The Vedanta Sutras, commentary by Sankaracharya (SBE34), tr. by George Thibaut  at sacred-texts.com
24. The 'light' (is Brahman), on account of the mention of feet (in a passage which is connected with the passage about the light).
Scripture says (Kh. Up. III, 13, 7), 'Now that light which shines above this heaven, higher than all, higher than everything,
in the highest worlds beyond which there are no other worlds that is the same light which is within man.' Here the doubt presents itself whether the word 'light' denotes the light of the sun and the like, or the highest Self. Under the preceding Sûtras we had shown that some words which ordinarily have different meanings yet in certain passages denote Brahman, since characteristic marks of the latter are mentioned. Here the question has to be discussed whether, in connexion with the passage quoted, characteristic marks of Brahman are mentioned or not.
The pûrvapakshin maintains that the word 'light' denotes nothing else but the light of the sun and the like, since that is the ordinary well-established meaning of the term. The common use of language, he says, teaches us that the two words 'light' and 'darkness' denote mutually opposite things, darkness being the term for whatever interferes with the function of the sense of sight, as, for instance, the gloom of the night, while sunshine and whatever else favours the action of the eye is called light. The word 'shines' also, which the text exhibits, is known ordinarily to refer to the sun and similar sources of light; while of Brahman, which is devoid of colour, it cannot be said, in the primary sense of the word, that it 'shines.' Further, the word gyotis must here denote light because it is said to be bounded by the sky ('that light which shines above this heaven'). For while it is impossible to consider the sky as being the boundary of Brahman, which is the Self of all and the source of all things movable or immovable, the sky may be looked upon as forming the boundary of light, which is a mere product and as such limited; accordingly the text says, 'the light beyond heaven.'--But light, although a mere product, is perceived everywhere; it would therefore be wrong to declare that it is bounded by the sky!--Well, then, the pûrvapakshin replies, let us assume that the light meant is the first-born (original) light which has not yet become tripartite 1. This explanation again cannot be
admitted, because the non-tripartite light does not serve any purpose.--But, the pûrvapakshin resumes, Why should its purpose not be found therein that it is the object of devout meditation?--That cannot be, we reply; for we see that only such things are represented as objects of devotion as have some other independent use of their own; so, for instance, the sun (which dispels darkness and so on). Moreover the scriptural passage, 'Let me make each of these three (fire, water, and earth) tripartite,' does not indicate any difference 1. And even of the non-tripartite light it is not known that the sky constitutes its boundary.--Well, then (the pûrvapakshin resumes, dropping the idea of the non-tripartite light), let us assume that the light of which the text speaks is the tripartite (ordinary) light. The objection that light is seen to exist also beneath the sky, viz. in the form of fire and the like, we invalidate by the remark that there is nothing contrary to reason in assigning a special locality to fire, although the latter is observed everywhere; while to assume a special place for Brahman, to which the idea of place does not apply at all, would be most unsuitable. Moreover, the clause 'higher than everything, in the highest worlds beyond which there are no other worlds,' which indicates a multiplicity of abodes, agrees much better with light, which is a mere product (than with Brahman). There is moreover that other clause, also, 'That is the same light which is within man,' in which the highest light is identified with the gastric fire (the fire within man). Now such identifications can be made only where there is a certain similarity of nature; as is seen, for instance, in the passage, 'Of that person Bhûh is the head, for the head is one and that syllable is one' (Bri. Up. V, 5, 3). But that the fire within the human body is not Brahman clearly appears from the passage, 'Of this we have visible and audible proof' (Kh. Up. III, 13, 7; 8), which declares that
the fire is characterised by the noise it makes, and by heat; and likewise from the following passage, 'Let a man meditate on this as that which is seen and heard.' The same conclusion may be drawn from the passage, 'He who knows this becomes conspicuous and celebrated,' which proclaims an inconsiderable reward only, while to the devout meditation on Brahman a high reward would have to be allotted. Nor is there mentioned in the entire passage about the light any other characteristic mark of Brahman, while such marks are set forth in the passages (discussed above) which refer to prâna and the ether. Nor, again, is Brahman indicated in the preceding section, 'the Gâyatrî is everything whatsoever exists,' &c. (III, 12); for that passage makes a statement about the Gâyatrî metre only. And even if that section did refer to Brahman, still Brahman would not be recognised in the passage at present under discussion; for there (in the section referred to) it is declared--in the clause, 'Three feet of it are the Immortal in heaven'--that heaven constitutes the abode; while in our passage the words 'the light above heaven' declare heaven to be a boundary. For all these reasons the word gyotis is here to be taken in its ordinary meaning, viz. light.
To this we make the following reply. The word gyotis must be held to denote Brahman. Why? On account of the feet (quarters) being mentioned. In a preceding passage Brahman had been spoken of as having four feet (quarters). 'Such is the greatness of it; greater than it is the Person (purusha). One foot of it are all the beings, three feet of it are the Immortal in heaven.' That which in this passage is said to constitute the three-quarter part, immortal and connected with heaven, of Brahman, which altogether comprises four quarters; this very same entity we recognise as again referred to in the passage under discussion, because there also it is said to be connected with heaven. If therefore we should set it aside in our interpretation of the passage and assume the latter to refer to the ordinary light, we should commit the mistake of dropping, without need, the topic started and introducing
a new subject. Brahman, in fact, continues to form the subject-matter, not only of the passage about the light, but likewise of the subsequent section, the so-called Sândilya-vidyâ (Kh. Up. III, 14). Hence we conclude that in our passage the word 'light' must be held to denote Brahman. The objection (raised above) that from common use the words 'light' and 'to shine' are known to denote effected (physical) light is without force; for as it is known from the general topic of the chapter that Brahman is meant, those two words do not necessarily denote physical light only to the exclusion of Brahman 1, but may also denote Brahman itself, in so far as it is characterised by the physical shining light which is its effect. Analogously another mantra declares, 'that by which the sun shines kindled with heat' (Taitt. Br. III, 12, 9, 7). Or else we may suppose that the word gyotis here does not denote at all that light on which the function of the eye depends. For we see that in other passages it has altogether different meanings; so, for instance, Bri. Up. IV, 3, 5, 'With speech only as light man sits,' and Taitt. Sa. I, 6, 3, 3, 'May the mind, the light, accept,' &c. It thus appears that whatever illuminates (in the different senses of the word) something else may be spoken of as 'light.' Hence to Brahman also, whose nature is intelligence, the term 'light' may be applied; for it gives light to the entire world. Similarly, other scriptural passages say, 'Him the shining one, everything shines after; by his light all this is lighted' (Kau. Up. II, 5, 15); and 'Him the gods worship as the light of lights, as the immortal' (Bri. Up. IV, 4, 16). Against the further objection that the omnipresent Brahman cannot be viewed as bounded by heaven we remark that the assignment, to Brahman, of a special locality is not contrary to reason because it subserves the purpose of devout meditation. Nor does it avail anything to say that it is impossible to assign any place to Brahman because Brahman is out of connexion with all place. For it is possible to make such
an assumption, because Brahman is connected with certain limiting adjuncts. Accordingly Scripture speaks of different kinds of devout meditation on Brahman as specially connected with certain localities, such as the sun, the eye, the heart. For the same reason it is also possible to attribute to Brahman a multiplicity of abodes, as is done in the clause (quoted above) 'higher than all.' The further objection that the light beyond heaven is the mere physical light because it is identified with the gastric fire, which itself is a mere effect and is inferred from perceptible marks such as the heat of the body and a certain sound, is equally devoid of force; for the gastric fire may be viewed as the outward appearance (or symbol) of Brahman, just as Brahman's name is a mere outward symbol. Similarly in the passage, 'Let a man meditate on it (the gastric light) as seen and heard,' the visibility and audibility (here implicitly ascribed to Brahman) must be considered as rendered possible through the gastric fire being the outward appearance of Brahman. Nor is there any force in the objection that Brahman cannot be meant because the text mentions an inconsiderable reward only; for there is no reason compelling us to have recourse to Brahman for the purpose of such and such a reward only, and not for the purpose of such and such another reward. Wherever the text represents the highest Brahman--which is free from all connexion with distinguishing attributes--as the universal Self, it is understood that the result of that instruction is one only, viz. final release. Wherever, on the other hand, Brahman is taught to be connected with distinguishing attributes or outward symbols, there, we see, all the various rewards which this world can offer are spoken of; cp. for instance, Bri. Up. IV, 4, 24, 'This is he who eats all food, the giver of wealth. He who knows this obtains wealth.' Although in the passage itself which treats of the light no characteristic mark of Brahman is mentioned, yet, as the Sûtra intimates, the mark stated in a preceding passage (viz. the mantra, 'Such is the greatness of it,' &c.) has to be taken in connexion with the passage about the light as well. The question how the mere circumstance of Brahman being
mentioned in a not distant passage can have the power of divorcing from its natural object and transferring to another object the direct statement about light implied in the word 'light,' may be answered without difficulty. The passage under discussion runs 1, 'which above this heaven, the light.' The relative pronoun with which this clause begins intimates, according to its grammatical force 2, the same Brahman which was mentioned in the previous passage, and which is here recognised (as being the same which was mentioned before) through its connexion with heaven; hence the word gyotis also--which stands in grammatical co-ordination to 'which'--must have Brahman for its object. From all this it follows that the word 'light' here denotes Brahman.
88:1 I. e. which has not been mixed with water and earth, according to Kh. Up. VI, 3, 3. Before that mixture took place light was p. 89 entirely separated from the other elements, and therefore bounded by the latter.
89:1 So as to justify the assumption that such a thing as non-tripartite light exists at all.
91:1 Brahmano vyavakkhidya tegahsamarpakatvam viseshakatvam, tadabhâvo#viseshakatvam. Ân. Gi.
93:1 If we strictly follow the order of words in the original.
93:2 Svasâmarthyena sarvanâmnah sannihitaparâmarsitvavasena.