This review appeared in Dialogue XXVII (1988), pp. 523-528.

Rigorous Regularism: Physical Laws Without Necessity
— MICHAEL RUSE, University of Guelph
Norman Swartz, The Concept of Physical Law (Cambridge: Cambridge University Press, 1985), pp. xi, 220, $29.95.
This is a book about laws. Not, however, about the laws of which we learned in science classes at school: "scientific laws". It is rather about those universalities which govern the world of facts, what Swartz calls "physical laws" – although this language is slightly misleading because the term is intended to cover the living as well as the non-living world. Of course, it may well be that a scientific law (like Boyle's law) does capture the essence of a physical law, but not necessarily or (apparently) usually. A physical law by definition can never be false, whereas (as Michael Scriven pointed out some years ago) almost paradoxically nearly all (if not all) scientific laws are false.

This distinction between physical law and scientific law, made at the beginning of the book and stuck to throughout, tells us much about what sort of book this is. It is not a book in the philosophy of science – at least, not the philosophy of science as we understand it today, where one sees oneself almost at one with the very scientists, digging into their laws and theories and trying to make sense of their ways of thinking. Swartz is basically not interested in this book in science and its problems. Rather, The Concept of Physical Law is an exercise in metaphysics, and it is metaphysical problems that Swartz tackles and tries to solve. It is the problem of miracles that interests him, and (particularly) the problem of free will and determinism.

What is a physical law? Swartz lists five conditions that he thinks necessary. They must, as we have seen, be true. They must be contingent. They are not logical truths, which have their value purely by virtue of their form. They must be "purely descriptive" in their terms. This means that there must be no reference to particulars. ("All planets go round Harry Truman in ellipses" could not be a physical law.) They must be conditional. And they must be general – universal or statistical.

But the question remains. Is more needed to make for sufficiency? Is there that about physical laws which takes us beyond the conditions just listed? Essentially, there are two answers, "no" and "yes", and Swartz identifies them with two traditions in thought about the nature of (physical) laws: Regularism and Necessitarianism. These options are characterized thus:
Regularist: Physical laws derive their truth from the actual (i.e., instanced) connections (between states and between events) in the world. Physical laws, therefore, express only what does occur.

Necessitarian: Physical laws (and antecedent conditions) determine which connections can and cannot occur; physical laws, that is, express what must occur in particular circumstances (38).
Swartz does note that there are varieties of necessitarianism, particularly between those who find the necessity in things themselves and those who find it in laws. But, ultimately, they boil down to the same position: "Events in the world must accord with a nomological, natural (ontological) necessity" (38).

How do you decide between the two views? Swartz acknowledges that our language seems to show a preference towards necessitarianism. We speak of things having to happen, or that they must occur. But, acknowledging that there can be no knockdown blows, Swartz decides solidly for the regularist view. There is nothing more to physical laws than what actually happens. There is no special necessity (often called "nomic" necessity) or whatever. Swartz allows a notion that he calls "physical necessity", but this seems to be no more than what actually happens – or rather, what happens according to physical laws. The point is that it is not regularity plus something. What you see is what you get.

How can you argue for this? To be honest, it is not altogether easy to follow Swartz's line of argument, not the least because he takes it upon himself to quote people, but putting his own terms in when he thinks the original less than helpful. In the mode of the late Imré Lakatos, what is really said is then relegated to the footnotes – leaving at least one reader slightly bemused. Add to this, some rather odd use of language – "It is a positive distortion of the Regularity Theory to take its assertion that something is physically impossible as equivalent to an assertion that that thing could not exist" (62) – and comprehension is really quite uphill.

However, the argument seems to go like this (I rely heavily here on the summary on 101-103). If something is nomologically necessary, then we are saying that something must follow – given A, then B is necessary. This means that something is physically impossible (A and not B). Likewise, but in reverse, if we have the notion of nomic necessity, then we must also have (by contrast) the notion of contingency: given A, then B, but quite by chance (or some such thing). In other words, it is physically possible that we have A and not B. Belief in the notion of nomic necessity can thus be seen as equivalent to belief in the notion of physical possibility. But, now, what are the truth-grounds for talking about "physical possibility"? Actual possibility is obviously a sufficient condition (Pierre Trudeau is physically possible because his [sic] is actually possible). But, what if Pierre Trudeau did not exist? What of the first female Prime Minister of Canada? How do we say that she is physically possible?

The usual way to answer questions like these is to invoke dispositions. We say that even though the wood may never actually burn, it is physically possible that it do so because it has the disposition to burn. However, here, Swartz brings up the problem of entities which never even existed, and yet which the nomological necessity supporter likewise feels have to be given dispositions.
It is of the essence of Necessitarianism to argue that some events or properties are physically possible even though there may be nothing in particular in the actual world that has a disposition to realize those properties. For example, the theory requires that it be physically possible that someone should run a four-minute mile even if everyone were (accidentally) lame; indeed, that someone should run a four-minute mile even if there never had been any persons at all. It is these latter kinds of cases of physical possibility that mark the sharpest dividing line between the two rival theories. The Regularist does not comprehend how such existential propositions might acquire their modality. He does not understand what it is about the world that these propositions are supposed to be describing (102).
So, it all seems to come down to rival intuitions (I say this in a non-evaluative sense). A regularist like Swartz simply cannot see how non-existent entities can have anything, let alone a bearing on the real world. They are invoked simply to shore up an untenable philosophical theory about the nature of the laws of the world.

With his position now staked out, as mentioned earlier Swartz is now into a position to move towards the metaphysical questions which motivate him. Take first the question of miracles. How does one deal with them, given the existence of physical laws? They are troublesome, we learn, only if you hold to a necessitarian view of law. If this is your position, then you have the problem of explaining how miracles come about. For the regularist, miracles are no such problem. Understood as a violation of a law of nature, miracles are simply conceptually impossible! What happens is what happens and that defines physical necessity. If Jesus really walked on water, then that was part of the natural order of things. It was undoubtedly a remarkable event – a marvel – but it was no miracle.

The free-will problem gets a likewise brisk treatment. Expectedly, as a regularist, Swartz inclines towards compatibilism, but it is compatibilism of a fairly rigorous kind. The problem of determinism and necessity is no real worry, because necessity all boils down to what happens anyway. Hence, real freedom is possible. If you do one thing, then it is part of the course of nature. Equally if you do the opposite thing, it is part of the course of nature. The choice is yours, for there is no nomic necessity to constrain you.
It is partly up to us to decide what the grand physical truths (physical laws) of the world are. By choosing to do this rather than that, we "make" it a timeless truth of this world that, in such and such circumstances, persons do this.

Beware, I am not saying that it lies entirely within our province to choose the grand physical truths of this world. But what I am saying is that it is not wholly outside our will, either. We are not straws driven by the irresistible winds of natural necessity (126).
(In fairness, I should say that Swartz puts this argument into the mouth of a philosophically inclined Martian. However, the extraterrestrial does seem to be speaking for the author – although quite why a Martian would go to British Columbia to do philosophy escapes me. Perhaps the student grants are higher than in Ontario.)

There is much more to Swartz's book, and I have failed in my exposition if you do not now understand what I mean when I say that that remaining shows the same careful subtlety as the earlier part of the book. But, enough has now been said in discussion of content, for with the just-quoted passage I confess I find Swartz asking us to embrace ideas no less paradoxical and implausible than those he rejected from the necessitarian. I think it slightly ridiculous to talk of creating the "grand physical truths" of the world. To me – or rather to my intuitions – it seems no more plausible to talk of creating laws, than of making 2+2=4. And yet, I think of myself as a regularist in a way – and certainly as a compatibilist.

Likewise, if Jesus really did walk on water, then it would seem to me a miracle. It is marvellous that my teenage son would teach a friend to drive in my car – and smash it up. But that was no miracle. I could – and can – reluctantly believe this happened (as it did!). Yet Jesus' walking on water is something quite different. It is not a logical impossibility like a square circle; but, it is something which, for all the good Humean reasons, I find impossible to accept.

What has gone wrong – or, rather, where might one unhitch oneself from Swartz's rather remorseless argumentation? I think the clue is to be found in an admission that Swartz himself makes, right in his preface. Remarkably (not miraculously!) he has written a book on laws making virtually no mention of causality.
One thing that surprised me as I reread what I had written in this book was how remarkably little I have said about the concept of causal laws and of causality in general. I did not set out consciously to avoid the latter subject; it is just that it did not naturally, of its own, come up very much. That, I think, is a pretty interesting philosophical discovery. If this book had not evolved in the way it did, I probably would not have believed that one could say as much as I have without also talking at length about causality. It seems to me, now, in judging this book, that the two notions – that of physical law and that of causal law (a specialization of the former) – can, profitably, be discussed apart. But, clearly, this claim is contentious, and I alert you to it so that you might judge for yourself (xi).
I will say the claim is contentious. It seems to me that the whole secret to Swartz's book is that he has the kind of thesis that does not need reference to causation. What do I mean by this? Simply, that (as Hume showed us) you can either think of causes as powers "out there" or not – and as Hume also showed us, "or not" is the more tenable position. (I realize that there are those who would disagree with Hume, but for the sake of trying to understand Swartz I will assume here that powers are not a viable option.) But, if you agree that there is no causal necessity out there, then you are pretty much stuck with (and no doubt happy to embrace) a regularity theory of laws.

So far with Swartz. But, because he does not talk of causes, he does not then go on to make Hume's next move, namely to locate the sense of causal necessity within us. For Swartz, necessity is out there or not at all. And having set up this stark dilemma, he is readily able to poke holes in his version of the necessitarian thesis. It is virtually contradictory, claiming a necessity when all such necessity has been disallowed. (At least, it has been if you think – as Swartz and I think – that the arguments against powers succeed.) Hence, with his stripped-down ontology, Swartz gets his kind of regularity.

But if you stay a moment longer with Hume on causation, then you see that there is a role for a notion of epistemic necessity (not the ontological necessity that Swartz denies). This is the sense that things happen necessarily because that is what we read into the world. And this necessity is nomic necessity – the feeling that if A happens then B simply must happen, even though there is no logical necessity. It is this that necessitarians today, those whom Swartz criticizes, assert about laws of nature. And once we have got this assertion, then I believe that although the solutions to the great metaphysical problems may not come as readily as Swartz suggests, they do come more plausibly. What we who embrace a mind-imposed necessity count as a miracle, depends on interpretation (although not necessarily on a conscious decision to interpret). Miracles really are miraculous because they are those things which go against that which we consider necessary. They are possible, even though I doubt they ever occur.

Likewise, free-will becomes more difficult to understand – but more plausible when understood. All I will say here, since I really have nothing very new to say on what so many have already said, is that without a fairly robust sense of necessity for nature's laws, I really fail to see how you can distinguish rational choice from capriciousness. If I create nature's laws, then who is ever to blame me? The connections between past, present and future are very tenuous. I believe, to the contrary, that Freedom lies more in being a sufficiently sophisticated entity to bring first-order desires to achieve (or not to achieve) second-order desires, than in being able to make nature after your own pattern.

One final point. Swartz is clearly not insensitive to the kind of counter-position I am endorsing. Towards the end of his book, in discussing Nicholas Rescher's pragmatic idealism, he touches on the notion that we read necessity into nature. However, he dismisses it as too implausible for words.
Could nomicity thus explicated, lying "in the eyes of the beholder," possibly do what Rescher requires of nomicity? Could it, for example, account for the modal aspect of the "having to" in "all cats have to have backbones"?

The suggestion strikes us as implausible. All cats, to the best of our knowledge, have backbones. But whether they do or not has nothing to do with us, not even with our engaging in scientific research and theory construction. Although we may discover this fact about cats, surely it was "there" antecedently to be discovered. Why should the case be any different when it comes to ascertain whether cats have to have backbones? Why should this be a question requiring for its answer a fact about us? Surely – like their having backbones – their having to have backbones, if indeed it is a fact at all, is a fact about cats, not about us.
At least, at this point, Swartz should consider the suggestion of the late John Mackie1 that perhaps it is part of our nature to "objectify" experience. It may be implausible to think that our sense of necessity comes from us – but it does, and part of its power is the very sense of implausibility that is engendered by thinking that necessity starts with us.

If we did not think that necessity is "out there" – if common sense were not with causal powers – then our belief systems would not work. They would not work for us as biological beings, for we would no longer think that fire really causes burning and we would no longer fear it.

Of course, this is only Mackie's intuition (and mine) although it is backed by Darwinian evolutionary theory. The being who "objectifies" necessity will be biologically fitter than the being who does not. However, since Swartz likewise appeals to intuitions, I will leave it to the reader of this review to decide which position is preferable.

1 J. Mackie, Hume's Moral Theory (London: Routledge and Kegan Paul, 1979).