Question: “What is the ontological status of scientific law?”
I’m going to provide an account of what I take scientific laws to be. I dimly remember starting a thread asking what scientific laws were a few months ago. I never got around to contributing to my own thread, so here’s a (delayed!) contribution.
First, we need some metaphysical commitments:
1)There exists a world in which there exists objects of some sort. We don’t need any assumptions about what these objects, or the world, are like.
2)We need the distinction between an object as it appears (appearance) and an object as it really is (reality). By this I mean no more than the view that just because I experience a tree as being an object of a certain sort that doesn’t mean that another being with a different cognitive make-up wouldn’t experience the tree as being an object of another sort. Therefore, whether the tree is like my experience of it is necessarily an open question, and so we need the concept of ‘tree as it really is’.
3)Human actions have an obvious effect in the world of appearance. We need to assume that there is a correlative effect of human action in the world of reality. I believe this assumption is prima facie warranted: if what ‘appears to be’ changes, and we have no knowledge of what is, the view that ‘what is’ changes as well seems most reasonable. We can express this (badly) by saying that the worlds of appearance and reality causally interact.
Lets think about the scientific process of measurement. When I determine that what appears to be a cat is 0.3m in length, what does this mean? It is at once subjective and objective. Subjective: the concept of a metre is just an arbitrary standard to which we are comparing our cat. If we suppose that to have a length of 0.3m is an attribute that the cat possesses we suppose that any object possesses an infinite number of attributes corresponding to whatever arbitrary standard of measurement we wish to apply to that object. Objective: at the same time, once we have our arbitrary unit of measurement a certain object at a certain time will have a certain measurement on our arbitrary scale no matter how you or I feel about it. The cat makes it the case that it is 0.3m in length without being 0.3m in length being a property of the cat.
We can now formulate a general story of what scientific measurement is all about. Its purpose is to probe into the world as it appears to us, compare it to a set of arbitrary standards, so producing a description of what appears in terms of these standards. What we get is a numerical description of the world as it appears to us. Note that, of course, we also apply this description to the ‘world as it is’ (reality), but whereas when we describe appearance in this way all we’re doing is trying to organise what we see into something coherent, when we apply this description to reality we are making claims that in all possibility may be completely wrong. Anyway, this description is equivalent to a description in a natural language, say english. To say that a cat is 0.3m long is, fundamentally, no different to say that it is ‘short’: one is a mathematical description of the world, the other uses words, but both are just descriptions.
So, we have a thesis:
“The experimental method in science (i.e. measuring things) produces the building blocks (i.e. the measurements) of a mathematical description of the world (as a description of appearance it describes what we actually see, as a description of reality it describes what we (rather groundlessly) infer to be there from what we see).”
So measurements are building blocks of a mathematical description of reality.
So we have two alternative types of description of the world. Neither description is the world, a description of the world can never be the world itself, no matter how precise. There is a tendency to think that mathematical statements must, somehow, be more characteristic of the world ‘as it is’ than descriptions using words alone. This tendency is utterly without ground. To think otherwise is just to think that our most precise interpretation of the world is the world itself.
Before answering my initial question, I’m going to offer a general picture of reality. Here’s a pointless statement: reality is what there is. Here’s a sometimes overlooked statement: we are a part of what there is, hence of reality. Where I want to diverge from someone like Kant is that I want to put real emphasis on something that we might forget. We cannot ever get into the way of thinking that holds that the worlds of appearance and reality are separate. As I said, we are a part of what there is. When I form a description of reality I don’t just sit there happy with my description. I go out and perform actions. Now, as before, we must assume that our actions have some correlate in reality for the same reason we have assumed that what we see as a cat must have some correlate in reality. As such, reality is in part shaped by the descriptions that we have formed of it based upon our experiences. The crucial point, that Kant never seems to touch upon, is that ‘reality’, if it is just what underlies what appears to us, should not just be assumed to be static: what appears to us isn’t static. Rather, there is a process of constant 2-way interaction between what we have called the worlds of appearance and reality. Reality affects what we see as appearance, this leads us to form a description of reality, then this description affects reality. As such, what we really have is something approaching a monism rather than the apparent dualism. Reality and our reconstruction of it are constantly shaping each other.
We need to note that there can be no coherent distinction between ‘internal’ and ‘external’ reality: we are no more in touch with us as we are in reality than we are with the cat in reality. We are always an appearance to ourselves. I’m not going to develop this, I just want to make it clear that I’m not going to privilege the human case.
Finally, now, to answer my initial question. Take an example: F=dp/dt. Remember that we have the beginnings of a mathematical description of reality in our measurements. We could say that measurements are like the nouns of our language: they are used to identify objects, mathematically speaking. But a language needs more than nouns. We use equations like our example to provide generalisations, i.e. statements that aren’t just about one object but about many. We could think of them as sentences. Obviously, we can have laws of greater or less generality (as, in our example, F=ma is a statement of less generality, Einstein’s equation of motion is a statement of greater generality).
So: a law is really just a component of our mathematical description of reality, just as a measurement is. As such, it is no more ‘reality itself’ than the measurement is. The language is always a description of reality, never reality itself, hence the law is never reality itself. My conclusion is that a scientific law has no more reality than a word: certainly, a particularly powerful word, with a great deal of descriptive power. But still no more ‘real’.
Of course, there are differences between a measurement and a law. But all I am claiming is that both have the same ontological status. We form laws by comparing measurements in different experiments and coming up with equations.
I’m aware that none of this is really an argument, but I do feel that the sheer simplicity of my solution, when compared to the problem that we have accounting for the existence (in whatever sense) of scientific law in a more usual way, is its chief merit.
If anybody is at all interested feel free to press me upon any of this. There’s a lot of stuff that I’ve left out, especially about scientific method.