The Vedanta Sutras of Badarayana, Commentary by Sankara (SBE38), tr. by George Thibaut  at sacred-texts.com
2. And on account of an indicatory mark.
An indicatory mark also gives to understand that repetition is required. For, in the section treating of meditation
on the udgîtha, the text rejects the meditation on the udgîtha viewed as the sun, because its result is one sun only, and (in the clause 'Do thou resolve his rays,' &c.) enjoins a meditation on his manifold rays as leading to the possession of many suns (Kh. Up. I, 5, 1; 2); which shows that the repetition of meditations is something well known. Now as other meditations are meditations no less than the one referred to, it follows that repetition holds good for all of them.
Here the following objection may be raised. With regard to those meditations whose fruit is something to be effected repetition may hold good, because thereby superior strength may be imparted to them. But of what use can repetition be with regard to the meditations having for their object the highest Brahman, which present to us Brahman as the universal Self characterised by eternal purity, thought, and freedom? Should it be said that repetition has to be allowed because the knowledge of Brahman being the Self cannot spring up on hearing a text once only, we reply that in that case it will not spring up even when it is heard repeatedly. For if a text such as 'Thou art that' does not originate the true notion of Brahman if heard once, what hope is there that the desired effect should be produced by its repetition?--Perhaps it will be said that a sentence alone is not able to lead to the intuition of a thing; but that a sentence assisted by reasoning may enable us to intuite Brahman as the universal Self. But even in that case repetition would be useless; for the reasoning will lead to the desired intuition even if gone through once only.--Again it will perhaps be said that the sentence and reasoning together effect only a cognition of the generic nature of the object known, not of its specific individual character. When, to exemplify this, a man says that he feels a pain in his heart another person can infer from this statement--and certain accompanying symptoms such as trembling of the limbs--only that there exists a pain in general but is unable to intuite its specific character; all he knows is 'This man suffers a pain.' But what removes ignorance is (not
a general knowledge but) the intuitive knowledge of the specific character of something. And repetition serves to produce such knowledge.--This also is not so. For if so much only is done repeatedly even, no specific knowledge can spring up. When a specific character is not cognized through scripture and reasoning being applied once, it will not be cognized through them if applied a hundred times even. Hence whether scripture and reasoning produce specific knowledge or general knowledge, in either case they will do so even if acting once only; and repetition therefore is of no use. Nor can it be laid down as a binding rule that scripture and reasoning, applied once, in no case produce intuitive knowledge; for their effect will after all depend on the various degrees of intelligence of those who wish to learn. Moreover a certain use of repetition may be admitted in the case of worldly things which consist of several parts and possess generic character as well as individual difference; for there the student may grasp by one act of attention one part of the object, and by another act another part; so e.g. in the case of long chapters to be studied. But in order to reach a true knowledge of Brahman whose Self is mere intelligence and which therefore is destitute of generic character as well as specific difference there clearly is no need of repetition.
To this we make the following reply. Repetition would indeed be useless for him who is able to cognize the true nature of Brahman even if enounced once only in the sentence 'Thou art that.' But he who is not able to do that, for him repetition is of use. For this reason the teacher in the Khândogya, having given instruction in the sentence 'Thou art that, O Svetaketu,' and being again and again asked by his pupil--'Please, sir, inform me still more'--removes his pupil's reasons for doubt, and again and again repeats the instruction 'Thou art that.' We have already given an analogous explanation of the passage 'The Self is to be heard, to be thought, to be reflected upon.'--But has not the pûrvapakshin declared that if the first enunciation of the sentence 'Thou art that' is not able to effect an intuition of its sense, repetition will likewise
fail of the desired effect?--This objection, we reply, is without force, because the alleged impossibility is not confirmed by observation. For we observe that men by again and again repeating a sentence which they, on the first hearing, had understood imperfectly only, gradually rid themselves of all misconceptions and arrive at a full understanding of the true sense.--Moreover the sentence 'Thou art that' teaches that what is denoted by the term 'thou' is identical with what is denoted by 'that.' Now the latter term denotes the subject of the entire section, viz. the thinking Brahman which is the cause of the origin and so on of the world; which is known from other passages such as 'Brahman which is true knowledge, infinite' (Taitt. Up. II, 1); 'Brahman that is knowledge and bliss' (Bri. Up. III, 9, 28); 'That Brahman is unseen, but seeing; unknown, but knowing' (Bri. Up. III, 8, 11); 'not produced' (Mu. Up. II, 1, 2); 'not subject to old age, not subject to death' (Bri. Up. IV, 4, 25); 'not coarse, not fine; not short, not long' (Bri. Up. III, 8, 8). In these passages terms such as 'not produced' deny the different phases of existence such as origination; such terms as 'not coarse' deny of it the qualities of substances such as coarseness; and such terms as 'knowledge' declare that the luminousness of intelligence constitutes its nature. The entity thus described--which is free from all the qualities of transmigratory existence, has consciousness for its Self and is called Brahman--is known, by all students of the Vedânta, as what is denoted by the term 'that.' They likewise know that what is denoted by the term 'thou' is the inward Self (pratyagâtman); which is the agent in seeing and hearing, is (successively) apprehended as the inward Self of all the outward involucra beginning with the gross body (cp. Taitt. Up.), and finally ascertained as of the nature of intelligence. Now in the case of those persons for whom the meaning of these two terms is obstructed by ignorance, doubt, and misconception, the sentence 'Thou art that' cannot produce a right knowledge of its sense, since the knowledge of the sense of a sentence presupposes the knowledge of the sense of the words; for them therefore the repetition of the scripture
text and of reasoning must be assumed to have a purpose, viz. the discernment of the true sense of the words.--And although the object to be known, viz. the Self, does not consist of parts, yet men wrongly superimpose upon it the attribute of being made up of many parts, such as the body, the senses, the manas, the buddhi, the objects of the senses, the sensations, and so on. Now by one act of attention we may discard one of these parts, and by another act of attention another part; so that a successively progressing cognition may very well take place. This however is merely an antecedent of the (true) knowledge of the Self (in which there can be no successive stages).
Those quick-witted persons, on the other hand, in whose mind the sense of the words is not obstructed by ignorance, doubt, and misconception, are able to intuite the sense of the sentence 'Thou art that' on its first enunciation even, and for them therefore repetition is not required. For the knowledge of the Self having once sprung up discards all ignorance; so that in this case no progressive process of cognition can be acknowledged.--All this might be so--an objection is raised--if cognition did spring up in any mind in the way described. (But this is not the case); for the cognition of the Self being subject to pain and so on has such strength that nobody ever reaches the cognition of all absence of pain and so on.--This objection, we reply, is without force; for it can be shown that the conceit of the Self being subject to pain, &c., is a wrong conceit, no less than the conceit of the body being the Self. For we clearly observe that when the body is cut or burned a wrong notion springs up, 'I am being cut,' 'I am being burned;' and similarly we observe that when sons, friends, &c. who are even more external to the Self than one's own body--suffer affliction, that affliction is wrongly attributed to the Self. Analogous to these cases is the conceit of the Self being subject to pain, &c.; for like the body and so on, the condition of being subject to pain is observed as something external to intelligence. This moreover follows from its not being continued in such states as dreamless sleep and the like; while scripture expressly declares that in deep
sleep intelligence suffers no interruption, 'And when there he does not see, yet he is seeing,' &c. (Bri. Up. IV, 3, 22). Hence the intuition of the Self consists in the knowledge, 'My Self is pure intelligence free from all pain.' For him who possesses that knowledge there remains no other work. Thus scripture says, 'What shall we do with offspring, we who have this Self and this world' (Bri. Up. IV, 4, 22). And Smriti also says, 'But that man who loves the Self, is satisfied by the Self and has all his longings stilled by the Self only, for him there is no further work' (Bha. Gîtâ III, 12).--For him, on the other hand, who does not reach that intuition all at once, we admit repetition, in order that the desired intuition may be brought about. He also, however, must not be moved towards repetition in such a way as to make him lose the true sense of the teaching, 'Thou art that.' In the mind of one on whom repetition is enjoined as a duty, there arise infallibly notions opposed to the true notion of Brahman, such as 'I have a claim on this (knowledge of the Self) as an agent; this is to be done by me 1.' But if a learner, naturally slow-minded, is about altogether to dismiss from his mind the purport of the sentence, because it does not reveal itself to him, it is permissible to fortify him in the understanding of that sense by means of reasoning on the texts relative to repetition and so on.--All this establishes the conclusion that, also in the case of cognitions of the highest Brahman, the instruction leading to such cognition may be repeated.
337:1 Care must be taken not to engender in the mind of such a learner the notion that the repeated acts of reflection are incumbent on him as a duty; for such notions would only obstruct the end aimed at, i.e. the intuition that the Self of the meditating man is identical with Brahman's Self, to which no notions of duty or action apply.