Vātsyāyana’s Guide to Liberation

In this essay, my aim is to explain Vātsyāyana’s solution to a problem that arises for his theory of liberation. For him and most Nyāya philosophers after him, liberation consists in the absolute cessation of pain (ātyantika-duḥkha-vimukti). Since this requires freedom from embodied existence, it also results in the absolute cessation of pleasure. How, then, can agents like us (who habitually seek pleasure) be rationally motivated to seek liberation? Vātsyāyana’s solution depends on what I will call the Pain Principle, i.e., the principle that we should treat all aspects of our embodied existence as pain. If we were to follow this advice, we would come to apply the label of pain (duḥkha-saṃjñā) to all aspects of our embodied existence, including pleasure. This would undermine our attachment to our own embodied existence. I show that this fits with Vātsyāyana’s general theory of motivation. According to this theory, by manipulating the labels (saṃjñā) using which we think about the world and ourselves, we can induce radical shifts in our patterns of motivation.

of these tools is necessary for inquiring into liberation and the means to achieve it. On the other hand, Nyāya supplies us with an awareness of the truth about the knowable entities (prameya), e.g, the self (ātman), the body (śarīra), etc. Discovering that truth liberates us. 1 Most Nyāya philosophers offer a purely negative characterisation of liberation. Liberation, for them, is the absolute cessation of pain (ātyantika-duḥkha-vimukti). That involves freedom from repeated birth (janman) and, therefore, all aspects of embodied existence. 2 This raises a problem. Since pleasure can only arise within a body, a liberated being couldn't experience pleasure. However, ordinary agents like us are motivated by (at least) two kinds of desires: the desire to obtain pleasure and things that give rise to pleasure, and the desire to avoid pain and things that bring about pain. But then it's not obvious whether we can, or should, be motivated to seek liberation at all.
In this essay, I will look at a solution to this problem, given by Vātsyāyana in his commentary Nyāya-bhāṣya (NB) on the Nyāya-sūtra (NS). Vātsyāyana's solution depends on what I shall call the Pain Principle, i.e., the principle that we should treat all aspects of embodied existence as pain. That explains why we should act solely from the motive of avoiding pain, and therefore should seek liberation in order to free ourselves from all pain involved in embodied existence. On Vātsyāyana's proposal, we should treat all aspects of our embodied existence as 1 Vātsyāyana says in his commentary Nyāya-bhāṣya (NB) on Nyāya-sūtra (NS) 1.1.1 (NB 3.11-3): "Moreover, what is reasoning (nyāya)? Reasoning is the examination of an object with the help of the means of knowing; it is inference that is based on perception and testimony. It is critical inquiry. The investigation of an object that has been apprehended by perception and scripture (āgama) is critical inquiry. That which proceeds by means of that is the science of critical inquiry, i.e., the science of Nyāya or the discipline of Nyāya'' (kaḥ punar ayaṃ nyāyaḥ? pramāṇair artha-parīkṣaṇaṃ nyāyaḥ| pratyakṣāgamāśritam anumānam| sānvīkṣā| pratyakṣāgamābhyām īkṣitasyānvīkṣaṇam anvīkṣā| tayā pravartata ity ānvīkṣikī nyāya-vidyā nyāya-śāstram|). He later says : "This very awareness of the truth and the attainment of the highest good are to be known in accordance with each science. However, here, in the science of the self, the awareness of the self and so on is the awareness of the truth, and the attainment of the highest good is liberation (apavarga)" (tad idaṃ tattva-jñānaṃ niḥśreyasādhigamaś ca yathā-vidyaṃ veditavyam| iha tv adhyātma-vidyāyām ātmādi-jñānaṃ tattva-jñānam| niḥśreyasādhigamo'pavarga-prāptiḥ|). In his sub-commentary Nyāya-vārttika (NV), Uddyotakara notes that the role that Nyāya plays with respect to highest goods that are unobserved (i.e., liberation) is different from the role it plays with respect to observed highest goods, e.g., a good harvest and the conquest of the world (which are promoted by economics and the science of politics respectively). In the latter case, it is useful insofar as it provides an awareness of the truth about epistemological tools, e. g., the means of knowing and so on. In the former case, it is useful insofar as it provides an awareness of the truth about the knowable entities . For a slightly different understanding of the purpose of Nyāya, see Jayanta's Nyāya-mañjarī (NM1 7.7-14) and Kataoka's (2006) discussion of this issue. 2 The locus classicus for this conception of liberation is NS 1.1.24 and Vātsyāyana's commentary on it (both of which we shall discuss soon). Amongst early Naiyāyikas, Uddyotakara, Vācaspati Miśra (9th century CE), Jayanta Bhaṫṫa (9th century CE) and Udayana (10th/11th century CE) expand on this very conception; see Uddyotakara's Nyāya-vārttika (NV) 81. ), Jayanta's Nyāya-mañjarī , and Udayana's Ātma-tattva-viveka ). For discussion of this theory, see Ram-Prasad (2001, pp. 57-101). A notable exception is Bhāsarvajña (9th century CE), who defends the thesis that liberation involves an awareness of pleasure; see his Nyāya-bhūṣaṇa (NBhū 594-8). Among later Naiyāyikas, Gaṅgeśa Upādhyāya (14th century CE), Harirāma Tarkavāgiśa (17th century CE), and Gadādhara Bhaṫṫacārya (17th century CE) revised this earlier conception of liberation; see Gaṅgeśa's Tattva-cintāmaṇi (TCM 397-442), Harirāma's Muktivāda-vicāra (MVV), and Gadādhara's Nava-mukti-vāda (NVM). pain, not because they are literally pain, but rather because they are inextricably connected to pain. This expansive application of the label of pain (duḥkha-saṃjñā) is supposed to undermine our attachment to embodied existence. I will explain how this idea connects up with Vātsyāyana's thesis that thinking about the world under certain labels (saṃjñā) or concepts-some of which may inaccurately represent the world-plays an important role in transforming us into agents who are capable of seeking liberation. 3 Here is the plan for this essay. I will begin by laying out Vātsyāyana's negative conception of liberation. Then, I will say why he rejects an alternative positive view according to which liberation involves the manifestation of permanent pleasure. This raises a problem for Vātsyāyana, namely that ordinary agents like us cannot be rationally motivated to seek liberation. After posing the problem, I will sketch Vātsyāyana's response to this problem, and show how it, as it stands, is inadequate. I will then argue that the response can succeed if we take Vātsyāyana to be relying on the Pain Principle. Finally, I will explain how this principle fits with Vātsyāyana's more general theory that cultivating different labels (saṃjñā) or concepts with respect to ourselves and the world can play an important role in transforming our patterns of motivation.

What Is Liberation?
that it is the unconscious body that persists even when one is in a state of deep sleep. In each case, Indra notices a problem. First, the consciousness that remains active in dreams can never be completely free from suffering. So, once again, there would be nothing satisfying about discovering that that is the self. Second, the unconscious person is incapable of having conscious thoughts and experiences about herself or other things, and thus lacks one of the essential characteristics of the self. At last, Prajāpati reveals the truth about the self to Indra: "This body, Maghavan, is mortal; it is in the grip of death. So, it is the abode of this immortal and nonbodily self. One who has a body is in the grip of joy and sorrow, and there is no freedom from joy and sorrow for one who has a body. Joy and sorrow, however, do not affect one who has no body." 5 Thus, in this Upaniṡadic picture, the connection between the self and hedonic states like pleasure and pain is merely contingent: when the self inhabits the body, it is subject to pleasure and pain, but, when it is disembodied, it is affected by neither. For Nyāya philosophers like Vātsyāyana, liberation is simply this disembodied state of the self where it is invulnerable to pain.
To understand the Nyāya account of liberation, let's start with a story that is laid out in NS as well as in Vātsyāyana's commentary on it. According to NS 1.1.17, a practical undertaking (pravṛtti) is any effort (ārambha) that gives rise to a mental, linguistic or physical action. Such practical undertakings are produced by defects (doṣa): namely, attachment (rāga), aversion (dveṣa) and delusion (moha). These motivate us towards both vicious (pāpa) and virtuous (puṇya) practical undertakings (NB 20.3-7 on NS 1.1.18). 6 In his commentary on NS 1.1.20, Vātsyāyana says that 5 Chāndogya Upaniṣad VIII.12.1 in Olivelle 1998, pp. 284-7: maghavan martyaṃ vā idaṃ śarīram āttaṃ mṛtyunā | tad asyāmṛtasyāśarīrasyātmano 'dhiṣṭhānam |ātto vai saśarīraḥ priyāpriyābhyām| na vai saśarīrasya sataḥ priyāpriyayor apahatir asti| 6 The claim that these defects give rise to both virtuous and vicious practical undertakings might seem surprising. However, there is some textual support for this claim in Vātsyāyana's own work. First, in his commentary on NS 1.1.2, he says : "Due to this false awareness, there is an attachment towards agreeable things, and an aversion towards disagreeable things. Moreover, due to the power of the attachment and the aversion, there are defects (doṣa) such as untruthfulness, envy, deceitfulness and greed. Motivated by the defects, a person who is undertaking an action by means of the body practises injury, theft and forbidden sex. By means of speech, [he practises making] false, harsh, slanderous, and incoherent [utterances]. By means of the manas [i.e., the inner sense], [he practises] malice towards others, craving for others' possessions, and nihilism (nāstikya). These very vicious practical undertakings give rise to demerit (adharma). As for wholesome (śubhā) [practical undertakings], by means of the body, [he practises] giving, rescuing, and serving; by means of speech, [he practises making] beneficial, and pleasing utterances as well as the study of the Veda (svādhyāya); by means of the manas, [he practises] kindness, absence of longing, and faith by means of the manas. These very [wholesome practical undertakings] give rise to merit" (etasmān mithyājñānād anukūleṣu rāgaḥ, pratikūleṣu dveṣaḥ| rāgadveṣādhikārāc cāsatyerṣyamāyālobhādayo doṣā bhavanti| doṣaiḥ prayuktaḥ śarīreṇa pravarttamāno hiṃsāsteyapratiṣiddhamaithunāny ācarati| vācā anṛtaparuṣasūcanāsambaddhāni| manasā paradrohaṃ paradravyābhīpsāṃ nāstikyaṃ ceti| seyaṃ pāpātmikā pravṛttir adharmāya| atha śubhā, śarīreṇa dānaṃ paritrāṇaṃ paricaraṇaṃ ca| vācā satyaṃ hitaṃ priyaṃ svādhyāyaṃ ceti| manasā dayām aspṛhāṃ śraddhāṃ ceti| seyaṃ dharmāya|). In his commentary on NS 1.1.18-"The defects have the defining characteristic of being an inciter of action (pravartanā)" (pravartanālakṣaṇā doṣāḥ)-Vātsyāyana himself later says: "Being an inciter of action is the property of being the cause of a practical undertaking. For attachment and so on cause a thinker (jñātṛ) to undertake virtuous or vicious [actions]. Where there is false awareness, there is attachment and aversion. [The opponent:] Why are these defects, which are indeed experienced within oneself, specified by appealing to a characteristic? [Reply:] Certainly, those who are attached, averse and deluded have the defining characteristic of action (karman). For an attached person performs that action by means of which he obtains pleasure or pain. The same is our experiences (saṃvedana) of pleasure and pain are the results of our past practical undertakings and the defects that underlie them . This idea is based on a background belief in the karmic law, the principle that our practical undertakings, depending on whether they are virtuous or vicious, give rise to an appropriate result, i.e., an experience of pain or pleasure. When an agent acts virtuously, her practical undertaking produces in her a dispositional property, i.e., merit (dharma), which gives rise to a future experience of pleasure. When an agent acts viciously, her practical undertaking produces in her a different dispositional property, i.e., demerit (adharma), which gives rise to a future experience of pain. Since merit and demerit are the immediate effects of practical undertakings, they count as practical undertakings in a derivative sense (NB 7.6-7 on NS 1.1.2). Since our embodied existence is just a sequence of one practical undertaking after another, we cannot possibly experience all the pleasures and pains that we deserve to experience in one life. Thus, in order to obtain these results, we must be born with the body, the senses, and so on. In this way, merit and demerit bind us in the cycle of rebirth (saṃsāra) where we acquire a body over and over again .
Liberation consists in an irreversible escape from this cycle of rebirth. According to NS 1.1.21, "pain is that which has the defining characteristic of distress (bādhanā)" (bādhanālakṣaṇaṃ duḥkham). In his commentary, Vātsyāyana says: Distress is synonymous with affliction (pīḍā) and sorrow (tāpa). Anything, which is present inextricably intertwined (anuviddha) and connected (anuṣakta) with it, is pain in virtue of being related to pain. This being, who perceives that everything is intertwined by pain and desires to avoid pain, becomes dejected (nirvidyate) with respect to this birth (janman). Having become dejected (nirviṇṇa), he becomes detached. Having become detached, he becomes free. 7 The argument of this passage depends on the premise that, when a person sees that this birth is inseparably connected to (an overwhelming amount of) pain, she will come to see all aspects of it as pain. But what is birth? In his commentary on NS 4.1.55, Vātsyāyana explains: "Since 'birth' means that which is born [rather than Footnote 6 continued true of someone who is averse and of someone who is deluded. When [merely the expression] 'attachment, aversion and delusion' is uttered, not a lot is said" (pravartanā pravṛttihetutvam, jñātāraṃ hi rāgādayaḥ pravartayanti puṇye pāpe vā| yatra mithyājñānaṃ tatra rāgadveṣāv iti| pratyātmavedanīyā hīme doṣāḥ kasmāl lakṣaṇato nirdiśyanta iti? karmalakṣaṇāḥ khalu raktadviṣṭamūḍhāḥ, rakto hi tat karma kurute yena karmaṇā sukhaṃ duḥkhaṃ vā labhate, tathā dviṣṭas tathā mūḍha iti| rāgadveṣamohā ity ucyamāne bahu noktaṃ bhavatīti|). This second passage suggests that vicious and virtuous practical undertakings arise from the defects. 7 NB 21.11-3: bādhanā pīḍā tāpa iti| tayānuviddham anuṣaktam avinirbhāgena vartamānaṃ duḥkhayogād duḥkham iti| so 'yaṃ sarvaṃ duḥkhenānuviddham iti paśyan duḥkhaṃ jihāsur janmani duḥkhadarśī nirvidyate| nirviṇṇo virajyate| virakto vimucyate| to the event of being born], it stands for the body, the senses, and awareness-events (buddhi)" (janma jāyate iti śarīrendriyabuddhayaḥ). 8 So, birth encompasses both physical and psychological aspects of our embodied existence. When an agent sees this birth as pain, she will lose all attachment towards these aspects of embodied existence, and will desire to be free from it. In NS 1.1.22, Gautama says, "Liberation is absolute freedom from that" (tad-atyanta-vimokṣo'pavargaḥ). Vātsyāyana expands on this.
The absolute freedom from that pain which is birth is liberation. Why? Because it involves relinquishing the birth that has been appropriated, and not appropriating another such birth. Those who are learned in matters of liberation know this limitless state to be liberation. 9 Thus, if liberation is just absolute freedom from birth, it is a permanent disembodied state of the self. 10 How can we attain this disembodied state? NS 1.1.2 sketches a story: "When suffering, embodied existence, practical undertakings [or, alternatively, immediate effects of practical undertakings, namely merit and demerit], defect, and false awareness cease one after another, due to the cessation of what immediately follows from them, there is liberation" (duḥkha-janma-pravṛtti-doṣa-mithyā-jñānānām uttarottarāpāye tad-anantarāpāyād apavargaḥ). Vātsyāyana unpacks the story as follows. Our false awareness about the world and ourselves gives rise to the three defects, on the basis of which we engage in virtuous or vicious practical undertakings. These practical undertakings, by producing merit and demerit, come to cause future births. Thus, false awareness indirectly gives rise to the cycle of rebirth. Therefore, by eliminating such false awareness, we can eliminate each link in this causal chain, thereby completely eliminating pain (NB 6.9-8.3). 8 My translation here follows Angot's (2009, p. 677) French translation and Tarkavāgīśa's (1988, p. 318) Bengali translation. Moreover, I am translating the terms "buddhi" and "jñāna" as "awareness" or "awareness-event." Standardly, these are translated as "cognition," but that is slightly misleading, since, in contemporary philosophy and cognitive science, cognition is often distinguished from perception. But perceptual experiences can count as buddhi or jñāna on the Indian view. 9 NB 22.1-2: tena duḥkhena janmanā atyantaṁ vimuktir apavargaḣ | katham? upāttasya janmano hānam, anyasya cānupādānam| etām avasthām aparyantam apavargaṃ vedayante 'pavargavidaḥ| 10 A similar account of liberation is put forward in Vaiśeṣika Sūtra 5.2.20 (VS 43.11): "Liberation is that which consists in the absence of conjunction (saṃyoga) and the absence of appearance (prādurbhāva) when that is absent" (tadabhāve saṃyogābhāvo 'prādurbhāvaḥ sa mokṣaḥ). The commentator, Candrānanda, explains the idea as follows : "Liberation is that which consists in (i) the absence of a conjunction between the self and the body, called "life", and (ii) the absence of the appearance of another body, when an unobserved karmic factor of this nature, i.e., the cause for the beginningless egress (apasarpaṇa) [of the manas out of the body at the time of death], etc. is absent." (evaṃ-rūpasyānādy-apasarpaṇādi-nimittasyādṛṣṭasyābhāve jīvanākhyasyātma-manaḥ-saṁ yogasyābhāvo'nyasya ca śarīrasyāprādurbhāvo yaḥ sa mokṡaḣ |). I am taking "apasarpaṇa" as referring to the egress of the manas out of the body at the time of death following Candrānanda's commentary on Vaiśeṣika Sūtra 5.2.19 (VS 43.7). The "ādi" (here, translated as "etc.") probably refers to upasarpaṇa, i.e., the ingress of the manas into a new body at the time of birth. For some discussion of whether this is the right interpretation of these terms as their occur in Vaiśeṣika Sūtra 5.2.19, see Honda (1992, pp. 296-7).

The Permanent Pleasure View
Vātsyāyana contrasts his conception of liberation with another view. According to this latter view, the state of liberation involves the manifestation (abhivyakti) of permanent pleasure (nitya-sukha). 11 Call this the Permanent Pleasure View. But Vātsyāyana thinks that there is no good evidence that this view is true. Some think: "In liberation, the permanent pleasure of the self is manifested (abhivyajyate) just like its extension (mahattva). In virtue of that manifested pleasure, the absolutely free being is pleased." They face a problem (anupapatti) due to the absence of any means of knowing. There exists neither perception nor inference nor any scriptural statement (āgama), which could show that, in liberation, just like the extension (mahattva) of the self, the permanent pleasure of the self is manifested. 12 Why? The argument is this. Suppose there is a manifestation of permanent pleasure in the state of liberation. Presumably, this is an experience (saṃvedana) or awareness of that pleasure. Either that awareness is non-permanent, or permanent. But it can't be either. So, the Permanent Pleasure View is false. The crucial premise of this argument is that the awareness of permanent pleasure can be neither permanent nor impermanent. Let's motivate that premise in light of what Vātsyāyana says.
If that awareness is non-permanent insofar as it has a beginning, the defender of the Permanent Pleasure View would have to say what causes it. 13 To avoid this line of questioning, she might argue that the awareness of permanent pleasure is also permanent i.e., without a beginning or an end. That is problematic, since it implies that the pleasure should also be always manifested in an ordinary state of embodied existence. This has two implausible consequences. 14 First of all, this would make it 11 The source of this view is likely Upaniṡadic statements that equate the Brahman with bliss. For example, in Taittirīya Upaniṣad 2.7.1, we find the statement, "[I]t is the essence, for only when one has grasped that essence does one attain bliss" (raso vai saḥ | raso hy evāyaṃ labdhvā'nandībhavati, Olivelle 1998, 205). Later Naiyāyikas seem clueless about the source of this view. Following Udayana (KA 6.3), Gadādhara ascribes this view to the Bhāṫṫa Mīmāṁ sakas (NVM 100.12). This seems wrong, since, in verse 105 of the chapter called "Sambandhākṣepa-parihāra" of Ślokavārttika, Kumārila says (Ś V 475.13-4): "Moreover, if liberation is postulated to have the nature of enjoyment of pleasure, then this would simply be heaven. And that is gradually destroyed" (sukhopabhoga-rūpaś ca yadi mokṣaḥ prakalpyate|svarga eva bhaved eṣa paryāyeṇa kṣayī ca saḥ||). In his commentary, Pārthasārathi takes this to be a Sāṁ khya view (which also seems wrong). 12 NB 22.4-7: nityaṃ sukham ātmano mahattvavan mokṣe'bhivyajyate, yenābhivyaktenātyantaṃ vimuktaḥ sukhī bhavatīti kecin manyante| teṣāṃ pramāṇābhāvād anupapattiḥ| na pratyakṣaṃ nānumānaṃ nāgamo vā vidyate nityaṃ sukham ātmano mahattvavan mokṣe 'bhivyajyata iti| 13 NB 22.8-9: "The manifestation of something permanent is an experience. Its cause is to be stated. [To explain:] Since the manifestation of permanent pleasure is an experience or an awareness-event, its cause, i.e., that from which it is produced, is to be stated" (nityasyābhivyaktiḥ saṃvedanam, tasya hetuvacanam| nityasyābhivyaktiḥ saṃvedanam jñānam iti tasya hetur vācyo yatas tad utpadyata iti|) 14 NB 22.10-15: "If it is permanent like the pleasure (sukhavat), then there would be no distinction between someone who is in the cycle of rebirth and someone who is liberated. Just as the liberated being possesses (upapanna) pleasure and its experience which are [both] permanent, someone who is in the cycle of rebirth will also end up being like this. For both [the pleasure and its experience] are permanent. And if this were admitted, then these would be apprehended as accompanied by and simultaneous with difficult to explain the distinction between someone who has achieved liberation and someone who hasn't. After all, both would experience permanent pleasure! If the distinguishing characteristic of the state of liberation is the manifestation of permanent pleasure, the presence of that defining characteristic in an ordinary embodied state implies that there is no difference between the two states. Second, even if an embodied being could experience such permanent pleasure, it would experience such pleasure at the same time as and together with other pleasures and pains that arise as a result of the merit and demerit left by her previous practical undertakings. But such simultaneity and co-presence aren't apprehended in ordinary experience. 15 The defender of the Permanent Pleasure might attempt to avoid these consequences by arguing that the awareness of permanent pleasure is impermanent, i.e., has a beginning. If it has a beginning, then it must be produced by something. For Vātsyāyana, our experiences of ordinary pleasures are produced by an internal monitoring mechanism-the manas-which serves as a faculty of introspective attention and makes us aware of our own hedonic states when it is conjoined to the self (NB 11.15-21 on NS 1.1.4). When the self possesses pleasure, the conjunction (saṃyoga) between the manas and the self (ātman) makes us aware of the pleasure. Suppose, for the sake of argument, that it's the conjunction (saṃyoga) between the manas and the self (ātman) that also produces the awareness of permanent pleasure. But, since the self is all-pervading, the manas is always conjoined to the self. So, there must be some enabling condition, or auxiliary cause (sahakāri-nimitta), which explains why this conjunction sometimes produces the awareness of permanent Footnote 14 continued the results of merit and demerit [i.e., ordinary pleasures and pains]. [To explain:] The pleasures or pains, which are results of merit or demerit and are experienced in a sequence in the places of production (utpatti-sthāna) [i.e., in beings that are born with a body, the senses, and awareness-events], would be apprehended as accompanied by and simultaneous with the the permanent experience [of pleasure]. There is no absence of pleasure or an absence of manifestation. For both are permanent" (sukhavan nityam iti cet? saṃsārasthasya muktenāviśeṣaḥ| yathā muktaḥ sukhena tat-saṃvedanena ca san nityenopapannaḥ, tathā saṃsārastho'pi prasajyata iti, ubhayasya nityatvāt| abhyanujñāne ca dharmādharma-phalena sāhacaryaṃ yaugapadyaṃ gṛhyeta| yad idam utpatti-sthāneṣu dharmādharma-phalaṃ sukhaṃ duḥkhaṃ vā saṃvedyate paryāyeṇa, tasya ca nitya-saṃvedanasya ca sahabhāvo yaugapadyaṃ gṛhyeta| na sukhābhāvo nānabhivyaktir asti, ubhayasya nityatvāt|) 15 In his Nyāya-vārttika (NV) Uddyotakara brings out the implausible consequences of this : "There would be no sequence in which pleasure and pain are experienced. Since this being would apprehend permanent pleasure, any effort for the sake of liberation would be futile. Moreover, this being doesn't avoid pleasure, since it is impossible to avoid it in separation (viveka) [from permanent pleasure]. For anyone who seeks to avoid pain also ends up avoiding pleasure. Furthermore, given that this being doesn't ever apprehend pain, for the sake of avoiding what does he undertake actions?" (sukha-duḥkhasaṃvedana-paryāyaś ca na syāt, nityam ayaṃ sukham upalabheta tataś ca mokṣārthaḥ prayāso vyarthaḥ syāt| na cāyaṃ khalu sukhaṃ jihāsati, viveka-hānasyāśakyatvāt| duḥkhaṃ jihāsamānaḥ sukham api jihāsati| na cāyaṃ kadācit duḥkham upalabhata iti kasya hānārthaṃ pravartate|) The worry is that, since ordinary pains and pleasures, which are aspects of embodied existence, are experienced at the same time as permanent pleasure, it's not possible to distinguish the ordinary pleasures from the permanent pleasure. So, it's impossible to avoid the aspects of embodied existence that cause pain without giving up on permanent pleasure. And, if the response is that this person never experiences any pain at all, then there would be no need to seek liberation at all. pleasure and not always. 16 More generally, the question is this: if the self does possess permanent pleasure, why is one only aware of it sometimes but not always? As Uddyotakara goes on to note, it's not easy to escape this line of questioning. For example, the opponent might argue that the mere presence of permanent pleasure itself is sufficient for the conjunction between the self and the manas to produce an awareness of pleasure. But if the awareness of pleasure could arise from this conjunction merely due to the presence of its intentional object, then the contact between the self and the manas could produce the awareness of intentional objects such as colour, etc. merely by depending on those intentional objects themselves. If that were to happen, we would always be flooded with various awareness-events about different intentional objects. Moreover, the agent would always undergo the awareness of permanent pleasure. But, as we've seen, that too doesn't happen. 17 Therefore, the opponent must explain why the conjunction between the manas and the self only sometimes produces the awareness of permanent pleasure.
When it comes to our ordinary pleasures, our awareness of such pleasures arises from the conjunction between the self and the manas due to the presence of previously accumulated merit. If the opponent now appeals to something like merit to explain how the awareness of permanent pleasure arises, she will face a different challenge. Where does that merit come from? Perhaps, the opponent could say that even a liberated self may have some merit left over from yogic meditative absorption (yoga-samādhi), and that allows us to become aware of this permanent [16][17][18][19]: "If it is impermanent, then a cause is to be stated. If the experience of permanent pleasure that arises in the state of liberation is impermanent, the cause-from which it is produced-is to be stated. The cause of the experience of ordinary pleasure is the conjunction between the self and the manas along with some other cause. Suppose you say, 'The conjunction between the self and the manas is the cause of the experience of permanent pleasure.' Even if this is right, the other auxiliary cause of that experience is to be stated"(anityatve hetuvacanam| atha mokṣe nityasya sukhasya saṃvedanam anityam? yata utpadyate sa hetur vācyaḥ|ātma-manaḥ-saṃyogasya nimittāntara-sahitasya hetutvam| ātmamanaḥsaṃyogo hetur iti cet? evam api tasya sahakāri nimittāntaraṃ vacanīyam iti|). As Uddyotakara explains the matter, this has to do with the nature of conjunction (saṃyoga) as a produced quality of substances (dravya) (NV 81.11-3): "If you think that the conjunction between the self and the manas is the cause of the awareness, then the cause that it depends on (apekṣā-kāraṇa) is to be stated. For, amongst substances, qualities, and movements that are produced (ārabdha), conjunction isn't an independent cause" (yadi manyase ātma-manaḥ-saṃyogo jñānasya kāraṇam, tasya tarhy apekṣā-kāraṇam vaktavyam| na hi dravya-guṇa-karmasv ārabdhavyeṣu saṃyogo nirapekṣam kāraṇam|) 17 NV 81.14-19: "If you think that the conjunction between the self and the manas is the cause of the awareness [of pleasure] insofar as it depends on the pleasure that is permanently established in the self, then that isn't reasonable. For it conflicts with the state of isolation (kaivalya). Just as [on your view] this conjunction between the self and the manas produces the awareness of pleasure depending on only the intentional object [i.e., pleasure] and without depending on any other cause, so also it would produce awareness-events about intentional objects like colour and so on merely by depending on them. And, if that is the cause, then the state of isolation will cease, since this self apprehends all objects. Moreover, this is also unreasonable because it leads to the undesirable consequence that there would be permanent apprehension" (atha manyase ātma-manaḥ-saṃyogaḥ nityam ātmani vyavasthitam sukham apekṣamāṇo jñāna-kāraṇaṃ bhavatīti, tan na yuktam, kaivalya-virodhāt| yathā'yam ātma-manaḥ-saṃyogo viṣayamātram apekṣamāṇaḥ anya-nimitta-nirapekṣaḥ sukha-jñānaṃ karoti, evaṃ rūpādīn api viṣayān apekṣamāṇaḥ tad-viṣayāṇi jñānāni kuryāt| tataś ca kaivalyaṃ nivartate, sarvān arthān ayam ātmā upalabhate iti| nityopalabdhi-prasaṅgāc ca|).
pleasure. 18 But, since practical undertakings cease completely in the state of liberation, the accumulated merit will run out at some point. At that stage, there will be no awareness of permanent pleasure. If that happens, there will be no difference between that state and the state where there is no permanent pleasure. 19 Moreover, if the defender of the Permanent Pleasure View insists that the merit doesn't run out, she needs to explain why that is so. We have strong evidence for the generalization that things that are produced are destroyed; on the basis of that evidence, we can infer that the merit, produced by yogic meditative absorption (yoga-samādhi), should also be destroyed. In contrast, there is no inference to the contrary. One cannot get out of this problem by claiming that the merit itself is permanent, i.e., something that isn't produced or destroyed. For that implies that the awareness of permanent pleasure is also permanent. Thus, the previously mentioned objection will apply again: namely, that the permanent pleasure will be experienced together with ordinary pleasures and pains. 20 In response, the defender of the Permanent Pleasure View might try to say that the relation with the body somehow serves as an impediment to the permanent pleasure being experienced at the same time as other pleasures and pains. This view would be extremely bizarre. In the Nyāya picture, the whole purpose of the body is to give rise to experiences of pleasure and pain. So, it's quite strange to say that the body prevents the experience of pleasure from arising, and, yet, a disembodied 18 This fits with the idea that the awareness of the truth that leads to liberation is produced by yogic meditative absorption. See NS 4.2.38. 19 NB 22.20-23.5: "The cause of merit is to be stated. If merit is the other cause, its cause-from which it arises-is to be stated. Since the [merit] which is produced by yogic meditative absorption conflicts with the cessation of actions, when it is destroyed, the experience will cease.[To explain:] if the merit produced by yogic meditation is the cause, then, due to its conflict with the cessation of actions, the experience would cease if it were to be destroyed. And if there is no experience, there is no distinction between this state and one where the experience is absent. If, due to the destruction of merit, there is a cessation of experience, then the permanent pleasure isn't experienced. As a result, there is no inference in favour of a specific view (viśiṣṭe), i.e., whether [the permanent pleasure] isn't experienced insofar as it is present or insofar as it is absent"(dharmasya kāraṇavacanam| yadi dharmo nimittāntaram? tasya hetur vācyo yataḥ utpadyata iti| yogasamādhijasya kāryāvasāyavirodhāt prakṣaye saṃvedananivṛttiḥ| yadi yogasamādhijo dharmo hetuḥ? tasya kāryāvasāyavirodhāt prakṣaye saṃvedanam atyantaṃ nivartate| asaṃvedane cāvidyamānenāviśeṣaḥ| yadi dharmakṣayāt saṃvedanoparamo nityaṃ sukhaṃ na saṃvedyata iti| kiṃ vidyamānaṃ na saṃvedyate, athāvidyamānam iti nānumānaṃ viśiṣṭe 'stīti|) 20 NB 23.6-12: "Moreover, the non-destruction of merit isn't supported by any inference. For merit has the property of being produced. There is no inference that shows that the merit that is produced by yogic meditative absorption isn't destroyed. Rather, there is an opposite inference that shows that, insofar as it has the property of being produced, it is impermanent. However, anyone, according to whom the experience of permanent pleasure doesn't cease, must infer that its cause is permanent. And it has been said that, if the experience were permanent, then there would be no distinction between someone who is liberated and someone who is caught in the cycle of rebirth. Just as, for the liberated being, there is permanent pleasure and the cause of its experience, but the experience doesn't cease due to the permanence of the cause, so also should be the case for someone who is caught in the cycle of rebirth. If this is right, then it would be accepted that [this experience] is accompanied by the experience of pleasure and pain produced by merit and demerit" (aprakṣayaś ca dharmasya niranumānam utpatti-dharmakatvāt| yoga-samādhijo dharmo na kṣīyata iti nāsty anumānam| utpatti-dharmakam anityam iti viparyayasya tv anumānam| yasya tu saṃvedanoparamo nāsti tena saṃvedana-hetur nitya ity anumeyam| nitye ca muktasaṃsārasthayor aviśeṣa ity uktam| yathā muktasya nityaṃ sukhaṃ tat-saṃvedana-hetuś ca, saṃvedanasya tūparamo nāsti, kāraṇasya nityatvāt, tathā saṃsārasthasyāpīti| evaṃ ca sati dharmādharma-phalena sukha-duḥkha-saṃvedanena sāhacaryaṃ gṛhyeteti|).
being can still experience permanent pleasure. 21 As Vātsyāyana notices, in order to avoid this inelegance, the opponent must say that the body itself is permanent. But that simply contradicts what is observed. 22 The upshot: the Permanent Pleasure View is hard to defend. 23

The Problem of Motivation
Even though the Permanent Pleasure View faces these problems, it has a virtue that Vātsyāyana's view lacks. As we have seen, for Vātsyāyana, liberation is complete freedom from embodied existence. But our embodied existence brings with it lots of different kinds of pleasure. Given that we are at least part-time pleasure-seekers, why should we seek to free ourselves completely from our embodied existence? At a certain juncture, Vātsyāyana himself expresses the thought (albeit as a false awareness) as follows, "With respect to liberation, [there is the false awareness]: 21 NB 23.13-17: "Suppose you say: 'The impediment to that is the connection with the body and so on.' [We reply:] No, since the body and so on are for the sake of enjoyment, and the opposite claim isn't inferred. Let the following be your view: 'The connection of a person caught in the cycle of rebirth with the body and so on serves as the impediment to the cause of the experience of permanent pleasure. So, there is no lack of distinction.' But this is unreasonable. The body and so on are for the sake of enjoyment. It is unreasonable that they will prevent enjoyment. And there is no inference to the conclusion that there is some enjoyment that a disembodied self can undergo" (śarīrādi-sambandhaḥ pratibandha-hetur iti cet? na, śarīrādīnām upabhogārthatvāt viparyayasya cānanumānāt| syān matam saṃsārāvasthasya śarīrādisambandho nitya-sukha-saṃvedana-hetoḥ pratibandhakaḥ, tenāviśeṣo nāstīti| etac cāyuktam, śarīrādayaḥ upabhogārthās te bhogapratibandhaṃ kariṣyantīty anupapannam; na cāsty anumānam aśarīrasyātmano bhogaḥ kaścid astīti|) 22 NB 24.1-6: "Moreover, one goes beyond what is observed in the same manner with respect to the body and so on. [To explain:] Just as one desires permanent pleasure having abandoned observed impermanent pleasure, so also must one posit a permanent body, permanent senses, and permanent awareness-events for the liberated being by going beyond the observed impermanent body, senses, and awareness-events. And, if this is right, it is better to posit that the liberated being also has the nature of being alone (aikātmya). If you say that this conflicts with reason (upapatti), the same is true [of the permanence of pleasure]. Since the permanence of the body and so on cannot be posited insofar as it conflicts with the means of knowing, the same sort of permanence of pleasure cannot be posited insofar as it conflicts with the means of knowing." (dṛṣṭātikramaś ca dehādiṣu tulyaḥ| yathā dṛṣṭam anityaṃ sukhaṃ parityajya nityaṃ sukhaṃ kāmayate, evaṃ dehendriyabuddhīr anityā dṛṣṭā atikramya muktasya nityā dehendriyabuddhayaḥ kalpayitavyāḥ, sādhīyaś caivaṃ muktasya caikātmyaṃ kalpitaṃ bhavatīti| upapattiviruddham iti cet? samānam| dehādīnāṃ nityatvaṃ pramāṇa-viruddhaṃ kalpayitum aśakyam iti? samānaṃ sukhasyāpi nityatvaṃ pramāṇa-viruddhaṃ kalpayitum aśakyam iti|) I am translating the term "aikātmya" as "the nature of being alone" taking "eka" to mean alone or solitary. Here, I am following Uddyotakara who takes it to refer to the state of isolation (kaivalya) (NV 82.10) that we encountered in footnote 17. The same interpretation is given by Tarkavāgīśa (1981, p. 246). 23 Later, Vātsyāyana revisits the question of how we should interpret the scriptural statements that seem to suggest that the state of liberation involves the manifestation of pleasure. His solution is simple: we should take them to be talking about the absence of pain. He says (NB 24.7-10): "Moreover, there is also no conflict with any scriptural statement, since such statements only speak about the absolute absence of the pain that is the cycle of rebirth. Even though there is some scriptural statement to the effect that the liberated being possesses absolute pleasure, it is reasonable that the term 'pleasure' is applied to mean the absence of pain. For it is observed that the term 'pleasure' often is applied to mean the absence of pain in common usage" (ātyantike ca saṃsāraduḥkhābhāve sukhavacanād āgame 'pi saty avirodhaḥ| yady api kaścid āgamaḥ syāt muktasyātyantikaṃ sukham iti? sukha-śabda ātyantike duḥkhābhāve prayukta ity evam upapadyate| dṛṣṭo hi duḥkhābhāve sukha-śabda-prayogo bahulaṃ loka iti|).
'This cessation of all effects (kārya) is certainly terrible. Given that liberation involves the loss of everything, many good things disappear. So, how could an intelligent person want that liberation which involves the destruction of all pleasure and is without consciousness?" 24 There's a more rigorous way of putting the point. We can be rationally motivated to bring about a state of affairs by means of actions that are themselves painful only if we rationally think (or expect) that realizing that state of affairs will result in some positive payoff, i.e., some desired outcome, which will outweigh the pain that we undergo in bringing it about. If this is true, then we cannot be motivated to realize the state of liberation. For, insofar as the state of liberation involves a perpetual disembodied existence, it involves a loss of all that we value about our lives. So, it doesn't bring about any desired outcome at all. In contrast, if we were to accept the Permanent Pleasure View, then this problem can be avoided. If the state of liberation involves the manifestation of permanent pleasure but no pain, then it contains an enormous amount of pleasure. Assuming that pleasure is at least one of the things we desire, we can be rationally motivated to seek liberation if we accept this view. Vātsyāyana is aware of at least a version of this problem. While discussing the Permanent Pleasure View, he considers the objection that "a practical undertaking is for the sake of attaining a desired object (iṣṭa)." 25 Uddyotakara glosses the objection as follows: "Here, this ordinary person undertakes action for the sake of attaining a desired object. And those who desire liberation undertake actions. For them too, the practical undertaking must be for the sake of attaining a desired object. This very practical undertaking has an end (artha) if there is permanent pleasure in liberation, not otherwise." 26 The argument is simply this. All (rational) practical undertakings are motivated by a desire to obtain a desired object. So, (rational) practical undertakings aimed at liberation must also be motivated by a desire to obtain a desired object, i.e., pleasure. That cannot be accounted for on Vātsyāyana's conception of liberation, since, for him, liberation involves no pleasure. But it can only be explained by appealing to something like the Permanent Pleasure View.
In response, Vātsyāyana rejects the claim that all practical undertakings are motivated by a desire to obtain a desired object, i.e., pleasure.
No, since it is for the sake of the cessation of an undesired object (aniṣṭa). This is the inference: "It is for the sake of attaining a desired object that liberation is taught and that people desirous of liberation undertake action. Both of these aren't without an end (artha)." And this is unreasonable. It is for the sake of the cessation of an undesired object that liberation is taught and that people desirous of liberation undertake action. Since the desired object isn't possible without being intertwined with the undesired object, even the desired object ends up being undesired. One who acts in order to avoid an undesired object also avoids the desired object, since it is not possible to avoid one in separation from the other. 27 In his gloss, Uddyotakara says, "In ordinary people, two kinds of practical undertakings are observed: those for the sake of attaining a desired object, and those for the sake of avoiding an undesired object." 28 This response depends on a background picture of motivation. So, let's expand on this.
As I have already said, within the Nyāya system, a practical undertaking (pravṛtti) is just the commencement (ārambha) of any mental, physical or linguistic action (NS 1.1.20). Vātsyāyana says that the motive (prayojana) underlying any such practical undertaking is an object that causes the practical undertaking in virtue of being determined by the relevant agent as a thing to be attained or avoided: "That, motivated by which the agent undertakes an action, is the motive" (yena prayuktaḥ pravartate tat prayojanam) (NB 3.9 on NS 1.1.1). The thought is explained further later in the commentary on NS 1.1.24, which says, "That in relation to which a person undertakes an action is the motive" (yam artham adhikṛtya pravartate tat prayojanam) (NB 26.14).
If, having determined an object to be something to be attained or avoided, a person performs the act that serves as a means to obtain or to avoid it, then that object is to be understood as the motive. For it is a cause of practical undertakings. The relation (adhikāra) with the object is the determination, "I shall obtain this object, or avoid it." An object, which is thus determined, is something that the agent is related to. 29 In his preamble to NS 1.1.1, while motivating the theoretical enterprise of Nyāya, Vātsyāyana seems to explain what the motive is: it is either pleasure and something that brings about pleasure, or pain and something that brings about pain.
The sub-commentator, Uddyotakara, makes this idea explicit in his Nyāyavārttika: What, then, is the motive? This is the ordinary meaning: "That, motivated by which a person undertakes an action, is the motive." What is a person motivated by? Some say, "By righteous conduct (dharma), profit (artha), pleasure (kāma) and liberation." In contrast, we observe that a person is motivated by the attainment of pleasure and the avoidance of pain. And all objects motivate a conscious being in virtue of being the means to pleasure and pain. 31,32 Slightly later in the same discussion, Uddyotakara refines this picture helpfully by isolating two different kinds of motivational profiles.
People either have attachment, etc. (rāgādi-mat), or are unattached (vīta-rāga). Among them, attachment has the defining characteristic of inclination (abhiṣaṅga) towards an object. Those who have this have attachment, etc. Detachment, in contrast, has the defining characteristic of disinclination (anabhiṣaṅga) from enjoyment (bhoga). Those who have that are unattached. Practical undertakings are of two kinds in accordance with the distinction amongst people. The practical undertakings of those people are of both kinds in accordance with the distinction amongst people. The practical undertakings of an unattached person are of one kind. Amongst those, the practical undertakings that belong to unattached people have as their end (artha) the avoidance of an undesired object. They undertake actions thinking, "We shall avoid something undesired." But they don't have any inclination towards anything. The practical undertakings of people who have attachment, etc. are of two kinds. The practical undertakings of these people who have attachment, etc. are of two kinds, either aimed at the acquisition of a desired object, or the avoidance of an undesired object. Thinking, "I shall obtain a desired object," [a person who has attachment, etc.] undertakes an action due to attachment. 31 NV 12.11-14: kiṃ punaḥ prayojanam iti? yena prayuktah pravartate tat prayojanam iti laukiko'yam arthaḥ| kena punaḥ prayujyate? dharmārtha-kāma-mokṣair iti kecit| vayaṃ tu paśyāmaḥ sukha-duḥkhāptihānibhyāṃ prayujyata iti| sukha-duḥkha-sādhana-bhāvāt tu sarve'rthāś cetanaṃ prayojayanti| 32 The same idea is repeated in Uddyotakara's sub-commentary on NS 1.1.24 : "In [the expression] 'that object in relation to which' (yam artham adhikṛtyeti), the relation is a determination.
A determination of what? Of the means to pleasure and pain. One attempts to obtain pleasure having undergone the awareness, 'This is a means to pleasure.' And one attempts to avoid pain having undergone the awareness, 'This is a means to pain.' Since this ordinary person is motivated by the acquisition of pain and avoidance of pain, the acquisition of pleasure and the acquisition of pain are the motives." (yam artham adhikṙ tyeti vyavasāyo'dhikāraḣ | kasya vyavasāyaḥ? sukha-duḥkha-sādhanānām| idaṃ sukhasādhanam iti buddhvā sukhāvāptaye yatate| idaṃ duḥkha-sādhanam iti cādhigamya duḥkha-hānāyeti| sukha-duḥkhayor avāpti-hānābhyām ayaṃ lokaḥ prayujyata iti sukha-duḥkhāpti-hanī prayojanam iti|) Thinking, "I shall avoid an undesired object," [such a person] abstains due to aversion. 33 For Uddyotakara, there are two kinds of motivational profiles: attached and unattached. Unattached agents don't perform any actions in order to obtain pleasure. This is because such unattached agents have no desire for pleasure at all; as Uddyotakara puts it, they are disinclined towards enjoyment. They only perform actions in order to minimize pain. By contrast, attached agents (i.e., those who have attachment, etc.) perform actions for two kinds of reasons: sometimes to obtain pleasure, and at other times merely to avoid pain. This idea is useful in understanding Vātsyāyana's reply to the problem that we considered above.
In response to that problem, Vātsyāyana pointed out that not all (rational) practical undertakings (even in attached agents like us) are for the sake of attaining a desired object: we are sometimes motivated to act in order to avoid an undesired object like pain, e.g., in the case where we avoid food mixed with honey and poison in order to avoid the deadly consequences of poison. We can use this to explain how someone can be rationally motivated to seek liberation. All of us face a choice between liberation (i.e., permanent disembodied existence) and repeated embodied existence. The pleasure that we undergo in virtue of having a body, the senses, etc. is unavoidably connected to pain. Since it's not possible to obtain that pleasure without also obtaining the pain that comes with it, the pleasure is to be avoided. The thought is anticipated in Vātsyāyana's commentary on NS 1.1.2, where he says, "Just as food mixed with honey and poison is not to be obtained, so also pleasure connected to pain is not to be obtained." 34 Since the pleasure that is part of embodied existence is connected with pain (either in virtue of being invariably accompanied by it, or in virtue of having the same causes, the same locus or the same subject 35 ), and we can be rationally motivated to act for the sake of avoiding pain, we can be rationally motivated to avoid embodied existence exactly in the same way as one avoids the food mixed with rice and honey. Thus, we have no reason to prefer the Permanent Pleasure View on these grounds. 33 NV 2.9-17: puruṣā rāgādi-manto vīta-rāgāś ca| tatra rāgo viṣayādiṣv abhiṣaṅga-lakṣaṇaḥ| sa yeṣām asti te rāgādi-mantaḥ| vairāgyaṃ punar bhogānabhiṣvaṅga-lakṣaṇam| tad yeṣām asti te vīta-rāgāḥ| pravṛtter dvaividhyam puruṣa-bhedānuvidhānena| teṣāṃ puruṣāṇāṃ yāḥ pravṛttayas tāḥ puruṣa-bhedam anuvidhīyamānā ubhayarūpa bhavanti| vīta-rāga-pravṛttir ekadhā| tatra yā vīta-rāgāṇāṃ pravṛttiḥ sā khalv eka-rūpā aniṣta-pratiṣedhārthā aniṣṭaṃ hāsyāma ity evam eva pravartate, na punar eṣāṃ kvacit abhiṣvaṇgo'stīti| rāgādi-mat-pravṛttis tu dvi-rūpa| ya ete rāgādi-mantas teṣāṃ yāḥ pravṛttayas tā dvividhā bhavanti, iṣṭāniṣṭa-viṣayādhigama-pratiṣedhārthāḥ| iṣtam āpsyāmīti rāgāt pravartate| aniṣṭaṃ hāsyāmīti dveṣān nivartate| 34 NB 8: 2-3: tad yathā madhu-viṣa-sampṛktānnam anādeyam iti evaṃ sukhaṃ duḥkhānuṣaktam anādeyam iti| 35 I am following Uddyotakara's gloss of the passage here. See NV 25.1-4: "The connection is a relation of invariable concomitance (avinābhāva): where there is one, there is the other. Alternatively, having the same cause is the connection: those very things which are the means to pleasure are the means to pain. Alternatively, having the same locus is the connection: where there is pleasure, there is pain. Alternatively, being apprehended by the same subject is the connection: he who apprehends pleasure also apprehends pain." (anuṣaṅgo'vinābhāvaḥ, yatraikaṃ tatretarad iti| samāna-nimittatā vānuṣaṅgaḥ, yāni vā sukha-sādhanāni tāny eva duḥkha-sādhanānīti| samānādhāratā vānuṣaṅgaḥ, yatra sukhaṃ tatra duḥkham iti| samānopalabhyatā vānuṣaṅgaḥ, yena sukham upalabhyate tena duḥkham apīti|) It's far from obvious whether this argument succeeds. Suppose, along the lines suggested by Uddyotakara, that we are agents who have attachment, etc. So, we are motivated to act by two kinds of desires: the desire to gain pleasure and things that give rise to pleasure and the desire to avoid pain and things that give rise to pain. These two desires will often pull us in different directions. Suppose, on a night out with friends, I am wondering whether I should order a third gin and tonic. On the one hand, I might be tempted to do so solely because of the breezy joy that gin brings me. On the other hand, I might hesitate, since I worry that I will be mildly hung-over tomorrow. Whether I order the gin and tonic depends on how much weight I attach to these desires. For example, there is no good reason to think that an attached agent like me is rationally required, under these circumstances, act on the desire to avoid the pain that hangovers involve, and refrain from ordering the gin and tonic. If I am predominantly a pleasure-seeker, I will perhaps attach a little less weight to my desire to avoid pain and go ahead and order the gin. And, intuitively, there's nothing irrational about doing so. Why can't the case of liberation be exactly like this? Typically, we are motivated by both the desire to attain pleasure and the desire to avoid pain. All of us have lives that are flawed in virtue of being inescapably painful at times. But interspersed between these pains are lots of pleasures that we care about. Provided that we rationally attach suitable weights to our desires to gain pleasure and to avoid pain, we may still rationally forgo liberation and seek to continue our embodied existence. So, unless Vātsyāyana tells us why the correct way of weighing our desires to gain pleasure and to avoid pain will always favour opting for liberation, his solution to the problem raised above cannot be convincing. 36 In the rest of the essay, I will explain how the solution can be made to work. 36 Vātsyāyana does have a second response to the worry that the defender of the Permanent Pleasure View raises. His thought is that if someone is motivated to attain liberation in virtue of being attached to pleasure, then that person cannot really be liberated, since attachment itself is well-known as bondage. He writes : "Moreover, [if this view is right,] given that the attachment to permanent pleasure won't be destroyed, there won't be any attainment of liberation. For attachment is well-known as bondage. [To explain:] This person, who is striving for liberation due to an attachment to permanent pleasure, thinking, "Permanent pleasure is manifested in liberation," wouldn't attain liberation, and cannot attain liberation. For attachment is well-known as bondage. And it is not reasonable that someone should be liberated even when there is bondage. [The opponent:] The attachment to permanent pleasure, when destroyed, isn't an impediment. The attachment of this person to permanent pleasure is destroyed. When that is destroyed, his attachment to permanent pleasure isn't an impediment. [Reply:] If this is right, then, whether or not the liberated being possesses permanent pleasure, the attainment of liberation isn't in doubt on either view" (nitya-sukha-rāgasyāprahāṇe mokṣādhigamābhāvaḥ, rāgasya bandhana-samājñānāt| yady ayaṃ mokṣe nityaṃ sukham abhivyajyata iti nitya-sukha-rāgeṇa mokṣāya ghaṭamāno na mokṣam adhi-gacchet, nādhi-gantum arhatīti| bandhana-samājñāto hi rāgaḥ| na ca bandhane saty api kaścin mukta ity upapadyata iti| prahīṇa-nitya-sukha-rāgasyāpratikūlatvam| athāsya nitya-sukha-rāgaḥ prahīyate, tasmin prahīṇe nāsya nitya-sukha-rāgaḥ pratikūlo bhavati? yady evam, muktasya nityaṃ sukhaṃ bhavati, athāpi na bhavati, nāsyobhayoḥ pakṣayor mokṣādhigamo vikalpyate iti|).

Shifting Motivational Profiles
For Vātsyāyana, there is something wrong about the motivational profile of attached agents, i.e., agents who are motivated, in equal if not greater measure, by the desire to obtain pleasure. This is suggested by his defence of a certain principle.
The Pain Principle. We should treat all aspects of embodied existence as pain. 37 When an attached agent fulfils the requirement laid down by the Pain Principle, she no longer sees anything as pleasant and sees everything as painful. As a result, she is no longer motivated to act for the sake of pleasure, and only desires to avoid pain. Thus, she undergoes a shift in her motivational profile, whereby she transforms into an unattached agent. From this perspective of detachment, she now sees that the only way of getting rid of pain once and for all is to free herself from repeated embodied existence. If I am right, then the Pain Principle gives us an indirect explanation of why even an attached agent should come to seek liberation. In a nutshell, the explanation is this. An attached agent should treat all aspects of embodied existence as pain. Once she does so, she should seek to avoid those aspects (since she also has the desire to avoid pain). That is why she should seek liberation.
Before moving on, it's worth explaining how this helps us address the worry that I raised at the end of the last section. Consider the gin and tonic example. In that scenario, it seems that I am rationally permitted to order a third gin and tonic because that will bring me pleasure. But Vātsyāyana would say that this is a mistake. If the Pain Principle is true, then I should not even think that there is any pleasure to be gained from drinking the gin. So, no matter how much weight I actually attach to my desire for pleasure, I cannot rationally let considerations about pleasure play any role in my decision-making. The same goes for our deliberation about whether to continue our embodied existence. If the Pain Principle is true, we should treat all aspects of embodied existence as painful. So, no matter how much weight we actually attach to our desire for pleasure, we cannot rationally continue our embodied existence on the basis of the consideration that it promises to yield pleasure.
Why is the Pain Principle true? In the rest of this section, we shall look at Vātsyāyana's defence of this principle.

A Contrast: The Naïve Pain Principle
An initial thought might be that the Pain Principle is true, because all aspects of embodied existence in fact are pain. This suggests: The Naïve Pain Principle. All aspects of embodied existence are pain by their own nature.
The Naïve Pain Principle is a descriptive claim, but the Pain Principle is a normative one. But, using a suitable assumption (e.g., the assumption that, if a claim is true, we should treat it as true), we can explain why the Pain Principle is true by appealing to the Naïve Pain Principle. 38 However, as Vātsyāyana notes in his commentary before NS 4.1.55, the Naïve Pain Principle is in tension with what we know by perception: all conscious living beings experience pleasure; it's not possible to deny its existence.
It has been said that "pain is that which has the defining characteristic of distress" (NS 1.1.21). Is this very claim a rejection of pleasure which is experienced within oneself (pratyātma-vedanīya) and is perceived by every living creature, or is it some other alternative? He [i.e., the author of the sūtra] has said that it's something else. Why? Surely, pleasure, witnessed by all ordinary people, cannot be rejected. Rather, this is a teaching of the cultivation (bhāvana) of the label of pain (duḥkha-saṃjñā), for the sake of avoiding pain, addressed to someone who has become dejected with regard to the pain caused by the experience (anubhava) of the uninterrupted series of births and deaths, and, therefore, desires to avoid pain. 39 When NS 1.1.21 defines pain as that which has the defining characteristic of distress, Vātsyāyana interprets it as saying that every aspect of embodied existence, including pleasure, counts as pain insofar as it is inextricably connected to pain. But this isn't supposed to be a literal denial of the existence of pleasure. Rather, it is meant to instruct someone who has become dejected with respect to the pain that is involved in our embodied existence and wishes to avoid it. This person is being taught to cultivate the label of pain with respect to all aspects of our embodied existence. In both Buddhist and non-Buddhist contexts, the term "bhāvana" or "bhāvanā"-which I am translating here as "cultivation"-often refers to the sustained development of a mental state. 40 When a person is taught to cultivate the 38 Note that I am not relying here on the contested principle (sometimes called Hume's law) that it is not possible to derive a normative conclusion (i.e., a conclusion about what should or ought to be the case) solely from a descriptive premise (i.e., a premise about what is the case). Defenders of this principle say that, in order to derive a normative conclusion from a descriptive premise (or a set of descriptive premises), we would need an additional (normative) premise which connects the descriptive premise to the normative claim. For example, to derive the claim that one should not set a cat on fire from the premise that setting a cat on fire causes it gratuitous pain, we need an additional premise, i.e., that one should not cause any living being any gratuitous pain. I don't have to accept this. My point is simply that, merely from the claim that every aspect of embodied existence is pain, nothing follows about what we should believe about various aspects of our embodied existence, or how we should treat various aspects of our embodied existence. But if we adopt a suitable assumption, e.g., that, if a claim is true, we should treat it as true, or that, if an object is F, we should treat it as F, we should be able to derive the Pain Principle from the Naive Pain Principle. 39 NB 244.11-245.2: uktaṃ ca bādhanālakṣaṇaṃ duḥkham iti| tat kim idaṃ praty-ātma-vedanīyasya sarva-jantu-pratyakṣasya sukhasya pratyākhyānam, āhosvid anyaḥ kalpa iti? anya ity āha| katham? na vai sarva-loka-sākṣikaṃ sukhaṃ śakyaṃ pratyākhyātum| ayaṃ tu janma-maraṇa-prabandhānubhavanimittād duḥkhān nirviṇṇasya duḥkhaṃ jihāsato duḥkha-saṃjñā-bhāvanopadeśo duḥkha-hānārtha iti| 40 For Buddhist uses of this term "bhāvanā" in this sense, see Dīgha Nikāya (DN xxxiii.1.10), Majjhima Nikāya (MN 36 and 44), especially the Cūḷavedalla Sutta, and Aṅguttara Nikāya (ANI I.iii.1 and 3 and I. vi.1-2). For relevantly similar uses of the term "bhāvanā" or "bhāvana" in the context of Yoga, see Yogasūtra 1.28, 1.33, 2.1, and 2.33 (YS 33.2,and 105.12). Amongst these passages, Yogasūtra 2.33-"When distressed by [wrong] thoughts, there should be cultivation of the opposite" (vitarkabādhane pratipakṣa-bhāvanam)-is particularly important, since it speaks of the cultivation of the label of pain, she is being taught to engage in a sustained development of a mental state where she applies the label of pain to every aspect of embodied existence and thereby perceives everything as pain.
As Vātsyāyana emphasises in his commentary on NS 4.1.58, this isn't a literal use of the term "pain." Vātsyāyana here entertains a worry that if one asserts the sentence, "Birth is just pain," one thereby conveys that there is no pleasure involved in ordinary existence. That is false. The response to the worry is that this is an example of metonymy (upacāra). Here's the exchange.
[The opponent:] If this is so, why is it not said, "Birth is pain"? If such is to be said, the person who has said the following, "Birth is just pain," conveys the absence of pleasure.
[Reply:] So, this expression "just" is certainly for the sake of the cessation of birth.
[Reply:] Birth isn't pain in virtue of its own nature, but rather in virtue of a metonymic use (upacāra) of "pain." The same is also true of pleasure. So, this [birth] is produced by this [pleasure] alone. However, birth is not simply pain. 41 In his commentary on NS 1.2.14, Vātsyāyana gives an example of metonymy while discussing verbal tricks (chala) in a debate. 42 While describing the reactions of an audience to high melodrama in a play, someone might say, "The benches are crying" (mañcāḥ krośanti). But a quibbler might take this utterance literally, and say, "The people on the benches are crying. But the benches aren't crying" (mañcasthāḥ puruṣāḥ krośanti, na tu mañcāḥ krośanti). The problem, as Vātsyāyana diagnoses it, is that the hearer construes literally (prādhānyena) an expression-Footnote 40 continued opposite (pratipakṣa-bhāvanā). Following this, in the 9th chapter (āhnika) of Nyāyamañjarī, Jayanta claims that practising the cultivation of the opposite (pratipakṣabhāvanābhyāsa) can help us get rid of the defects (doṣa) by uprooting the underlying false awareness (NM II 449.1-451.4); for discussion, see Slaje (1995). 41 NB 247.13.16: yady evaṃ kasmād duḥkhaṃ janmeti nocyate? so 'yam evaṃ vācye yad evam āha duḥkham eva janmeti tena sukhābhāvaṃ jñāpayatīti janma-vinigrahārthīyo vai khalv ayam eva-śabdaḥ| katham? na duḥkhaṃ janma svarūpataḥ, kiṃ tu duḥkhopacārāt; evaṃ sukham apīti etad anenaiva nirvarttyate na tu duḥkham eva janmeti| 42 Tzohar (2018) translates "upacāra" as "metaphor." This seems inaccurate. Both are non-literal uses of expressions: a metaphorical use of an expression involves the application of an expression to designate something similar to the primary referent of the expression, while a metonymic use of an expression involves the application of an expression to designate something that is an attribute of the primary referent of that expression. Just to understand the distinction, consider two examples: (1) Juliet is the sun.
(2) The pen is mightier than the sword.
On the Nyāya explanation of these sentences, when (1) is uttered by Romeo, the expression "the sun" doesn't designate the primary referent of the expression, i.e., the sun, but rather something that is similar to the sun; in (2), the term "the pen" stands for the written word, which, in virtue of some relation, is an attribute of the primary referent, i.e., the pen. namely, "benches"-that is used in a secondary sense (bhaktyā prayoge). As Vātsyāyana goes on to explain, "Metonymy is characterised by a meaning that is derived (nītārtha) on the basis of accompaniment (saha-caraṇa), etc. The designation (abhidhāna) of an object that possesses something (tad-vat), when it isn't that thing (atadbhāve), is metonymy." 43 Since the benches here are accompanied by the people sitting on them, the people, who possess the benches in virtue of the relation of sitting on them, are referred to or designated by means of the term "benches." That is why this is an instance of metonymy. The same is true of birth. The term "pain" doesn't literally refer to "birth." But pain is an attribute of any birth. So, we can refer to "birth" by using the term "pain" as a metonym. The same goes for pleasure insofar as it is invariably connected to pain.
In his sub-commentary on NS 1.1.21, Uddyotakara gives us some further reasons for thinking that the Naïve Pain Principle is false. The first is an argument from linguistic usage.
Some say that everything is pain by nature. This isn't right, since it conflicts with perception. For it's not possible to deny the existence of perceptible pleasure. Suppose it is said that it's a variety (vikalpa) of pain. Someone might say, "Pleasure is simply a variety (vikalpa) of pain. However, it doesn't exist [as pain] in virtue of its own nature." No, since the negative particle (nañ) cannot be applied to a variety [of something]. For it isn't observed that the negative particle (nañ) is being applied to a variety [of something]. Indeed, with respect to a variety of brahmin, no one applies the term "non-brahmin." In the same way, the term "non-pain" shouldn't be applied to a kind of pain. 44 Pleasure is called "non-pain." If it were a variety of pain, then we wouldn't call it "non-pain," just as we don't call a brahmin of a certain kind "non-brahmin." So, pleasure isn't a kind of pain.
The second argument is more complex.
Moreover, if there were no pleasure, then merit (dharma) would be futile. Since merit is the means to pleasure, merit would be futile in the absence of pleasure. It is not reasonable that the result of merit is the cessation of pain. For, then, given that there would be the undesirable consequence that merit has an absence as its result, merit would have an absence as its result. 45 If there were no pleasure (but only pain), then there couldn't be any experience of pleasure, so the merit we may have accumulated by performing virtuous acts would have no effect. And if the result of such actions were only the removal of existing 43 NB 49.17-18: upacāro nītārthaḥ sahacaraṇādi-nimittena| atad-bhāve tad-vad-abhidhānam upacāraḥ| 44 NV 80.9-13: sarvam svarūpato duḥkham iti kecit| na, pratyakṣa-virodhāt| na hi pratyakṣaṃ sukhaṃ śakyaṃ pratyākhyātum iti| duḥkha-vikalpa iti cet? athāpīdaṃ syād duḥkha-vikalpa eva sukham iti, na punaḥ svarūpato'stīti? na, vikalpe nañ-prayogāsambhavāt| na hi vikalpe nañ pravartamāno dṛṣṭaḥ| na hi brāhmaṇa-viśeṣe bhavaty abrāhmana iti| evaṃ duḥkha-viśeṣe'duḥkham iti na syāt| 45 NV 80.13-16: yadi ca sukhaṃ na syād dharma-vaiyarthyam| kiṃ kāraṇam sukha-sādhanaṃ dharma iti, sukhaṃ ca nāstīti vyartho dharmaḥ| duḥkha-pratiṣedhaḥ phalam asyeti na yuktam, dharmasyābhāvaphalatva-prasaṅgāt abhāva-phalo dharma iti syāt| pain, then merit would have an absence as an effect. This contradicts the ordinary manner in which merit is supposed to work: for example, rituals like the new-moon and full-moon sacrifices (darśa-paurṇamāsa) are supposed to produce merit, which, in turn, is supposed to produce a positive effect, e.g., heaven.
The third argument is that, according to Uddyotakara's theory of practical motivation, there are two different kinds of practical undertakings depending on the underlying motivation.
Moreover, the two-fold-ness of practical undertakings, which is ordinarily observed, wouldn't exist. One person undertakes action, thinking, "I shall attain a benefit." The other undertakes action, thinking, "I shall avoid something undesired." Given that there would be no benefit [i.e., pleasure], there would be no two-fold-ness of practical undertakings in ordinary practice. 46 Some practical undertakings are motivated by the desire to obtain pleasure, while others are motivated by the desire to avoid pain. This distinction between the two kinds of practical undertakings wouldn't make sense if everything were just pain by its own nature. Why should we want to keep the distinction? The distinction at least helps us explain the difference between an attached agent and an unattached one. The attached person acts out of both kinds of desires, while the unattached acts only out of the second.
The fourth argument is the most significant. The thought is that the advice that various sages give-namely, that one should treat every aspect of one's embodied existence as pain-wouldn't make sense if everything were actually pain.
Furthermore, the cultivation of pain (duḥkḥa-bhāvanā) wouldn't be taught. For, given that the opposite (pratipakṣa) [i.e., pleasure] would be absent, there would also not be any attachment (sakti) [to pleasure]. Since no one is attached to pain, the cultivation of pain wouldn't be taught. Therefore, everything is not "pain" in the primary sense. Rather, everything is said to be "pain" by way of teaching the cultivation of pain. 47 The reason why the advice makes sense is that we in fact experience pleasure, and this experience leads to bad consequences. To counteract those bad consequences, it's necessary for us to focus on pain. As Uddyotakara notes, if everything were pain, then its opposite, i.e., pleasure, wouldn't exist. So, there would be no attachment to pleasure. Since no one is attached to things other than pleasure, e.g., pain, Vātsyāyana's advice that we cultivate the label of pain wouldn't make much sense either.
While these arguments may show that the Naïve Pain Principle is false, it still doesn't explain why the Pain Principle is correct.

Three Arguments for the Pain Principle
Vātsyāyana gives at least three arguments for the Pain Principle in his commentary on NS 4.1.55-8.
The first argument is the Argument from Connection. In his commentary on NS 4.1.55-"The production of embodied existence is simply pain due to a connection with various kinds of distress" (vividha-bādhanā-yogād duḥkham eva janmotpattiḥ) -Vātsyāyana argues that, if a person were to notice the many different kinds of distress that embodied beings are subject to, she would (rationally) come to think of both pleasure and the means to pleasure as pain.
Since 'birth' means that which is born [rather than the event of being born], it stands for the body, the senses, and awareness-events. The appearance (prādurbhava) of the body and so on, which involve an arrangement of parts, is production. Distress is of various kinds: slight, moderate, and acute. Acute pain belongs to beings in hell, whereas moderate pain belongs to animals. Human beings undergo slight pain, and gods and unattached beings undergo even slighter pain. In this way, for a person who perceives these places of production [i.e., beings born with a body, etc.] as connected to pain, the label (saṃjñā) of pain is established with respect to pleasure and with respect to the body, the senses, and the awareness-events that serve as a means to pleasure. Due to the establishment of the label of pain, a label of disinterestedness (anabhirati) is established with respect to all the realms. 48 For the person who devotes himself to the label of disinterestedness, thirst (tṛṣṇā) with respect to all the realms is broken. Due to the abandonment of thirst, he is freed from all pain. This is just as in the case of a person, who, taking milk to be poison due to a connection with poison, doesn't acquire it. As he doesn't acquire it, he doesn't attain the pain of death. 49 The thought depends on the previously discussed analogy of food mixed with poison. Suppose one knows that there is poison in a saucer of milk. One may desire the pleasure of drinking milk, but one also knows that the deadly consequences of poison outweigh the benefits of drinking the milk. In such a scenario, regarding the milk mixed with poison merely as poison and not as milk (and therefore disregarding the benefits of milk) might help one steer clear of it, and thus might prevent one from undergoing the pain of death. In the same way, embodied existence contains both pleasure and pain, but the pain outweighs (both in variety and amount) the pleasure. So, treating all aspects of embodied existence as pain 48 For a parallel discussion of the label of disinterestedness with respect to all realms (sabbaloke anabhiratasaññā) in a Buddhist context, see Aṅguttara-Nikāya (AN IV XLVI.9-10). 49 NB 245.7-14: janma jcaāyate iti śarīrendriyabuddhayaḥ| śarīrādīnāṃ ca saṃsthāna-viśiṣṭānāṃ prādurbhāva utpattiḣ | vividhā ca bādhanā hīnā madhyamā utkṛṣṭā ceti| utkṛṣṭā nārakiṇām, tiraścāṃ tu madhyamā, manuṣyāṇāṃ tu hīnā, devānāṃ hīnatarā vītarāgāṇāṃ ca| evaṃ sarvam utpatti-sthānaṃ vividha-bādhanānuṣaktaṃ paśyataḥ sukhe tat-sādhaneṣu ca śarīrendriyabuddhiṣu duḥkha-saṃjñā vyavatiṣṭhate| duḥkha-saṃjñā-vyavasthānāt sarva-lokeṣv anabhirati-saṃjñā bhavati| anabhirati-saṃjñām upāsīnasya sarva-loka-viṣayā tṛṣṇā vicchidyate, tṛṣṇā-prahāṇāt sarva-duḥkhād vimucyate iti| yathā viṣayogāt payo viṣam iti budhyamāno nopādatte, anupādadāno maraṇaduḥkhaṃ nāpnoti|| helps us disregard their attractions, and thus allows us to effectively avoid pain that accompanies embodied existence. Regarding all these pleasures as pain loosens our attachment to them. This paves the way for freedom from all suffering. That is why one should treat all aspects of embodied existence as pain.
The second argument is the Argument from Desire. It proceeds from the idea that, if we apprehend pleasure as pleasure, we will end up in more pain than we otherwise would. In his commentary on NS 4.1.57-"For someone who experiences, there is no absence of the cessation of distress due to the defect of seeking" (bādhanānivṛtter vedayataḥ paryeṣaṇa-doṣād apratiṣedhaḥ)-Vātsyāyana explains the thought as follows.
Seeking (paryeṣaṇa) is craving (prārthanā), the thirst for obtaining an object. The defect of seeking is this. Someone who experiences [pleasure or a means to pleasure as such] seeks it. That thing which this person seeks either isn't obtained, or, having been obtained, is endangered. Or, something lesser is obtained, or the object is obtained along with many obstacles. Due to this defect of seeking, there are many kinds of mental sorrow (mānasa-santāpa). In this way, for someone who experiences [pleasure or a means to pleasure as such], there is no cessation of distress due to the defect of seeking. Since the cessation of distress is absent, the cultivation of the label of pain is taught. For this reason, birth is pain, not due to the absence of pleasure. 50 The argument is this. The experience of pleasure, when one regards it as pleasure, gives rise to more suffering than pleasure. This is because, when one apprehends pleasure as pleasure (or a means to pleasure as a means to pleasure), that apprehension gives rise to cravings, and cravings are never fully satisfied, thus giving rise to many different kinds of mental pain. So, the only way to avoid such pain is to not to apprehend pleasure as pleasure (or a means to pleasure as a means to pleasure) but rather as pain, so that no desire for re-experiencing such pleasure arises again.
The third argument is the Argument from Error. It proceeds from the idea that if we apprehend the pleasure involved in embodied experience as pleasure, we will end up falsely treating even the pain involved in embodied existence as pleasure insofar as it serves as a means to pleasure. In his commentary on NS 4.1.57-"And since there is erroneous awareness of pleasure with respect to a variety of pain" (duḥkhavikalpe sukhābhimānāc ca)-Vātsyāyana explains the thought as follows.
And this person, who abides in pleasure, takes pleasure to be the final aim of human existence, thinking, "There is no highest good other than pleasure, and, when pleasure is attained, one attains one's end and accomplishes that which was to be accomplished." On the basis of a false awareness (saṅkalpa), he is attached to pleasure and to the objects that serve as the means to pleasure. Being so attached, he strives for pleasure. For this person who is striving, there arise all the different kinds of pain which are caused by birth, old age, disease, death, contact with undesired objects, separation from desired objects, and the dissatisfaction of cravings. Yet, he is erroneously aware (abhimanyate) of that variety of pain as pleasure. Pain is a constituent element (aṅga) of pleasure. It is not possible to attain pleasure without attaining pain. The person, whose intellect (buddhi) is struck by the perception of pleasure that takes the form, "This is indeed pleasure, since it is a means for attaining that pleasure", rushes on and doesn't surpass the cycle of rebirth, following what is said [in the Chāndogya Upaniṣad 5.10.8], "Be born! Die!" The cultivation of the label of pain is taught as an antidote (pratipakṣa) to this label of pleasure. Birth (janman) counts as "pain" due to a connection with pain, not due to an absence of pleasure. 51 The thought is that we tend to treat even pain as pleasure insofar as it serves as a means to pleasure. That is counterproductive, since it inures us to the painfulness of embodied existence, thereby preventing us from getting out of the cycle of rebirth. So, if we want to avoid this error that perpetuates the pain involved in embodied existence, the best policy for us will be to treat all aspects of embodied existence as pain. We therefore need to cultivate the label of pain with respect to embodied existence, and treat the pleasure involved in embodied existence as pain.
Let's take stock. If these arguments are sound, then the Pain Principle is true. If we accept the Pain Principle, then we should treat all aspects of embodied existence as pain and therefore must lose all desire for pleasure and objects that give rise to pleasure. As a result, we should only retain our desire to avoid pain. So, our motivational profile should come to match that of an unattached agent. Since liberation involves permanent freedom from pain, we would then be required by rationality to seek liberation. That is how we would be rationally motivated to seek liberation.

The False Origins of Attachment and Aversion
Vātsyāyana argues that the three defects-attachment, aversion, and delusion-are all based on or involve false awareness of some sort. 52 Why? Following NS 3.1.26 -"No, since attachment and so on have a saṅkalpa as their cause" (na, saṅkalpanimittatvād rāgādīnām)-Vātsyāyana's story appeals to a mental state called saṅkalpa. He repeats the story at a number of places.
Indeed, it is found that this attachment of living beings that enjoy intentional objects (viṣaya) is produced by a saṅkalpa. And a saṅkalpa has as its origin (yoni) the recollection (anucintana) of intentional objects that experienced earlier. 53 Attachment and aversion don't arise in anyone who isn't deluded. But, in a deluded person, they arise in accordance with a saṅkalpa. The saṅkalpas that are agreeable (rañjanīya) with respect to intentional objects are causes of attachment. The saṅkalpas that are disagreeable (kopanīya) are causes of aversion. And both these kinds of saṅkalpas are nothing other than delusions, since they have the characteristic of being false awareness. These very attachment and aversion have delusion as their origin. 54 From false saṅkalpa-s that are agreeable, disagreeable and deluding (mohanīya), desire, attachment and delusion arise. 55 The intentional objects of desire are objects of the senses. That is why they are said to be colour, etc. And, when those are made into objects of false saṅkalpas (mithyāsaṅkalpyamāna), they give rise to attachment, aversion, and delusion. 56 From these passages, it is clear that, for Vātsyāyana, a saṅkalpa is a false awareness, 57 which can be of three kinds: agreeable, disagreeable, and deluding. 52 Compare Nāgārjuna's Mūla-madhyamaka-kārikā 23.1 : "Attachment, aversion, and delusion are said to be produced by a saṅkalpa; they arise depending indeed on an error that is either good or bad (saṅkalpa-prabhavo rāgo dveṣo mohaś ca kathyate | śubhāśubhaviparyāsān saṃbhavanti pratītya hi ||) In his commentary, Candrakīrti says that"saṅkalpa" is simply thought (vitarka). 53 NB 150.11-2 on NS 3.1.26: ayaṃ khalu prāṇināṃ viṣayān āsevamānānāṃ saṅkalpa-janito rāgo gṛhyate| saṅkalpaś ca pūrvānubhūta-viṣayānucintana-yoniḥ| 54 NB 221.9-11 on NS 4.1.6: amūḍhasya rāga-dveṣau notpadyete| mūḍhasya tu yathā-saṅkalpam utpattiḥ| viṣayeṣu rañjanīyāḥ saṅkalpā rāga-hetavaḥ| kopanīyāḥ saṅkalpā dveṣa-hetavaḥ| ubhaye ca saṅkalpā na mithyā-pratipatti-lakṣaṇatvān mohād anye| tāv imau moha-yonī rāgadveṣāv iti| 55 NB 256.11-2 on NS 4.1.68: mithyā-saṅkalpebhyo rañjanīya-kopanīya-mohanīyebhyo rāga-dveṣamohā utpadyante… 56 NB 259.14-5 on NS 4.2.2: kāma-viṣayā indriyārthā iti rūpādaya ucyante| te ca mithyā-saṅkalpyamānā rāga-dveṣa-mohān pravartayanti| 57 Uddyotakara doesn't seem to agree with Vātsyāyana here. In his sub-commentary on NS 3.1.24, while explaining Vātsyāyana's claim that the recollection (anu-cintana) of a previously experienced object is the origin of attachment, he says: "The craving (prārthanā) for an intentional object that was experienced earlier is the saṅkalpa" (pūrvānubhūta-viṣaya-prārthanā saṅkalpaḥ) (NV 347.14). The problem with this interpretation is that it makes it hard to understand how saṅkalpa could be a kind of delusion or a false awareness. In his commentary, Vācaspati glosses this remark by saying that: "The meaning of the sentence 'pūrvānubhūta-viṣaya-prārthanā saṅkalpaḥ' is that the saṅkalpa that is for the sake of a craving is the saṅkalpa behind that craving, and that has as its intentional object something that was experienced Presumably, when an agent recollects a previously experienced object, she undergoes an awareness-event that portrays that object in a certain way. That is a saṅkalpa. When the saṅkalpa is agreeable, it portrays the object in a pleasing way. Then, it gives rise to attachment towards that and other similar objects. When it is disagreeable, it portrays the object in an unpleasant way. That gives rise to aversion. Finally, when it is deluding, it perhaps portrays the object in an inaccurate way, which in turn gives rise to a delusion. But, since, for Vātsyāyana, any false awareness is a kind of delusion, any saṅkalpa must also count as a delusion. So, attachment and aversion only arise in beings who are deluded in some way. That is why he takes delusion to be the most vicious of the three defects (NB 221.7-8 and 256.14-6 on NS 4.1.6 and NS 4.1.68).
This seems like a radical claim, so let me explain why it might make sense. For Vātsyāyana, delusions are of four kinds: false awareness (mithyā-jñāna), uncertainty (vicikitsā), pride (māna), and confusion (pramāda) (NB 220.7-8 on NS 4.1.3). Amongst these, saṅkalpas fall under the category of false awareness. But do all our attachments and aversion depend on some kind of false awareness? The typology of desire that Vātsyāyana puts forward does seem to suggest this.
Amongst those, lust is the desire with respect to women; the craving that arises with respect to women is lust. And thus, they have said, "He who doesn't lust doesn't rejoice." The desire not to give up something that isn't diminished is stinginess; stinginess is the desire not to give up something which, when given away or enjoyed, isn't diminished, e.g., the desire of the form, "May he not drink the water from the king's reservoir." The desire to acquire an unowned object is longing; the desire to obtain an object that isn't owned is longing. Thirst is the cause of being connected to rebirth; the craving for rebirth is thirst. Greed is the desire to unjustifiably (pramāṇaviruddha) acquire someone else's possession; a person who acquires someone else's possession unjustifiably is called greedy. 58 Vātsyāyana goes on to explain that the defining characteristic of all these different kinds of attachment is clinging (āsakti), which Uddyotakara glosses as inclination (abhiṣaṅga) towards various objects (NB 220.10; NV 426.1-2).
Anger is the cause of the agitation of the substrata of the body and the senses; anger is that upon the production of which the substrata of the body and the senses are agitated. Envy is the desire to thwart the adherence (abhiniveśa) of others to a common object; the desire to thwart the adherence of others to that which is common, i.e., unacquired by anyone, is envy. Indignation is impatience with respect to others' virtues; the impatience that arises after one hears about others' virtues is indignation. Malice is the desire to harm others; the desire of an incapable person to harm others is malice. Vengefulness is the intolerance of someone who has been harmed; the intolerance of someone who has been harmed is vengefulness. 59 Vātsyāyana goes on to explain that the defining characteristic of all these different kinds of aversion is non-endurance (amarṣa), which Uddyotakara takes to be a form of intolerance with respect to pain and the means to pain (NB 220.10; NV 426.2). 60 For Vātsyāyana, all these attachments and aversions are based on a mistake either about the objects of the senses that we are attached or averse to, or about ourselves. In some of these cases, the explanation might be obvious. For example, in the case of stinginess or greed or jealousy, the desire may be driven by a false belief about who should be or is the rightful owner of an object; similarly, indignation may be based on a false belief about one's own superiority, malice may be based on the false belief that one can harm another person, and vengefulness may be based on the false idea that one harmful act can compensate or undo the effects of another harmful act. What I want to focus on is Vātsyāyana's treatment of two kinds of attachment: the first is thirst (tṛṣṇā), i.e., the desire for rebirth, and the second is lust (kāma), i.e., sexual desire.

Thirst and Lust
Let's begin with thirst. In his preamble to NS 4.2.1, Vātsyāyana claims that the kind of true awareness that helps us escape the cycle of rebirth can't be about everything. 59 NV 425.5-10: śarīrendriyādhiṣṭhāna-vaikṛtya-hetuh krodhaḣ | saṃjāte yasmin śarīrendriyādhiṣṭhānāni vikṛtāni bhavanti sa krodha iti| sādhāraṇe vastuni parābhiniveśa-pratiṣedheccha īrṡyā| yad-aparigṛhītaṃ sādhāraṇaṃ vastu, tasmin yaḥ parābhiniveśa-pratiṣedhābhiprāyaḥ sā īrṣyā| para-guṇākṣamatā asū yā| yā para-guṇān śrutvā akṣamatopajāyate, sā asūyā| parāpakārecchā drohaḣ | aśaktasyāpi yā paraṃ praty apacikīrśā sā drohaḥ| apakārāsahiṣṇutā amarṡaḣ | yā kṛtāpakārasyāsahiṣṇutā so'marṣaḥ| An anonymous referee suggested that I read as the compound "śarīrendriyādhiṣṭhāna-vaikṛtya-hetu" as a bahuvrīhi compound and translate it as "the cause of the agitation that has the body and the senses as its substratum." I think this reading doesn't fit the next sentence very well: that seems to be saying that, when anger is produced, the substrata of the body and the senses (śarīrendriyādhiṣṭhānāni) are agitated. 60 I have translated the term "amarṣa" once as "vengefulness" and then as "non-endurance." The reason for this is that Vātsyāyana himself seems to be using the term in two different senses in the two contexts: in the first context, it's a specific kind of aversion, while, in the second, it's a common character shared by all kinds of aversion.
It is directed at a certain object whose nature, when concealed from us by some false awareness, gives rise to the cycle of rebirth and therefore to all our suffering. 61 What is that false awareness? Vātsyāyana's reply: It's a false awareness about the self. When the nature of the self is hidden from us by this false awareness, we come to have the desire to be born again. This leads to rebirth. Vātsyāyana describes the nature of this false awareness as follows.
It is the ego-construction (ahaṅkāra), i.e., the delusion that is the apprehension of the self in what is not the self and takes the form, "I exist." The egoconstruction is the view of someone who undergoes a perception, "I exist," with respect to what is not the self.
[The opponent:] What are the objects about which there is ego-construction?
[Reply:] The body, the senses, the manas, the hedonic states (vedanā), and the awareness-events.
[The opponent:] How is ego-construction about them the root of the cycle of rebirth?
[Reply:] Indeed, this person who determines, "I exist," with respect to things such as the body, etc., is overwhelmed by a thirst for their non-destruction insofar as he thinks that their destruction is his own destruction, and appropriates them over and over again. Having appropriated them, he proceeds to be born and to die. Since he can't surpass that, he isn't completely freed from pain. 62 On Vātsyāyana's view, the kind of false awareness that gives rise to the cycle of rebirth is ego-construction, i.e., a false awareness about oneself. Since we identify ourselves with the body, the senses, the manas, the hedonic states, awareness-61 NB 250.4-9: "[The opponent:] But indeed, sir, amongst all the intentional objects that there are, does the awareness of the truth [that gives rise to liberation] arise with respect to each one of them, or does it arise with respect to some? [Reply:] What is the distinction here? [The opponent:] It doesn't arise, one by one, with respect to all intentional objects, since the objects of awareness are infinite. It also doesn't arise with respect to some of them. For, given that the delusion won't cease with respect to anything with respect to which it [i.e., the awareness of the truth] doesn't arise, there will be the undesirable consequence that some delusion will be left over. And it is not possible to destroy the delusion with respect to one intentional object by means of an awareness of the truth about another. [Reply:] Only a certain false awareness counts as delusion; not just any non-production of the awareness of the truth. The intentional object-a false awareness arising with respect to which becomes the origin of the cycle of rebirth-is to be apprehended truly." (kiṃ nu khalu bhoḥ yāvanto viṣayās tāvatsu pratyekaṃ tattvajñānam utpadyate, atha kvacid utpadyata iti| kaś cātra viśeṣaḥ? na tāvad ekaikatra yāvad viṣayam utpadyate jñeyānām ānantyāt| nāpi kvacid utpadyate, yatra notpadyate tatrānivṛtto moha iti mohaśeṣaprasaṅgaḥ| na cānyaviṣayeṇa tattvajñānenānyaviṣayo mohaḥ śakyaḥ pratiṣeddhum iti|mithyājñānaṃ vai khalu moho na tattvajñānasyānutpatti-mātram| tac ca mithyā-jñānaṃ yatra viṣaye pravartamānaṃ saṃsāra-bījaṃ bhavati sa viṣayas tattvato jñeya iti|) 62 NB 250.10-16: anātmany ātma-grahaḥ, aham asmīti moho'haṅkāra iti| anātmānaṃ khalv aham asmīti paśyato dṛṣṭir ahaṅkāra iti| kiṃ punas tad artha-jātaṃ yadviṣayo'haṅkāraḥ? śarīrendriya-mano-vedanābuddhy-ādayaḥ| kathaṃ tad-viṣayo 'haṅkāraḥ saṃsāra-bījaṃ bhavati? ayaṃ khalu śarīrādy-arthajātam aham asmīti vyavasitaḥ tad-ucchedenātmocchedaṃ manyamāno 'nucchedatṛṣṇayā pariplutaḥ punaḥ punas tad upādatte| tad upādadāno janmamaraṇāya yatate| tenāviyogān nātyantaṃ duḥkhād vimucyata iti| events, etc., we take our own existence to be tied up with their existence. Now, we don't want our own destruction. This, presumably, is explained by our desire for pleasure: if we want pleasure and if our own existence is our only means of getting it, we should want our own existence to continue. Then, given that we take ourselves to be our body, etc., we also want the body, etc. not to be destroyed. This gives rise to thirst, i.e., the desire for embodied existence, which causes us to be born again and again.
Consider now the second example: lust or sexual desire. In this case, the desire arises from a failure to see the object of desire for what it is. In his commentary on NS 4.2.3-"The cause of those [defects] is an erroneous awareness with respect to partite objects" (tan-nimittaṃ tv avayavy-abhimānaḥ)-Vātsyāyana writes: The cause of those defects, then, is an erroneous awareness with respect to partite objects. Indeed, in men, that consists in the label "woman", along with all its dressings (sa-pariṣkara). And in women, it's the label "man." In contrast, the dressing consists in a label based on marks (nimitta-saṃjñā) and a label based on secondary characteristics (anuvyañjana-saṃjñā). A label based on a mark is of the following sort: "tongue" and "ears", "teeth" and "lips", "eye" and "nose." A label based on secondary characteristics takes the form, "The teeth are like this," "The lips are like this." Since this very label enhances the attachment that is lust, and gives rise to defects that are connected to it, it is to be abandoned. 63 The idea seems to be this. Sexual desire arises from applying certain concepts or labels to certain material objects. The person may apply the label "man" or "woman" to a certain body, along with a number of other concepts. 64 Uddyotakara explains: Moreover, even in that case, there are two labels: a label based on a mark and a label based on a secondary characteristic. A label that is based on a mark is of the following sort: "teeth" and "lips," etc. The label that applies to teeth, etc. on the basis of tooth-hood, etc. is a label based on a mark. And, similarly, a label that is based on a secondary characteristic involves the projection (adhyāropa) of something on to the mark in the form, "It is thus": "The teeth are thus," "The two lips are thus," etc. This very label, which is cultivated by means of a projection, is a delusion. 65 Some of these may be labels based on marks (nimitta) of a man's or a woman's body, e.g., their teeth, their lips, etc., which the subject identifies on the basis of their defining characteristics or marks, such as tooth-hood. But some of the others may be labels based on secondary characteristics (anuvyañjana). For example, one takes one's lover's teeth to be a string of pearls, the application of the concept "string of pearls" will be a label based on a secondary characteristic, i.e., a similarity with other objects. The application of these latter labels, as Uddyotakara notes, always involve delusion, since they involve projection of properties that are in fact absent from the relevant objects. Since these labels allegedly create sexual desire, they are to be abandoned.

Cultivating Labels
How can we get rid of these kinds of attachment and aversion? In his commentary on NS 4.2.2, Vātsyāyana explains the process as follows.
First, meditate on them [the objects of the senses]. And for one who reflects on them, the false saṅkalpa about colour, etc. ceases. When that ceases, reflect on the body, etc. in relation to the self. From the meditative awareness (prasaṃkhyāna) of that, the ego-construction with respect to oneself ceases. This person, who roams with a mind that detached from himself and the world outside, is said to be free. 66 In his explanation of this passage, Uddyotakara says that our ordinary attachments towards our own material possessions are based on the thought that there is something special about these material possessions insofar as they belong only to us. We think, "These are just mine." But the awareness that undermines that false judgement has the content: "These are not mine; they are shared just like something to be used for purposes of worship, or something that is stolen, or fire, or something that is inherited, etc." 67 The thought, then, is that, by discovering the nature of objects of the senses, we can get rid of some of our attachments. However, in order to undermine other attachments e.g., the thirst for rebirth, we will have to meditate on the nature of the body, the senses, the manas, etc. Here, we need a different kind of meditative awareness. As Uddyotakara puts it, "The meditative awareness is the perception of distinctness with respect to the body, etc. in the form, 'These are not selves. '" 68 In relation to lust, Vātsyāyana offers the following advice. We can overcome the negative effects of the labels that give rise to sexual desire by cultivating foul labels (aśubha-saṁjñā). 69 Foul labels raise to salience certain features of the body that elicit disgust.
However, their abandonment consists in a label based on parts by means of distinguishing them, e.g., a label such as "hair on the head", "body-hair", "flesh", "blood", "bones", "ligaments", "veins", "phlegm", "bile", "excrement", etc. That label is said to be a foul label (aśubha-saṁjñā). A person who cultivates it loses the attachment that is lust. 70 By cultivating foul labels with regard to the body, one comes to see the body really to be the foul object that it is, i.e., a mass of hair, flesh, blood, bones, ligaments, veins, etc. Since these labels elicit disgust instead of lust, they can serve as the antidote to lust.
Interestingly, the work that foul labels do in undermining lust can be carried out by the label of pain with respect to thirst. As we saw earlier, Vātsyāyana thinks that we should apply the label "pain" to all aspects of embodied existence in order to induce a motivational shift, such that we first become disinterested with respect to all the realms, and then lose our thirst with respect to all the realms. Recall what he says in his commentary on NS 4.1.55: In this way, for a person who perceives these places of production [i.e., beings born with a body, etc.] as connected with pain, the label (saṃjñā) of pain is established with respect to pleasure and with respect to the body, the senses, and the awareness-events that serve as a means to pleasure. Due to the establishment of the label of pain, there will be a label of disinterestedness (anabhirati) with respect to all the realms. For the person who devotes himself to the label of detachment, thirst with respect to all the realms is broken. Due to the abandonment of thirst, he is freed from all pain. 71 This passage suggests that the primary function of applying the label of pain to all aspects of embodied existence is to undermine a specific kind of attachment, namely thirst. Vātsyāyana in fact revisits the point again in his preamble to NS 4.2.1. After arguing that the conflation of the self with the body and so on generates thirst, he says: However, he who perceives pain, the basis of pain, and the pleasure that is connected to pain in the form, "This is all pain," is comprehensively aware of (parijānāti) pain. And the pain, when comprehensively apprehended (parijñāta), ceases. For it isn't obtained, like food mixed with poison. In this way, he perceives the defects and the karmic factors as causes of pain. Moreover, since it isn't possible to destroy the uninterrupted series of pains unless the defects cease, he avoids the defects. Furthermore, when the defects have ceased, it has been said, "there is no practical undertaking for the sake of rebirth (pratisandhāna) in a being whose defilements have been destroyed" (NS 4.1.64). 72 The lesson is supposed to be this. Just as the foul labels are useful for undermining the false awareness that underlies sexual desire, so also the label of pain is useful for undermining a number of errors about pain. An ordinary human being labours under the misconception that she is her body, her senses, etc. Insofar as she desires her own continued existence, she desires to be born again and again. This is thirst. However, a person who is subject to thirst misunderstands the nature of the pain that she undergoes. She takes the pain that necessarily accompanies embodied existence to be a necessary evil that she must accept in order to obtain the pleasure that she truly values. As we saw in our discussion of the arguments for the Pain Principle, either this person fails to realize how our attachment to pleasure gives rise to pain that outweighs the pleasure in variety and amount, or how our experiences of pleasure as pleasure gives rise to more pain, or how our attachment to pleasure misleads us into thinking of pain as pleasure. However, applying the label "pain" to all aspects of embodied existence serves as an antidote to all these errors. This, in turn, gives rise to a comprehensive awareness (parijñāna) of pain. This ultimately undermines our desire for rebirth.
The contrast between sexual desire and thirst lies in this. While the antidote to sexual desire requires us to focus on foul labels that correctly describe the objects of sexual desire, the label of pain doesn't correctly describe all aspects of embodied existence. In fact, Vātsyāyana himself thinks that the application of the label of pain Footnote 71 continued anabhirati-saṁ jñā bhavati| anabhirati-saṁ jñām upāsīnasya sarva-loka-viṡayā tṙṡṅā vicchidyate, tṙṡṅāprahāṅāt sarva-duḣkhād vimucyate iti| 72 NB 250.15-yas tu duḥkhaṃ duḥkhāyatanaṃ duḥkhānuṣaktaṃ sukhaṃ ca sarvam idaṃ duḥkham iti paśyati sa duḥkhaṃ parijānati| parijñātaṃ ca duḥkhaṃ prahīṇaṃ bhavaty anupādānāt saviṣānna-vat| evaṃ doṣān karma ca duḥkha-hetur iti paśyati| na cāprahīṇeṣu doṣeṣu duḥkha-prabandhocchedena śakyaṃ bhavitum iti doṣān jahāti, prahīṇeṣu ca doṣeṣu na pravṛttiḥ pratisandhānāya hīna-kleśasya (NS 4.1.64) ity uktam| My translation of "pratisandhāna" as "rebirth" might seem strange, but this follows Uddyotakara's gloss under NS 4.1.64 : "However, pratisandhi is being born again when the previous birth has ceased" (pratisandhistu pūrva-janma-nivṛttau punar-janma|) to all aspects of embodied existence involves treating something that isn't literally pain as pain. So, this is simply meant to be a useful delusion that can undo the foundations of our attachment to pleasure. This suggests that, on this view, a false awareness can be a guide to the truth about pain.
The upshot is this. Vātsyāyana believes that some of our desires-which are either forms of attachment or of aversion-are invariably based on some false awareness which obscures or conceals the true nature of the object of desire. The only way to get rid of them is to meditate on (i.e., repeatedly think about or focus on) these objects by means of certain labels. By manipulating the labels using which we think about such objects, we can get rid of those desires.

Conclusion
In this essay, I have explained Vātsyāyana's solution to a problem that arises for his theory of liberation. For him, liberation is a permanent state of disembodiment: it involves absolute cessation of pain, but, as a result, also involves absolute cessation of pleasure. The problem was this: How can agents like us-who habitually seek pleasure-be rationally motivated to seek liberation? Vātsyāyana's solution depends on what I called the Pain Principle, the principle that we should treat all aspects of our embodied existence as pain. If we follow this advice, we will come to apply the label of "pain" to all aspects of our embodied existence, including pleasure. This in turn is supposed to undermine our attachment to our own embodied existence. As I went on to argue, this fits with a general theory of human motivation that Vātsyāyana defends, namely that, by manipulating the labels using which we think about the world and ourselves, we can induce radical shifts in our patterns of motivation.