Albert Hirschman spent almost five years in Colombia were he said he had acquired a point of view. This point of view led him to formulate an original approach to development, far from grand theories and mere empiricism. What he called a strategy of development relied not only on linkages between productive sectors but mostly on skills, beliefs and expectations necessary, and many times unexploited, to make a proactive decision process possible. He was convinced individuals were more than recipients of development policies, and saw them as active agents of their own development processes. In this paper, after exploring this point of view, I present what I see as Hirschman's possible lessons for modern day development economics, in general, and randomized field experiments, in particular.