The present-day epistemology of social science resembles a picture puzzle whose pieces are scattered to and fro across the vast domain of philosophical inquiry. This study attempts to assemble them in what appears to be a common thread of thinking for a necessary epistemic reconstruction, the historical specificity of social sciences. This understanding reveals itself as a method of validating truth in acknowledgement of three logical principles: (1) causality indeterminately becomes embedded in spatial—temporal distortions; (2) linear time is replaced by (...) multiple, overlapping timescales, ‘multiple’ being a cultural rather than numerical concept; and (3) prediction remains associated with the least historical events, the particulars; that is, event-regularities normally specific for short periods of time. (shrink)