Quite apart from all metaphysical questions, the notion of cause is a complex one which needs a fairly elaborate and subtle logical analysis. It would be inappropriate to enter in detail into this here and now; it will suffice for our present purpose to say that the statement that C causes E sometimes means that C is a necessary though perhaps not sufficient condition of E, sometimes that C is a sufficient though not perhaps necessary condition of E, and sometimes that C is a set of conditions which are severally necessary and jointly sufficient to produce E. Now popular talk about this causing that does not clearly distinguish these alternatives. It is very common, e.g., to start from the fact, which may be quite trivial and even tautologous, that C causes E in the sense that it is a necessary condition of E; then to take for granted that C causes E in the important and doubtful sense that it is necessary and sufficient to produce E; and then to infer various far-reaching practical conclusions from this.
An example is the assertion, often made with a great flourish of trumpets by pacifists, that armaments cause war. Since war involves, by definition, a conflict between the armed forces of nations, it is a tautological proposition that armaments are a necessary condition of wars. From this nothing follows except the platitude that, if all nations simultaneously disarmed and remained disarmed, there would be no more wars. This does not give the slightest guidance as to what a particular nation should do, if it is practically certain that at least one fairly strong nation will retain its armaments. It is obvious that there are situations in which a diminution of armaments by a certain nation or group of nations increases the chances of war, whilst an increase in their armaments diminishes it.
(C. D. Broad, "Some Common Fallacies in Political Thinking," Philosophy 25 [April 1950]: 99-113, at 104-5 [italics in original])
