19.11: pravrajiṣyato ’yanam idaṃ manya iti dhānañjayyas tad eva manuṣyebhas tirobhavatīti. “Dhānañjayya [says] ‘I think this course is for the man wanting to depart [from the world;’ for a brāhmaṇa (PB 25.11.4) states] ‘There he disappears from the (eyes of) men.’” The translation of the quoted section from the PB is Caland’s (1931, p. 639). Parpola (1968a, p. 75) argues, however, that “the wording of the quotations, including those with iti, must have been formulated by the sūtrakāra…”, though he also notes that the LŚS may have “wanted the quotations with iti to be taken as genuine, for in three cases the wording is distinctly marked to be original.” Here he refers us to quotes that include the word manye.

  77. For example, NidS 6.8, p. 104: ṣaḍṛce sarvasminn ity ācāryāḥ | arddharceṣv anugānāny evam asaṃrodhataro gītānām iti | dvādaśarca iti śāṇḍilyaḥ and LŚS 10.9.8-9: ṣaḍṛce sarvasminn ity ācāryā arddharceṣv anugānāny evam saṃrodhataro gītānām | dvādaśarca iti śāṇḍilyaḥ …. It is of course possible that one of these texts simply makes use of the other; this is the opinion of Parpola (1968b, p. 1:133–134), who bases his conclusions on the examination of one set of corresponding sections. Perhaps here too we should pay attention to Wezler’s methodological caution when it comes to texts. Both Caland (1908, p. xviii) and Bhatnagar (1971, p. 28) conclude that the LŚS and the Nidānasūtra do not know each other, though both scholars suggest that in a number of instances the two texts appear to have drawn from similar sources.

  78. Excluding the GDhS, which contains no verses.

  79. This suggestion was proposed by the anonymous reviewer of this article. It should be noted that Brick (2006, pp. 299–301) discusses this idea, stating that, at least in early contexts, smṛti should not be equated with the verses. Later, he (2006, p. 301) argues that the frequent use of verses “could facilitate a collapsing of the distinction between tradition and fixed expressions of tradition.” Yet he does not, at least insofar as I can tell, specifically suggest that smṛti ever referred to individual verses as opposed to a body of literature.

  80. Though, as Brick (2006, p. 287) informs us, “medieval Sanskrit commentators” often took smṛti to refer to a number of non-Vedic genres “such as dharmaśāstra, itihāsa, and purāṇa.”

  81. In a personal communication (Jan 13, 2021), Asko Parpola tentatively suggested that this statement might mean: “They chant (the sacrificial songs) knowing on how many verses (three or one) the sāman is to be sung and the yoni verse. For in the Smṛti (i.e., singing after the Ūhagānas), what use would one have of the knowledge of these two things [as that knowledge has already been used and the singer is in the Ūhagāna provided with the ready result of applying that knowledge]?”

  82. NidS 2.1, p. 23.

  83. The discussion begins with NidS 1.1, p. 2: athāto vṛttipradeśa iti yatra hrasvam akṣaram upottamaṃ pādasya bhavati sā jāgatī vṛttir iti. The subsequent passages discuss various types and conditions.

  84. As in Ṛgvedaprātiśākhya 13.46: tisro vṛttīr upadiśanti vāco vilambitāṃ madhyamāṃ ca drutāṃ ca. But compare NidS 2.10, p. 36: athāpi bahulaiṣā chāndasī vṛttiḥ, which seems to point specifically to the rhythmic cadences of the various meters—or does chandas point instead to the ṛks? See also NidS 2.12, p. 39: athāpy uddhriyamāṇeṣūpasargākṣareṣu nevārtho hīyate na vṛttir duṣyati. This deals with the problem of how to determine which syllables serve as the liturgical interpolations (upasargas) of the Gaurīvita sāman.

  85. The reading pārthikā vṛttir yoni kalpante is difficult to understand. The editor of the edition Bhatnagar (1971, p. 70) suggests emending to “vṛttir yonir iti kalpante,” though unfortunately he does not provide a translation of any sort. For the meaning of chāndasa-, see NidS 2.2, p. 24: chāndasenādhyāyenaikarcān bhūyiṣṭāñ chandasy adhīmahe. This refers to the fact that most sāmans in the chāndasa-reading are recited over a single ṛk.

  86. NidS 2.1, p. 24.

  87. NidS 2.1, p. 24.

  88. See also NidS 2.1, p. 23: yad v etad yājñikā vipratipadyanta iti.

Abbreviations

ĀpDhS:

Āpastambadharmasūtra

ĀpŚS:

Āpastambaśrautasūtra

BDhS:

Baudhāyanadharmasūtra

DŚS:

Drāhyāyaṇaśrautasūtra

GDhS:

Gautamadharmasūtra

LŚS:

Lāṭyāyanaśrautasūtra

MīSū:

Mīmāṃsāsūtra

NidS:

Nidānasūtra

RPrŚ:

Ṛgvedaprātiśākhya

Rvidh:

Ṛgvidhāna

SVB:

Sāmavidhānabrāhmaṇa

References

Sanskrit Texts

  • Anekārthasaṃgraha, i.e., Sānekārthanāmamālātmakaḥ koṣavaraḥ śubhaḥ hemacandrapraṇītābhidānacintāmaṇir maṇiḥ. Calcutta: 1807.

  • Aṣṭādhyāyī of Pāṇini, see Cardona (1997).

  • Āpastambadharmasūtra, i.e. The Dharma Sutra by Apastamba with the Commentary (-ujjvala) of Haradatta, ed. R. Halasyanatha Sastry. Kumbakonam: Sri Vidya Press, 1895.

  • Ārṣeyakalpa, i.e., Der Ārṣeyakalpa des Sāmaveda, ed. W. Caland. Leipzig: Deutschen Morgenländischen Gesellschaft, 1908.

  • Aitareyabrāhmaṇa, i.e., Das Aitareya Brāhmaṇa, ed. Theodor Aufrecht. Bonn: Adolph Marcus 1879.

  • Ūhagānam and Ūhyagānam with Uttarārchika and Padapāṭha of the Kauthuma Śākhā, ed. A. M. Ramanath Dikshit. Varanasi: Vedic Research Committee, Banaras Hindu University, 1967.

  • Kaṭhopaniṣad, see Olivelle (1998).

  • Kṣudrasūtra, see Ārṣeyakalpa.

  • Gautamadharmasūtra, i.e., Gautamapraṇītadharmasūtrāṇi with the Commentary (-mitākṣarāvṛtti) of Haradatta, ed. Ganeśa Śāstrī Gokhale. Pune: Ānandāśrama Press, 1931.

  • Drāhyāyaṇaśrautasūtra with the Commentary (-bhāṣya) of Dhanvin, ed. M. Ramanatha Dikshitar. Madras: Sir C. P. Ramasvami Iyer Foundation, 1982.

  • Nidānasūtra, i.e., Nidāna-Sūtra of Patañjali, ed. K. N. Bhatnagar. Delhi: Meharchand Lachhmandas, 1971.

  • Bṛhatī of Prabhākara Miśra with the Commentary (-ṛjuvimalā pañcikā) of Śālikanātha, ed. Sastri, S. K. Ramanatha and S. Subrahmanya Sastri. 5 vols. Madras: University of Madras, 1934-1967.

  • Baudhāyanadharmasūtra, i.e. The Baudhāyana-Dharmasūtra with the Commentary (-vivaraṇa) of Śrī Govinda Svāmī. Ed. Umeśa Chandra Pāṇḍeya. Varanasi: Chowkhamba Sanskrit Series Office, 1972.

  • Mīmāṃsāsūtra, i.e., Śrīmajjaiminipraṇīte Mīmāṃsādarśane with the Commentaries of (-bhāṣya) Śabara, (-prabhā) of Śrī Vaidyanātha Śāstri, (-tantravārttika) and (-ṭupṭīkā) of Kumārila Bhaṭṭa, eds. Kāśinātha Vāsudeva Abhyankar and Ganeśaśāstrī Jośī. 7 vols. Poona: Ānandāśrama Sanskrit Series, 1970.

  • Ṛgvidhāna, see Bhat (1987).

  • Ṛgveda, i.e., Die Hymnen des Ṛigveda, ed. Theodor Aufrecht. 2 vols. Bonn: Adolph Marcus, 1877.

  • Ṛgvedaprātiśākhya, i.e. The Rgveda-Pratisakhya with the Commentary of Uvaṭa, ed. Mangal Deva Shastri. Allahabad: The Indian Press, 1959.

  • Lāṭyāynaśrautasūtra, i.e., Śrautasūtra of Lāṭyāyana with the Commentary (-bhāṣya) of Agnisvāmin, ed. Ānandachandra Vedāntavāgīśa. Calcutta: Asiatic Society of Bengal, 1872.

  • Sāmavidhānabrāhmaṇa with the Commentaries (-vedārthaprakāśa) of Sāyaṇa and (-padārthamātravivṛti) Bharatasvāmin, ed. B. R. Sharma. Tirupati: Kendriya Sanskrit Vidyapeetha, 1980.

  • Sāmavedasaṃhitā with the Commentary (-vedārthaprakāśa) of Sāyaṇa, ed. Satyavrata Sāmaśramī Bhattacharyya. 5 vols. Calcutta: Bibliotheca Indica, 1871-78.

  • Hiraṇyakeśidharmasūtra, see Hiraṇyakeśi-Satyāṣāḍhaśrautasūtra, vol 10.

  • Hiraṇyakeśi-Satyāṣāḍhaśrautasūtra, i.e., Satyāṣāḍhaviracitaṃ Śrautasūtram, ed. Kāśīnāthaśāstrī Āgāśe and Śaṃkaraśāsrtī Mārulakara. 10 vols. Poona: Ānandāśrama Publishing House, 1907-1932.

Other References

  • Bhat, M. S. (1987). Vedic Tantrism: A study of Ṛgvidhāna of Śaunaka with text and translation. Motilal Banarsidass.

    Google Scholar 

  • Brick, D. (2006). Transforming tradition into texts: The early development of “smṛti.”. Journal of Indian Philosophy, 34(3), 287–302.

    Article  Google Scholar 

  • Bronkhorst, J. (2007). Greater Magadha: Studies in the culture of early India. Brill.

    Book  Google Scholar 

  • Bühler, G. (Trans.). (1898). Sacred laws of the Aryas as taught in the Schools of Apastamba, Gautama, Vāsiṣṭha, and Baudhāyana. The Christian Literature Company.

  • Burnell, A. C. (Ed.). (1873). The Sāmavidhānabrāhmaṇa (Being the Third Brāhmaṇa of the Sāma Veda) (Vol. 1). Trübner & Co.

  • Caland, W. (Trans.). (1919). Das Jaiminīya-Brāhmaṇa in Auswahl: Text, Übersetzung, Indices. Johannes Müller.

  • Caland, W. (Trans.) (1931). Pañcaviṃśa-Brāhmaṇa: The Brāhmaṇa of Twenty-Five Chapters. Asiatic Society of Bengal.

  • Cardona, G. (1997). Pāṇini, his work and its traditions: Volume I, background and introduction; second edition, revised and enlarged. Motilal Banarsidass.

    Google Scholar 

  • Gonda, J. (1975). A history of Indian literature: Vedic literature (Saṃhitās and Brāhmaṇas) Volume I, Fasc I. Otto Harrassowitz.

    Google Scholar 

  • Howard, W. (1977). Sāmavedic Chant. Yale University Press.

  • Kane, P. V. (1946). History of Dharmaśāstra: Ancient and Medieval Religious and Civil Law (Vol. 3). Bhandarkar Oriental Research Institute.

    Google Scholar 

  • Klaus, K. (1992). On the meaning of the root smṛ in Vedic literature. Wiener Zeitschrift für die Kunde Südasiens, 36, 77–86.

    Google Scholar 

  • Konow, S. (Ed. and Trans.). (1893). Das Sāmavidhānabrāhmaṇa: ein altindisches Handbuch der Zauberei. Max Niemeyer.

  • Lubin, T. (2016). Custom in the Vedic ritual codes as an emergent legal principle. Journal of the American Oriental Society, 136(4), 669–687.

    Article  Google Scholar 

  • Shastri, M. D. (1959). see Ṛgvedaprātiśākhya.

  • McCrea, L. (2010). Hindu jurisprudence and scriptural hermeneutics. In D. D. Timothy Lubin & J. Krishnan (Eds.), Hinduism and law: An introduction (pp. 123–136). Cambridge University Press.

    Chapter  Google Scholar 

  • Olivelle, P. (Ed and Trans.). (1998). The early Upaniṣads: Annotated text and translation. Oxford University Press.

  • Olivelle, P. (Trans.). (1999). Dharmasūtras: The law codes of Āpastamba, Gautama, Baudhāyana and Vasiṣṭha. Oxford University Press.

  • Olivelle, P. (2002). Abhakṣya and abhojya: An exploration in dietary language. Journal of the American Oriental Society, 122(2), 345–354.

    Article  Google Scholar 

  • Olivelle, P. (2018). Epistemology of dharma. In P. Olivelle D. R. Davis Jr. (Ed.), Hindu law: A new history of Dharmaśāstra (pp. 49–59). Oxford University Press.

  • Parpola, A. (1968a). On the quotations of ritualistic teachers in the śrautasūtras of Lāṭyāyana and Drāhyāyaṇa. In S. Denis (Ed.), Studies in South, East and Central Asia: Presented as a memorial volume to the late Professor Raghu Vira (pp. 69–85). International Academy of Indian Culture.

    Google Scholar 

  • Parpola, A. (1968b). The Śrautasūtras of Lāṭyāyana and Drāhyāyaṇa and their commentaries, vol. I. Commentationes Humanarum Litterarurm, 42(2), 1–141.

    Google Scholar 

  • Ranade, H. G. (Ed. and Trans.). (1998). Lāṭyāyana-Śrauta-Sūtra (Vol. 3). Indira Gandhi National Centre for the Arts and Motilal Banarsidass Publishers Pvt. Ltd.

  • Renou, L. (1939). Review of Nidāna-sūtra of Patañjali. Journal Asiatique, 231(1), 151–158.

    Google Scholar 

  • Scharfe, H. (2002). Education in ancient India. Brill.

    Book  Google Scholar 

  • Tarlekar, G. H. (2001). The Puṣpasūtra: A Prātiśākhya of the Sāmaveda (Vol. 2). Indira Gandhi National Centre for the Arts and Motilal Banarsidass Publishers Pvt. Ltd.

    Google Scholar 

  • Weber, A. (1886). Verzeichniss der Sanskrit- und Prākrit-handschriften der Königlichen bibliothek zu Berlin (Vol. 2). A. W. Schade’s Buchdruckerei.

    Google Scholar 

  • Wezler, A. (2004). Dharma in the Veda and the dharmaśāstras. Journal of Indian Philosophy, 32(5/6), 629–654.

    Article  Google Scholar 

Download references

Acknowledgements

I would like to thank Professors Sheldon Pollock and John Hawley for their insightful comments on an earlier draft of this paper. I would also like to thank Professor Asko Parpola for kindly answering questions about one of the sources. Finally, I would like to thank the anonymous reviewer for a number of criticisms that fundamentally altered the shape of this paper.

Funding

No funding was received to assist with the preparation of this manuscript.

Author information

Authors and Affiliations

  1. Department of Religion, Columbia University, 80 Claremont Avenue, New York, NY, 10027, USA

    Guy St. Amant

Authors
  1. Guy St. Amant
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Guy St. Amant.

Ethics declarations

Conflict of interest

The author has no relevant financial or non-financial interests to disclose.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix: Nidānasūtra 2.1

Appendix: Nidānasūtra 2.1

NidS 2.1, p. 22 contains two additional instances of the word smṛti. Both appear in the course of an extended pūrvapakṣa:

ṛṣikṛtāḥ svid ūhā3 anṛṣikṛtā iti | anṛṣikṛtā iti vai khalv āhuḥ | apy ādyā viratā evohanāt | no evam ārṣeyo bhavati | athāpy ūha ity enam ācakṣate | kartavya iti vaitad bhavati | athāpi yājñikā vipratipadyante | yajñārthaḥ khalu punar ūho bhavati | athāpi vṛttir yonir iti gāyanti | kim u khalu smṛtau vrttyā kiṃ yonyā bhaviṣyatīti | athāpi nātyantaṃ gāyanti | tatra yad agītaṃ kartavyam ity evāpannaṃ bhavati | athāpy anyasaṃhitasya saṃhitām avekṣya gāyanti | tad etat smṛtau nopapadyata iti.

I cannot at present offer a convincing interpretation of this passage. What may be useful, however, is to compare the specific pūrvapakṣa (P) statements from the above passage that include the word smṛtau with the corresponding refutations provided by the siddhānta (S). The first set of corresponding arguments run as follows:

P: athāpi vṛttir yonir iti gāyanti | kim u khalu smṛtau vrttyā kiṃ yonyā bhaviṣyatīti.Footnote 81

S: yad v etad vṛttir yonir iti gāyantīti chāndasasya pārthikā vṛttir yoni kalpante.Footnote 82

In the Sāmavedic context, yoni generally refers to the underlying ṛk-verse to which a sāman is set. It can also mean, as Tarlekar (2001, p. 2:620) explains in his study of the Puṣpasūtra, “the first stotrīyā of the sāman, the chant of which is employed in the remaining two stotrīyās.” The meaning of vṛtti seems uncertain. The opening section of the NidS defines vṛtti as a sort of rhythmic cadence determined by the weight of a pāda’s penultimate syllable.Footnote 83 Does vṛtti really mean this here? Or does it mean something like “recitational mode?”Footnote 84 In any event, the pūrvapakṣa seems to make the point that modified sāmans are performed after having considered the vṛtti and the yoni.

The subsequent part of the argument seems to assume an unstated ūhasya that construes with smṛtau. The idea seems to be that if there were a smṛti (“memory?”) of an ūha, there would be no point in referring to or using the vṛtti and the yoni for the performance of that ūha. The siddhānta appears to argue that people think that the form found in the Prakṛtigāna (chāndasasya) also has a separate (pārthika?) vṛtti and yoni.Footnote 85 The point seems to be that making use of or reference to these two things in the course of performance does not vitiate the possibility of smṛti, whatever it might mean. The second set of corresponding arguments is perhaps more helpful:

P: athāpy anyasaṃhitasya saṃhitām avekṣya gāyanti | tad etat smṛtau nopapadyata iti.

S: yad v etad anyasaṃhitasya saṃhitām avekṣya gāyantīti chandasy api vṛttijño ’nyasaṃhitasya saṃhitām avekṣya gāyet na tāvatā chando ’smṛtaṃ manyeta.Footnote 86

Here too we are confronted with a number of unclear phrases, including anyasaṃhitasya saṃhitām, for which I can provide no satisfactory translation. The pūrvapakṣin’s point seems to be that taking account of particular type of collection (anyasaṃhitasya saṃhitām avekṣya) would be unreasonable if there were a smṛti (memory?) of an ūha. The siddhāntin argues from analogy that even those who know the vṛttis of the underlying verses (or is it meters?) take account of the same type of collection (anyasaṃhitasya saṃhitā) before singing. This much, however, does not mean that the chandas has been forgotten (asmṛta). A clearer picture of the meaning of smṛta is provided by a subsequent statement:

S: athāpi ya upavādād asmṛtaṃ manyetāpi nūnaṃ sa sarvām evāpi trayīvidyām asmṛtāṃ manyeta tāṃ hy apy eka upavadantīti.Footnote 87

Now, were someone to think that [ūha] is asmṛta because people disparage it, then, to be sure, he would have to think that the entire triple-knowledge [of the Veda] was asmṛta; for there are some who disparage even that.

This passage makes clear, I think, that smṛta means, in this particular context, “remembered [rightly]” or perhaps “transmitted [authoritatively],” especially as it is used with reference to the Vedic texts themselves.Footnote 88 Given the correspondence between smṛtau nopapadyate and na tāvatā chando ’smṛtam, I think it makes the most sense to conjecture that smṛti means “a memory” in the specific sense of “an authoritative transmission.” Alternatively, it might be conjectured that smṛtau should be read as smṛte, which could then be construed by supplying a locative ūhe: “were ūha transmitted authoritatively.”

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

St. Amant, G. “Memory” Revisited: What Sāmavedic Technical Literature Tells Us About Smṛti’s Early Meaning. J Indian Philos 49, 699–724 (2021). https://doi.org/10.1007/s10781-021-09481-2

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10781-021-09481-2

Keywords

Search

Navigation