Adhik. V (11-21) is, according to Sankara, taken up with the question as to the nature of the highest Brahman in which the individual soul is merged in the state of deep sleep. Sfitra 11 declares that twofold characteristics (viz. absence and presence of distinctive attributes, nirvLeshatva and savLeshatva) cannot belong to the highest Brahman even through its stations, i. e. its limiting adjuncts; since all passages which aim at setting forth Brahman’s nature declare it to be destitute of all distinctive attributes.—The fact, Sutra 1 % continues, that in many passages Brahman is spoken of as possessing distinctive attributes is of no relevancy, since wherever there are mentioned limiting ad¬ juncts, on which all distinction depends, it is specially stated Ixii VEDANTA-S^TRAS.
that Brahman in itself is free from all diversity; and—Sutra 13 adds—in some places the assumption of diversity is spe¬ cially objected to.—That Brahman is devoid of all form (Sfitra 14), is the pre-eminent meaning of all Ved 4 nta-texts setting forth Brahman’s nature.—That Brahman is repre¬ sented as having different forms, as it were, is due to its connexion with its (unreal) limiting adjuncts; just as the light of the sun appears straight or crooked, as it were, according to the nature of the things he illuminates (15).— The BWhad&ra/zyaka expressly declares that Brahman is one uniform mass of intelligence (16); and the same is taught in other scriptural passages and in Smnti (17).—At the unreality of the apparent manifoldness of the Self, caused by the limiting adjuncts, aim those scriptural passages in which the Self is compared to the sun, which remains one although his reflections on the surface of the water are many (18),—Nor must the objection be raised that that comparison is unsuitable, because the Self is not material like the sun, and there are no real upadhis separate from it as the water is from the sun; for the comparison merely means to indicate that, as the reflected image of the sun participates in the changes, increase, decrease, &c., which the water undergoes while the sun himself remains unaffected thereby, so the true Self is not affected by the attributes of the upadhis, while, in so far as it is limited by the latter, it is affected by them as it were (19, 20).—That the Self is within the upadhis, Scripture declares (21).
From the above explanation of this important adhikara^a the one given in the Sn-bhashya differs totally. According to Raminu^a the adhikara^a raises the question whether the imperfections clinging to the individual soul (the dis¬ cussion of which has now come to an end) affect also the highest Lord who, according to Scripture, abides within the soul as antaryamin. ‘ Notwithstanding the abode (of the highest Self within the soul) (it is) not (affected by the soul’s imperfections) because everywhere (the highest Self is repre¬ sented) as having twofold characteristics (viz. being, on one hand, free from all evil, apahatapapman, wig ara, vimrztyu, &c., and, on the other hand, endowed with all auspicious INTRODUCTION.
Ixiii qualities, satyak&ma, satyasa/^kalpa, &c.) (n).—Should it be objected that, just as the soul although essentially free from evil—according to the Pra^cipativakya in the Kh an- dogya—yet is liable to imperfections owing to its connexion with a variety of bodies, so the antaryamin also is affected by abiding within bodies; we deny this because in every section of the chapter referring to the antaryamin (in the IWhad&ra/zyaka) he is expressly called the Immortal, the ruler within; which shows him to be free from the short¬ comings of the ^iva (is).—Some, moreover, expressly assert that, although the Lord and the soul are within one body, the soul only is imperfect, not the Lord (dvfi supar/za sayu^a sakhctyd) (13).—Should it be said that, according to the .Oandogya, Brahman entered together with the souls into the elements previously to the evolution of names and forms, and hence participates in the latter, thus becoming implicated in the sa;^sara; we reply that Brahman, although connected with such and such forms, is in itself devoid of form, since it is the principal element (agent; pradhana) in the bringing about of names and forms (according to ‘ aka.ro ha vai namarupayor nirvahita’) (14).—But does not the pas¬ sage ‘ satya m ^anam anantam brahma ’ teach that Brah¬ man is nothing but light (intelligence) without any difference, and does not the passage ‘ neti neti' deny of it all qualities?
—As in order, we reply, not to deprive passages as the one quoted from the Taittiriya of their purport, we admit that Brahman’s nature is light, so we must also admit that Brahman is satyasa^kalpa, and so on; for if not, the pas¬ sages in which those qualities are asserted would become purportless (15).—Moreover the Taittiriya passage only asserts so much, viz. the praka^arupata of Brahman, and does not deny other qualities (16).—And the passage ‘neti neti 5 will be discussed later on.—The ubhayalingatva of Brahman in the sense assigned above is asserted in many places of 5 ruti and Smrzti (17)-—Because Brahman although abiding in many places is not touched by their imperfec¬ tions, the similes of the reflected sun, of the ether limited by jars, &c., are applicable to it (18).—Should it be said that the illustration is not an appropriate one, because the IxiV VEDANTA-stjTRAS.
sun is apprehended in the water erroneously only while the antarydmin really abides within all things, and therefore must be viewed as sharing their defects (19); we reply that what the simile means to negative is merely that Brahman should, owing to its inherence in many places, participate in the increase, decrease, and so on, of its abodes. On this view both similes are appropriate (20).—Analogous similes we observe to be employed in ordinary life, as when we compare a man to a lion (21).
Sutras 22-30 constitute, according to.Sankara, a new adhikara^a (VI), whose object it is to show that the clause f not so, not so'(neti neti; Brzhadar.) negatives, not Brah¬ man itself, but only the two forms of Brahman described in the preceding part of the chapter. Sutras 23-26 further dwell on Brahman being in reality devoid of all distinctive attributes which are altogether due to the upadhis. The last four Sfttras return to the question how, Brahman being one only, the souls are in so many places spoken of as different from it, and, two explanatory hypotheses having been rejected, the conclusion is arrived at that all difference is unreal, due to fictitious limiting adjuncts.
According to Ramanuja, Sutras 22 ff. continue the dis¬ cussion started in Sutra 11. How, the question is asked, can the ubhayalingatva of Brahman be maintained considering that the ‘ not so, not so ’ of the Brzhadara^yaka denies of Brahman all the previously mentioned modes (prakara), $0 that it can only be called that which is (sanmatra)?—The reply given in Sutra 22 is that ‘ not so, not so ’ does not deny of Brahman the distinctive qualities or modes declared previously (for it would be senseless at first to teach them, and finally to deny them again 1 ), but merely denies the prakrztaitavattva, the previously stated limited nature of Brahman, i.e. it denies that Brahman possesses only the previously mentioned qualifications. With this agrees, that subsequently to ‘ neti neti ’ Scripture itself enunciates further qualifications of Brahman.—That Brahman as stated 1 All the mentioned modes of Brahman are known from Scripture only, not from ordinary experience. If the latter were the case, then, and then only, Scripture might at first refer to them ‘ anuvadena,’ and finally negative them.
INTRODUCTION.
Ixv above is not the object of any other means of proof but Scrip¬ ture is confirmed in Sutra 23, ‘ Scripture declares Brahman to be the non-manifest/—And the intuition (sdkshatkara) of Brahman ensues only upon its sa^radhana, i.e. upon its being perfectly pleased by the worshipper’s devotion, as Scripture and Smrz'ti declare (24).—That this interpretation of ‘neti ’ is the right one, is likewise shown by the fact that in the same way as prakai’a, luminousness, ^/zana, intelligence, &c., so also the quality of being differentiated by the world (prapa^avirish/ata) is intuited as non-different, i. e. as like¬ wise qualifying Brahman; and that prakai’a, and so on, characterise Brahman, is known through repeated practice (on the part of rishis like Vamadeva) in the work of sawridhana mentioned before (25).—For all these reasons Brahman is connected with the infinite, i.e. the infinite number of auspicious qualities; for thus the twofold indica¬ tions (linga) met with in Scripture are fully justified (26 ).— In what relation, then, does the a£id vastu, i.e. the non- sentient matter, which, according to the Brzhadara/zyaka, is one of the forms of Brahman, stand to the latter?—Non- sentient beings might, in the first place, be viewed as special arrangements (sa/^sthanavLesha^) of Brahman, as the coils are of the body of the snake; for Brahman is designated as both, i.e. sometimes as one with the world (Brahman is all this, &c.), sometimes as different from it (Let me enter into those elements, &c.) (27).—Or, in the second place, the relation of the two might be viewed as analogous to that of light and the luminous object which are two and yet one, both being fire (38).—Or, in the third place, the relation is like that stated before, i.e. the material world is, like the individual souls (whose case was discussed in II, 3, 43), a part— ainsz.—of Brahman (29, 30).
Adhik. VII (31-37) explains how some metaphorical expressions, seemingly implying that there is something different from Brahman, have to be truly understood.
Adhik. VIII (38-41) teaches that the reward of works is not, as Gaimini opines, the independent result of the works acting through the so-called apurva, but is allotted by the Lord.
PADA III.
With the third pada of the second adhyaya a new section of the work begins, whose task it is to describe how the individual soul is enabled by meditation on Brah¬ man to obtain final release. The first point to be deter¬ mined here is what constitutes a meditation on Brahman, and, more particularly, in what relation those parts of the Upanishads stand to each other which enjoin identical or partly identical meditations. The reader of the Upa¬ nishads cannot fail to observe that the texts of the different jakhas contain many chapters of similar, often nearly iden¬ tical, contents, and that in some cases the text of even one and the same ^akha exhibits the same matter in more or less varied forms. The reason of this clearly is that the common stock of religious and philosophical ideas which were in circulation at the time of the composition of the Upanishads found separate expression in the different priestly communities; hence the same speculations, legends, &c. reappear in various places of the sacred Scriptures in more or less differing dress. Originally, when we may suppose the members of each Vedic school to have confined themselves to the study of their own sacred texts, the fact that the texts of other schools contained chapters of similar contents would hardly appear to call for special note or comment; not any more than the circumstance that the sacrificial performances enjoined on the followers of some particular jakha were found described with greater or smaller modifications in the books of other ^akhas also. But already at a very early period, at any rate long before the composition of the Vedanta-sutras in their present form, the Vedic theologians must have apprehended the truth that, in whatever regards sacrificial acts, one sakha, may indeed safely follow its own texts, disregarding the texts of all other Vakhas; that, however, all texts which aim at throwing light on the nature of Brahman and the relation to it of the human soul must somehow or other be com¬ bined into one consistent systematical whole equally valid for the followers of all Vedic schools. For, as we have had occasion to remark above, while acts may be performed INTRODUCTION.
lxvii by different individuals in different ways, cognition is de¬ fined by the nature of the object cognised, and hence can be one only, unless it ceases to be true cognition. Hence the attempts, on the one hand, of discarding by skilful interpretation all contradictions met with in the sacred text, and, on the other hand, of showing what sections of the different Upanishads have to be viewed as teaching the same matter, and therefore must be combined in one medi¬ tation. The latter is the special task of the present pada.
Adhik. I and II (1-4; 5) are concerned with the question whether those vidyas, which are met with in identical or similar form in more than one sacred text, are to be con¬ sidered as constituting several vidyas, or one vidya only. Sankara remarks that the question affects only those vidyas whose object is the qualified Brahman; for the knowledge of the non-qualified Brahman, which is of an absolutely uniform nature, can of course be one only wherever it is set forth. But things lie differently in those cases where the object of knowledge is the saguzzam brahma or some outward manifestation of Brahman; for the qualities as well as manifestations of Brahman are many. Anticipating the subject of a later adhikarazza, we may take for an example the so-called Sazzafilyavidya which is met with in Kh. Up. Ill, 14, again—in an abridged form—in B ri. Up. V, 6, and, moreover, in the tenth book of the *Satapatha- brahmazza (X, 6, 3). The three passages enjoin a medita¬ tion on Brahman as possessing certain attributes, some of which are specified in all the three texts (as, for instance, manomayatva, bharupatva), while others are peculiar to each separate passage (pra/za^ariratva and satyasa/zzkalpatva, for instance, being mentioned in the AVzandogya Upanishad and Satapatha-br&hmazza, but not in the BrzTiadarazzyaka Upanishad, which, on its part, specifies sarvavaritva, not referred to in the two other texts). Here, then, there is room for a doubt whether the three passages refer to one object of knowledge or not. To the devout Vedantin the question is not a purely theoretical one, but of immediate practical interest. For if the three texts are to be held apart, there are three different meditations to be gone through; if, on the e 2 lxviii vedAnta-s6tras.
other hand, the vidya is one only, all the different qualities of Brahman mentioned in the three passages have to be combined into one meditation.—The decision is here, as in all similar cases, in favour of the latter alternative. A careful examination of the three passages shows that the object of meditation is one only; hence the meditation also is one only, comprehending all the attributes mentioned in the three texts.
Adhik. Ill (6-8) discusses the case of vidyas being really separate, although apparently identical. The examples selected are the udgithavidyas of the.Oandogya Upanishad (I, 1-3) and the Brzhadara/zyaka Upanishad (I, 3), which, although showing certain similarities—such as bearing the same name and the udgitha being in both identified with prazza—yet are to be held apart, because the subject of the Khk.ndogya vidya is not the whole udgitha but only the sacred syllable Om, while the Brzhadarazzyaka Upanishad represents the whole udgitha as the object of meditation.
Sfitra 9 constitutes in.Sankara’s view a new adhikarazza (IV), proving that in the passage, ‘ Let a man meditate ’ ( Kh. Up. I, 1, 1), the Oz^kara and the udgitha stand in the relation of one specifying the other, the meaning being, f Let a man meditate on that Ozzzkara which/ &c.—Ac¬ cording to Ramanuja’s interpretation, which seems to fall in more satisfactorily with the form and the wording of the Sutra, the latter merely furnishes an additional argument for the conclusion arrived at in the preceding adhikarazza.— Adhik. V (10) determines the unity of the so-called prazza- vidy&s and the consequent comprehension of the different qualities of the prazza, which are mentioned in the different texts, within one meditation.
Adhik. VI comprises, according to.Sankara, the Sutras 11-13. The point to be settled is whether in all the medi¬ tations on Brahman all its qualities are to be included or only those mentioned in the special vidya. The decision is that the essential and unalterable attributes of Brahman, such as bliss and knowledge, are to be taken into account everywhere, while those which admit of a more or less (as, for instance, the attribute of having joy for its head, men- INTRODUCTION.
lxix tioned in the Taitt. Up.) are confined to special medita¬ tions.—Adhik. VII (14, 15), according to.Sankara, aims at proving that the object of Ka/^a Up. Ill, 10, 11 is one only, viz. to show that the highest Self is higher than everything, so that the passage constitutes one vidya only. —Adhik. VIII (16, 17) determines, according to.Sankara, that the Self spoken of in Ait. Ar. II, 4, 1, 1 is not a lower form of the Self (the so-called sutr^tman), but the highest Self; the discussion of that point in this place being due to the wish to prove that the attributes of the highest Self have to be comprehended in the Aitareyaka meditation.
According to Rdmanu^ci the Sutras 11-17 constitute a single adhikara/za whose subject is the same as that of .Sankara’s sixth adhikara/za. Sfitras 11-13 are > on the whole, explained as by.Sankara; Sutra 12, however, is said to mean, ‘ Such attributes as having joy for its head, &c. are not to be viewed as qualities of Brahman, and therefore not to be included in every meditation; for if they were admitted as qualities, difference would be intro¬ duced into Brahman’s nature, and that would involve a more or less on Brahman’s part.’ Sfitras 14-17 continue the discussion of the passage about the priya.nrastva.—If priya^irastva, &c. are not to be viewed as real qualities of Brahman, for what purpose does the text mention them?— ‘ Because,’ Sutra 14 replies, ‘ there is no other purpose, Scripture mentions them for the purpose of pious medita¬ tion.’—But how is it known that the Self of delight is the highest Self? (owing to which you maintain that having limbs, head, &c. cannot belong to it as attributes.)—‘ Be¬ cause,’ Sutra 15 replies/ the term “ Self” (atma anandamaya) is applied to it.’—But in the previous parts of the chapter the term Self (in atm& pra/zamaya, &c.) is applied to non- Selfs also; how then do you know that in atma dnanda- maya it denotes the real Self?—‘The term Self,’ Sutra 16 replies, ‘ is employed here to denote the highest Self as in many other passages (atmi va idam eka, &c.), as we con¬ clude from the subsequent passage, viz. he wished, May I be many.’—But, an objection is raised, does not the con¬ text show that the term ‘ Self,’ which in all the preceding lxx VEDANTA-S<JTRAS.
clauses about the prazzamaya, &c. denoted something other than the Self, does the same in anandamaya atman, and is not the context of greater weight than a subsequent passage?—To this question asked in the former half of 17 (anvayad iti kdt) the latter half replies, £ Still it denotes the Self, owing to the affirmatory statement,’ i.e. the fact of the highest Self having been affirmed in a previous passage also, viz. II, 1, £ From that Self sprang ether/ Adhik. IX (18) discusses a minor point connected with the prazzasaz/zvada.—The subject of Adhik. X (19) has been indicated already above under Adhik. I.—Adhik. XI (20-22) treats of a case of a contrary nature; in Brz. Up. V, 5, Brahman is represented first as abiding in the sphere of the sun, and then as abiding within the eye; we therefore, in spite of certain counter-indications, have to do with two separate vidyas.—Adhik. XII (23) refers to a similar case; certain attributes of Brahman mentioned in the Razzaya- niya-khila have not to be introduced into the corresponding AVzandogya vidya, because the stated difference of Brah¬ man’s abode involves difference of vidya.—Adhik. XIII (24) treats of another instance of two vidyas having to be held apart.
Adhik. XIV (25) decides that certain detached mantras and brahmazza passages met with in the beginning of some Upanishads—as, for instance, a brahmazza about the maha- vrata ceremony at the beginning of the Aitareya-arazzyaka —do, notwithstanding their position which seems to connect them with the brahmavidya, not belong to the latter, since they show unmistakable signs of being connected with sacrificial acts.
Adhik. XV (26) treats of the passages stating that the man dying in the possession of true knowledge shakes off all his good and evil deeds, and affirms that a statement, made in some of those passages only, to the effect that the good and evil deeds pass over to the friends and enemies of the deceased, is valid for all the passages.
Sutras 27-30 constitute, according to /Sankara, two adhi- karazzas of which the former (XVI; 27, 28) decides that the shaking off of the good and evil deeds takes place—not, as INTRODUCTION.
lxxl the Kaush. Up. states, on the road to Brahman’s world— but at the moment of the soul’s departure from the body; the Kaushitaki statement is therefore not to be taken literally.—The latter adhikarazza (XVII; 29, 30) treats of the cognate question whether the soul that has freed itself from its deeds proceeds in all cases on the road of the gods (as said in the Kaush. Up.), or not. The decision is that he only whose knowledge does not pass beyond the saguzzam brahma proceeds on that road, while the soul of him who knows the nirguzzam brahma becomes one with it without moving to any other place.
The.Sri-bhashya treats the four Sutras as one adhikarazza whose two first Sutras are explained as by.Sankara, while Sutra 29 raises an objection to the conclusion arrived at, ‘ the going (of the soul on the path of the gods) has a sense only if the soul’s freeing itself from its works takes place in both ways, i. e. partly at the moment of death, partly on the road to Brahman 5 for otherwise there would be a con¬ tradiction 5 (the contradiction being that, if the soul’s works were all shaken off at the moment of death, the subtle body would likewise perish at that moment, and then the bodi¬ less soul would be unable to proceed on the path of the gods).—To this Sutra 30 replies, ‘The complete shaking off of the works at the moment of death is possible, since matters of that kind are observed in Scripture,’ i. e. since scriptural passages show that even he whose works are entirely annihilated, and who has manifested himself in his true shape, is yet connected with some kind of body; com¬ pare the passage, ‘para m ^yotir upasampadya svena rupe- zzabhinishpadyate sa tatra paryeti krn/an ramamana/^ sa svar kd bhavati tasya sarveshu lokeshu kama^aro bhavati.’ That subtle body is not due to karman, but to the soul’s vidyamahatmya.—That the explanation of the.Sri-bhashya agrees with the text as well as.Sankara’s, a comparison of the two will show; especially forced is.Sankara’s explana¬ tion of ‘ arthavattvam ubhayatha,’ which is said to mean that there is arthavattva in one case, and non-arthavattva in the other case.
The next Sutra (31) constitutes an adhikarazza (XVIII) VEDANTA-S^TRAS.
lxxil deciding that the road of the gods is followed not only by those knowing the vidyas which specially mention the going on that road, but by all who are acquainted with the saguzza-vidyas of Brahman.—The explanation given in the •Sri-bhashya (in which Stitras 31 and 33 have exchanged places) is similar, with the difference however that all who meditate on Brahman—without any reference to the dis¬ tinction of nirguzza and saguzza—proceed after death on the road of the gods. (The 6ri-bhashya reads ‘sarveshdm,’ i. e. all worshippers, not ‘ sarvasam/ all saguzza-vidyas.)
Adhik. XIX (32) decides that, although the general effect of true knowledge is release from all forms of body, yet even such beings as have reached perfect knowledge may retain a body for the purpose of discharging certain offices. —In the Sri-bhashya, where the Sutra follows immediately on Sfitra 30, the adhikarazza determines, in close connexion with 30, that, although those who know Brahman as a rule divest themselves of the gross body—there remaining only a subtle body which enables them to move—and no longer experience pleasure and pain, yet certain beings, although having reached the cognition of Brahman, remain invested with a gross body, and hence liable to pleasure and pain until they have fully performed certain duties.
Adhik. XX (33) teaches that the negative attributes of Brahman mentioned in some vidyas—such as its being not gross, not subtle, &c.—are to be included in all meditations on Brahman.—Adhik. XXI (34) determines that Ka/fe Up. Ill, 1, and Mu. Up. Ill, 1, constitute one vidya only, because both passages refer to the highest Brahman. According to Ramanuja the Sfitra contains a reply to an objection raised against the conclusion arrived at in the preceding Sutra.—Adhik. XXII (35, 36) maintains that the two passages, Bri. Up. Ill, 4 and III, 5, constitute one vidya only, the object of knowledge being in both cases Brahman viewed as the inner Self of all.—Adhik. XXIII (37) on the contrary decides that the passage Ait. Ar. II, 2, 4, 6 con¬ stitutes not one but two meditations.—Adhik. XXIV (38) again determines that the vidyi of the True contained in Bri. Up. V, 4, 5, is one only.—According to Rimanu^a, INTRODUCTION/ Sfitras 35-38 constitute one adhikara^a only whose subject is the same as that of XXII according to Sankara.
Adhik. XXV (39) proves that the passages Kh. Up. VIII, 1 and Bri. Up. IV, 4, 22 cannot constitute one vidya, since the former refers to Brahman as possessing qualities, while the latter is concerned with Brahman as destitute of qualities.—Adhik. XXVI (40, 41) treats, according to.San¬ kara, of a minor question connected with Kh. Up. V, 11 ff. —According to the.Srt-bhashya, Sutras 39-41 form one adhikara^a whose first Sfitra reaches essentially the same conclusion as.Sankara under 39. Sutras 40, 41 thereupon discuss a general question concerning the meditations on Brahman. The qualities, an opponent is supposed to re¬ mark, which in the two passages discussed are predicated of Brahman—such as va^itva, satyakamatva, &c.—cannot be considered real (paramarthika), since other passages (sa esha neti neti, and the like) declare Brahman to be devoid of all qualities. Hence those qualities cannot be admitted into meditations whose purpose is final release.—To this objec¬ tion Stitra 40 replies, ‘(Those qualities) are not to be left out (from the meditations on Brahman), since (in the passages under discussion as well as in other passages) they are stated with emphasis 1 /—But, another objection is raised, Scrip¬ ture says that he who meditates on Brahman as satyakama, &c. obtains a mere perishable reward, viz. the world of the fathers, and similar results specified in Kh. Up. VIII, 2; hence, he who is desirous of final release, must not include those qualities of Brahman in his meditation.—To this ob¬ jection Sutra 41 replies, ‘Because that (i.e. the free roaming in all the worlds, the world of the fathers, &c.) is stated as proceeding therefrom (i. e. the approach to Brahman which is final release) in the case of (the soul) which has approached Brahman;' (therefore a person desirous of release, may include satyakamatva, &c. in his meditations.)
1 Ramanuja has here some strong remarks on the improbability of qualities emphatically attributed to Brahman, in more than one passage, having to be set aside in any meditation: ‘ Na £a matapitWsahasrebhyo s pi vatsalataraw jastraw pratarakavad aparamarthikau nirasaniyau guwau pramawantaraprati- pannau adarewopadlrya sawsara^akraparivartanena purvam eva bambhramya- m&nan mumukshhn bhftyo * pi bhramayitum alam,’ Ixxiv VEDANTA-SUTRAS.
Adhik. XXVII (42) decides that those meditations which are connected with certain matters forming constituent parts of sacrificial actions, are not to be considered as perma¬ nently requisite parts of the latter.—Adhik. XXVIII (43) teaches that, in a Bri. Up. passage and a similar Kh. Up. passage, Vayu and Pr&;za are not to be identified, but to be held apart.—Adhik. XXIX (44-52) decides that the fire- altars made of mind, &c., which are mentioned in the Agni- rahasya, do not constitute parts of the sacrificial action (so that the mental, &c. construction of the altar could optionally be substituted for the actual one), but merely subjects of meditations.