This is wrong, though I empathize. Social scientists come up with useful models that help explain events and behaviours, that aren't disproven by empirical evi

This is wrong, though I empathize. Social scientists come up with useful models that help explain events and behaviours, that aren't disproven by empirical evidence. There are plenty of ways to falsify a social science theory: wait long enough, or study enough data, and you'll amass enough empirical evidence to determine whether your theory holds up or not. Of course, this does not mean it is proven - there may be a falsifiable data point some day. But that's why we call it a theory. Environmental science is somewhat harder because the timescales are much longer than we have existed. But we also have well established ways of estimating the climate of ancient eras. This is why global warming is considered a scientific theory. In economics, there is, for example, a lot of evidence these days in favor of the standard Keynesian IS-LM model for a macroeconomy in liquidity trap conditions, and a dearth of evidence supporting neoclassical models and Real Business Cycle theory. This is almost a paradigm re-shift, in that a previously abandoned paradigm (by many), thought by some to have been falsified, is actually proving more predictive with respect to actual events than previously recognized. Unfortunately economics is closely intertwined with politics and ideology which doesn't mix well with science -- thus the profession is in crisis. An alternative, I suppose, is to go "full Austrian" and presume economics is an inherently unfalsifiable subject, based on fundamental logical truths. Thus it is justified to ignore all empirical evidence and mathematical analysis in favor of deductive praxeology. This is what Von Mises and Rothbard (founder of the US Libertarian party) believed, and they've been a common source of contra-Krugman arguments.