Is Paul’s inferential view a viable response to the self-knowledge problem? One question is whether Paul fully preserves the first-personal/third-personal asymmetry that was emphasized at the beginning of this paper […]. For on an account like Paul’s while there will clearly be fundamental differences between your and my respective ways of knowing about my intentions, it appears that once that knowledge is in place, you may be in every bit as good a position as I am to know what I am doing on the basis of the knowledge of those intentions plus my general tendencies and the evident favorability of my circumstances: this is an inference that you are every bit as capable of making as I am, and it seems that you will be no less justified in it. But it seems as if it is knowledge of one’s actions, and not just of the intentions that underlie them, that ought to be characterized by first-personal privilege, and it is a significant defect of Paul’s position if it fails to ensure this.
Schwenkler, J. (2011) “Perception and Practical Knowledge”: p.143
I have a few remarks about Schwenkler’s remark about Sarah Paul’s view.
 I agree that a proper theory for practical knowledge must be able to account for the first-personal/third-personal asymmetry. But it is a mistake to construe it as a task to understand first-personal privilege. What we care about is not epistemic privilege of the first person. There is no contradiction to the idea that there is something special about practical knowledge to accept that there is an all knowing God, who knows about our actions better than ourselves. It is the first-personal access, not the privilege that interests us.
 Schwenkler’s objection to Paul is not entirely fair. Look, you can’t accuse Paul’s theory for not being able to account for the difference between two kinds of knowledge by abstracting everything else away and leaving only the inferential patterns behind. If you do that, then of course there is no difference (left) between my knowledge about my own action and your knowledge of my action. Paul would argue that that is only because you have chosen to abstract away all the factors that are supposed to make the difference: the introspective data about my own intentions that I have and you do not have. In the quote, Schwenkler opines that one must find something special about our knowledge of our own actions beyond whatever makes our access to our own intentions special in order to properly address the first-persona/third-personal asymmetry. It is not clear to me why Schwenkler accepts that further condition.
 Despite my disagreement with Schwenkler’s objection to Paul’s view, I believe he is right that her view does not properly explain the asymmetry, but for a different reason. Consider this analogy. Say I believe I must have eaten something spoilt, because I have really bad stomachache. My access to my stomachache is introspective. And I infer from this introspective datum to the conclusion that I’ve eaten something spoilt. It is clear that you can also believe that I have eaten something spoilt but you would not have the introspective datum as part of your inferential base. In that sense, your access to the fact is different from mine. Paul can use the phrase “non-observational knowledge” however she likes. But even if there is a twisted, artificial definition of the phrase such that my knowledge that I have eaten something spoilt counts as non-observational knowledge, that can’t be the kind of non-observational knowledge that we are trying to highlight in practical knowledge. Our practical knowledge is non-observational in a special way that is not shared by my knowledge about my diet. And I believe the key is the active/passive distinction. My knowledge about my actions is, in some sense, not passively received. And the activity of practical knowledge cannot be captured by the idea of introspection at all.
 Here is a second reason to think that the inferential account does not successfully capture what we intend to capture about practical knowledge. Surely I have my data-base about my own intentional actions. You data-base about my actions is different from mine. But notice that Susan’s data-base about my actions is also different from mine and yours. Everyone has a different set of data about my actions. (In fact, if everyone has his own data set for everything.) If an account of the first-personal/third-personal distinction just has the distinction boils down to the difference in the data set at the bottom, the account fails to capture anything of special interest. My knowledge of my own actions is meant to be not just different, but different in a unique way. It is different in a way that cuts deep. Surely, some of my data is introspective. So what? Say some of Susan’s knowledge about my actions is based on testimony from my therapist. And say some of your knowledge about my actions is based on your direct observation. So, we each have our own source of data. So what? Heck, part of Susan’s data set might even be introspectively available: she might think that I am trying to poison her because she feels dizzy after drinking something I offered her. What we have seen is just that there are countless differences among people’s knowledge of the same fact. The inferential account is yet to illuminate on the difference that makes first-personal knowledge of action a unique kind of knowledge.