Topics Physics

Question Is it true that "Things fall because of gravity?" "Gravity" is just a placeholder word for the tendency of things to fall. So to say "Things fall because of gravity", is to say "Things fall because of their tendency to fall." Which is vacuous. A better explanation would be "Things fall because they have mass and are nearby another massive object (the earth)." Am I right here?

Accepted:
January 27, 2009

Comments

This sounds like an accusation that was regularly thrown at Medieval Aristotelian physicists. Aristotelian physics was built around the "teleological" principle that things have natural tendencies to strive to achieve certain goals or destinations. Why does a stone fall? Aristotle would say that the explanation for this rests on the fact that it is in the nature of an earthy body to move towards the natural place of such bodies, which (he believed) is in the centre of the cosmos. But you're quite right, this does sound rather vacuous, to say that it moves as it does because it has a natural tendency to do so. The Aristotelians sought to explain natural phenomena in terms of what came to be known as "occult qualities" and, although the term "occult" might not have carried quite the connotations it has now, it was used perjoratively by many non-Aristotelians to point to the fact that these supposed qualities really weren't explanatory at all. The Aristotelian approach was famously lampooned by Molière, the seventeenth century French playwright, who had one of his characters ask why it was that opium should put people to sleep. Because it has a dormitive virtue, came the reply. Which is just a fancy way of saying that it puts people to sleep because it has the power to put people to sleep. Gee, thanks for that: now everything's clear!

But let's skip ahead to Isaac Newton. The one thing that everyone knows about Newton is that he explained how gravity works, right? Well, not quite. Newton's achievement -- and it certainly was a colossal achievement -- was to discover a law that would enable one to calculate the effect the proximity of one massive body would have on the motion of another massive body. But, as to how it produced this effect, on this Newton remained deliberately silent. In the General Scholium to the 1713 edition of his Principia (wherein he had presented this law), Newton wrote: "I have not as yet been able to deduce from phenomena the reason for these properties of gravity, and I do not feign hypotheses. For whatever is not deduced from the phenomena must be called a hypothesis; and hypotheses, whether metaphysical or physical, or based on occult qualities, or mechanical, have no place in experimental philosophy."

Newton's attitude to the nature and scope of experimental science ("natural philosophy", in his day) would subsequently come to dominate the field. But does this mean that scientists never truly explain anything after all, but merely subsume phenomena under laws of nature? Well, perhaps we should take a narrower view of the nature of explanation itself: many modern philosophers of science would say that to subsume a phenomenon under a law of nature is to explain it. They have come up with something they call the "D-N" ("deductive-nomological") model of scientific explanation. Although there are some notorious problem-cases that this model can't neatly handle, which are extensively discussed in the literature, for present purposes we will just put those to one side. In this model, if you want to explain some phenomenon, what you need to do is (i) specify some initial conditions and (ii) find a law of nature ("nomological" means "pertaining to laws"), such that the phenomenon is (deductively) entailed by those initial conditions in conjunction with that law. It is not quite enough just to say that a stone falls because it has mass and is near another massive object: you additionally need to point to a law (such as Newton's) to connect these various concepts up together. Why is A the case? Because B was the case, and it's a law of nature that, whenever you get B, A will surely follow.

Such an explanation is not vacuous, because it really does add some genuinely new information-content that goes beyond the phenomenon to be explained. It is set out in different terms (at least in part), and it refers to a principle that encompasses a very much wider domain than just that one specific phenomenon. But is it enough? Can we regard this as a complete explanation of the phenomenon? No, unfortunately, we cannot. As Newton recognised, there are still further questions to ask. How is the gravitational influence of the Earth propagated across the intervening distance? Why should the laws of nature be like that at all? These were the issues that Newton declined to feign hypotheses about, on the grounds that he didn't (yet) have enough experimental data to settle them. But the important thing to appreciate is that there are always going to be further questions to ask. As small children tend to discover fairly early in life, they can quite effectively infuriate their parents by repeatedly asking "Why?" Whatever answer their parents might give them, explaining one thing in terms of something else, it will always invite a further question: "But why is that the case?" And this process can go on indefinitely, for as long as the child -- or the scientist -- has the patience to keep pressing. Equally, though, it can stop at any time. Explanation is always tied to specific, subjective interests, and, sooner or later, the questioner will be satisfied. "Why A?" "Because of B." "But why B?" "Because of C." "Fine, but why C?" "Because of D." "Ah, now I see: thank you." The questioner could perfectly well carry on: but they don't, because things have now been sufficiently explained to their own satisfaction.

Is there such a thing as an "ultimate" explanation, one that would close off all further "Why?" questions once and for all? I doubt it. Although it is not absolutely guaranteed, it does seem fair to assume that the natural world, as far as both its individual facts and its universal laws are concerned, is contingent through and through. It is one way, but it could have been another way -- which invites the question of why it is this way rather than that. But does the fact that the process of explanation might never reach an end-point mean that it cannot get anywhere at all? Certainly not. If we manage to subsume a phenomenon under some general law, as in the D-N model, we have provided some genuine information. If, let's say, we then manage to unify that law with others, and show that this phenomenon is actually an instance of a still-wider class of phenomena than we had previously suspected, this would be a major advance, pushing our explanation to a more fundamental level and increasing its information-content still further. Cutting-edge physicists right now are trying to come up with a set of equations that will enable them to connect gravitation up with the other three fundamental forces they countenance (electro-magnetism, and the strong and weak nuclear forces). And success in this project would be a major breakthrough, providing a tremendous new insight into the nature of the universe, even though it would then just invite a new question: why should those be the equations that govern how things work, rather than some others?