Veena - thanks for your response - it is good to discuss these issues. I fully agree with you that the participation of poor people is important (where deep and genuine, it fulfils an obligation that goes alongside any claim to an exercise of legitimate power)and I take your point that poverty is very often experienced as humiliation and powerlessness. I agree that there should be a serious discussion of cash transfers with and without strings and what their advantages and drawbacks might be, and if my comments have any merit, it is in their relevance for the metric and consequences of such an assessment.
I am not sure that cash transfers to the poor offer an alternative to aid or rights discourses (were you saying this?). Just for clarity, I take your point that responsibilities can be used as 'strings attached' in practice, but for something to be a human right is to say that access to it should not be conditional: so conditionality is hostile to human rights thought - "you can have your human right fulfilled, but only if you do X" is the wrong message.
More basically, rights or charity offer answers to the question "why help the poor" whether that be in terms of cash transfers or alternatives (e.g. political empowerment). So I was thinking they are present in the background even of the cash transfer policy (I mean, if you don't think it is either morally good or morally required to give money to the poor, why give money to the poor?), and if they are in the background, they might be relevant to assessing whether it works. To sharpen this, giving money (or any aid) to the poor requires justification because (a) that money comes from somewhere or someone else and (b) it could be spent in a number of ways that advanced human welfare.
It would be interesting to know, when poor people ask "why are you giving us money?" what the answer is - how would anyone give an answer that didn't sound patronising, and without echoes of either (a) feeling sorry for them (b) having to, regardless of whether we wanted to or (c) calculating our selfish interest to do so. All of these, I would guess, would have some effect on how the poor construct themselves - I'm genuinely interested how this goes in practice.
I agree that caution about goals is required, but we may disagree about how much. It is worth distinguishing between ends/objectives/goals and potential means here. The retrospective assessment of goals you mention looks at consequences - but aren't the opportunities to flourish offered by education, health care etc. good in themselves - and maybe in a way that cash isn't? To put this more sharply, it seems odd to say "it will only be retrospectively clear whether preventing easily preventable child deaths was correct". And it seems odd to view the consequences of things that are morally required in themselves as "distortions" (but your point is well taken with respect to any particular means of realising an end).
So, when it comes to means, caution should surely be applied to cash transfers too - but I wouldn't pretend to have any knowledge of how the arguments for and against stack up in the end. Even if unconditional cash transfers are what the poor want (and, facetiously, everyone wants them of course), this doesn't seem decisive, on my view - maybe we differ here?