On the one hand, Newton seemed to have banished a priori explanation from natural philosophy, and established experiment as the sole basis for scientific truth; but on the other hand, he used concepts that seemed to have no foundation in sense experience. He had shown that gravitational force acted at a distance without any direct physical contact. But as the scientists of his day regarded “natural” and “mechanical” virtually as synonyms, and believed that a “natural” explanation had to be in terms of particles of matter in motion, Newton seemed to have suggested that matter was moved by a non-material, supernatural force. Moreover, one of his central concepts in the Principia was “absolute space and time” for which he gave no experimental evidence. Newton never discussed the metaphysical status of absolute space and time, but he suggested that it might be the sensorium of the Deity, and sometimes spoke as if it described a reality. Certainly, many of his followers took the reality for granted.
The continental admirers of the experimental method, Leibnitz, Huygens, Bernoulli, declared that Newton had left the straight and narrow path of empiricism and had taken refuge in scholastic, occult qualities to explain natural phenomena. As gravitational force was not given by experience, Newton had made a causal inference that was inadmissible in science. Their suspicions were confirmed by the defenders of religion, who eagerly claimed Newton’s theory as evidence for the existence of an immaterial mind and welcomed the notions of “absolute time and space” as proof of man’s ability to discover realities beyond experience, to grasp “occult qualities,” and therefore to know God. Or else, like Clarke, the pious saw a direct relation between infinite space and infinite intelligence.1 Newton himself preferred “to avoid all questions about the nature or quality of this force,” insisting that he had merely described what he had observed, and did not pretend to explain it. He was sufficiently disturbed by the dangerous implications, however, to try, unsuccessfully, to explain gravitation by means of an ethereal fluid.
Certainly, in the context of mechanically oriented science, both of Newton’s concepts carried intimations of divinity. They moved his contemporaries to consider whether science had to go beyond experience and have recourse to metaphysical realities to explain natural phenomena. They brought into question the whole nature and validity of the causal inference and of abstract ideas.
Hume’s explanation of space and time followed the lines suggested by Newton’s critics. Leibnitz, as well as Toland, argued that time and space were not absolute realities but were meaningful only in connection with objects either coexisting or in succession. We do not understand the manner of the existence of objects in these two distinct ways, they said. We simply find ourselves aware of certain relations of situation or an order of objects. This solution was especially satisfactory for Hume because it enabled him also to reduce the notion of infinity to a similar status. For had he instead resolved space and time into sensation, it would have been at the cost of giving reality to the even more dangerous idea of an infinite being.
That he was mainly concerned to undermine the dependence of abstract ideas on creative reason (without, however, denying their basis in reality), he made perfectly clear. In the first book of the Treatise, he had taken over Berkeley’s criticism of abstract ideas and shown that they were “in themselves individual, however they may become general in their representation. The image in the mind is only that of a particular object, tho’ the application of it in our reasoning be the same as if it were universal.”2 But he took Berkeley’s analysis one step further and explained how a particular idea came to stand for other resembling particular ideas. This was, he said, the result of a custom of associating a whole number of objects with the name of one of them: “The word raises up an individual idea, along with a certain custom; and that custom produces any other individual for which we may have occasion.”1 His hypothesis, Hume declared, was utterly “contrary to that, which has hitherto prevail’d in philosophy.” It was founded on the impossibility of general ideas, as usually understood.
We must certainly seek some new system on this head, and there plainly is none beside what I have propos’d. If ideas be particular in their nature, and at the same time finite in their number, ’tis only by custom they can become general in their representation, and contain an infinite number of other ideas under them.2
His analysis of space and time was entirely parallel to his analysis of abstract ideas, and he carefully underscored the connection. Space and time, as separate realities, were conceived of, he explained, by the imagination. We perceive only patches or points of colour and touch. After experiencing many such coexistences, we can separate the space of these different perceptions from them and think of space in the abstract. The idea of time, because it was derived from a number of different kinds of impressions, afforded, he said, “an instance of an abstract idea which comprehends a still greater variety than that of space, and yet is represented in the fancy by some particular idea of a determinate quantity and quality.”3 The conception of space and time is exactly like the conception of relations, substances and universals. In all these cases, we are misled into turning the fixed names for relations into absolute entities. Such indefinite application of ideas which are originally relative to certain limited perceptions is characteristic of the human mind. It is the work of imagination, and is the result of custom.
As Hume came to recognize in the course of the Treatise, the custom from which, he had said, abstract ideas arise must itself be dependent on a capacity to form general concepts and to make “distinctions of reason” between, for instance, figure and colour, motion and the body moved. It was a difficulty that haunted his whole enterprise—in order to explain how a general term comes to be applied, or how it comes to be, he had to allow that the mind could recognize the identity of an object through time and could apprehend a resemblance between particular images. Indeed, if a particular idea becomes general by being annexed to a general term, the recognition of a resemblance between particulars must occur before that use of the general term upon which the custom rests. How these operations were possible to a mind whose every idea is derived from an impression, Hume never explained. He merely took the relation of resemblance for granted as an ultimate fact of experience. Philosophers have since suggested that he might have escaped these difficulties by acknowledging and developing a supplementing and synthetic activity of the imagination at which he hints. But he could not have gone much further than he did without running a danger he was most anxious to avoid, of letting in by another door the kind of power he was trying to exclude.
Custom played an even more remarkable role in Hume’s theory of causation. And far from denying the radicalism of his theory, he made certain his readers noticed that the notion of causality was at the heart of all discussions of human intelligence and that his views were most unorthodox:
I doubt not but these consequences will at first sight be receiv’d without difficulty, as being evident deductions from principles, which we have already establish’d, and which we have often employ’d in our reasonings. This evidence … may seduce us unwarily into the conclusion, and make us imagine it contains nothing extraordinary, nor worthy of our curiosity … for which reason I think it proper to give warning, that I have just now examin’d one of the most sublime questions in philosophy, viz., that concerning the power and efficacy of causes, or that quality which makes them be followed by their effects. ”1
Hume had no wish to deny that there was a real connection between cause and effect or that all our reasoning depended on our being convinced that the connection was a necessary one. Quite the contrary, the very necessity of the causal relation was crucial for his purpose. That cause and effect were connected by contiguity and succession, Hume granted at once. But that did not explain, he pointed out, what was really distinctive about the causal relation, that it seemed to be a necessary connection: “Shall we then rest contented with these two relations of contiguity and succession as affording a compleat idea of causation? By no means. An object may be contiguous and prior to another without being considered its cause. There is also a necessary connexion to be taken into consideration.”2 We can see that the dog moves and that the stone does not; we see the facts and can conceive of their contrary; yet we say that the dog must move and the stone can not. In the same way, we do not say simply that every change has a cause, but that it must have a cause. What gives us assurance of that necessity?
If