The three main schools of Vedānta in South India—Advaita, Viśiṣṭādvaita, and Dvaita all center themselves around a similar canon of sacred texts, including the Vedas, Upaniṣads, Purāṇas, etc. Of such canonical scriptures, the Bhagavadgītā (BG) is regarded as authoritative in all three traditions and cited frequently in commentaries on other sacred texts. Śaṅkara and Rāmānuja have each authored a commentary on the BG, while Madhva has authored two.
Accepting the BG as authoritative poses to all these schools an important problem of reconciling the BG’s statements on the Vedas, since all three writers are committed to regarding the Vedas and BG as sacred texts that are consistent with each other. In the BG, Kṛṣṇa refers to the Vedas in ways that appear to undermine their authority, often by comparing them unfavorably with the BG itself. In this post, I look at BG 2.45-6, which speak negatively about the Vedas and see how Śaṅkara, Rāmānuja, and Madhva attempt to re-interpret the verses in conformity with their own doctrinal commitments. The verses read as follows:
The Vedas have that which is related to the three guṇas [sattva, rajas, and tamas] as their subject. O Arjuna! You be one who is without the three guṇas…
How much is the use of a well with/when there are waters flooded from all directions? That much is the use of all the Vedas for a brāhmaṇa who knows.[1]
The guṇas, which Kṛṣṇa explains in detail elsewhere in the BG, are the three forces or modes of material nature. There are multiple problems in the verses here, but I am interested in the conundrum for the Vedānta commentators—Kṛṣṇa clearly states that the Vedas deal (only) with the three guṇas, instructing Arjuna to rise above them. The Vedas, then, are distinguished from Kṛṣṇa’s own teachings in the BG and have only a limited applicability, which is problematic for all three commentators. The commentators use the metaphor in the verse to create ambiguity and defend the Vedas or some part of the Vedas from the charge of being superfluous.
Śaṅkara does this by narrowing the scope of the BG’s statement about the Vedas to the specific sections of the Vedas that enjoin ritual actions:
In the world, all the purposes that can be served by a well, etc., that has limited water, such as bathing and drinking, can also be served by water flooded in all directions, i.e., all the uses of a well are subsumed within the uses of a larger quantity of water. In the same way, all the purposes served by all the Vedas, i.e., by the actions enjoined in the Vedas, are subsumed within the purposes served by the knowledge of the ultimate principle in the case of the brāhmaṇa, i.e., celibate monk who possesses such knowledge… Therefore, even actions that are similar to the well, etc., must be performed by one who is eligible for such action, before one obtains the eligibility for knowledge.[2]
In other words, Śaṅkara reads the metaphor as saying that a well is useless when you have plenty of water. By limiting the statement to certain sections of the Vedas, Śaṅkara uses this verse to argue for the superiority of knowledge (jñāna) over ritual action (karma), but still manages to defend the Vedic statements that enjoin rituals as having a limited and preliminary value. While the metaphor tells us that the Veda has very limited use to a knowledgeable brāhmaṇa, Śaṅkara reads it as advocating for ritual until one has reached such a state of knowledge.
Rāmānuja reads the metaphor very differently. To Rāmānuja, the primary question that the verse answers is: “How can it be that the injunctions of the Vedas must be discarded?”[3] Rāmānuja is then making an assumption that his ideal reader shares—that the Vedas cannot be discarded, and consequently, the BG, which is an authoritative text, cannot instruct one to discard them. He proceeds to elaborate as follows: “Why would the Vedas, which are more affectionate than thousands of mothers and fathers, which are engaged always in the welfare of the self, instruct us to perform trivial actions which lead to being born again and again?” This is the question that Rāmānuja sees the verse of the Gītā as answering.
All that is enjoined by the Vedas is not to be undertaken by everyone, just as a thirsty person only obtains the water needed to drink from a well filled with flooded waters from all directions. Similar is the case of a brāhmaṇawho knows, with regard to all the Vedas. A vaidika who desires liberation must only undertake that [those actions laid down in the Vedas] which is the means for achieving liberation, and not others [actions which are not].[4]
As we can see, Rāmānuja understands the metaphor of the well and flooded waters in a manner that conforms to his own philosophical commitment to the primacy of the Vedas. The meaning of the metaphor, to Rāmānuja, is not that the flooded waters cause the well to lose its utility in any way. Rather, Kṛṣṇa is implying that one person cannot use all the water of the flooded water in the well but must use only a limited amount of water to fulfil their needs.
Madhva offers two different interpretations of the same two verses. Both these interpretations assume that these verses as communicating something other than their apparent meaning, since it is impossible that the Vedas be unauthoritative or even inapplicable. I will look at the Gītābhāṣya here, where Madhva draws elements from Śaṅkara and Rāmānuja, while disagreeing with both:
All the purposes served by the well with limited water are indeed subsumed within the purposes served by water flooded from all directions. Similarly, the results of gained by the actions performed out of desire (kāmyakarma) as enjoined by the Vedas are subsumed within the results achieved by a brāhmaṇa who is knowledgeable. Here, the word brāhmaṇa is to be understood [etymologically] as the one who reaches (aṇati) God (brahma) and therefore refers to one who has seen God directly (aparokṣajñānī).[5]
Madhva’s reading does not oppose the two paths of action and knowledge against each other as alternatives, since he believes that both must be performed in tandem in order to achieve liberation, arguing against Śaṅkara’s view that action can cease after acquiring knowledge. It would go against Madhva’s doctrine to argue that all ritual actions are unnecessary upon reaching a state of knowledge. Therefore, Madhva argues that the reference to the Vedas must be construed narrowly, not to refer to ritual actions within the Vedas, but to refer only to the optional ritual actions (kāmyakarmas) enjoined therein. Such optional ritual actions are performed when one has a desire to be fulfilled, and do not result in any penalty when not performed. Madhva does not clarify in what way the knowledgeable person obtains the results of such rituals without performing them. Jayatīrtha, Madhva’s fourteenth century commentator, understands this to mean that the results achieved by the one with knowledge reduces the results of such optional rituals to insignificance. This would be consistent with the general idea in all three schools of Vedānta that true seekers would not want to gain material results, since they would seek only liberation.
Another interesting point is that none of these three commentators understand the word ‘brāhmaṇa’ in the verse as referring to the jāti or varṇa. Śaṅkara glosses the word as ‘monk’ (sannyāsī), an even smaller subset of renunciates, Rāmānuja as ‘a vaidika who desires liberation’, and Madhva as ‘one who has perceived God directly’ (aparokṣajñānī), thereby allowing it to refer to different groups of people.
[1] traiguṇyaviṣayā vedā nistraiguṇyo bhavārjuna /
nirdvandvo nityasattvastho niryogakṣema ātmavān //
yāvān artha udapāne sarvataḥ samplutodake /
tāvān sarveṣu vedeṣu brāhmaṇasya vijānataḥ //
Bhagavadgītā 2.45-46.
[2] yathā loke kūpataḍāgādyanekasmin udapāne paricchinnodake yāvān yāvatparimāṇaḥ snānapānādir arthaḥ phalaṃ prayojanaṃ sa sarvo ‘rthaḥ sarvataḥ saṃplutodake ‘pi yo ‘rthaḥ tāvān eva sampadyate tatra antarbhavatīty arthaḥ / evaṃ tāvān tāvatparimāṇa eva sampadyate sarveṣu vedeṣu vedokteṣu karmasu yo ‘rthaḥ yat karma-phalaṃ so ‘rthaḥ brāhmaṇasya sannyāsinaḥ paramārth-tattvaṃ vijānataḥ yo ‘rthaḥ yat vijñānaphalaṃ sarvataḥ saṃplutodakasthānīyaṃ tasmin tāvān eva sampadyate tatraivāntarbhavatīty arthaḥ /… tasmāt prāk jñānaniṣṭhādhikāraprāpteḥ karmaṇy adhikṛtena kūpataḍāgādyarthasthānīyam api karma kartavyam /
Śaṅkara on BG 2.46
[3] evam atyalpaphalāni punarjanmaprasavāni karmāṇi mātāpitṛsahasrebhyo ‘pi vatsalataratayā ātmojjīvane pravṛttā vedāḥ kimarthaṃ vadanti kathaṃ vā vedoditaṃ tyājyatayocyate ity ata āha…
Rāmānuja’s prelude to commentary on BG 2.45
[4] na ca vedoditaṃ sarvaṃ sarvasyopādeyam yathā sarvārthaparikalpite sarvataḥ saṃplutodake udapāne pipāsor yāvān arthaḥ yāvad eva prayojanam tāvad eva tenopādīyate na sarvam evaṃ sarveṣu ca vedeṣu brāhmaṇasya vijānataḥ vaidikasya mumukṣoḥ yad eva mokṣasādhanaṃ tad evopādeyam nānyat /
Rāmānuja on BG 2.46
[5] tathāpi kāmyakarmiṇāṃ phalaṃ jñānināṃ na bhavatiti sāmyam evety ata āha yāvān artha iti / yathā yāvān arthaḥ prayojanam udāpāne kūpe bhavati tāvān sarvataḥ samplutodake ‘ntarbhavaty eva / evaṃ sarveṣu vedeṣu yatphalaṃ tadvijānato ‘pi jñānino brāhmaṇasya phale ‘ntarbhavati / brahma aṇatīti brāhmaṇo ‘parokṣajñānī / sa hi brahma gacchati / vijānata iti jñānaphalatvaṃ tasya darśayati /
Madhva’s Gītābhāṣya on BG 2.46
For Shankara, the authority of the Vedas are only in the ream of duality and ignorance. They are valid means operating in a phenomenal world of mithya. Shankara is very clear that once the Vedas complete their aim of giving Self Knowledge to an individual, their authority and validity is relegated to the realm of ignorance and duality, because in Self there is no distinction between the knower-knowledge-known.
So we have from Shankara the following words of Brahmasutra Bhashya:
With this teaching of the Veda in mind, the revered Commentator raises the question, ‘But in what sense do we mean that perception and the other secular means of knowledge, together with Vedic tradition, belong to those in the realm of Ignorance?’ And he replies: ‘What we say here is this. Without seif-identification with the body and senses expressed in the feelings “I” and “mine” there can be no empirical knower, and so the processes of empirical knowledge cannot begin’ (B.S.Bh. I.i.1, intro.).
And the quote from Bhagavad Gita (2.46), you have taken up for discussion refers exactly to this
Thank you for this post, Anusha. I found it really interesting. I thought the latter part of chapter 2 was one of the parts of the Gītā I was most familiar with (in part since Gandhi liked it so much) and yet I had somehow managed to miss the fact that it seems to make a somewhat revolutionary denial of the authority of the Veda. The fact that that denial is in metaphor is probably why I missed it on previous readings – as well as being what enables the commentators to say it’s not really a denial!
I looked up what the Marathi bhakti philosopher Dnyaneshwar (Jñāneśvara) has to say about this verse, and I think he takes Rāmānuja’s interpretation a bit more strongly: one only needs a part of the Vedas, and specifically the part that has to do with “the eternal”. Which, I suspect, might turn out to be in order to say you don’t need the parts that have to do with caste distinctions.
Hi Amod,
Thank you so much for the comments. I completely agree that the metaphors in the Gītā allow for a lot of variation in interpretation. Even in verses like Gītā 15.7, when there is no metaphor in the text, commentators assume some non-literal/metaphorical meaning of some kind just to allow themselves such leeway.
Thanks for pointing me to Dnyaneshwar’s reading! The classifications of which portions of the Vedas are to be accepted/rejected, or taken literally/metaphorically within different traditions are very interesting.
Concerning the overfull well of Madhva: I think that the background assumption could be that kāmyakarman give you paricchinna sukha ‘limited happiness’, whereas bhakti gives you niratiśaya ānanda ‘joy beyond limit’, so that trying to reach the former would be silly, given that it the latter is just the enhanced version of the former. (I found this formulation explicitly only in a Dharmaśāstra, though). What do you think?
Hi Elisa, that makes sense, and I think that is exactly what Jayatīrtha understands Madhva to be saying. He says that the results of the jñānī are great (mahat) like an overflowing ocean, and the results of kāmyakarma are small (alpa) like a well, etc., and therefore the two are not equal, and seekers should strive for the greater one. I was just wondering whether Madhva could mean something further, such as that the aparokṣa jñānī actually obtains the results of optional rituals without performing them. I have not found anything to indicate that so far.
I would be very interested to hear where this idea appears in dharmaśāstra. Thanks!
I have a comment on it, it might look out of context but how i understand this were (i am no scholar or sanskritiyan to do so, kam just sharing my perspective) If we use the common sense for vedas, the Vedas contains the knowledge whichis apaurusheya, rishis in their meditative state heard the word of knowledge, which is in our vedas (i am taking my logic giving the testimony to Vivekananda’s commentary on the case of divine word). a person (brahamana of the verse) who have recieves the ulltimate knowledge would also be able to hear the word of the same vedas just like the ancient rishis (afterall this knowledge is eternal and all pervasive in divine form) , then the written form of vedas or hearing vedic words from someone wouldn’t make sense to the learnt brahmana. Otherwise it would be like buying a new book which i already have.