During the course of a series of lectures delivered at the Indian Institute of Advanced Study in 2005–in an attempt to explicate what he saw as one of the primary distinctions between the ‘Western’ and the ‘Indian’ conceptions of the relationship of the individual to society–Daya Krishna noted:
The idea that one may be responsible for actions that have not been taken by one’s own self and that one may be redeemed by someone else’s action [as man’s by Christ] may seem positively outrageous to a sensibility that treats the individual as essentially apart for his relationships with others, relationships in which he may happen to be accidentally involved. The doctrine of karma in traditional Hindu thought primarily reflects this basic presupposition that it would be an immoral world indeed if one were to reap the fruits of someone else’s actions. The monadic morality of the Hindu is thus conceived of in an essentially asocial manner. It does not derive from an other-centered consciousness in which the consequences of one’s actions on others are the subject of one’s focus of attention. Rather, it is the consequences of one’s actions upon oneself which provides the main ground for morality in Hindu thought and thus paves the way for a very different kind of perspective on the entire issue of action and one’s relations with others. At the deepest level, not merely does what one does have consequences upon oneself but, conversely whatever happens to one could only be the result of one’s own actions. Thus, not only do one’s own actions have consequences on oneself, but also, if the world is to be a moral world, nothing else could. [Civilizations: Nostalgia and Utopia, Sage Publications, 2012, pp. 13-14]
This explication of the doctrine of karma raises, I think, a pair of vexing questions.
To see this, consider some actor A that takes action X. The consequences of X can only accrue to A. But Y may have–visibly, in this world–effects that impinge on another being B. For instance, I may tip a cup of hot water–as a cruel joke–on a waiter at a restaurant. According to Krishna, what has happened to B–the waiter–is a consequence of B‘s actions, perhaps in a past life, perhaps in the present one. But A is the actor, the agent, that brought about those effects on B, so are the causes of A‘s actions B‘s actions?
Furthermore, actions are, I presume, within the karmic doctrine, reckoned as good or bad, moral or immoral, a calculus which then plays out in the effects that will take place in the future on their actor. A‘s action X of tipping a glass of hot water–purely for jest–on a waiter is, let’s say, an immoral one: it is gratuitously cruel. A has then presumably accrued a ‘negative credit’ of sorts in the karmic ledger, one which will presumably result in some negative effect on A in the future–again, either in this life, or in some future one. But will A then be chastised for an action, X, that was the result of B‘s actions?
I hope the puzzles of responsibility and action that I had in mind are visible. I welcome clarification if I’ve misunderstood Krishna in any way. (Perhaps Y‘s effects on B are only apparent etc.)
Samir Chopra 
I like this. I think you’re right that it’s tricky to say who is really responsible. But I’ll just give it a quick thought. You say that karma is working, in the case of B, to bring retribution for a past action, Y, which B had previously inflicted on another, and that A is therefore potentially free of guilt/responsibility for having spilled the hot water on B. A would just be the agent who is ‘used’ by karma to bring about this retribution. But I think the paradox of such a situation is pretty clear: that there is really no such thing as freedom in action, despite the appearance of free will which we would all intuitively think exists. It’s an unfree freedom, since karma informs or utilizes our motivations to bring about the actions which we otherwise think we are freely performing.
But I’d probably say that karma isn’t only being exercised against B through A’s act of spilling the cup, because we know that A deliberately chose to do it. I’d think that, if karma is a true description of agency, there would be a potential argument of infinite regression, since B’s act in his previous or present life also wouldn’t strictly speaking be his act; it’d be another act of retribution directed at some third agent, C, which was caused by C’s act of Z-ing. So it seems that there isn’t a way out of this circular argument with its intrinsic paradox unless we either at least admit to a rendition of the theory of compatiblism or reject the theory of karma outright.
I also believe in karma, though it’s hard to digest the idea that we only are able to freely choose what we were necessarily determined to do. The latter is definitely a bullet that you have to bite. Or do you see some way of getting around this disjunction?
Hi Samar… This is a very thought provoking post indeed. Two key questions you have posted:
Question: But A is the actor, the agent, that brought about those effects on B, so are the causes of A‘s actions B‘s actions?
My take: I don’t think it is either A or B that is the actor. I think time does not flow forward or backward. Hence time and causality dont really exist.
Question: But will A then be chastised for an action, X, that was the result of B‘s actions?
My take: Action is only a link. Undirected association (not uni or bi). The only ones who are in trouble are those that think this as a directed graph and that they are the antecedents. That is 99.999999999999% of us.