Śrīvidyā practices and the nature of initiatory experience

Gopinātha Kavirāja, a twentieth-century scholar of Sanskrit texts and a well-known Tantric practitioner, describes the process of initiation as follows:

An author in the Śrīvidyā tradition, Lakṣmīdhara, describes the experience of initiation as followsFootnote ¹:

Bhāskararāya, another figure within Śrīvidyā, writes, ‘The Guru, looking into the sight of the disciple with his own sight, through the powers of Yoga, draws out his own awareness through his eyes and through the eyes of the disciple, with himself as the awareness of the disciple, pierces his mind’ (Bhāskararāya 1908, 156).

Such experiences are central to the Śrīvidyā tradition. Śrīvidyā is a Tantric religious tradition where the primary deity is Goddess Lalita Tripurasundari. She is regarded as the active aspect of the divine.Footnote ² Initiation in Śaiva, Śākta, or Tantric traditions can occur in different ways.Footnote ³ Often, it will involve at least the whispering of the mantra into the ear of the discipleFootnote ⁴ (as we saw above in Lakṣmīdhara’s discussion). Sometimes, it may involve the initiator (the guru) looking intensely into the eyes of the disciple or initiand (as in Bhāskararāya’s example) (Gonda Reference Gonda1965, 431). In some kinds of initiation, we are told that ‘A person … realizes at once the oneness of his own self with that of the teacher, the mantra and the deity. He becomes “the very self of Śiva (God)”’ (Gonda Reference Gonda1965, 443).

One crucial aspect in Śrīvidyā initiatory experiences is best expressed as the initiand ‘sharing’ the experience of the guru, as in the accounts above.Footnote ⁵ Sometimes it can be difficult, especially for those not initiated into such traditions, to make sense of the phenomena reported during initiation in Śrīvidyā. The accounts sometimes seem to describe some kind of a supersensuous conduit which allows access to the experiences of the other person. The initiand seems to be able to literally access the experiences of the guru during the process of initiation. Such access, on the face of it, goes against our entrenched view of the scope and limits of our psychological powers.

Nevertheless, the presence of such experiences has been documented within anthropological reports. One example is Fred Smith’s work on possession in South Asia. Smith suggests that ideas of possession are taken seriously in contexts where the self is considered to be ‘porous’ (Smith Reference Smith2006, 594). In Śrīvidyā, too, there seems to be a kind of porosity between two people – a guru (initiator) and the disciple (initiand). In these initiatory experiences, the guru is able to ‘transmit’ some experiences to the disciple. When we read these accounts, they seem to point to what Reich and Williams observe: In certain Goddess-based lineages, selves are ‘porous and interpenetrating’ (Reich and Williams Reference Reich and Williams2024, 163). Śrīvidyā – the tradition from which our opening examples come – is associated with some of the traditions Reich and Williams have in mind.

From the study of Śrīvidyā initiatory experiences, the following can serve as a sufficient description of the process of initiation as it appears in these accounts:

1. a) There are two subjects – guru (initiator) and disciple (initiand).

2. b) The whole framework of what transpires between the initiator and the initiand can be understood or explained as part of the practice of Śrīvidyā.Footnote ⁶

3. c) Either as an effect or as part of the process, the guru and the disciple can be said to undergo the same token experiences together.Footnote ⁷

4. d) To possess experiences together entails that somehow the selves of the guru and the disciple are porous in the sense that even if it is for a brief moment, the disciple experiences or has transparent accessFootnote ⁸ to the different instantiated states of the guru in a real way.

1. 1. a)–d) have been testified by practitioners and third-person accounts (insofar as it is possible) to be an adequate account of what transpires within initiatory rites. We suggest that c) and d) entail the possibility of subject combination (SC). By SC, we mean that it is possible for either a) two subjects to combine to be one subject or b) one subject to be subsumed or become a part of another subject. We will explain and defend SC in the section ‘Subject unity and unity of consciousness’.

In what follows, we will discuss two issues concerning the initiatory experience: One, the epistemic value of the initiatory experience, and two, an account of subjects or consciousness where such a sharing of token experiences is possible.

The first issue arises because, given everything we know about consciousness and the limits of first-person experiences, initiatory experiences may sound counter-intuitive or, at best, a façon de parler for a phenomenon that can be explained through psychological or physical conceptual apparatuses. These experiences – especially how the initiand can ‘transparently’ hold within herself the experiential properties or events of another person (initiator) – seem to transcend the ambit of our ordinary perception. If at all they are ‘real’, they must be supersensuous. Therefore, to argue that such experiences are possible, we must first defend the epistemic value of some kind of supersensuous perception. Religious literature often talks about experiences which cannot be explained in the same terms as our ordinary everyday experiences. In fact, such literature often acknowledges that such experience is supersensuous. Psychological and neurological studies have also demonstrated that these experiences are not of the garden variety, and if they exist, they cannot be explained reductively using psychological concepts.Footnote ⁹

The second issue is that even if we can establish that these supersensuous experiences have epistemic value, it does not necessarily follow that there is real sharing of token mental states between two subjects. This is to say that even if they have epistemic value as to their content, it does not automatically entail that such sharing is real. It could be that these experiences were based on something entirely different, and Śrīvidyā authorities are mistaken about what they understand to be a sharing of token mental states. This seems like an attractive option, especially because such sharing goes against much of what we intuitively understand as constituting the range of conscious phenomena.

To address these issues, we will begin by presenting an argument for the epistemic validity of initiatory experience. Through this argument, we will suggest that initiatory experiences can give credence to the possibility of sharing token experiences. The idea seems to run counter to our basic intuitions about subjects and consciousness. However, through our examination of some arguments on the unity of consciousness, we will demonstrate that the positions which ground our common-sense intuitions have some problems. By pointing to those problems, we will argue that our intuitions against SC are mistaken. If that is indeed the case, then we need not write off the possibility of sharing token experiences.Footnote ¹⁰

In the next section, we will present an argument to prove the possibility of such experiences and to make a case for giving such initiatory experiences epistemic consideration.

Allowing for Śrīvidyā experiences

Śrīvidyā involves a soteriology and a metaphysics regarding the ultimate constituents of the world, including subjects, as well as a set of ritualistic practices. For our purposes, we will stick to only one phenomenon within the Śrīvidyā tradition: the initiatory experience. To be more specific, we will only look at what we might call the ‘experience-sharing’ aspect between the initiator and the initiand within the Śrīvidyā tradition. As we have noted, this ‘experience-sharing’ feature entails that there is subject-sharing or SC (we use the two terms interchangeably) in a way that is to be explicated yet. But we assume a ready sense of the idea of subject-sharing. In the last section, we suggested that this kind of instantiated experiences of porous selves is not ordinary and thereby requires a supersensuous capacity. The reason why we want to acknowledge the supersensuous nature of such experiences is to do justice to the specialized form (or context) through which such experiences are possessed.

Firstly, we will present an overall argument to show that such supersensuous perception is possible. For this, we will use Swami Medhananda’s master argument for the epistemic value of supersensuous experience (AEV). We will then adapt this argument to show that in initiatory experiences, supersensuous experience is modally allowed and epistemically valid. To do so, let us first present some background information that will help us set up the argument. Different systems within classical Indian philosophy have different understandings of whether we should, by default, regard a cognition as pramāṇa (source of knowledge) or not. Medhananda holds (drawing partially on the Mīmāṃsā system) that we should regard cognitions as pramāṇa unless there is reason to believe otherwise. In his defence of supersensuous perception, he focuses on two pramāṇas in particular: perception and testimony (Medhananda Reference Medhananda2022, 163–170).

In response to the question ‘If God exists, is it possible for us to know God?’ Medhananda says ‘Yes’. To justify this, he uses the following argument, which he calls AEV. It runs as follows:

The above argument is intended as proof for mystics’ experience of God as having epistemic value. We take this argument at face value, and we refer the reader to Medhananda’s defence (Medhananda Reference Medhananda2022, chapters 5 and 6). We will adapt it to the case of Śrīvidyā initiatory experiences. There are some self-evident parallels. In the case of the initiatory experience, too, perception and testimony can be used as primary means of cognition. Initiation is a form of supersensuous perception, like experiencing God. If the parallel exists as we think it does, the following argument should hold.

1. 1. Śrīvidyā initiators and initiands exist.

2. 2. Experiences during initiation are perceptual in nature,Footnote ¹² since they are sufficiently similar to mystical experiences (already established as perceptual in Medhananda’s argument).

3. 3. Perception as Epistemic Justification (PEJ): When a subject S has a perceptual experience as of F, S is thereby rationally justified in believing that she has a veridical perceptual experience of F, unless she has good reasons to doubt the veridicality of her perceptual experience as of F.

4. 4. At least some Śrīvidyā initiands and initiators have no good reason to doubt the veridicality of all or most of their experiences.

Therefore,

1. 5. At least some Śrīvidyā practitioners are rationally justified in believing that they are having the same experience (from 1–4).

2. 6. Perceptual Testimony as Epistemic Justification (PTEJ): People other than Śrīvidyā practitioners are rationally justified in believing that sets of initiators and initiands have the same experiences, unless they have good reasons to doubt their sincerity or trustworthiness.

3. 7. People other than Śrīvidyā practitioners have no good reasons to doubt the sincerity or trustworthiness of all or most sets of initiators and initiands who claim to have the same experience.

Therefore,

1. 8. People other than Śrīvidyā initiators and initiands are rationally justified in believing that at least some initiators and initiands have had at least some identical experiences (from 5–7).

It might be argued that premise 4 above, unlike in Medhananda’s argument, is itself in need of support. If so, it cannot be taken for granted as a premise. After all, unlike a mystical experience, whose validity is predicated on whether the experience is revelatory of a supersensory or transcendental realm, the point of initiatory experiences is not whether it is revelatory of a different transcendental realm or not. At least, we are not making such a claim in our paper. Moreover, in our argument, premise 4 hints at a kind of a subject-melding or combining. If such melding or combining is logically incoherent, one could easily say that the initiator and initiand do, in fact, have good reason to doubt the veridicality of the initiatory experience. Later in the paper, we will address such incoherence through theoretical tools available within consciousness studies and philosophy of mind.

Building a framework

In the last section, we used Swami Medhananda’s argument to defend the epistemic value of initiatory experiences. However, Medhananda’s argument does not necessarily entail that the experience that the guru and disciple have is token-identical. In this section and the next, we will work towards making room for a framework that can accommodate the sharing of token experiences. By token-identical shared experiences, we mean that two people have the self-same experience. In such an experience, one person has transparent access to the instantiated states of the other. We had said in the section ‘Śrīvidyā practices and the nature of initiatory experience’ that c) and d) (that there is a shared token experience, which suggests a porous self) represent a phenomenon that opens the possibility of token experiences being shared such that one person can possess the same token experience that another person has. This kind of sharing experiences between subjects in initiatory episodes is not like two people having say coffee together. They may be drinking the same coffee and having the same experience of that coffee. However, the experience of sharing coffee, which has the same object, is not token-identical. We can say it is of the same type but not literally identical. In the case of Śrīvidyā, what reports testify is that there is a sharing of the same token experience such that we can say that two subjects possess the same token experience in the same way at the same time. In the case of Śrīvidyā, two subjects retain their subject-hood while still being able to share token experiences with each other.

An example from fiction might help us make sense of this. In the later Harry Potter movies, the protagonist, Harry Potter, can ‘possess’ or ‘access’ what Voldemort, his arch-nemesis, experiences. This happens in a few different ways. The way that is of relevance to us is how Harry Potter is presented with what Voldemort is perceiving. Potter is here, seeing things as Voldemort would see. If Voldemort is seeing a cat, Potter also sees the cat as Voldemort would see it. An easy way of making sense of this is to talk about what in cinema technique is called a Point-of-View (POV) shot. In movies, to represent how a character is seeing something from a first-person perspective, the camera is directly positioned in place of the character’s eyes. Directors use techniques like helmet-mounted cameras to show what the character is seeing from their perspective. As a member of the audience, one gains an indirect way of perceiving what the character is perceiving. However, when such a technique is used, one is still aware that it is the character’s perspective that is being shown. Also, as a member of the audience, one is perceiving another character’s perception of things.

In the case of Harry Potter, especially in the Potter movies, this POV shot is used to show how Potter is seeing things from Voldemort’s perspective. In the movies, whenever Voldemort’s mental states ‘invade’ Potter’s, the camera is, as it were, mounted on Potter’s subjective location. We, as audience members, know that the camera is figuratively located within the mind of Potter. We are seeing the mental states play themselves out within Potter’s inner eye, as it were, through which he can see things as Voldemort would. As an audience member, one realizes immediately that a) Harry Potter is seeing things from Voldemort’s perspective and b) that Potter is not merely an audience member to Voldemort’s sensory field (as we would be if a POV shot was used on a character. The reason for a) is because Potter is presented as seeing things that are not in the immediate sensory vicinity of himself. Potter, as he is going about his daily business, suddenly gets a ‘vision’ of something else going on, and he sees things that are not visible to him or others. This is not a hallucinatory state either. In the movie, Potter immediately knows that what he is seeing is someone else’s perspective (Potter comes to realize that it is Voldemort’s perspective only later on, but nevertheless he immediately knows, and so do we as audience members, that it is a veridical mental state which is not indexed to his own subjective position). In the movie, we as the audience come to know that b) Potter is not merely watching in his mental screen Voldemort’s sensory field, but that he is immediately instantiating it. There is no second-order perception in Potter, unlike us, in the case of watching a POV shot. Potter can be seen to occupy the same first-person perspective. It is because he can see it as Voldemort would himself see it (in the sense of instantiating the same subjective position) that Potter is able to feel a certain kind of dissonance when he has those mental states. When we are watching a POV shot in a movie, there is no room for dissonance because we already know that what the camera is showing is not what we would see but merely represents what the character sees from their subjective perspective. Importantly, Harry Potter also undergoes, as a result of having these mental states, a temporary brain-freeze or a kind of mental discomfort (possibly a migraine) partly because he knows that these are verdical mental states that he is instantiating, but they are not his own. All of this shows that there is a kind of mind-sharing that is going on between Potter and Voldemort.

It must also be said that in both the movies and the books, Harry is still aware of being Harry, and experiences in his own body (say, the burning pain in his scar) continue, but he is also aware that he is seeing things as Voldemort does. When Harry’s recollects these experiences, he recollects them as his (i.e., Harry’s) experiences. However, Harry’s experience has the added phenomenal property of being experienced from Voldemort’s perspective. That is, he recollects them much the same way as he would recollect his experience of flying a broom, only with the added phenomenal property of having been from Voldemort’s perspective. The Śrīvidyā experience is something similar.

When we are presented with a set of initiatory experiences, there are three explanatory options. One is to deny that such kinds of sharing of token experiences are possible across subjects. We have, however, argued against this option through our defence of the epistemic value of supersensuous experience and its extension to the initiatory experience. Part of the phenomenal content of the experience is that the intiand experiences it as an experience of the initiator. Since such experiences have epistemic value, we cannot dismiss them out of hand. The second explanatory option is to say that if it is the same token experience possessed across two subjects, then there simply are not two subjects. Rather, there is only one subject. One may argue that it is a case of a split-brain or schizophrenic personality where one part of the personality is occluded from the awareness of the other such that ‘both persons’ could be having the same experience at the same time. If this happens, we can say two subjects possess the same token experience. However, unlike the Schizophrenic patients, the initiator and initiand are clearly two subjects with two different causal histories. In the case of the Schizophrenic patient, the life history is the same. In contrast, in Śrīvidyā, the initiator and initiand have led different lives; they exist in different places both before and after the initiatory experience, and are clearly two different people. Importantly, these two people are embodied beings with different nerve centres and brains and so on.

This leaves us with the third possibility, which is to affirm that token experiences can be shared across two subjects.Footnote ¹³ This is possible only if we affirm that, at least during the event of such sharing, the two different subjects combine to form one subject. In other words, it is the possibility of SC that entails the possibility of such sharing of experiences. Our claim is that within the Śrīvidyā tradition, subjects combine with each other.

Let us first present how we understand combination. The idea of combination can be understood from different perspectives. For the purposes of our discussion, we will try and interpret it from a metaphysical perspective. In that sense, the idea of combination denotes how one property is grounded fully in another set properties.Footnote ¹⁴ Let us illustrate it with an example. A table is a real composite (or is combined) when the property of being a table is fully grounded in a) all the properties of the atoms that compose it, b) all the relations that obtain between the atoms, and c) the existence of the table can be fully explained through the properties of the atoms. This is not the most sophisticated illustration of combination, but this will do for our purposes. To say that subjects can combine is to affirm some kind of metaphysical combination in the way we just explained.Footnote ¹⁵

In what follows, we will first address an obvious obstacle to SC, i.e., the unity of consciousness. However, since raising issues with unity of consciousness does not by itself mean that SC is possible, we will consider some metaphysical principles which, together, seem to foreclose the possibility of SC: the principles of subsumption, ownership, and exclusivity. We propose that there are ways to undermine the subsumption and ownership principles such that SC is no longer impossible. We will tackle that in the section ‘The metaphysics of subject combination’.

Subject unity and unity of consciousness

Before we discuss these principles, we will discuss another way in which such kinds of SC can be resisted. Tim Bayne, for example, makes a case for what he calls the Unity of Consciousness thesis (We will call it the Unity thesis). According to Bayne, consciousness is one total conscious state.Footnote ¹⁶ Bayne says, ‘a total conscious state is a state that is subsumed by nothing but itself, where one conscious state subsumes another if the former includes the latter as a “part” or “component”. This total state will capture what it is like to be the subject at the time in question’ (Bayne Reference Bayne2010, 15). For Bayne, this unity of consciousness is a feature of our lives – both waking and sleeping. According to Bayne, there is a unity relation that permeates experiences such that different experiences that a subject has can be said to be unified as one. He takes this experience to be quite common. Here is Bayne:

He uses the term ‘phenomenal unity’ for such a feature of our consciousness.Footnote ¹⁷ Phenomenal unity is a primitive for Bayne and, in some sense, acts as the ground for all kinds of conscious experience for an individual. He says

Though Bayne does not discuss SC directly, we can extract an argument against SC from his remarks on the unity of consciousness. For him, whenever there are two experiences of one conscious subject, those experiences are co-subsumed. The total consciousness is homoeomerous in the sense that all the parts of that consciousness share their phenomenal quality. Bayne says that each subject has a total unified consciousness such that they each have a) a single unified phenomenal state and b) the content of that state is identical to the conjunction of each of their phenomenal parts. This would mean that each of us as subjects occupies a single phenomenal field which differs from every other insofar as parts of our consciousness cannot be subsumed into another subject’s phenomenal field.Footnote ¹⁹

If Bayne is right, then it would seem that all conscious subjects are unified insofar as they have a bounded phenomenal field. From this, it seems straightforward to conclude that SC is not possible since each subject necessarily possesses a unified phenomenal field and no two subjects can share experiences across phenomenal fields. To further see how Bayne’s Unity thesis can help build a case against SC, let us unpack a bit what his Unity thesis entails. A conscious subject (Subject₁) for Bayne would have a field of phenomenal experiences a₁, a₂ … a_n. This subject is differentiated from another subject (Subject₂) whose phenomenal experiences would contain another set of experiences b₁, b₂ … b₃. a₁ … a_n is unified and experientially accessed by subject₁. Subject₂ has a phenomenal field b₁ … b_n which is unified and experientially accessed by subject₂. Given the unity thesis, it would entail that Subject₁ can never have in her field any of the experiential tokens b₁ … b_n within her phenomenal field.

In this way, the unity thesis of Bayne can be used to resist SC. However, there are some problems with Bayne’s view that we will discuss below. Let us recall that for Bayne, the unity of consciousness is a phenomenal unity. This means that for a subject, there is something to experience a₁ and a₂ and also to experience a₁ and a₂ together (at time t), which is over and above the individual phenomenal experience of each of those experiences. For Bayne, this is supposed to be revealed through introspection. However, recent empirical work done by Peter Langland-Hassan has shown that in the case of craniopagus twins – twins conjoined at the head and the brain – it is possible for one of the twins to be introspectively aware of a mental state that is not in her own mind. One twin is called Krista, and the other is called Tatiana. It is remarkable that

Langland-Hassan also confirms that each of the girls has a cerebral hemisphere connected by a corpus callosum. Their neural overlap does not go beyond their thalami. It seems that this phenomenon, Langland-Hassan notes, forces us to keep the option alive that the twins, in fact, have immediate access to experiences that occur in the other person’s mind. If the above case is possible, then conscious experiences need not be bounded to one person such that a subject can experience another person’s first-hand experience.Footnote ²⁰

Nevertheless, one could defend the unity thesis by pointing out that even if Krista is able to access Tatiana’s experiences, Tatiana’s phenomenal token continues to be Tatiana’s and not Krista’s. If we hold that every state that occurs within one’s stream of consciousness is ipso facto a part of one’s mind and cannot be part of another mind, then we have to admit that Tatiana’s experience is her own and not Krista’s. For us to get around this, we need to show that somehow the experience that Krista is having (which in some sense ‘belongs to’ Tatiana) is both Krista’s and Tatiana’s. Let us also recall that it is part of our claim that it is possible that two people (in initiatory experiences) have the same token experience. In our interpretation, this situation exemplifies the possibility that Krista and Tatiana have the same token experience.

How does one go about establishing that Krista and Tatiana have the same token? We know that the experience token v occurs in the brain of Tatiana and not in the brain of Krista. However, Krista is aware of the experience v (as is Tatiana). In this case, what is happening is that Krista and Tatiana are both having/sharing the same token experience v, but only one of them can say that it is mine. As Langland-Hassan points out, the logical possibility that Krista and Tatiana are having the same experience can be saved by positing the idea that both of them have the same experience, though only one of them can, with justification, self-ascribe it. As he says, ‘Krista might become introspectively aware of v and erroneously self-ascribe it. That is, she might judge, through introspection, “I am having a visual experience as of a doll,” and be wrong for the sole reason that she has misidentified the owner of the state (her sister) as herself’ (Langland-Hassan Reference Langland-Hassan2015, 1746).

The phenomenon of sharing token experiences in the case of cranial twins does not mean that two people cannot have the same token experience. Two people can have the same token experience, even if they cannot self-ascribe that experience with equal justification. This means that one can say that two people can have the same experience while only one of them can own it. Thus, the case of the twins suggests that we can draw a distinction between having a phenomenal experience and owning it. Thus, Krista is able to have the phenomenal experience, which is owned by Tatiana. This means that Tatiana’s experiential token is also within Krista’s phenomenal field. What Krista cannot do is to say it is her own experiential token; this is to say, Krista cannot say it is her first-person experience. However, she can still possess Tatiana’s experiential token and thereby share experiences in the literal sense.

The metaphysics of subject combination

At this point, the task of arguing that our resistance to SC (and shared token experiences in the initiatory experience) is grounded in assumptions which are demonstrably ridden with problems has been carried out. What follows is an extra step that only makes sense within an Śrīvidyā framework. Thus, the discussion in the section ‘Subject unity and unity of consciousness’ should suffice to remove resistance to the very possibility of SC and shared token experiences.

Baker (Reference Baker, Heil and Mele1995, Reference Baker2014) argues that there are two distinct kinds of unity: unity of consciousness (or what we called above the unity thesis) and the unity of the subject. The unity of consciousness is phenomenological and dependent upon experience, while the unity of subject is dependent upon the metaphysical nature of the subject. The unity thesis talks only of phenomenal unity, which is concerned with what it is like to be in that state. According to Baker (Reference Baker2014), the metaphysical nature of subject is not always tied to the phenomenal aspect of consciousness. The metaphysics of subjects involves issues like mereology, the set of essential properties and whether a subject is a natural kind and so on. Thus, for Baker, settling the issue of the unity thesis does not necessarily settle the question of the metaphysical nature of the subject. It is possible that a subject, for metaphysical reasons, cannot combine, regardless of the verdict on the unity thesis and its consequence for SC.

In the last section, we argued that the unity thesis (i.e., accounts of the unity of consciousness) does not foreclose SC. However, it does not follow from our argument that SC is in fact possible.

If we think of the phenomenal features of a subject as the states of consciousness, then clearly the subject of consciousness need not be co-terminus with the states of consciousness. The subject could be something metaphysically different from the states that it possesses.Footnote ²¹

The following is a straightforward metaphysical argument against SC:

1. 1) Subjects are metaphysically simple.

2. 2) What is simple cannot have substantial parts.

3. 3) Combination entails that there are substantial parts.

4. 4) Therefore, SC is not possible (follows from 1–3).

SC says that subjects can combine insofar as one can be subsumed under the other. To unpack this further, let us go to Luke Roelofs. He has a good formulation for why there cannot be SC. Roelofs builds his argument for anti-combination through three premises:

1. 1. Principle of Subsumption. If there are two experiences e₁ and e₂ which are unified, then they are subsumed in another experience e₃ in which e₁ and e₂ become parts.

2. 2. Ownership principle. For any experience, there must be a subject who experiences it. In other words, if there is an experience e, it has to have a subject S of which it is a part.

3. 3. Exclusivity Principle: No token experience can be possessed by two subjects. In other words, if there is a token experience possessed by two subjects S₁ and S₂, then S₁ and S₂ are identical (Roelofs Reference Roelofs2016, 3205).

Putting these together, we get the following argument.

1. 1. When there are two experiences, e₁ and e₂, combining, there is another experience, e₃, under which are subsumed e₁ and e₂ (Principle of subsumption).

2. 2. e₃ has to have a subject. Call it S₃ (Ownership principle)

3. 3. S₃ experiences both e₁ and e₂ (Principle of subsumption).

4. 4. However, S₃ has to be identical to S₁ and S₂ (Exclusivity principle).

5. 5. Therefore, S₁ and S₂ are identical, which goes against the initial assumption (in SC) that S₁ and S₂ are not identical.

The conclusion (5) is opposed to SC because if 5) is true, then the subjects that combine are identical to each other. Clearly, that is not what a combinationist would want since she would like to show that there can be subjects who can combine but are not identical to each other. If 5) is true, combination becomes a trivial statement because that would warrant the statement that we are all combining with ourselves since we are self-identical with each other. Combination becomes just another semantic term for self-identity. For a robust SC to work, we would want S₁ and S₂ not to be identical.

Of the three premises for this argument, the ownership principle can be rejected without rejecting SC: Ownership principle is not necessary for one to hold SC. That is, it is possible to deny ownership while still holding SC. Thus, the anti-combinationist argument can work without this premise as well. We have also suggested in our discussion above that two subjects can have an experience even if only one of them owns it – that is to say, there is a difference between owning and having an experience. Thus, we can have a kind of SC even if we admit the ownership principle. Since both the combinationist and the anti-combinationist positions can stand regardless of whether the ownership principle is true, we will not attempt to address it in detail here.Footnote ²²

Roelofs deflects the argument against SC by rejecting the exclusivity principle. If we deny the exclusivity principle, it would allow for experience-sharing across (different, non-identical) subjects. This is, in fact, the tactic that Roelofs uses. Anti-combinationism gets its force from the idea that experiential properties are individuated by subjects, such that it is difficult to make sense of the idea that one token experiential property can be individuated by two subjects. However, this rule can be modified to accommodate the sharing of experiences as well. It is sometimes the case that some properties are shared by both the part and the whole. Let us say there is a physical object which is blue in colour. This colour is a physical property of the whole object and is individuated by that object (in that context). Even if it is individuated by that object, it does not mean that physical property cannot be shared by part and whole. The blueness is a property of the whole as well as the part. Roelofs can then argue that at the very least, wholes with non-discrete parts (like parts of one person’s brain) can share token experiences (Roelofs Reference Roelofs2016, 3207–3208). A similar argument could be extended to the case of the twins that we discussed above. Roelofs develops at length the argument based on rejection of exclusivity (see Roelofs Reference Roelofs2019).

However, we wish to take a route that Roelofs does not take – and which is perhaps not available to Roelofs. We are going to argue now that it is possible to admit the principle of subsumption and still preserve the idea of SC, by admitting a third, transcendent subjectFootnote ²³ in a loosely Kantian sense.

The principle of subsumption is a metaphysical principle which says that whenever there are two entities (S₁ and S₂) which share any property, they are said to be subsumed under a third entity (S₃), in which the first two entities become parts. This is problematic for SC because if we have two subjects combining with each other, they thereby become a part of a third subject. If they (S₁ and S₂) become part of a third subject, S₃, then those two subjects (S₁ and S₂) are no longer independent but become dependent parts of the third. This conclusion would undermine the idea that there can be independent subjects even after combining, ontologically speaking. However, the phenomenon of subsumption does not necessarily entail that there is a loss of independence. We must keep in mind that if we don’t specify further what the metaphysical status of this third subject is, it would be presumptuous of us to say that the third subject (S₃) can thereby completely subsume S₁ and S₂. The move from thinking that S₁ and S₂ will naturally become parts of S₃ is illegitimate without looking at the possibility that S₃ could be a fundamentally different kind of subject (from S₁ and S₂) such that the kind of subsumption that S₃ undertakes is not erosive of the independence of S₁ and S₂.

In fact, in systems like Śrīvidyā, some subjects (like us) are homogeneous when it comes to essential properties, whereas some subjects – like the subject of God – are transcendent and may share some properties with us but are fundamentally different from each one of us. Let us call this (third) subject a super-subject.Footnote ²⁴ If we take the validity of supersensuous experiences like Śrīvidyā initiatory practices, then we can at least provisionally entertain the possibility that there can be super-subjects. In fact, Śrīvidyā does posit such a subject.Footnote ²⁵ We said above that Śrīvidyā is focused on the worship of Lalitā Tripurasundarī (whom we will call the Goddess). The Goddess is part of a ‘dyadic divinity’ (which we will call God) (Brooks Reference Brooks1992, 60). According to Śrīvidyā, God transforms into the world – including all the subjects. The whole metaphysics of Śrīvidyā involves the idea that initiation is supposed to happen between guru and initiand on the one hand and the Goddess on the other hand (Brooks Reference Brooks1992, 100; Padoux Reference Padoux and White2000, 42). We cannot understand the initiatory experience fully without invoking the idea that there is a super-subject, in this case the Goddess, which is the ultimate object of their experience (Singh Reference Singh1990, 34; Flood Reference Flood1993, 28–33).

In Śrīvidyā, this super-subject is the subject of every token experience. Any token experience possessed by a human subject is also experienced by the super-subject. If we dig deeper into the metaphysics of Śrīvidyā, the only reason human subjects do not have access to all experiences that the super-subject is having is because of a self-imposed contraction – a form of metaphysical and/or epistemic contraction which stops them from accessing the super-subject. The subsumption of the subjects in a super-subject, then, is not a problem as such for Śrīvidyā, which accepts this from the beginning.

We need not examine in detail the features of this super-subject. However, given the metaphysics of Śrīvidyā, we can safely remark that in initiatory experiences – i.e., when two subjects share a token experience – there can be subsumption of those two subjects under a super-subject such that there is sharing of some minimal properties without a complete sharing of all properties. This would entail that we can have the subsumptive principle and still hold onto subject-independence.

In fact, the description of how this is supposed to work within Śrīvidyā shows that something similar is what is going on in these initiatory practices. In initiatory practices, the shared experience is subsumed under a super-subject (the Goddess) who may or may not be epistemically accessible to those two subjects. Even if it is accessible, given the metaphysical nature of the super-subject, it would follow that those two subjects do not thereby lose their metaphysical independence while being subsumed under the super-subject.

The above set of arguments, especially within the context of Śrīvidyā, should help make a good case for SC. Someone might object by saying the burden of having to accept a super-subject is too high. Our answer is that the burden is not heavy once we accept the possibility of Śrīvidyā experiences.