The Vedanta Sutras of Badarayana, Commentary by Sankara (SBE38), tr. by George Thibaut [1896] at sacred-texts.com
11. Not on account of (difference of) place also twofold characteristics can belong to the highest; for everywhere (scripture teaches it to be without any difference).
We now attempt to ascertain, on the ground of Sruti, the nature of that Brahman with which the individual soul becomes united in the state of deep sleep and so on, in consequence of the cessation of the limiting adjuncts.--The scriptural passages which refer to Brahman are of a double character; some indicate that Brahman is affected by difference, so, e.g. 'He to whom belong all works, all desires, all sweet odours and tastes' (Kh. Up. III, 14,2); others, that it is without difference, so, e.g. 'It is neither coarse nor fine, neither short nor long,' &c. (Bri. Up. III, 8, 8). Have we, on the ground of these passages, to assume that Brahman
has a double nature, or either nature, and, if either, that it is affected with difference, or without difference? This is the point to be discussed.
The pûrvapakshin maintains that, in conformity with the scriptural passages which indicate a double nature, a double nature is to be ascribed to Brahman.
To this we reply as follows.--At any rate the highest Brahman cannot, by itself, possess double characteristics; for on account of the contradiction implied therein, it is impossible to admit that one and the same thing should by itself possess certain qualities, such as colour, &c.. and should not possess them.--Nor is it possible that Brahman should possess double characteristics 'on account of place,' i.e. on account of its conjunction with its limiting adjuncts, such as earth, &c. For the connexion with limiting adjuncts is unavailing to impart to a thing of a certain nature an altogether different nature. The crystal, e.g. which is in itself clear, does not become dim through its conjunction with a limiting adjunct in the form of red colour; for that it is pervaded by the quality of dimness is an altogether erroneous notion. In the case of Brahman the limiting adjuncts are, moreover, presented by Nescience merely 1. Hence (as the upâdhis are the product of Nescience) if we embrace either of the two alternatives, we must decide in favour of that according to which Brahma is absolutely devoid of all difference, not in favour of the opposite one. For all passages whose aim it is to represent the nature of Brahman (such as, 'It is without sound, without touch, without form, without decay,' Ka. Up. I, 3, 15) teach that it is free from all difference.