Empiricism, as a theory of knowledge rests upon the false proposition that perception and sensation constitute the only material and source of knowledge. Marx as a materialist, of course, never denied that the material world, existing prior to and independently of consciousness, is the only source of sensation. But he knew that such a statement, if left at that point, could not provide the basis for a consistent materialism, but at best a mechanical form of materialism, which always left open a loop-hole for idealism. It is true that empiricism lay at the foundation of seventeenth- and eighteenth-century materialism in England and France. But at the same time this very empiricist point of view provided the basis for both the subjective idealism of Berkeley and the agnosticism of Hume. How is it possible, starting with the proposition that sensation is the sole source and material of knowledge, to end up either denying the objectivity of the external world (subjective idealism) or denying the possibility of an exhaustive knowledge of that external world (scepticism)? To take the latter case, the argument runs as follows: to men are given directly perceptions and sensations; they provide the only legitimate source of knowledge. But in these perceptions are to be found no internal necessary connections. How do we know that one thing is the cause of another? We see only one thing followed by another; if this is constantly repeated we come to expect the second whenever the first occurs. This is merely a psychological expectation, not a causal connection. These were essentially the conclusions drawn by Hume from the empiricist theory of knowledge. It followed that any statements about the objectivity of the categories of philosophy or science (causality, interaction, law, etc.) are purely metaphysical, reflecting nothing in the sensed material of knowledge. On this view, logical categories are only schemes which we use (purely out of convention and habit) for the organisation of sense-data. But such schemes remain, necessarily, wholly subjective. They are subjective first in relation to the external world, the existence of which, according to scepticism, can never be established; second in relation to the very sense data themselves, since they are determined by the very constitution of the subject – that is by the aggregate of the individual’s former psychical experiences.

Marx’s objection to empiricism rests upon this: that its attention is directed exclusively to the source of knowledge, but not the form of that knowledge. For empiricism the form assumed by our knowledge tends always to be ignored as something having no inherent, necessary, connection with the content, the source of our knowledge. To return again to a previous example in the light of this: Ricardo saw in labour the source and measure of value, capital, etc. But he failed to consider the form assumed by this labour. Here was an expression not so much of the weakness of his economic theory as of his philosophical stance, empiricism. Here we can see why Marx considered it vital to examine economic forms and why political economy ignored this matter. (It must be said that this neglect is unfortunately to be found in much Marxist writing on Capital.)