Posted on

Permissions, rights and adhikāra

As discussed in previous blogposts and articles, it is established that in Mīmāṃsā and Mīmāṃsā-following Dharmaśāstra all commands are dyadic; prescriptions, prohibitions and permissions are not interdefinable; permissions are always exceptions to previous prohibitions or negative obligations, and they are better-not permissions.

Permissions in Medhātithi: Two examples

Case 1:

Manu:

etān dvijātayo deśān saṃśrayeran prayatnataḥ |
śūdras tu yasmiṃs tasmin vā nivased vṛttikarśitaḥ || 2.24 ||

Medhātithi thereon:
śūdrasya dvijātiśuśrūṣāyā vihitatvāt taddeśanivāse sarvadā prāpte tatrājīvato deśāntaranivāso ’bhyanujñāyate.

So, living in another place (deśāntaranivāsaḥ) for a śūdra is permitted, if he cannot get a living where the twice-born ones live, because a śūdra is prescribed (vihita) to obey the twice-born ones. What we see is:

—the permission is a better-not option

—a specific permission is always parasitical on a general (sarvadā prāpte) prohibition or negative obligation (in this case: it is prohibited to live elsewhere, in turn depending on the duty to serve).

Case 2:

Manu:

strīṇāṃ sukhodyam akrūraṃ vispaṣṭārthaṃ manoharam | maṅgalyaṃ dīrghavarṇāntam āśīrvādābhidhānavat || 2.33 ||

Medhātithi:
puṃsa ity adhikṛtatvāt strīṇām aprāptau niyamyate | sukhenodyate sukhodyam | strībālair api yat sukhenoccārayituṃ śakyate tat strīṇāṃ nāma kartavyam | bāhulyena strīṇāṃ strībhir bālaiś ca vyavahāras teṣāṃ ca svakaraṇasauṣṭavāb- hāvān na sarvaṃ saṃskṛtaṃ śabdam uccārayituṃ śaktir asti | ato viśeṣeṇopadiśy- ate | na tu puṃsām asukhodyam abhyanujñāyate |

So, girls lack the śakti to pronounce Sanskrit words, hence they need easy- to-pronounce names. This command is taught explicitly with regard to them because of their inability, but it does not mean that difficult names are permitted for men.

Noteworthy here:

—The opponent is suggesting that F(x/y)—>P(x/¬y) —Medhātithi explains that this is wrong. It is true that P(x/¬y)—>F(x/(y∧¬y)) but the opposite is not true.

Rights and adhikāra


Having permissions just as exceptions means that they cannot be used to ground the notion of right (as in Hansson 2013). What else can correspond to “rights”?

• 1. There does not need to be a corresponding term. The deontic horizon is, like any other partition of the cognitive world, arbitrary.

• 2. There can be functional equivalents, one of which is adhikāra, I think.

adhikāra is connected to ability (sāmarthya and śakti), in the sense that unless there is ability, there is no adhikāra. Differences: adhikāra also implies duty. Contexts in which adhikāra is discussed: poor people having adhikāra, because they still have the śakti, although they momentarily lack the sāmarthya (all in ŚBh ad 6.1.1–3); disabled people lacking adhikāra for sacrifices but having adhikāra for svādhyāya (TV ad 1.3.4).

Posted on

Rights in Mīmāṃsā and further steps in mapping the deontic horizon—Updated

I have been working for years on mapping the deontic space of Mīmāṃsā authors. In order to do that, I tried to find a balance between systematicity, for the purpose of which I need as many information as possible and I often take whatever I can from whatever source, including many different authors, and historical attention to individual authors.

In order to strike this balance, I tend to assume that by default the same deontic concepts are shared by all authors, unless and until the opposite is proven, either because someone has an explicitly competing theory (e.g., Maṇḍana’s iṣṭasādhanatā) or because they de facto do something different (like Medhātithi’s approach to permissions if compared to Kumārila’s).

This ongoing attempt has led my coauthors and myself to understand some basic elements applying to all deontic concepts, such as their dyadic nature (each command is always of the form O(x/y) and not just O(x) or F(x)). We also examined the non-identity relation between prohibitions and obligations (in short: F(x/y) is not tantamount to O(~x/y), but an ideal system will not have F(x/y) and O(~x/y) ceteris paribus) and that between permissions and other commands (again P(x/y) is not tantamount to ~F(x/y), but P(x/y) presupposes a previous either O(~x/T) or F(x/T), with T being more general than y). In other words, if something is permitted it means that one would naturally be inclined to do it, but that its performance had been prohibited (or that refraining from it has been prescribed). The state of affairs of something being already available to one as something one is inclined to do is generally captured by the expression “prāpti“. So, there is a permission only with regard to something prāpta and niṣiddha (or the opposite of which has been vihita). Furthermore, the analysis of permissions as being always “better-not” permissions has also made us able to discuss supererogations, which occur when, although a permission to do x is present, one still avoids doing it.

Now, some authors have suggested that P(x) with no prior negative obligations or prohibitions is needed because of making sense of rights. This led me to think about rights and their seeming absence in the deontic landscape of Mīmāṃsā and Dharmaśāstra authors. The starting point is that we should not expect the same concepts to be present in each deontic landscape. Hence, we shouldn’t be looking for an equivalent of a given concept in a different setting. Rather, we should be looking at the different systems that each deontic author construes. In Mīmāṃsā, there are no “rights” as Euro-American authors know them (and, after all, this might be a good idea and we may suggest that the category of “rights” is muddy and unclear!), but there are other concepts that are times overlapping. One of them is the idea of adhikāra.

This is discussed at length in relation to rituals in the context of PMS 6.1 and its commentaries and sub-commentaries. Basically, Śabara etc. observe that adhikāra presupposes ability, but have then to explain that temporary inability does not infringe on adhikāra. In this context, they speak of an external (bahirbhūta) inability (like, being poor or having a sprained ankle) and an intrinsic (ātmavarttin) one, which blocks the adhikāra (like an inborn disability). They seem to oppose therefore śakti (for the intrinsic ability) and sāmarthya (that can be temporary).

UPDATE (thanks to MS for raising the issue): Another thing that we have not looked into yet are the various types of prescriptions (utpatti-, viniyoga-, adhikāra– and prayogavidhi) and how they behave deontically.

Posted on

Read more primary texts to unsettle your philosophical views

I tend to think that the history of philosophy is not (just) of historical significance, but that it is an integral part of philosophy.

I think that it would be delusional to think that philosophy of the last, say, twenty years, is, by coincidence the best philosophy ever produced by humans and that therefore exposing oneself to just those twenty years means blindfolding oneself and risking to mistake one’s prejudices for genuine philosophical “intuitions”. Therefore, I think that it is key to read texts written in different time-periods and coming from different philosophical traditions. With “texts” I mean primary texts (at least in translation), not just summaries of them in English.

To make an easy example: A contemporary scholar of, say, cosmopsychism who has never read Abhinavagupta or Utpaladeva may have heard about their philosophical positions and incorporate a paragraph on one or the other in their last article. But given that this engagement is based only on hearsay and not on a direct engagement with Utpaladeva or Abhinavagupta’s thought, the potential of Utpaladeva or Abhinavagupta’s thought to trigger a real rethinking of cosmopsychism is lost and they are used only to confirm the contemporary scholar’s own views. 

For instance, “material cause” and “efficient cause” are Aristotelian categories and superimposing them on Vedānta in an uncritical way risks leading to fundamental misunderstandings. The antidote? Reading Sāṅkhya, Nyāya and Vedānta theories of causation, so as to. understand similarities, but also differences with the Aristotelian paradigm.

There might come a time when secondary literature on, e.g., causation in Sanskrit philosophy or on specific authors will be enough. But we are not there yet. Only a few texts have been edited among the millions of manuscripts that await an edition, a small percentage of these have been translated, and even less have been properly studied. At this stage, any conclusion reached by secondary literature is preliminary at best.

Posted on

Further notes on Mīmāṃsā permissions UPDATED 11.3.2024

The following are a few random notes on permissions which have not (yet) found a place in an article.

1. A colleague wondered whether the command my co-authors and I have been discussing in several articles on Mīmāṃsā and called the “better-not permission” can be meaningfully described as permission at all. In fact, the term `permission’ in Euro-American philosophy or in Deontic Logic is strongly polysemic, covering, among others, acts that are not normed as well as acts that were previously prohibited and are permitted, and even rights.

Philosophers of the Mīmāṃsā school, by contrast, adopt the standard Sanskrit terms for permission, anujñā and anumati), but focus on only one aspect among better-not, encouraged or normatively indifferent, and use different terms for the others, thus offering a way out of the polysemy of `permissions’ (for instance, adhikāra comes close to rights, as I will discuss in a separate text).
Using the term `permission’ thus highlights a single shared aspect and suggests a way out of the polysemy of `permissions’.

2. Within the Mītākṣara commentary on Yājñavalkya, Vijñāneśvara discusses better-not permissions for Brāhmaṇas in distress. For instance, the general prohibition to steal is overrun by the permission to steal if one has not eaten for three days, provided one is stealing only from a non-brāhmaṇa (thus presupposing that the prohibition to stealing from a brāhmaṇa holds unconditionally).

The Mitākṣarā commentary further explains that stealing from a non-brāhmaṇa is permitted, provided one has not eaten for three days and one only takes enough for one meal and not additional supplies

Now, if one goes on like that for a long time, one might eventually die of starvation (because one is stealing only enough for one meal and only once every three days). Verse 44 suggests the solution (the king should take care of one), but this is not a solution one can count on in every case. Hence, Mitākṣarā verse 43 does not rule out a situation in which, in order to avoid violating the prohibition at stake (that of stealing in general), though weakened by the permission to steal from a non-brāhmaṇa if one has not eaten for 3 days,) and F(stealing from a brāhmaṇa/in general)) one ends up actually dying.

This further strengthens the point that there is no obligation to avoid starvation as the result of the general prohibition to perform any violence.

3. I use the term `command’ to cover all sorts of deontic statements. Thus, a command can convey an obligation, a prohibition or a permission. This terminology is different than the one adopted in Guhe, pp. 422–423, where `permission’ covers elective obligations. This distinction is due to the fact that E. Guhe relies on the definitions found in a late Nyāya primer, the Nyāyakośa, whereas I focus on Mīmāṃsā. It is historically interesting that later Nyāya authors considered elective rituals as permissions (they aren’t Mīmāṃsā better-not permissions, but they share with Euro-American permissions the fact that there is no risk connected with their lack of performance).

4. Permissions are not easily recognised by their linguistic form, because linguistic forms can be misleading in Sanskrit, as it can be in English (for instance, within a series of instructions, one can encounter statements in the indicative that are, however, to be analysed as commands, e.g. “One takes two cups of flour”, or “One threshes grains”). Therefore, Mīmāṃsā authors identify commands through a semantic and contextual analysis.

5. The Mīmāṃsā position is neatly distinguished from the one of, e.g., Hansson (2013), who thinks that introducing permissions even in the absence of general prohibitions is useful to define rights.
In fact, Mīmāṃsā authors deal with the category of rights differently, and I am currently working on its possible functional equivalent, namely adhikāra.

6. In general, the Mīmāṃsā analysis of commands focuses on commands occurring within a closed corpus of norms, such as the Veda, since only a closed corpus of norms provides the ideal setting for applying nyāyas.

(cross-posted on the Indian Philosophy Blog)

Posted on

Recommendation letters: Dos and Donts

So, you are a student and you want to ask a professor for a recommendation letter? Consider doing the following:

—Explain who you are (e.g., “My name is Elisa Freschi, I was in your PHL 100 and PHL 401 classes”)

—Explain what you want their recommendation for (e.g., “I am writing to ask whether you might want to write a recommendation letter for me for my application to an MA in Gender Studies at Western University”)

—Be specific: Do you need a recommendation letter or just to put down their name as a reference?

—Give all the details about yourself: Attach your transcripts, letter of intent, writing sample, cv…

—Give all the details about the position you are applying for (a graduate program? A summer course? A conference?…)

—ASK before giving the name of a certain professor as reference (it is annoying to know that you have no choice)

—Explain when the deadline is and REMIND the professor one week in advance (I have never received an automatic reminder, but I know that it looks bad on an application if a professor did not submit their application)

—Don’t aim too high: It’s better to have an engaged letter by a TA or postdoc than to have a lukewarm letter by a well-known professor who does not really know you and will only say vague things about you.

—Connected with the above: Be sure that the professor will be happy to write a strong letter for you (not just a letter). Some letters are not really of help.

—Optional: Explain why you are asking exactly this professor (e.g., “I am asking you because you are the one who engaged the most with my writing assignments and could therefore speak about them/because your class is the one during which I have been improving the most and I hope you can attest to that/because of the many 1-1 discussions we had, which makes me confident that you have a clearer opinion about my philosophical ideas than most other professors/…”).

—INFORM your letter-writers of what happened. It’s nice to hear back from you!

Am I forgetting something?

(Updated, thanks to @neuroyogacara.bsky.social)

Posted on

Kumārila conference

The first Kumārila Conference will be held at the Department of Philosophy at the University of Toronto, Mississauga, from May 27th to June 1st, 2024. Venue: Maanjiwe nendamowinan building, Room 3230, UTM.

The conference will bring together experts who will lead two-hour reading sessions on key passages of Kumārila’s texts and provide participants with the necessary tools to understand the hidden gems of Kumārila’s philosophy. More in detail, these sessions will include the reading and commenting on selected passages on a given topic (e.g., adhikāra in Ṭupṭīkā 6.1) and a talk on the topic itself (e.g., mapping the intersection of adhikāra and sāmarthya) and then a discussion session. Besides, there will be opportunities for scholars and advanced students to present their research related to Kumārila in shorter sessions (60′ and 30′).

The conference will be coordinated by Elisa Freschi and Nilanjan Das and will see the participation of other experts in Sanskrit philosophy and philology. Confirmed participants (so far) include Dan Arnold, Tarinee Awasthi, Purushottama Bilimoria, Hugo David, Alessandro Graheli, Ham Hyoung Seok, Kei Kataoka, Malcolm Keating, Lawrence McCrea, Sudipta Munsi, John Nemec, Monika Nowakowska, Andrew Ollett, Parimal Patil, Akane Saito, Wintor Scott, Taisei Shida, Elliot Stern, Angela Vettikkal, Alex Watson and Kiyotaka Yoshimizu.

This will be an in-person only event, since we believe in the power of collective intelligence and collaboration and these are hardly replicable when some participants speak per Zoom and others are in the room. Please consider that we are happy to assist you with invitation letters etc. if you need them for visa purposes, just let us know!

Posted on

Thinking about rights in Sanskrit philosophy

I started thinking about rights while working on permissions, because some deontic logicians think that permissions need to be also independent of prohibitions, in order to ground rights. Now, as I argued elsewhere, Mīmāṃsā permissions are always exceptions to previous negative obligations or prohibitions, so what happens to rights?


There is not a directly correspondent concept (how could there be one, given how historically loaded ‘right’ is?), but an interesting parallel is adhikāra. How does this work?


Let me look, to begin with, at the discussion in the apaśūdrādhikaraṇa by Rāmānuja. There, he explains that śūdras (humans belonging to the lowest class) do not have the adhikāra to perform worship (upāsana), because they are unable (asamartha), even though they desire the result of the action. Why are they unable? Because the do not have the adhikāra to perform a part of it, namely vedādhyayana.


So, as a preliminary understanding:

adhikāra to do x (e.g., upāsana) presupposes:

  1. desire to obtain the result of action x
  2. ability to perform action x, which, in turn, presupposes the ability to perform each of its parts
  3. [not yet determined, but see below]

The interesting part is that according to “as much as possible” principle, one does not need to perform each part of x in order to perform x, if x is a compulsory action (e.g., if you have the duty to each single day, you might leave out polishing each slide every single time). However, you still need to be “able” (samartha) to perform each part, even the ones you might have left out.

asāmarthya ‘inability’ can be due to physical or economical inability, but could also be the result of the lack of a predecing adhikāra. Thus, adhikāra has a double relation to sāmarthya: a) Lack of sāmarthya precludes the possibility adhikāra (adhikāra presupposes can); b) Lack of adhikāra to do x means that one will not get to do x and will therefore be asamartha with regard to y, if x is a part of y.

This leads to the adhikāra for vedādhyayana. This presupposes:

  1. desire to obtain the result of the action
  2. ability to perform the action (see PMS, book 6.1 on animals not being able to perform sacrifices)
  3. sanction by an authoritative text

Which kind of sanction? In the case of vedādhyayana, it seems to be a positive obligation (or vidhi). What would be the distinction between adhikāra and any other enjoined action, then? That the vidhi which lies at the basis of adhikāra establishes one’s general entitlement to do x, and not just one’s duty to perform x in a given case.

Posted on

Sarvagatatva in Nyāya and Vaiśeṣika: ātman, aether and materiality (mūrtatva)

The Sanskrit philosophical school called Vaiśeṣika is the one most directly dealing with ontology. Its fundamental text is the Vaiśeṣikasūtra, which is commented upon by Prāśastapada in the Pādarthadharmasaṅgraha (from now one PDhS) (the following is a summary of Padārthadharmasaṅgraha ad 8.7).

The school distinguishes substances and qualities. The first group includes four types of atoms (earth, water, fire, air) and then aether, time, space, ātmans and internal organs (manas). The latter are needed as a separate category, because they are point-sized and therefore not made of atoms, unlike the external sense faculties.
Among the 17 qualities, it recognises parimāṇa or `dimension’. This encompasses at first two possibilities, namely atomic (aṇu), or extended (mahat). The former covers partless entities that have allegedly no spatial dimension, like points in Euclidean geometry and atoms themselves. These are considered to be without extension and permanent through time (nitya). The latter is subdivided into mahat and paramahat. The first covers all objects one encounters in normal life, from triads of atoms (imagined to be of the size of a particle of dust, the first level of atomic structure to be extended) to the biggest mountain. These entities have parts and extension and have an origin and an end in time. The second subdivision covers special substances, listed as ākāśa `aether’, space, time and ātmans, which need to be imagined to be present at each location. Such entities are also imagined to be nitya, that is permanent through time. In other words, they are present at each location of time and space.
The above also implies that entities considered to be permanent through time can only be either atomic or all-pervasive.

However, space, time, aether and selves (ātman) are present at all locations in different ways.

About aether, to begin with, texts like Jayanta’s Nyāyamañjarī say that it needs to be accepted as a fifth substance in order to justify the diffusion of sound across multiple media. Texts of the Vaiśeṣika school, and of the allied school of Nyāya specify that aether does not occupy all locations, but rather is in contact with each individual atom):

[The aether’s] all-pervasiveness consists in the fact that it is in contact with each corporal (mūrta) substance.
(sarvamūrtadravyasaṃyogitvam vibhutvam (Tarkasaṃgrahadīpikā ad 14).)

This means that aether does not pervade atoms, but is in contact (saṃyoga) with each one of them.

This point is already explicit in the allied school of Nyāya, the Nyāyabhāṣya, and is needed because of the point-sized nature of atoms. If these were pervaded by aether, then they would have parts, and thus not be permanent. These undesired consequences are examined in the following:

This is impossible, because of the penetration through aether || NS 4.2.18 ||

It is impossible for an atom [to be] partless and permanent. Why? Because of the penetration through ether, that is, because an atom, if it were permeated, that is `penetrated’ by aether, within and outside, then, because of this penetration it would have parts, and due to having parts it would be impermanent.

Or, the aether is not all-located} || 4.2.19 ||

Alternatively, we don’t accept that. There is no aether within the atoms and therefore aether ends up not being all-located

(ākāśavyatibhedāt tadanupapattiḥ || 4.2.18 ||
tasyāṇor niravayasya nityasyānupapattiḥ. kasmāt. ākāśavyatibhedāt. antarbahiścāṇur ākāśena samāviṣṭo vyatibhinno vyatibhedāt sāvayavaḥ sāvayavatvād anitya iti.
ākāśāsarvagatatvaṃ vā || 4.2.19 ||
athaitan neṣyate paramāṇor antar nāsty ākāśam ity asarvagatatvaṃ prasajyeta iti.)

Aether is postulated as a substrate of sound (which can move through solids, liquids and air, thus proving that it has neither earth, nor water, nor air as substrate). Thus, it needs to be unitary (multiple aethers would not explain the propagation of sound, sound would stop at the end of the respective aether) and it needs to be present at all locations (for the same reason). More in detail: Only because of the unitary nature of aether is it possible for sound to travel between different loci. Otherwise, one would have to posit some mechanism to explain how the sound encountered in one aether travels to another one. Instead, the simpler solution is to posit that aether is necessarily both single (eka) and present at all locations (vibhu).

As for ātman, the self is by definition permanent (otherwise, no afterlife nor cycle of rebirths would be possible). It cannot be atomic, though, because the ātman is the principle of awareness and people become aware of things potentially everywhere. The fact that they don’t become perceptually aware of things being, e.g., behind a wall, by contrast, is only due to the fact that the ātman needs to be in touch (via the internal sense organ, manas, which is believed to be atomic and to move quickly from one to the other sense-faculty) to the sense faculties (indriya) in order for perceptual awareness to take place. Yogins are able to perceive things their bodies are not in contact with because their ātmans are omnipresent, like our ātman, and are able, unlike our ātman, to connect with other bodies’ sense faculties.
Within Sanskrit philosophy, Jaina philosophers suggested that the ātman is co-extensive with the body, since it can experience whatever the body can experience. Vaiśeṣika and other non-Jaina authors disagree, because this would lead to the absurd consequence of an ātman changing in size through one’s life.

A further element to be taken into account with regard to theories of location, and in particular while adjudicating whether they are about occupation or non-occupation is materiality.
Occupation of space seems to occur only from the level of atomic triads up to big, but not all-located, objects. Atoms are said to be mūrta and mūrta is usually translated as `material’, but taken in isolations, atom do not have parts and are only point-sized. In this sense, their being mūrta refers more about their being fundamental for material entities, rather than being material if taken in isolation. The distinction is theoretically relevant, but less evident at the pragmatic level, given that atoms are never found in isolation. Being mūrta is attributed to atoms of the four elements (not to aether) as well as to the inner sense organ (Nyāyakośa, s.v.), but not to ātman neither to aether.

Posted on

Quick summary of Maṇḍana’s Vidhiviveka, uttarabhāga (siddhānta view), on nitya vs kāmya sacrifices

We concluded today a great workshop on Maṇḍana’s Vidhiviveka and these are my first comments on what we could establish. My deepest gratitude goes to all participants (Akane Saito, Andrew Ollett, Elliot Stern, Kei Kataoka, Lawrence McCrea, Nilanjan Das, Parimal Patil, Patrick Cummins, Wintor Scott and others). (For more on the workshop, read here: https://philosophy.utoronto.ca/event/workshop-maṇḍana-on-ritual-duties/)

Structure:
vv. 2.1–2.6: Maṇḍana’s siddhānta on iṣṭābhyupāyatā (chapter 11 in Elliot Stern’s forthcoming edition)

v. 2.7: summary of chapter 11, opening the new topic (chapter 12)

v. 2.8: yathāśakti provision applies to nitya sacrifices, sarvāṅgopasaṃhāra in kāmya ones (chapter 13)

v. 2.9: time needs to be the distinguishing factor between nitya/namimittika and kāmya rituals. For the former, kāla is included in nimitta, that ensures avaśyakartavyatā (chapter 14)

Summary of Chapters 13–14:
Chapter 13 tries to find a suitable candidate as phala for nitya sacrifices. The Prābhākara opponent insists on akaraṇe pratyavāya, whereas the siddhāntin (or quasi-siddhāntin) prefers pāpakṣaya. A first proposal in this sense, however, is refuted, because it would lead to just vāstava nityatva, and this would not guarantee the application of the yathāśakti provision. This can only be guaranteed by the ought-implies-can metarule and thus by a śābda nityatva, prescribed as such by the Veda. Hence, Maṇḍana wants both levels.

The akaraṇe-pratyavāya alternative is discussed at length. A preliminary possibility saying that akaraṇe there is pratyavāya, and that the sacrifice, being duḥkha, leads in an ānuṣaṅgika way to the production of pāpakṣaya.
This is finally refuted because there is no Vedic text justifying this view (it would be just inflicting whimsical pain to oneself).

At this point, the discussion becomes more technical. The acceptance of the yathāśakti provision implies that one contracts (saṅkoca) the meaning of either the mention of the adhikāra (so that only a samartha adhikārin is meant) or of the various aṅgavidhis. Which one is better? The constant risk to be avoided is that the same view could be applied to nitya and kāmya rituals, thus ending up in their being non-different.

The opponent suggests the application of Kāmsyabhojinyāya and of bādha based on bhūyastva in order to get to the result that the single adhikāraśruti should be contracted and not the many aṅgavidhis. By contrast, Maṇḍana (atha matam, p, 671 in ES’ edition, 17.5.2023) thinks that the aṅgavidhis are restricted, bc otherwise two unwanted consequences would follow (vaicitrya-risk in the prayoga) and because a single śruti is not evidence, since words are used in context (naitat sāram, p. 672).

Thus, this solution is putting much wait on yāvajjīvam-śruti, whereas chapter 14 will put in the context of naimittika sacrifices.

In this chapter, the new risk is adhikārātikrānti and the additional hurdle are naimittika sacrifices, which are not performed every day, but are otherwise identical to the nitya ones. To the atha matam and naitat sāram positions mentioned above something else is added, namely the definition of nimitta as not being a viśeṣaṇa of the adhikārin (this is what the Prābhākara opponent argues for), but just a nimitta. Time (kāla) also belongs to the nimitta in the case of nitya and naimittika sacrifices, although it is an aṅga for kāmya ones and can hence be a distinguishing factor (v. 2.9).

One point we did not have the time to discuss: Vācaspati (on 13.5) discusses a tantra vs prasaṅga approach and concludes that the first one should be preferred. What two things are centrally performed via tantra? Vācaspati clearly says that one is a kāmya and the other one is a nitya ritual (e.g. “ubhayor api kāmyanityayoḥ karmaṇoḥ […] tantram anuṣṭhānam” or “tasmān na kāmye ‘nuṣṭhīyamāne prāsaṅgikatvaṃ nityasya”, p. 660), but what are these referring to? A suggestion: There is a nitya sacrifice corresponding to the śābda nityatva, and a kāmya sacrifice corresponding to the vāstava nityatva and the two are performed at the same time.