An action born together with oneself, arising with one's very birth, that natural action one should not give up, Kaunteya, even though it is faulty, since it is of the nature of the three qualities. All undertakings, that is, by the context, all actions whatever, whether one's own duty or another's, are wrapped about by fault, because they are of the nature of the three qualities, as fire is wrapped by the smoke that is born with it; that they are of the nature of the three qualities is here the reason. Even if one abandons the natural action called one's own duty and takes up another's duty instead, one is not freed of fault, and another's duty brings fear; and the ignorant man cannot in any case give up action without remainder; therefore one should not give it up. Is it because action cannot be given up without remainder by the ignorant that one should not give it up, or is there a fault in the very act of giving up one's natural action? And what follows from this? If action is not to be given up because it cannot be given up without remainder, then it follows that giving it up completely would be all to the good. True, if the complete giving up does not hold at all. Does it not? Is a man, as the Sankhyas hold of the qualities, ever in motion; or is the action itself the agent, as the Buddhists hold the aggregates to be momentary and perishing? On either view a complete giving up of action does not hold. Then there is a third position: when the thing acts it is active, when it does not act that same thing is inactive; and on that view action can be given up without remainder. The peculiarity of this third position is this: the thing is not ever in motion, nor is the action itself the agent; rather, in a stable substance an action that was not there arises, and one that was there perishes, while the substance itself abides, pure and possessed of power; so the followers of Kanada say, and that substance is the agent. What is the fault in this position? Just this fault: it is a doctrine not of the Lord's party. How is that known? Because the Lord says 'of the non-existent there is no coming-to-be', and the rest. The Kanada doctrine, that the non-existent comes to be and the existent ceases to be, is not of the Lord's party. If even a doctrine not of the Lord's party were sound in reasoning, what would be the fault? The answer is that this one is faulty, for it conflicts with all the means of knowledge. How? If a substance such as a dyad of atoms is, before its origination, utterly non-existent, and, having arisen and stood for some time, again becomes utterly non-existent, then the non-existent comes to be existent and the existent comes to be non-existent; absence becomes presence and presence becomes absence. In that case the absence, while coming to be, before its origination is like a hare's horn, and yet it arises in dependence on a cause called the inherent, the non-inherent, and the instrumental cause. But it cannot be said that absence arises and depends on a cause, for non-existent things like a hare's horn are never seen to do so. If pots and the like were of the nature of being while coming to be, it could be understood that they arise in dependence on some cause that is a mere making-manifest. Further, if the existent and the non-existent could each become the other, no one would ever have any confidence in the dealings of means and objects of knowledge, since the certainty 'the existent is just existent, the non-existent just non-existent' would not hold. Further, the Kanada says that a substance such as a dyad of atoms, when it 'arises', comes into connection with the existence belonging to its cause. Before its origination it is non-existent; afterward, in dependence on the working of the cause, it is connected with its own causes, the atoms, and with existence, by the connection called inherence; and being so connected, it becomes existent, inhering in its cause. Here it must be told how the non-existent could have a cause of its own, or could be connected with anything. For no one can, by any means of knowledge, suppose a cause of its own, or a connection with anything, for the son of a barren woman. It may be replied that the Vaisheshikas do not posit a connection of an absence; for the connection called inherence in its own cause is said only of substances such as the dyad that are existent. No, for they do not admit existence before the connection. The Vaisheshikas do not hold that pots and the like exist before the working of potter, staff, wheel, and the rest; nor do they hold that it is the clay itself that takes on the form of the pot. Hence, by elimination, a connection of what is non-existent is, against their will, what they are committed to. It may be replied that a connection of inherence even of the non-existent is not contradictory, since, unlike the son of a barren woman, it is not unseen; the prior absence of the pot itself, not of the son of a barren woman, comes into connection with its own cause, the difference of absences having to be told even though they are alike in being absences. But by their defining marks no difference can be shown among the absence of one thing, the absence of two, the absence of all, the prior absence, the absence-by-destruction, the mutual absence, and the absolute absence. And if there is no difference, then it is incongruous to say that the prior absence of the pot alone is brought by potter and the rest into the state of being a pot and is connected with an existent thing called a potsherd and so becomes fit for all dealings, while the absence-by-destruction of the pot, though equally an absence, is not; for then absence-by-destruction and the rest would never be fit for dealings, and only the prior absence, called a substance such as the dyad of atoms, would be fit for the dealings of origination and the rest, the two being indistinguishable as absences, like the absolute absence and the absence-by-destruction. It may be said: we do not assert that a prior absence comes to be a being. Then it is that a being comes to be a being, as the pot becomes a pot, or the cloth a cloth. This too, like the coming-to-be of a being out of an absence, conflicts with the means of knowledge. And the Sankhya's doctrine of transformation, since it admits the arising and perishing of new attributes, is not different from the Vaisheshika position. Even if it admits only manifestation and concealment, when the existence or non-existence of the manifestation and concealment is examined, the conflict with the means of knowledge is the same as before. By this, the view that origination and the rest are just the configuration of the cause is also answered. By elimination, then, the one existent thing is, by ignorance, fancied in many ways, with the attributes of origination, destruction, and the rest, like an actor. This is the doctrine of the Lord's party, stated in the verse 'of the non-existent there is no coming-to-be', for the cognition of the existent never fails while the others do. Then how, given the changelessness of the self, can the complete giving up of action not hold? Whether the qualities are real things or are fashioned by ignorance, action is their attribute; in that case it is only superimposed on the self by ignorance, and so it was said that the ignorant man cannot, even for a moment, give up action completely. But the knower, when by knowledge ignorance has ceased, is able to give up action completely without remainder, for what was superimposed by ignorance can have no residue. When the disorder of vision is gone, no residue of the second moon superimposed by it remains. And this being so, the words 'renouncing all actions with the mind', and 'devoted to his own action a man attains consummation', and 'worshipping Him with his own action a man finds consummation', all hold good. The consummation born of action, which is the fitness for the standing in knowledge, has been told; now its fruit, the consummation-in-actionlessness which is the standing in knowledge, is to be told, and the verse is begun.
Contemporary English rendering of the Sanskrit bhāṣya, pending scholar review.