theory-laden dependent on theory; specifically, involving a theoretical interpretation of what is perceived or recorded. In the heyday of logical empiricism it was thought, by Carnap and others, that a rigid distinction could be drawn between observational and theoretical terms. Later, N. R. Hanson, Paul Feyerabend, and others questioned this distinction, arguing that perhaps all observations are theory-laden either because our perception of the world is colored by perceptual, linguistic, and cultural differences or because no attempt to distinguish sharply between observation and theory has been successful. This shift brings a host of philosophical problems. If we accept the idea of radical theoryladenness, relativism of theory choice becomes possible, for, given rival theories each of which conditions its own observational evidence, the choice between them would seem to have to be made on extra-evidential grounds, since no theory-neutral observations are available. In its most perplexing form, relativism holds that, theory-ladenness being granted, one theory is as good as any other, so far as the relationship of theory to evidence is concerned. Relativists couple the thesis of theory-ladenness with the alleged fact of the underdetermination of a theory by its observational evidence, which yields the idea that any number of alternative theories can be supported by the same evidence. The question becomes one of what it is that constrains choices between theories. If theory-laden observations cannot constrain such choices, the individual subjective preferences of scientists, or rules of fraternal behavior agreed upon by groups of scientists, become the operative constraints. The logic of confirmation seems to be intrinsically contaminated by both idiosyncratic and social factors, posing a threat to the very idea of scientific rationality. See also CONFIRMA – TION , EPISTEMOLOGY , HYPOTHETICO -DEDUC – TIVE METHOD , INCOMMENSURABILITY , TEST- ABILIT. R.E.B.