According to instrumentalists, a successful scientific theory reveals nothing known either true or false about nature's unobservable objects, properties or processes.
[8] For Berkeley, a scientific theory does not state causes or explanations, but simply identifies perceived types of objects and traces their typical regularities.
[9] The last great British empiricist, David Hume, posed a number of challenges to Francis Bacon's inductivism, which had been the prevailing, or at least the professed view concerning the attainment of scientific knowledge.
Upon reading Hume's work, Immanuel Kant was "awakened from dogmatic slumber", and thus sought to neutralise any threat to science posed by Humean empiricism.
Kant's metaphysics, then, transcendental idealism, secured science from doubt—in that it was a case of "synthetic a priori" knowledge ("universal, necessary and informative")—and yet discarded hope of scientific realism.
Since the mind has virtually no power to know anything beyond direct sensory experience, Ernst Mach's early version of logical positivism (empirio-criticism) verged on idealism.
Moritz Schlick debated with Otto Neurath over foundationalism—the traditional view traced to Descartes as founder of modern Western philosophy—whereupon only nonfoundationalism was found tenable.
And since science aims to reveal not private but public truths, verificationists switched from phenomenalism to physicalism, whereby scientific theory refers to objects observable in space and at least in principle already recognizable by physicists.
Suddenly striking Western society, then, was Kuhn's landmark thesis, introduced by none other than Carnap, verificationism's greatest firebrand.
One scientific realist, Karl Popper, rejected all variants of positivism via its focus on sensations rather than realism, and developed critical rationalism instead.
[11] The British physicist David Deutsch, in his much later 1997 book The Fabric of Reality, followed Popper's critique of instrumentalism and argued that a scientific theory stripped of its explanatory content would be of strictly limited utility.
[12] Bas van Fraassen's (1980)[13] project of constructive empiricism focuses on belief in the domain of the observable, so for this reason it is described as a form of instrumentalism.