Because Brahman, being raised above all contact with the senses, is not an object of perception and the other means of proof, but to be known through Scripture only; therefore the text 'Whence these creatures are born,' &c., has to be accepted as instructing us regarding the true nature of Brahman.--But, our opponent points out, Scripture cannot be the source of our knowledge of Brahman, because Brahman is to be known through other means. For it is an acknowledged principle that Scripture has meaning only with regard to what is not established by other sources of knowledge.--But what, to raise a primâ facie counter objection, are those other sources of knowledge? It cannot, in the first place, be Perception. Perception is twofold, being based either on the sense- organs or on extraordinary concentration of mind (yoga). Of Perception of the former kind there are again two sub-species, according as Perception takes place either through the outer sense-organs or the internal organ (manas). Now the outer sense-organs produce knowledge of their respective objects, in so far as the latter are in actual contact with the organs, but are quite unable to give rise to the knowledge of the special object constituted by a supreme Self that is capable of being conscious of and creating the whole aggregate of things. Nor can internal perception give rise to such knowledge; for only purely internal things, such as pleasure and pain, fall within its cognisance, and it is incapable of relating itself to external objects apart from the outer sense-organs. Nor, again, perception based on Yoga; for although such perception--which springs from intense imagination-- implies a vivid presentation of things, it is, after all, nothing more than a reproduction of objects perceived previously, and does not therefore rank as an instrument of knowledge; for it has no means of applying itself to objects other than those perceived previously. And if, after all, it does so, it is (not a means of knowledge but) a source of error.--Nor also inference either of the kind which proceeds on the observation of special cases or of the kind which rests on generalizations (cp. Nyâya Sû. I, 1,5,). Not inference of the former kind, because such inference is not known to relate to anything lying beyond the reach of the senses. Nor inference of the latter kind, because we do not observe any characteristic feature that is invariably accompanied by the presence of a supreme Self capable of being conscious of, and constructing, the universe of things.--But there _is_ such a feature, viz. the world's being an effected thing; it being a matter of common experience that whatever is an effect or product, is due to an agent who possesses a knowledge of the material cause, the instrumental cause, the final end, and the person meant to make use of the thing produced. It further is matter of experience that whatever consists of non-sentient matter is dependent on, or ruled by, a single intelligent