The Vedanta Sutras of Badarayana, Commentary by Sankara (SBE38), tr. by George Thibaut  at sacred-texts.com
17. Should it be said that on account of the connected meaning (of the whole passage) (the highest Self cannot be meant); (we reply that) it is so, on account of the assertion.
We now have to refute the objection, made above by the pûrvapakshin, that the highest Self cannot be meant 'on account of the connected meaning of the passage.'--The Sûtrakâra remarks, 'It is so, on account of the assertion.' That means: It is appropriate to understand the passage as referring to the highest Self, because thus the assertion that the Self, previously to the creation, was one only, gives a fully satisfactory sense, while on the other interpretation it would be far from doing so. The creation of the worlds recorded in the Aitareyaka we connect with the creation of the elements recorded in other Vedic texts, in that way that we understand the worlds to have been created subsequently to the elements; just as we showed above (II, 4, 1) that the passage, 'It sent forth fire,' must be understood to say that the creation of fire followed on the creation of ether
and air as known from other texts. For, as proved by us before, particulars mentioned in one scriptural text have to be combined with particulars mentioned in other texts, if only the chief subject of the passages is the same.--The details about the activity of the Self referred to by the pûrvapakshin have likewise to be understood in such a way as to agree with the general matter about which the text desires to make assertions. For we must by no means assume that the text is interested in setting forth all the details of the story on their own account; the knowledge of them would be in no way beneficial to man. The only thing the text really means to teach is the truth that Brahman is the Self of everything. Hence it first relates how the different worlds and the guardians of the worlds, viz. Agni and so on, were created; explains thereupon the origination of the organs and the body, their abode; and shows how the creator having thought, 'How can all this be without me?' (II, 4, 3, 4), entered into this body, 'Opening the suture of the skull he got in by that door' (7). Then again the text relates how the Self after having considered the activities of all the organs ('if speech names,' &c.; 6) asked himself the question, 'What am I?' and thereupon 'saw this person as the widely spread Brahman' (10). The aim of all which is to declare that Brahman is the universal Self. The same truth is inculcated in a subsequent passage also, viz. II, 6, 1, 5; 6, where the text at first enumerates the whole aggregate of individual existences together with the elements, and then continues, 'All this is led by knowledge (i.e. the highest Self); it rests on knowledge. The world is led by knowledge, knowledge is its rest, knowledge is Brahman."--For all these reasons the view that the highest Brahman is meant in the Aitareyaka is not open to any objections.
The two preceding Sûtras may also be explained with reference to some other Vedic passages. We read in the Vâgasaneyaka (Bri. Up. IV, 3, 7), 'Who is that Self?--He who is within the heart, surrounded by the prânas, consisting of knowledge, the person of light.' Of the Self here first mentioned the text goes on to show that it is free from all contact and thus proves it to have Brahman
for its Self, the concluding statement being, 'This great unborn Self undecaying, undying, immortal, fearless is indeed Brahman' (IV, 4, 25).--In the Khândogya again we have a chapter in which the introductory statement does not use the term 'Self ('Being only this was in the beginning, one, without a second'), while at the conclusion the term 'Self' is used in the declaration of identity ('That is the Self. Thou art that').--A doubt here arises whether these two scriptural texts treat of the same matter or not.
They do not, the pûrvapakshin maintains, since they are not equal. Since the determination of the sense depends on the letter of the text, we have no right to maintain equality of sense where the texts differ. In the Vâgasaneyaka the initial statement about the Self shows that the whole passage conveys instruction about the true nature of the Self. In the Khândogya, on the other hand, the initial clause is of a different kind, and we therefore must assume that the whole passage imparts instruction differing in nature from that of the Vâgasaneyaka.--But has it not been said that the Khândogya-passage also teaches in the end the doctrine of universal identity with the Self?--That has been said indeed (but wrongly); for as the concluding passage must be made to agree with the initial passage (which latter does not say anything about the identity of the Self and Brahman), we assume that the concluding passage merely enjoins an imaginative combination (sampatti) of the Self and Brahman.
To this we reply that also the passage, 'Being only this was in the beginning,' has to be understood as referring to the Self; 'as other places,' i.e. in the same way as the passage quoted from the Vâgasaneyaka. For what reason?--'On account of the subsequent (statement),' viz. the statement as to identity. And if it be said that 'on account of the connected meaning' of the initial passage in which no mention is made of the Self, the chapter cannot be understood to refer to the Self; we reply 'that it may be so understood on account of the assertion' made in the passage about that 'by which we hear what is not heard, perceive what is not perceived, know what is not
known.' For this passage asserts that through the knowledge of one thing all things become known, and to make good this assertion the text later on declares that 'Being only this was,' &c. Now this knowledge of all things through one thing is possible only if we understand the passage last quoted to refer to the Self; for if the principal Self were not known, how could all things be known? Moreover the assertion that, before creation, there existed one thing only, and the reference to the individual soul by means of the word 'Self,' and the statement that in deep sleep the soul becomes united with the True, and the repeated inquiries on the part of Svetaketu, and the repeated assertions, 'Thou art that,'--all this is appropriate only if the aim of the whole section is not to enjoin an imaginative meditation on all things as identical with the Self, but to teach that the Self really is everything.--Nor must it be said that, in the section under discussion, the concluding passage must be interpreted so as to agree with the introductory clause (and cannot on that account teach anything about the Self); for the introductory passage declares neither that the Self is everything, nor that the Non-self is everything (but merely makes a statement regarding what is in general), and such an altogether general statement cannot be in conflict with any particular statement made in a supplementary passage, but rather is in want of some such particular statement whereby to define itself 1.--And moreover (to view the matter from a different point of view), the word 'Being' if looked into closely can denote nothing else but the principal Self, since we have proved, under II, 1, 14, the unreality of the whole aggregate of being different from the Self.--Nor, finally, does a difference of expression necessarily imply a difference of sense; not any more than in ordinary language the to phrases, 'Bring that vessel
over there,' and, 'That vessel over there, bring it,' have different meanings.--It therefore remains a settled conclusion that in texts such as discussed above, the matter of instruction is the same, however much the mode may vary in which the instruction is conveyed.
210:1 I.e. the definite statement about the Self in the concluding passage may be used for defining the sense of the indefinite initial statement about that which is. 'That which is' comprises the Self as well as the Not-Self.