The World Philosophy Made

by Scott Soames

  With this in mind, let us now ask how a psychological theory of the sort Fodor envisions might yield the kinds of explanations of behavior illustrated by (10).

  10.   X performed action A because (i) X desired it to be the case that S, (ii) X believed that performing A would bring it about that S, and (iii) X believed X could perform A.

  It must, he suggests, do so by formulating and confirming general psychological laws of (roughly, and subject to various qualifications) the following form.12

  11.   If any individual x believes that A is an action that x can perform, and if x believes that performing A will bring it about that S, and if x wants it to be the case that S, then x will act in a fashion intended to be a performance of A.

  Although there may be something important and correct about this, it is worth noting that claims of the form (11) cannot, in general, be expected to be true, exceptionless generalizations.

  The reason for this is easy to see. One can believe that doing A will bring it about that S by affirming (or being disposed to affirm) the mental representation (12) in which that S is the propositional content of the mental representation M1.

  12.   IF I DO A, THEN M1

  One can, simultaneously, desire that S in virtue of bearing an appropriate motivational relation to a mental representation M2 that has the same propositional content as M1, despite differing symbolically and computationally from M1. Since beliefs and desires relate agents to people and things in the world around them, the propositional contents of their beliefs and desires depend on environmental and other factors external to the agents themselves.13 Because of this, it can, and sometimes does, happen that mental representations with the same propositional content aren’t recognized as having the same content, and so are treated differently.14 Thus, our agent may feel no motivation at all to do A, despite taking A to be doable and believing that doing it would bring it about that S, while at the same time intensely desiring it to be the case that S. Cases like this are violations of (11), which show that cognitive psychology cannot reasonably aspire to formulating true, exceptionless, universal laws of this general form.

  This should not be cause for alarm. Generalizations like (11) require all-other-things-being-equal clauses in any case. Our example suggests that some of these clauses should specify that the agent apprehends the content that S in the same way (i.e., via the same mental representations) in the cognitive states that give rise to the relevant beliefs and desires. In addition, whatever questions may arise about the details of the more specific generalizations that are needed, they don’t undermine the correctness of many explanations of particular actions caused by beliefs and desires. John may have done A because he wanted it to be the case that S, and he believed that doing A would bring that about, even if not everyone with those beliefs and desires would do A. This is no more mysterious than the fact that John may fall and break his leg because he steps on a banana peel, even though not everyone who steps on a banana peel suffers a similar fate. Thus, the burgeoning enterprise of cognitive psychology today, which owes so much to philosophers and psychologists like Jerry Fodor and Philip Johnson-Laird, will have little trouble surviving the partial breakdown of what some may once have seen as the preestablished harmony between internal representations—needed (Fodor believes) to track and explain the role played by computational processes in our thought and action—and the propositional contents of our beliefs and other cognitive attitudes, which reflect our cognitive relationships to things in the world.

  COGNITIVE PSYCHOLOGY, NEUROSCIENCE, AND THE MIND-BODY PROBLEM

  As I have emphasized, the contents of the laws and hypotheses of cognitive psychology are not the contents of the laws and hypotheses of neuroscience. The vocabularies used, as well as the concepts and propositions expressed, in the two sciences are different. The same can be said about the relationship between neuroscience and chemistry, and between chemistry and physics. One can think of these sciences as different, but hierarchically related. The objects talked about by each science are real, and the properties ascribed to them are, when all goes well, ones the objects really have. But it is plausible that objects and properties at higher levels of abstraction (somehow) are grounded in those at lower levels, and that all may ultimately be grounded in the kinds of objects and properties recognized by physics.

  Sometimes we can discover what is grounded in what—for example, the biological concept of a gene is grounded in the chemical structure DNA. But that is unusual. The individual sciences are, though related, largely autonomous, with their own domains of productive theorizing. Cognitive psychology primarily studies the human mind. But the fact that the mind is an independent domain of study doesn’t mean that the things it talks about—beliefs, desires, perceptions (to say nothing of pains)—are unrelated to, or fundamentally different from, those we freely call ‘physical’. This is just another way of saying that we still have no final solution to the classical mind-body problem. The good news is that we don’t need one in order to continue the progress currently being made by philosophers, psychologists, neuroscientists, and others.

  CHAPTER 10

  PHILOSOPHY AND PHYSICS

  The continuity of physical categories over centuries; the collaboration of philosophically minded physicists with philosophers trained in physics; Einstein on the importance of philosophy; the key question: What do the unobservable parts of physical theories tell us about the universe?; the eighteenth-century debate over absolute space and the nineteenth-century statement of Newtonian theory without it; the abandonment of absolute time in special relativity; light, gravity, and general relativity; puzzles and interpretations of quantum mechanics.

  Physics is our most fundamental science. Its task is to explain what the universe is like, including when, how, and why macro- and microscopic events happen as they do. Much of this task requires elucidating the central notions needed for such explanations—time, space, matter, and motion. These were central to the physics of Aristotle, which remained dominant until the sixteenth century. They remained central, with important modifications, in Newton’s Principia Mathematica at the end of the seventeenth century. By the early twentieth century, they were still central, despite being radically rethought.

  The philosophy of physics is about physics; it attempts to explain what physics tells us. That may sound strange. Why do we need a separate study to explain what another study says? The question, though a good one, is somewhat misguided. The philosophy of physics is not an area of study distinct from physics. It is a philosophically self-conscious way of doing physics itself. In one way or another, this has always been so. Recall the first sentence of Aristotle’s metaphysics, “All men by nature desire to know.” Among the things Homo sapiens have always most wanted to know is what the vast universe, of which we are such seemingly insignificant inhabitants, is really like, including what it was like before we were here, and will be like after we are gone.

  Nearly 2400 years ago Aristotle systematized his thoughts on the subject in his Physics. He was followed centuries later by such men as Roger Bacon in the thirteenth century, William of Ockham, John Buridan, and Nicholas Oresme in the fourteenth century, Nicolaus Copernicus, Johannes Kepler, Galileo Galilei, and René Descartes in the sixteenth and seventeenth, and, of course, Isaac Newton in the late seventeenth and early eighteenth century. Some of these men were monks or theologians, some were mathematicians, some were astute observers or experimentalists, and some were all three. But, whatever else they were, all were, in part, philosophers, as the title of Newton’s great work, Philosophiae Naturalis Principia Mathematica, reminds us. The same is true of Albert Einstein (1879–1955) and several other great physicists of our era. Hence the original name of the subject, natural philosophy, still fits.

  There are of course very significant differences between professionals whose primary appointments today are in physics labs or departments and those whose primary appointments are in philosophy departments. But this does
n’t erase the overlap between philosophically minded physicists and scientifically informed philosophers of physics. The philosophers use tools of conceptual clarification and rigorous evaluation of arguments to reveal potential flaws and presuppositions of important scientific reasoning that, if left unaddressed, may cloud our understanding of physical theories and inhibit their further development. That is why today’s philosophers of physics are typically trained philosophers who are also physicists, and why some of the greatest physicists, like Einstein, were self-consciously philosophically minded.

  Einstein himself always recognized this. Writing in his autobiography about the importance of his philosophical studies, he says:

  Today everyone knows, of course, that all attempts to clarify this paradox [involving the nature of light that lead to special relativity] satisfactorily were condemned to failure as long as the axiom of the absolute character of time, or of simultaneity, was rooted unrecognized in the unconscious. To recognize clearly this axiom and its arbitrary character already implies the essentials of the solution of the problem. The type of critical reasoning required for the discovery of this central point was decisively furthered, in my case, especially by the reading of David Hume’s and Ernst Mach’s philosophical writings.1

  Much earlier, in 1915, Einstein had written something similar in a letter to the scientifically educated philosopher and founder of logical positivism, Moritz Schlick.

  Berlin, 14 December 1915

  I received your paper yesterday, and have studied it thoroughly. It’s among the best yet of what’s been written about relativity. Nothing nearly as clear has previously been written about its philosophical aspects. At the same time you have full command of the theory itself.… Truly masterful is your discussion of relativity theory’s relationship to the philosophy of Kant and his disciples. Their trust in the “incontrovertible certainty” of “a priori synthetic judgments” is badly shaken by the recognition that even a single one of those judgments is invalid. Your exposition is also quite right that positivism suggested relativity theory, without requiring it. Also you have correctly seen that this line of thought was of great influence on my efforts and indeed Mach and still much more Hume, whose Treatise on Human Nature I studied with eagerness and admiration shortly before finding relativity theory.2

  What Einstein learned from Hume and Mach was not anything specifically about space, time, or motion (though both had many ideas about them, and Hume held that there is no notion of time separate from the motion of bodies). Rather, what Einstein found illuminating was a gap these thinkers made vivid to him—the gap between our habitual ways of thinking, directly but often uncritically derived from sense experience, and the proper concepts needed to truly describe reality. Einstein’s debt to these philosophers was not to the contents of their doctrines, but to the inspiration provided by their willingness to rethink and revise even our most basic—seemingly rock-solid—commonsense notions, if doing so would increase our knowledge.3

  In Einstein’s case, the notions to be revised involved our pre-theoretic conceptions of space, time, and simultaneity. What the philosopher in him realized was that no matter how great their everyday utility to us, and no matter how deeply embedded they are in our biologically determined perceptual and cognitive architecture, there is no guarantee that these ordinary notions are well suited to understanding the fundamental structure of the universe. What the great scientist in him realized was that the universe was telling us that these notions had to be revised. It is a tribute to his genius that he saw how to do it.

  The relationship between physics and philosophy of physics today is, if anything, closer than it was in Einstein’s day. One can get a sense of why this should be so by thinking of physical theories as collections of abstract, mathematically sophisticated representations of reality which, when combined with attested observations, allow us to predict further observable events. When these events occur, the theory is partially confirmed (though not conclusively proved); when they don’t occur, it is often necessary to modify the theory that led to false predictions. This way of thinking of theories—as prediction-generating representations of reality—raises three natural questions. Which aspects of the theoretical representation of reality are merely conventional devices adopted to smooth the calculations needed to make observational predictions? Which aspects of the theory are genuinely representational, and so make claims (beyond the directly observable) about the nature of reality? What (beyond the directly observable) do our best physical theories tell us about reality?

  One view, formerly far more popular than it is today, dismissed the second and third of these questions by taking theoretical claims about non-observational matters to be mere calculating devices, with no representational content beyond the observational predictions to which they contribute. In the decades since this view has fallen from favor, philosophers of physics and philosophically minded physicists have struggled to answer questions about what our physical theories are telling us about the universe. Physicists themselves differ in the degree to which they are engaged in this enterprise. Some are understandably more concerned with using physical theories to calculate precise solutions to clear empirically stateable problems, while others place an equal priority on conceptualizing what, exactly, the non-directly verifiable aspects of their theories tell us about the world. These are the physicists who interact most deeply with today’s philosophers of physics—not, of course, by looking to philosophers for new physical theories, but by working with philosophers of physics to clarify what their own physical theories are telling us (which in turn may spark further improvements).

  To illustrate this, I will return for a moment to Newton. In chapter 3 we saw that he accepted absolute space and time in part because it was deeply intuitive and in part because stating physical laws in terms of them allowed him to explain the otherwise puzzling observed behavior of water in a spinning bucket. Thus absolute space and time were not empirically gratuitous constructs. They did, however, give him more structure than he needed, and so generated further puzzles. In Newton’s 3-D Euclidean space, a distribution of matter in one portion of absolute space could (in principle) be relocated in a straight line to another location, preserving all relative spatial positions and sizes, without having any effect on the laws of physics. So the question Where are we in absolute space? seems to be inherently impossible to answer; similarly for questions about absolute velocities.

  Newton realized this. He recognized that as long as we don’t introduce new circular motions (like the spinning bucket), or eliminate such motions in the initial state, we can imagine the entire collection of matter moving in a straight line in one direction at a constant speed without changing any of the physical relationships between any of the bodies. This is puzzling because it suggests that no empirical evidence could be brought to bear on the question of which of these states our universe is in.

  This was one of the scenarios that generated a spirited debate between the German philosopher and mathematician Gottfried Leibniz (1646–1716) and the British philosopher Samuel Clarke (1675–1729), a younger contemporary of Newton, and one of his defenders. Leibniz dismissed absolute space as an empirically empty fantasy, favoring his own relative conception. Although his system was metaphysical rather than scientific, and so not really a rival to Newton’s, and although assuming absolute space helped Newton provide explanations of some observed events, Leibnitz had a point. If certain questions about position in absolute space, and about absolute velocities, can’t be answered, or even supported by empirical data, then we might reasonably wonder whether it might be possible for us to replace absolute space in our theories, thereby avoiding unanswerable questions, without loss of explanatory power. If so, perhaps we should.

  It is now known that Newton’s laws can be translated into a theory that retains absolute time while giving up absolute space.4 Absolute time is maintained by preserving Newton’s linear, numerically measurable structure of moments of time,
one following another at a constant rate. Although space remains 3-D Euclidean, there are no points of absolute space persisting through time. Rather, the universe through time is conceived of as a series of simultaneity slabs (one following another in time) each of which consists of a set of space-time moments or events (allowing for events at places in which nothing happens) occurring at a given moment. The space-time points on such a slab stand in measurable 3-D Euclidean relations to one another, even though those on one slab bear no spatial relation to those on other slabs. In other words, there are no common (absolute) spatial locations through time.