‪Unlike the physical sciences, sociology is frequently described as an interpretative non-experimental science. Comparative epistemology sheds new light on this claim. 1. Experimentation is not a constant character of the physical sciences; 2. Experimental hypothetical-deductive reasoning, including the test of predictions, is also practicable in sociology. The argument is developed by a detailed step-wise comparison of the prediction of light ray deviation within the Sun’s gravitational field made in 1919 (physics) and the prediction of 8% cosmopolitanism of Cambridge University between 1250 and 1350 (sociology). Extensive analysis of regent masters’ prosopographic notes yields the result of 9.5 percent cosmopolitanism, which is close to the theoretical prediction.