Around 900 CE, somewhere in South India (or perhaps Kashmir — historians disagree, Selvalingam thinks not Kashmir despite his work being popular there), Sadyojyoti — also called Khetapāla or Khaṇḍa-Sadyojyoti — sat down to write the founding philosophical works of Shaiva Siddhānta.1 He produced the Mokṣakārikā (~900 CE, later commented on by the great South Indian master Aghora Śiva), the Bhogakārikā, the Tattvasaṃgraha, the Nareśvarapariksā, and a text Selvalingam calls his favorite: the Paramokṣa-nirāsa-kārikā — "Verses Refuting [the False Views of] Supreme Liberation."2
The text is a tour through the rival theories of liberation circulating in the Shaiva world at the time. Sadyojyoti refutes them one by one. The point of the refutations is not to win arguments — it is to set up a positive doctrine that the refutations clear space for: the soul is innately divine, and liberation is recognition of what was always already the case.
The three theories Sadyojyoti targets are still live theories today. Each one fails for the same underlying reason: each treats liberation as something gained, and what is gained can be lost.
The first view: liberation is the arising of qualities like omniscience and omnipotence as a result of liberation. Before liberation, you didn't have these qualities; after liberation, you do. The liberation itself is the moment of acquisition — you gain something you didn't have, and that's what makes you liberated.
This view marks an explicit break from the older Pāśupata position (and from the Buddhist Theravāda position before that), which held that liberation meant only duḥkha-nivṛtti — the cessation of suffering, with nothing positive added. Lākulīśa and some other later Shaivas argued the negative-only view was inadequate: liberation must mean the attainment of something, not just the removal of something.3
Sadyojyoti's refutation is structurally the Buddhist anti-state argument. Anything that has a beginning must have an end. If you didn't have omniscience before liberation and you have it after, then your omniscience has a starting moment. Anything with a starting moment is in time; anything in time is impermanent; impermanent omniscience can be lost. How can it be liberation if it's a state that comes and goes?4
The Buddhist comfortable with this: yes, exactly — that's why Nirvāṇa cannot be the gaining of anything. If your liberation is merely a state, it must pass, like all states. The Paramokṣa-nirāsa-kārikā uses this Buddhist logic against the Lākulīśan Shaiva to clear the ground for a different positive doctrine.
The second view, held by Lākulīśa and his lineage: liberation is like fragrance. Get close enough to a fragrant object and the fragrance transfers to you. Similarly, by getting closer and closer to Shiva through devotion and discipline, his qualities of omniscience and omnipotence transfer to you. The liberated being doesn't generate these qualities from nothing — they are transmitted from the source.5
Sadyojyoti has three refutations of this:
First, the Theory-1 critique still applies. If liberation is something transmitted to you, it is still something you didn't have and now have — a beginning, therefore an end. Contagion can be lost as easily as it can be gained. You moved close enough to catch the fragrance; you can move away and lose it.
Second, the metaphor itself is incoherent. They criticize it by saying it's kind of incoherent — the fragrance metaphor doesn't translate into a clean account of how the qualities transfer. What is the actual mechanism? The text demands more than the metaphor delivers.
Third — the elegant refutation — in every possible metaphor of this transference, the transmitter loses the quality and the transmittee gains the quality. Take heat: when you transmit heat from a hot object to a cold one, the hot one cools and the cold one warms. Transmission is zero-sum in the donor-recipient relationship. So if Lord Shiva is giving omniscience and omnipotence to the soul by transmission, then Shiva is also losing those qualities to give them. And how can that be true of Shiva, who is the source of the qualities and cannot diminish?6
The fragrance theory doesn't survive once you press the transmission mechanism. Either Shiva loses what he gives — which is impossible — or the transmission metaphor is actually doing something else (which the theory doesn't articulate).
The third view, common in temple traditions and especially Pentecostal-style devotional Shaivism: liberation is being possessed by Shiva. The way someone in a Tantric trance can be possessed by a deity, the liberated being is permanently possessed by Shiva. In many temples that's the goal — to be possessed by the deity being worshiped.7
Selvalingam notes the Paramokṣa-nirāsa-kārikā engages this view directly because it remains a live option for many devotional Shaivas. The refutation:
Possession is not freedom. In āveśa, you are being possessed by a deity. Your own agency is displaced; the deity's agency operates through you. So you have lost freedom, not gained it. In possession you don't gain freedom, you lose freedom. See, God is free. If God becomes free in me, am I free? No — I am just now a vehicle for God's freedom. I have not attained liberation actually.8
This is the cleanest of the three refutations. Mokṣa means freedom. Possession means subordination of one agency to another. So possession cannot be mokṣa, no matter how exalted the possessing entity. The liberated state must be the soul's own freedom, not the soul as a vehicle for someone else's freedom.
Having cleared the ground of the three competing theories, Sadyojyoti proposes the positive view that defines Shaiva Siddhānta: the soul is innately divine. It always had the qualities of omniscience and omnipotence. These are not things it attains as a result of mokṣa — it has them already.9
This is what survives all three refutations:
What we lack is not the qualities themselves but the recognition of them. Selvalingam's metaphor: cataract surgery. The eye is fundamentally healthy — it can see. The cataract is what blocks the seeing. Liberation is the removal of the cataract (called mala — impurity), not the addition of vision. The soul is innately pure; it's only a matter of removing the obstruction that prevents you from seeing that. Once the obstruction is removed, the innate omniscience and omnipotence of the soul shines forth.10
This is the Shaiva Siddhānta version of what later non-dual Shaivism will name as pratyabhijñā — recognition not attainment. The structural argument is the same on both sides of the dualism/non-dualism divide. The dualist's recognition leaves the soul as a paśu permanently distinct from Shiva but innately divine; the non-dualist's recognition collapses the soul-Shiva distinction. Both refuse to call liberation an acquisition.
In the fuller telling, Selvalingam labels the three rivals by their schools and runs each to a one-line death. The first is the Naiyāyika move: omniscience and omnipotence arise at the moment of liberation, switched on like a light that wasn't burning before. Refutation: what begins burning can go out — a quality that starts in time ends in time, so it can't be the deathless thing liberation is supposed to be. The second is the Sāṅkhya / Lākulīśa fragrance-transmission move, and Selvalingam doubles the metaphor to make it bite: it works like scent and like heat — but in both cases the giver is depleted. The rose spends its scent; the fire spends its heat into the cold pot. So if Śiva's qualities cross over into the soul, Śiva is drained of them, which is absurd for the source. The third is āveśa, divine possession, and it dies on a single question: possession is loss of freedom, not gain of it. "If God becomes free in me, am I free? No — I'm a vehicle for God's freedom." The possessed devotee is a car God drives, not a free agent.11
What's left standing after all three fall is one positive claim, stated flatly: omniscience and omnipotence are innate. They were never switched on, so they can't switch off; never transmitted, so the source loses nothing; never on loan from a possessor, so the freedom is the soul's own. Liberation gains the soul nothing it didn't already hold — it only takes away the film over the eye. This is the recognition-not-acquisition habit running at the level of mokṣa itself: the work isn't to get the divine qualities but to stop mistaking yourself for something that lacks them.
Sadyojyoti's three refutations all rest on the same logical move: anything you can gain you can lose. Apply this filter to any liberation theory and you get a forced choice — either the liberation is what you already are (and the work is recognition), or the liberation is impermanent and therefore not actually liberation. There is no third option once the filter is applied.
This filter explains why the major Indian liberation traditions — Shaiva Siddhānta, Trika, Advaita Vedānta, Mahāyāna Buddhism, Dzogchen — all converge on some form of recognition not acquisition doctrine, despite their disagreement on what is being recognized. The convergence is not coincidence. It is what survives once the anything-gained-can-be-lost filter eliminates the alternatives.
The same structural move appears in modern depth-psychology lineages, especially Jungian individuation and Internal Family Systems work. The Self in Jung is not produced by analysis — it is recognized through analysis; the work is removing the complexes, projections, and persona-identifications that obscure what was already there. IFS treats the Self as an innate competent center that exists prior to and independent of all the parts; therapy is parts-unburdening that allows the Self to lead.
Both frameworks are structurally Sadyojyoti's move. What you fundamentally are cannot be the product of the therapeutic process, because anything the process produces can be undone by another process. The therapeutic gain that survives must already have been there. The work is removal of obstruction (cataract surgery), not construction of capacity.
The handshake produces an insight neither domain generates alone: the recognition model in Indian liberation theory and the innate-self model in depth psychology both depend on the anything-gained-can-be-lost filter. Neither could survive without that filter; both arrive at it independently. This is one of the cleanest cross-civilizational convergences in the recognition vs. acquisition question.
Sadyojyoti's distinction between innate divinity and being possessed by a deity maps onto a real structural distinction in influence and bonding patterns. The follower who is told they are innately divine and the work is recognition retains agency — they can leave, dissent, refuse, recover. The follower who is told their liberation is possession by the master/deity/group has been structured into a relationship where their agency is on loan; leaving means losing the freedom-on-loan.
This is not a coincidence — it is why coercive religious groups and high-control movements consistently teach possession-style liberation models. You need us / the master / the group / the practice or you'll lose what you have. The possession model is what makes the follower structurally retainable. The recognition model is what allows the practitioner to graduate from any specific teacher.
The handshake to the cult-recruitment corpus: Sadyojyoti's refutation of āveśa-pakṣa is doing philosophical work that has direct implications for what makes a religious tradition recruitment-stable vs. coercion-stable. Recognition-doctrine is institutionally weaker (the practitioner can leave) but ethically cleaner. Possession-doctrine is institutionally stronger (the practitioner stays bound) but ethically compromised. Sadyojyoti's argument that possession cannot be mokṣa is also, structurally, an argument against coercive religious institutions — though Sadyojyoti himself frames it as a purely philosophical point.
Selvalingam clearly admires the Paramokṣa-nirāsa-kārikā — calls it his favorite Sadyojyoti text. But he also has a position that complicates the unqualified endorsement: in the same lecture he is arguing for religious pluralism (all schools as equally valid readings of one reality). The Paramokṣa-nirāsa-kārikā is a refutation text — it tells you that three other liberation theories are wrong. How can a pluralist endorse a refutation text?
The most defensible reading: Sadyojyoti's refutations operate at the internal coherence level, not at the which school is right about reality level. The fragrance theory is internally incoherent because the transmitter would lose what the transmittee gains is a philosophical argument that any pluralist can accept — it doesn't say the Lākulīśans are wrong about ultimate reality, it says the fragrance metaphor doesn't survive its own logic.
This is the clearest example of how pluralism can take philosophical critique seriously without endorsing the ranking conclusion the critique seems to support. The critique reveals an internal incoherence; the pluralist response is that specific account doesn't work, find a better account within that tradition rather than therefore that tradition is wrong.
Whether Sadyojyoti himself would have accepted this reading is unclear. He probably intended the refutations to be ranking moves, since he was building Shaiva Siddhānta's positive doctrine in their place. The pluralist appropriation of the refutations strips off the ranking.
Innate-divinity vs. eternal-distinction. Shaiva Siddhānta's positive doctrine has an internal tension Selvalingam doesn't fully unpack. The soul is innately omniscient and omnipotent — but the soul also eternally remains distinct from Shiva. Why doesn't the soul's innate omniscience and omnipotence make it identical with Shiva, who also has these qualities? The Shaiva Siddhāntin must hold that being innately divine is consistent with being not-Shiva. The non-dual Shaiva will press: if the soul has all the qualities of Shiva, what makes it not Shiva?
The cataract metaphor cuts both ways. Selvalingam's "cataract surgery" image works smoothly for the recognition-not-attainment doctrine — the eye is innately healthy, surgery removes the obstruction. But the metaphor also implies that the cataract is real. Mala is not illusory; the bondage is genuine; the obscuration genuinely blocks recognition until removed. This is what makes Shaiva Siddhānta a dualist recognition doctrine rather than a non-dualist one. The cataract model preserves the seriousness of bondage that some non-dualist accounts (where bondage is illusory) collapse.
Sadyojyoti's third refutation — possession is not freedom — has direct implications for how the contemporary spiritual marketplace operates. Many present-day teachers offer transmission, possession, or śaktipāt as the central liberation event. Sadyojyoti's argument: even if these phenomena are real, they are not mokṣa. They are the deity becoming free in you, which is structurally distinct from you becoming free.
The harder implication: any spiritual practice that requires the continued presence of the teacher / lineage / institution to sustain the realization may be giving the practitioner āveśa (possession) rather than mokṣa (freedom). The test Sadyojyoti's argument hands the practitioner: if I leave the teacher, the lineage, the institution — does the realization survive? If yes, it's the soul's own freedom (mokṣa). If no, it was the teacher / lineage / institution being free through the practitioner (āveśa). The latter is real but it is not what the soul came for.