[1] By convention current is actually said to flow in the opposite direction from the movement of electrons.

 

 

There are those who say that we aren’t justified in believing that there really are the sorts of things that are spoken of in our best theories, and for that reason we should look at theories not as giving us a description of the world but as a sort of tool or instrument for prediction. We input the initial conditions into the theory as an instrument and out pop predictions of what will happen. The worth of a theory is not to be found in how closely its description of the world matches the actual state of the world, but in how closely its predictions are matched by the observed outcomes. This way of looking at theories is called Instrumentalism.

 

The principal reason why instrumentalists deny that we should believe in the things that our best theories talk about is because there have been so many revisions to our theories of the world in the past. Not so long ago, for example, our best theory of light was that it was a kind of wave that propagated in the not-yet-detectable ‘ether’ in the same way that a longitudinal wave moves through water, or that a compression wave moves through metal. To accept this theory as a description of the world was to accept that there really is such a thing as the ‘ether’ which was required as the medium in which light waves existed. Now, of course, we don’t think of light as being that sort of wave at all and our theories don’t claim that anything like the ether exists. At one time also, doctors worked on the theory that illness was due to an imbalance in the levels of four ‘humours’ in the body. (They were the sanguine, choleric, phlegmatic, and melancholic humours.) (We still talk about a person being in a bad humour if they seem out of sorts.) In order to relieve the symptoms and therefore to cure the patient of his ailment it was necessary to find a way to restore the balance of these humours. But if you’re going to believe that this theory describes the real state of the human body, then you’re going to need to believe in the real existence of these humours. Now, of course, we can see that there are really no humours and illnesses have a vast range of different causes. Finally, we can note that at one time it the best astronomical theory claimed that the Sun and the planets and the Moon all went around the Earth, which remained fixed and immovable. According to this Ptolemaic theory of the universe, those celestial bodies were fixed to crystalline spheres centered on the Earth. They needed to be fixed to these spheres because there was no other way that they could be kept in their places; and the spheres had to be crystalline because otherwise we would see them. Those had to be real objects in the universe according to this theory; but once again we now know better: there are no such spheres, crystalline or otherwise, in our current theories.

 

Instrumentalists make a pessimistic induction from these and other episodes in the development of our knowledge

 

Etheric theory was our best theory, but the things it described didn’t exist

Humoral theory was our best theory, but the things it described didn’t exist

Ptolemaic theory was our best theory, but the things it described didn’t exist

…

----------------------------------------

The things that our best theories describe don’t exist

 

But if our best theories are not to be interpreted as telling us how things are in the world – as they certainly seemed to – then what is the point of them? The instrumentalist at this point answers that they are useful as instruments for predictions. After all, it was felt at the time that these theories were held that they performed well as instruments: the etheric theory told us how light beams could interfere with each other and how they could bend around objects, and so on; the humoural theory told us how we could relieve some distresses through purgings and emetics and bloodletting, and how some foods and drugs could affect us; and the Ptolemaic theory enabled us to predict with great accuracy most of the events in the heavens, such as eclipses and progressions and planetary alignments. They could perform an optimistic instrumental induction such as the following

 

Etheric theory was valuable as a predictive tool

Humoral theory was valuable as a predictive tool

Ptolemaic theory was valuable as a predictive tool

…

----------------------------------------

Our best theories are valuable as predictive tools

 

Which would have been supported by two further considerations: first, that in many cases what we meant by ‘best’ in ‘best theory’ seems to be no more than that the theory made good predictions; and, secondly, that we didn’t begin to doubt the worth of the theory until (a) we found too many places in which it couldn’t make good predictions, or, (b) a ‘better’ theory (in the sense just mentioned) was offered. For example, the Michelson-Morley experiment was intended to determine how fast the Earth was moving through the ether by measuring the different speed of light in two different directions. The unexpected outcome was that there was no difference at all, which had to mean either that the Earth wasn’t moving (which was obviously untrue) or that light did not move through an ether in the way theorized. For another example, when Galilei used his telescope to observe the moons of Jupiter orbiting that planet, it was then clear that (1) not all bodies in the heavens moved about the Earth’s centre, and (2) that Jupiter wasn’t fixed on a crystalline sphere, else the moons would have been unable to orbit it. These problems motivated the adoption of better – more accurately predictive theories; Einstein’s theory of relativity in the case of light, and the heliocentric Copernican theory in the case of astronomy.

 

Explanations

 

Yet, even granted the history of error in scientific theories, the instrumentalist view of theories seems unsatisfactory to many. The primary reason that many are inclined to dismiss it is that from the time of Aristotle we have looked for explanations of why the world is the way it is, and theories are supposed to be those explanations – but the instrumentalist view doesn’t seem to treat them as explanations at all, leaving our natural desire for explanation unfulfilled. So although we might say that the reason that the statue is brown is because the statue is made of bronze and bronze is brown, this is not, on the instrumentalist view, an explanation, but merely a shorthand for the use of the ‘theoretical’ instrument of prediction. But this leaves the facts about bronze statues actually unexplained, and since any explanation is going to look like a theory and the instrumentalists say all theories are merely instruments, it looks like they have to insist that explanations are just impossible in principle!

 

Surely, however, it is reasonable to accept that there are explanations for the way the world is – it is hard to imagine what it would even mean for the world to be inexplicable in principle – and it would follow, too that it is rational to believe the best explanations that we can come up with for the way that the world works, unless we have firm grounds to think that they are in fact false; and to believe an explanation means that when it says that there are electrons in the world then we should believe that there are electrons in the world, and when it says that there are not crystalline spheres in the world then we should believe that there are not crystalline spheres in the world. This is not to say that we should not approach these things with a certain degree of skepticism, but the declared intention of an explanation to describe the world should at least be accepted at face value and criticized on those terms.

 

At this point, however, an Instrumentalist might simply deny that explanations are fundamental to Science. The point of Science might perfectly well be taken to be the mere description of how the world behaves, without bothering with the further extra-scientific question of why it behaves in that way. This isn’t quite as absurd as it sounds at first, because we don’t typically think that scientists who are working at the very edges of our theories and describing the fundamental particles (quarks) and constants (such as the gravitational constant have ceased to do science.) On the other hand, to extend that approach to all of must Science seem a radical departure from our naïve understanding of what we’re doing when we study Nature. Aristotle himself, as we saw, accepted that there would be certain fundamental truths knowable only by non-scientific means, but we also saw that the point of episteme was to provide explanations for the rest of the facts about the world in terms of deductions from these fundamental facts. Without stronger arguments for the anti-explicative view of Science, we will remain unconvinced.

 

So much for the principal motivation for rejecting instrumentalism, but there are other more direct arguments in favour of realism. As we go through them, note that none of them are deductively valid, but are the sorts of arguments that urge one to consider the plausibility or likelihood of alternatives. We will see that it is always possible to reject them, but it’s probably more rational to accept them.

 

 

Best Explanations and Best Interpretations

 

The most common direct argument appeals to a kind of inductive argument that we’ve seen earlier: the inference to best explanation. Let us recall how this works. We begin with a certain set of phenomena and we also have a number of theories that might be used to explain them. We choose the best of these theories and declare that because it’s the best theory, that’s the one we should believe. In introducing this, I gave the example:

 

You return home to find your door broken and some valuable items missing.

Possible explanations include:         

                1.             A meteorite struck your door and vaporised your valuables.

                2.             Friends are playing a joke on you.

                3.             A police Tactical Response Group entered your house mistakenly.

                4.             You were robbed.

                                Explanation 4 seems the best, so you conclude

                                -------------------------------------------------------------------------------------------------------

                                You were robbed.

 

More generally, we observed that inferences to best explanation take the following form:

 

                                Phenomenon C is observed

                                A explains C and does so better than any rival explanation

                                -------------------------------------------------------------------------

                                A

 

But to apply this to the dispute between instrumentalism and realism we’ll have to make a few adjustments, because what we’re interested in now is not the ‘truth’ of one theory or another, of Copernican or Ptolemaic astronomy, of Humoral or Modern medicine, of Etheric or Quantum theories of light, but how we should interpret the rival claims to truth of those theories. What we are looking for now is the most rational approach to the interpretation of scientific theories. The appropriate argument will therefore look something like this:

 

The pursuit of scientific knowledge shows the following characteristics: A, B, C, …

The following interpretations of scientific theories make sense of those characteristics:     

                1.             Instrumentalism

                2.             Realism

                                Interpretation 2 seems to make the best sense, so you conclude

                                -------------------------------------------------------------------------------------------------------

                                Realism is right

 

To complete this argument it only remains to outline the characteristics of science that are relevant to the argument (the A, B, C, … above,)  and to show how it is that realism makes more sense of them than instrumentalism does. That’s what we’re going to do now. (Note that we have ignored the possibility of there being other interpretations. This too would need to be justified.)

1. No Miracles

 

The most obviously impressive thing about science is just how successful it is. By using the theories that science comes up with we’re able to provide an explanation of phenomena that allows amazingly accurate predictions of future or events, and that explains or is consistent with past events. We rely upon this accuracy every day of our lives. Bridges and buildings that are built according to the accepted physical theories are safe (and we don’t need to apply ‘rules of thumb’ to know it as ancient architects had to do.) Airplanes and spaceships work reliably. We find that common GPS devices – which get their data from satellites and have to take account of relativistic phenomena – are fantastically accurate. And so on. But how can this be so? What is it that makes these theories just so very useful? All would be explained if it really was the case that these theories are describing how the world actually, really is. The buildings and bridges are safe because there really are forces that are acting on masses in just the way that the theories describe. Similarly for planes and rockets. And the GPS devices work because time really does slow down and mass really does increase for objects moving at very high relative velocities. Is there any other way that these things could be explained? There’s a famous quote from Hilary Putnam that Realism “is the only philosophy that doesn't make the success of science a miracle”[1] and so this argument for the Realist interpretation of science is referred to as the No-Miracles Argument.

 

The Instrumentalist response to this argument has often been that the reason that scientific theories are so successful is that they are selected from all possible theories on the basis of their successfulness. There is really no mystery about it. Whether that’s a satisfactory response, however, has to be doubted: the fact that theories are chosen for success does not give us an answer to the question of why they are successful. The Instrumentalist simply declares that there is no place for such further explanations in Science.

2. Corroborations

 

It is a further notable feature of scientific theories that the unobservables that they propose can be approached from several different directions. Take the example of the existence of germs: they were proposed as the basic requirement of the germ theory of disease, and their existence was established by scientists through such techniques as epidemiology, a standard method in medical science. And then it was possible to direct the aim of optical instruments towards where the germ theory said that these entities would exist in abundance. When this was done an image appeared in the relevant microscopes. Accepting that microscopes do actually show things that are too small to be observed by the naked eye, which is a theoretical claim from a the science of optics, we can say that the microscopic observations corroborate the epidemiological inference that there are such things as germs. But note that epidemiology and medical science and optics are very remote from each other, so how is it that such corroborations can occur – and occur so often as they do? If each theory was merely an independent instrument for predictions in its own field of application, then there’s no good reason to think that the instruments will behave in this way at all. Only Realism can make sense of this characteristic of Science.

3. The Project of Unification

 

The last of the characteristics of Science that we’ll look at (though not the last of the characteristics that have been proposed for the Inference to Best Explanation argument for Realism) is the project of unification. It is a noted characteristic of Science that there is a tendency for scientists to try to create theories that unify several different fields of study that are covered by different theories. We’ve seen that chemistry has been united with particle physics, and particle physics with theories of light, and theories of light with theories of electromagnetism and radiation general, and so on. In fact the great project – very far still from completion – is the creation of the so-called Grand Unified Theory in which all the sciences will be unified and shown to be derivable from a single statement describing a fundamental truth about the world. At the very least, theories that contradict each other with respect to the supposed entities of the universe are rejected. Again, we can see why this would be well-motivated if we take a Realist attitude to theories, because they are all understood to be descriptions of different aspects the same world, and eventually they all have to agree with each other to state the unique truth about the world. For the Instrumentalist, on the other hand, there seems to be no reason to think that the project of unification of the sciences is achievable or even desirable.