4.2 Naturalism and reductive explanation
There is a widespread commitment among contemporary philosophers and scientists to a naturalistic view of the world. In broad terms, naturalism is the view that everything is scientifically explicable – to put it crudely, that there are no miracles. (Note that I am using ‘naturalism’ here for a metaphysical position – a view about the nature of the world. It is also used for a methodological position – a view about how the world, or some aspect of it, should be studied. Indeed, the word has a variety of meanings and should be used with care.) Thus, naturalists deny the existence of supernatural entities and powers and assume that everything that happens is causally explicable by reference to scientific principles and laws. Some naturalists also make a further assumption. They assume that natural phenomena form a hierarchy and that higher-level ones can be explained by reference to more basic ones, right down to the level of chemistry and physics. Reproduction, for example, counts as a high-level phenomenon, which can be explained in terms of more basic genetic and cellular processes, which can themselves be explained in chemical and physical terms. This sort of explanation, where a phenomenon at one level is explained in terms of those at a lower level, is called reductive explanation. The notion of reductive explanation is a key one in the modern debate about consciousness and it is important to understand how reductive explanations work. An example may be useful.
Suppose that a person who has had no previous contact with modern civilisation is shown a working television set, displaying images of events taking place many miles away. They are astonished by the device and declare that it must be magic. How would we convince them otherwise? The answer, of course, is by explaining how a television set works. We might begin by describing a television camera – explaining that it uses a lens to focus a moving image onto a light-sensitive plate, which then generates a stream of electronic pulses, corresponding to the pattern on the plate. We would then explain how this electronic signal is amplified and broadcast – explaining what radio waves are and how they can be used to carry an electronic signal. Finally, we would turn to the television set itself and explain that it detects radio waves via an antenna, decodes the signal and uses it to modulate the beam of a cathode ray tube, causing the tube to emit patterns of light which correspond to the images in the camera and which are perceived by the human eye as a moving picture. Of course, in order to make all this comprehensible we would have to provide a lot of further information about the underlying physical processes – about light, optics, electricity, radio waves and so on – but with time and access to reference books we could surely satisfy our hearer that there was nothing magical about the television.
In doing all this we would have reductively explained the television's power to display moving images of distant events. That is, we would have shown that this property follows from more basic, lower-level properties of the television – its possession of various mechanical and electronic components. These properties explain the television's power to display moving images of distant events because it is obvious that they are sufficient for it. Nothing more is needed in order for the television to have that power than for it to possess those properties. In a widely-used phrase, the lower-level properties realise the higher-level one: the television possesses the latter in virtue of the fact that it possesses the former.
It is important to distinguish reductive explanation from reduction. To say that a property can be reduced to a lower-level one is to say that it can be identified with it across the board – that they are in fact the same property, under different names. (Or at least that is one common meaning of ‘reduction’.) For example, the property of being water reduces to that of having the molecular structure H2O. Reductions like this are quite rare, however, since most properties can be realised in more than one way (‘multiply realised’). Different kinds of television, for example, work in different ways and are made of different materials (some have plasma screens instead of cathode ray tubes, some receive the signal by cable instead of aerial, older models use vacuum tubes or transistors instead of integrated circuits and so on). So we cannot identify the property of being a television with that of having a particular set of components. Any components will do, provided they do the job. However, the fact that a phenomenon cannot be reduced to a lower-level one does not mean that it cannot be reductively explained in lower-level terms. Each instance of the phenomenon may be realised in lower-level properties and explicable in terms of them, even if these properties are not the same in every case.
Now as I said, many philosophers and scientists assume that all phenomena above the level of basic physics can, in principle, be reductively explained. They view the natural world as a unified structure, whose higher levels of organisation emerge in a thoroughly comprehensible way from lower-level ones and ultimately from basic physical states and processes. I shall refer to this view as strong naturalism.
Strong naturalism has considerable plausibility. It is a remarkable fact that just about every phenomenon scientists have studied has turned out to yield to reductive explanation. Take life, for example. Until the middle of the nineteenth century it was common for biologists to maintain that life was not the product of more basic inorganic processes, but dependent on a special vital spirit or force – a view known as vitalism. It is easy to see why they thought this. Inanimate structures tend to decay steadily, whereas living things are able to sustain, repair and reproduce themselves. Given the undeveloped state of biological knowledge, it was not implausible to think that this amazing regenerative ability could not be the product of mere physical processes. In fact, of course, this was quite wrong. As biologists studied organic processes in more detail, they discovered that they were nothing more than complex chemical reactions, which could be replicated in the laboratory. With time, more and more biological phenomena yielded to reductive explanation, and today vitalism is wholly discredited. What is special about living things, it turned out, is not that they possess a non-physical ingredient but that they involve a unique and very complex organisation of physical elements.
What proved true in biology has proved true in the other sciences, too. Almost everywhere scientists have been able to explain higher-level phenomena in terms of lower-level ones.
Can you think of any properties that seem unlikely to yield to reductive explanation? (Set aside mental ones for the moment.)
The most obvious candidates, I think, are moral and aesthetic properties. Can the rightness and wrongness of our actions be reductively explained by reference to their physical characteristics – when and where and in what manner they were performed? It seems unlikely: the very same action, physically characterised, might be disloyal, say, in one context but not in another. Similarly, can we explain why objects have the aesthetic properties they do – why they are graceful or elegant or ugly, for example – by reference to their physical properties – their colours and shapes and so on? Again, many would say no: we cannot read off an object's aesthetic properties from its physical ones.
It may be, then, that moral and aesthetic properties cannot be reductively explained. But even if this is so, there is no fatal objection here to strong naturalism. For defenders of the doctrine may simply deny that moral and aesthetic properties are real properties of actions and objects and claim instead that they are just projections of our own responses to them. Indeed, for some people, the very resistance of these properties to reductive explanation is a reason for denying their reality. If a phenomenon cannot be reductively explained – if we cannot see how it could arise from lower-level processes – then, these people would say, that is a good reason for thinking that it is not real or has at least been seriously mischaracterised.
(It is worth stressing at this point that a strong naturalist need not claim that reductive explanation is the only legitimate kind of explanation. A reductive explanation shows how a phenomenon is constituted, but there are other types of explanation with different functions. For example, much scientific explanation involves explaining processes at a high level of description without going into the details of how they are constituted. This is the case with explanations in the so-called special sciences – the sciences devoted to specific phenomena above the level of basic physics – biology, chemistry, psychology and so on. Strong naturalists need not deny the legitimacy or usefulness of these other types of explanation, though they will claim that there are reductive explanations of why they hold.)
Here is an activity to reinforce the points just made.
Which of the following claims would strong naturalists endorse?
Everything that exists is natural.
Everything that happens can be scientifically explained.
Science can only deal with natural processes; supernatural ones are beyond it.
All phenomena above the level of basic physics can be explained in lower-level terms.
All phenomena can be reduced to physical ones.
Reductive explanation is the only legitimate kind of explanation.
Strong naturalists would endorse (2) and (4). (1) is ambiguous. If ‘natural’ means ‘not supernatural’, then strong naturalists would endorse it. If it means ‘not man-made’ then of course they would not. Naturalism has nothing to do with the contrast between the natural and the man-made. (3) is a misunderstanding of the naturalist position, as I have characterised it. Naturalists do not claim that supernatural processes are beyond science; they claim that there are no supernatural processes. As for claims (5) and (6), I have already explained that strong naturalists need not endorse them. The strong naturalist need not maintain that higher-level phenomena can be reduced to physical ones, merely that they can be reductively explained in physical terms. Nor need they deny the legitimacy of other kinds of explanation.