- university home
- a-z index
- help
- terms and conditions
- privacy and cookie policy
- © 2002-2012 University of Bristol
The philosophy of physics is a relatively new field, although it is also, in a way, the most ancient of all philosophical disciplines. Human reflection about the physical world has historically preceded reflection about the nature of our own thoughts and our social interactions with other human beings. This is not surprising, for it is of great evolutionary advantage for humans to know their way around their physical environment, and to exert control over nature. Thus questions about the reliability of our presumed knowledge of the physical world emerge early in the history of human thought, and become prominent during the scientific revolution in the sixteenth and seventeenth centuries. (See also History of Science).
The philosophy of physics involves a combination of conceptual, metaphysical, methodological and epistemological issues. Firstly, it has often been thought that because physics deals with the ultimately building blocks of the Universe, it should be possible to ‘reduce’ every other field to it. According to this thesis (which is one form of the doctrine known as ‘physicalism’) human behaviour can be understood as a consequence of evolutionary adaptation, which in its turn is a result of natural selection acting on human variation, which is a biological process ultimately reducible to physics. And if physics is the most fundamental discipline, it may be that some metaphysical questions (e.g. about time, existence, the origins of the Universe) should properly be addressed to physics. Whatever physics says about time, or about causation, or about identity, is thus to be taken very seriously indeed. For if our traditional metaphysical notions clashed with well confirmed physics, then either the traditional notions are bankrupt, or we must question the validity of our knowledge of the physical world. Thus philosophers of physics have been keen to investigate any possible such clash between metaphysics and physics, a project nowadays known as ‘experimental metaphysics’. (See also Philosophy of Science, Metaphysics).
Secondly, methodological views –i.e. views as to what the optimal procedure is in order to gain knowledge— are routinely tested against physics. This is because physical science has often been considered to be the model for science as a whole, by philosophers and scientists alike. And indeed, often physics encapsulates the best methods we have. But it is also arguable that physics has its own methods, different from those of other sciences, and particularly suited to the discipline itself. Even within physics, methods may well vary or be at odds with each other. Recent literature on the distinctive features of experimental physics bears witness to this methodological variability. (See also Philosophy of Science).
Thirdly, philosophers of physics collaborate closely with physicists, trying hard to understand the novel concepts that physicists employ. Physics occupies a privileged position among the sciences in that it deals with the most entrenched amongst our ordinary concepts, and throughout history it has invariably proceeded to refine and increase our conceptual repertoire. Our views about our own place in the Universe were dramatically changed with the Copernican revolution; and our concepts of motion, impulse, and even God were correspondingly altered too. Our understanding of ‘force’ changed with the advent of Maxwell’s electromagnetic theory, a change deeply felt in the social as well as the natural sciences. Our understanding of time has been profoundly altered by Einstein’s theory of relativity (you would probably never have heard of ‘time-travel’ were it not for Einstein’s theory). Similarly quantum theory has greatly affected our concepts of chance, and causation, and philosophers are still trying to understand its conceptual implications.
This field has seen an extremely rapid growth in the last twenty years or so. Methodological, metaphysical and conceptual topics have been broached in a variety of ways by different authors. The most important issues concern the ‘measurement problem’ of quantum mechanics, and the so-called Einstein-Podolski-Rosen (EPR) correlations.
The measurement problem is best understood as an inconsistency between the usual quantum dynamical law (the Schrödinger equation), and the fact that quantum experiments have outcomes. Quantum theory is supposedly the fundamental theory of microscopic matter and radiation, and as such it is supposed to provide foundations for the whole of classical physics. But if it is impossible to measure anything in quantum mechanics, then this must also be the case in all of physics, and therefore, given the hypothesis of physicalism, it would be impossible to achieve any knowledge whatsoever of the world around us.
There are two possible replies. The first is simply to deny physicalism, which still leaves us with a conceptual conundrum: what exactly does ‘measurement’ mean in quantum theory? The second reply confronts the problem head-on, by providing an interpretation of quantum theory. At this point philosophers of quantum mechanics divide into two groups: those who favour ‘collapse’ interpretations, and those who prefer ‘no-collapse’ interpretations. The former supplement the Schrödinger equation with another law specifically designed to solve the problem. But this seems ad-hoc: how does nature decide which interactions are measurements and which are not? On the other hand, ‘no-collapse’ interpretations postulate an ontology radically different to the classical, ordinary one. For instance, quantum properties are said to be ‘contextual’, in the sense that they depend on the presence or absence of further properties, etc. Or, an infinity of minds for every experimenter is postulated, each mind ‘recording’ one among all the possible outcomes of the experiment.
In 1935 Einstein and two of his collaborators, Podolsky and Rosen, showed that distant quantum particles, unmediated by any sort of physical contact, can exhibit statistical correlation. (This is what the import of the paper is nowadays taken to be - whether EPR actually intended to show this is also a matter of debate among philosophers.) The phenomenon, detected in laboratories across the world, has been given a variety of names: ‘quantum entanglement’, ‘non-locality’, passion-at-a-distance’, etc. In 1966, John Bell proved a theorem to the effect that any empirically adequate theory, not just quantum theory, would have to exhibit this kind of correlation at-a-distance. Ever since philosophers have been focusing on the implications for the concepts of locality, and of causation.
Some philosophers have claimed that Bell’s theorem shows how a metaphysical thesis (in this case, the thesis that all physical influences are mediated by spatial contact, that there is no physical action-at-a-distance) can be refuted by experimental evidence. Others argue that as a matter of fact the evidence doesn’t show that much; and that there are serious loopholes in these ‘metaphysical refutations’. The debate is a very lively one, involving metaphysical issues (How could a particle be in two places at once? What novel notion of space and time is required by quantum mechanics?); conceptual issues (What exactly do we mean by ‘locality’, or by ‘influence’?); and methodology (How much experimental evidence is needed to establish a theoretical claim?)
Bell’s theorem also has deep implications for philosophical accounts of causality. For instance, Bas van Fraassen has argued that Bell’s theorem constitutes empirical proof that causal realism (the thesis that a causal mechanism must underlie any established statistical dependence) is false. Others have objected: van Fraassen, they argue, uses a very specific account of causality; on other accounts, Bell’s result presents no problem. The debate links metaphysics (can a cause act probabilistically?), conceptual analysis (is there only one concept of ‘cause’?), and methodology (can a cause be inferred from a statistical correlation?)
Philosophers of physics prefer to discuss the traditional issues of space and time in the more concrete contexts of physical space-time theories. These theories include: Aristotelianism, Galilean relativity, Newtonian space-time, Einstein’s Special Theory of Relativity, and his General Theory of Relativity.
The debate between so-called substantivalists and relationalists goes back, in its present form, to a notorious exchange of letters between Gotfried Leibniz and Samuel Clarke (Isaac Newton’s spokesman). Clarke and Newton argued that space is a substance separate and independent from the objects that lie within it. This substance determines an absolute frame of reference (namely, the one that is at rest with respect to this very substance). In addition, Clarke and Newton argued that time is continuous and infinite, and would exist in the absence of any objects or events (your clock would continue to tick even if nothing at all happened in the Universe). By contrast, Leibniz argued that spatial properties are strictly relational: space does not exist independently of the objects that make up the Universe. And, similarly, time would cease to flow if nothing at all happened anywhere. The debate was not resolved.
Some philosophers (notoriously Hans Reichenbach) saw in Einstein’s relativity the final endorsement of the Leibnizian view. But more recent scholarship, in particular John Earman and John Norton’s widely discussed ‘hole argument’, has shown that this ancient debate is still very much alive. Additional arguments of a methodological type in favour of substantivalism have been provided by Michael Friedman.
Another important issue in contemporary space-time philosophy concerns the nature of simultaneity. It was well known since Einstein’s work that simultaneity is not absolute in the special theory of relativity. Events that are simultaneous in one frame of reference may not be so in another. Adolf Grünbaum further argued, following Reichenbach, that the definition of simultaneity in special relativity is conventional. This is because the one-way speed of light is not fixed by the theory, only the two-way (or ‘return’) speed is. Winnie derived space-time transformations more general still than the Lorentz transformations of relativity theory. But is simultaneity conventional also in the General Theory of Relativity?
Philosophers of statistical physics work together with statisticians and physicists, making progress on the fundamental conceptual issues of the discipline. There are two areas of intense research: interpretations of probability, and the problem of irreversibility and the direction of time.
A probability is a mathematical function that maps events to real numbers between 0 and 1. But what exactly does this mean? There are two competing interpretations of what probability is. The subjective interpretation states that a probability ascription is a measure of our ignorance of the real value of the quantity. This captures many of our everyday uses of the word ‘probable’. For instance, if you ask me whether the Euro devalued again this morning, I would say probably yes. Although I can’t tell you exactly, I can nonetheless estimate on the basis of what I know has been happening this week that its probability is about .75. There is a matter of fact here: the Euro either devalued yesterday or it didn’t. The probability hence only expresses my own subjective, partial knowledge of this fact. On the other hand, not all probabilities seem to be subjective in this way. In physics, many processes are inherently random: a radioactive atom either decays or not with a certain probability –and that probability is independent of how much we could come to know about the state of the atom. Statistical physics is the theory of probabilistic processes, so this issue of interpretation is central to its philosophical understanding.
Ludwig Boltzmann once proposed that the arrow of time (from the past towards the future) is determined by the local increase of entropy. He thought that psychological time could be reduced to the experience of entropy-increasing physical systems. Ever since, Boltzmann’s thesis has been the subject of intense debate among philosophers. Those philosophers that oppose the thesis claim that as long as thermodynamics remains a theory of the behaviour of macroscopic systems, without any deeper connection to microscopic physical laws, the arrow of time that it defines is without any metaphysical foundation. Defenders of Boltzmann’s thesis claim that thermodynamic entropy has a deeper explanation in the laws of statistical physics, and hence the direction of entropy is of fundamental significance in defining the direction of physical time.
Philosophers of physics have traditionally been concerned with scientific theories. This is in part due to the central role that epistemology has played in twentieth-century philosophy of science. The object of a belief is a theory, so it is natural that a concern for scientific belief has led to an obsession with theory. But just as epistemology can move away from the notion of belief, so can the philosophy of physics move away from the discussion of scientific theories. In recent years, a number of philosophers and historians of physics have begun to pay careful attention to experimental physics, and have argued that experiment has its own methods, and its own practice.