When we begin, however, to look at the second of these foundations more closely, it soon appears, that it is not in reality distinct from the first. For what is testimony? It is itself an event. When we believe any thing, therefore, in consequence of testimony, we only believe one event in consequence of another. But this is the general account of our belief in events. It is the union of the ideas, of an antecedent, and a consequent, by a strong association. I believe it is one o'clock. Why? I have just heard the clock strike. _Striking of the clock_, antecedent; _one o'clock_, consequent; the _second_ closely associated with the _first_. The striking of the clock is in fact a species of testimony. What does it testify? Not one event, but an infinite number of events, of which the term "one o'clock" is the name. At every instant in the course of the day, a number of events are taking place, some known to us, some unknown. The term one o'clock, is the name of those which take place at a particular point of the diurnal revolution. I believe in them all upon the testimony of the clock. Why? From experience;--every one would directly and {383} truly reply. I have found the events constantly, or at least very regularly, conjoined. From junction of the events, junction of the ideas; in other words, belief. If proof, only, were wanted, this would suffice. For the purpose, however, of instruction, tuition, training, a more minute developement of this important case of belief seems too useful to be dispensed with, notwithstanding the tediousness which so many repetitions of the same process are too likely to produce. The watchman calling the hour, is a case of human testimony. That the account of our belief, in this case, is precisely the same as that in the case of the striking of the clock, it is wholly unnecessary to prove. But if our reliance on testimony in one case is pure experience, it may reasonably be inferred that it is so in all. The forms of expression, which we apply to this case of belief, are very misleading. We say, "we believe a man," or, "we believe his testimony." "We attach belief to the man," or, "to his testimony." In these expressions, the name belief is applied to the wrong event; to the antecedent, instead of the consequent. What we mean to say is, that we believe the consequent, the thing testified, not the antecedent, the speaking of the words. The words the man uses, are, to us, sensations: belief that he uses the words, is not what is meant by belief in his testimony. The same form of expression is perfectly absurd, when applied to other cases. We never say that we believe the flame of the candle, or we attach belief to the flame of the candle, when we mean to state the belief, that a finger will be burnt if it is put into the flame; {384} we never say we believe the spark, when we mean to express our belief of an explosion when the spark falls upon the gunpowder.