Sarvadarshanasangraha 13.51

13.51

āvṛttiparipākāyāṃ buddhau śabdo ’vadhāryate//

All this long oration is really only like a drowning man's catching at a straw;[21] for either of the alternatives is impossible, whether you hold that it is the single letters or their aggregation which conveys the meaning of the word. It cannot be the former, because a collection of separate letters, without any one pervading cause,[22] could never produce the idea of a word any more than a collection of separate flowers would form a garland without a string. Nor can it be the latter, because the letters, being separately pronounced and done with, cannot combine into an aggregate. For we use the term "aggregate" where a number of objects are perceived to be united together in one place; thus we apply it to a Grislea tomentosa, an Acacia catechu, a Butea frondosa, &c., or to an elephant, a man, a horse, &c., seen together in one place; but these letters are not perceived thus united together, as they are severally produced and pass away; and even on the hypothesis of their having a "manifesting" power, they can have no power to form an aggregate, as they can only manifest a meaning successively and not simultaneously. Nor can you imagine an artificial aggregate in the letters, because this would involve a "mutual dependence" (or reasoning in a circle); for, on the one hand, the letters would only become a word when their power to convey one meaning had been established; and, on the other hand, their power to convey one meaning would only follow when the fact of their being a word was settled. Therefore, since it is impossible that letters should express the meaning, we must accept the hypothesis of sphoṭa. "But even on your own hypothesis that there is a certain thing called sphoṭa which expresses the meaning, the same untenable alternative will recur which we discussed before; and therefore it will only be a case of the proverb that 'the dawn finds the smuggler with the revenue-officer's house close by.'"[23] This, however, is only the inflation of the world of fancy from the wide difference between the two cases. For the first letter, in its manifesting power, reveals the invisible sphoṭa, and each successive letter makes this sphoṭa more and more manifest, just as the Veda, after one reading, is not retained, but is made sure by repetition; or as the real nature of a jewel is not clearly seen at the first glance, but is definitely manifested at the final examination. This is in accordance with the authoritative saying (of the teacher): "The seed is implanted by the sounds, and, when the idea is ripened by the successive repetition, the word is finally ascertained simultaneously with the last uttered letter." Therefore, since Bhartṛhari has shown in his first book that the letters of a word [being many and successive] cannot manifest the meaning of the word, as is implied by the very phrase, "We gain such and such a meaning from such and such a word," we are forced to assume the existence[24] of an indivisible sphoṭa as a distinct category, which has the power to manifest the word's meaning. All this has been established in the discussion (in the Mahābhāṣya) on "genus" (jāti), which aims at proving that the meaning of all words is ultimately that summum genus, i.e., that existence whose characteristic is perfect knowledge of the supreme reality[25] (Brahman).