In one of his Grammar of Assent‘s more famous lines, Newman remarks: “As to Logic, its chain of conclusions hangs loose at both ends; both the point from which the proof should start, and the points at which it should arrive, are beyond its reach; it comes short both of first principles and of concrete issues” (284). This is perhaps the clearest statement in his entire oeuvre of Newman’s anti-rationalism; he is in no way opposed to reason as such, but simply a realist about its limitations, and confident that those limitations in no way affect the verisimilitude of our knowledge. He could have quoted Pascal on this score, who argues that it is an “excess,” either “to exclude reason, [or] to admit nothing but reason” (Pensees, Sec. 183).
I suspect that Newman drew the above image from Hume’s Inquiry Concerning Human Understanding, but his alterations of this earlier material are as significant as his debts to it. Consider:
If we proceed not upon some fact, present to the memory or senses, our reasonings would be merely hypothetical; and however the particular links might be connected with each other, the whole chain of inferences would have nothing to support it, nor could we ever, by its means, arrive at the knowledge of any real existence. If I ask why you believe any particular matter of fact, which you relate, you must tell me some reason; and this reason will be some other fact, connected with it. But as you cannot proceed after this manner, in infinitum, you must at last terminate in some fact, which is present to your memory or senses; or must allow that your belief is entirely without foundation (5.1).
Hume is reflecting on the problem of induction (i.e., our inferences from the past to the future, as in our judgment that fire will burn me (since it did in the past), or that the sun will rise tomorrow). His point here seems modest: if an induction doesn’t terminate in some state of affairs about which we have empirical knowledge (i.e., that fire burned me last week), our inference (i.e., that fire will burn me today) is “without foundation, purely hypothetical.” And yet, he is quick to sever that connection as well, leaving his chain of conclusions flapping as aimlessly as Newman’s: one cannot, Hume insists, “by any reasoning…reach the idea of cause and effect, since the particular powers by which all natural operations are performed never appear to the senses.” He suggests instead that it is “custom or habit” that motivates our inferences (5.1). He concludes: “All belief of matter of fact or real existence is derived merely from some object present to the memory or senses and a customary conjunction between that and some other object…All these operations are a species of natural thought that instinct, which no reasoning or process of the thought and understanding is able either to produce or to prevent” (5.1).
In one sense, Newman and Hume profoundly agree: they both oppose in Locke an epistemology that attempts to claim far too much for reason, and which in consequence “seems theoretical and unreal” (Grammar, 164). But where Hume dwells on the frailty of reason in the service of a brooding skepticism, Newman remains unflappably convinced that we know ourselves and the world, that our ordinary means of knowing — induction ranking high among them — are well-founded, or (to use Alvin Plantinga’s phrase) “properly basic.” We don’t worry about whether the objects we perceive as green are actually “grue” (Nelson Goodman’s term for an object that is green now, and blue after 2050 or so), because it doesn’t belong to our epistemological constitution (or “design plan,” to use another Plantingism) to discern the future color-states of the objects we perceive. Does anyone think that our inability reliably to distinguish green from grue objects keeps us from reliable green-identification?
So yes, Hume is right to say that custom is part of our ability to infer the future from the past, as is (almost certainly) particular aspects of our biological nature. That doesn’t, of course, yield anything like a demonstration that any induction is valid. But then again, the epistemological rule that only claims susceptible of logical demonstration are trustworthy is itself not susceptible of logical demonstration. We’re not angels, but we aren’t beasts either, so that’s something.