I no longer remember where, but Gustavo Bontadini (a fine philosophical mind of some years back), on the problem of freedom, used to say that the question is so tangled that one does not even know where to begin. Before even debating whether human actions are free or not, the busillis would in short be to grasp what human freedom consists in: if the starting point is not clear, it obviously becomes difficult to see where one wishes to arrive. Is being free perhaps equivalent to maintaining that our actions are not caused by facts external to the acting itself? But granting that the sense in which acting is not merely a physical process, a biological one, say, is already clear, that it is clear, that is, why the pursuit of ends is an intentional and not a physical matter; granting, in short, that the impossibility of reducing human behaviour to natural processes is clear, a presupposition without which the whole discussion is meaningless (in what sense could a causal process, a physiological one, for instance, ever be free or unfree?), which is no small thing, it would follow that whatever is not caused would be random, merely contingent. But what sort of freedom would that ever be, in which our actions were wholly random? Finally, against many of our intuitions, dulcis in fundo, within the ranks of the theorists of freedom a prominent place is occupied by the so-called “compatibilists”. In the wake of Hobbes and Hume, many maintain that I ought blithely not to give a damn about the antecedents of my actions. If now, weary of racking my brains, I form the intention of going into the kitchen to make myself a coffee, I would be free so long as my wife had not decided to lock me in the study, even if my life were already settled in its particulars like that of a Homeric character. As if to say that freedom would have to do only with the possibility of realizing one’s own intentions, a possibility, precisely, compatible with the most iron-clad determinism of any event whatsoever.

Let us suppose, in any case, that there are good arguments against compatibilism. After all, if, driving like a madman, I run over a poor cat because my friend Giorgio has secretly slipped a hearty dose of some drug into my glass of wine, the cat’s owner ought to forgive me. Giorgio would in fact have rendered unusable the clause “Luigi could have done otherwise”, the clause of choice, which indeed holds sway in the courtroom as well. And yet, even granting a clause of that kind, the question goes on being confused.

Freedom, it is often asserted, is freedom of choice. And the perplexities go on looming. Suppose I have to choose whether to eat a chocolate or a custard ice cream. What kind of freedom do I possess if the two flavours are entirely indifferent to me? In that case my hand would try to seize one of the two ice creams without being guided by any precise intention, as if to say, precisely, that the choice would coincide with a random movement. If, on the other hand, I choose the custard ice cream because I prefer its flavour, that flavour seems precisely not to be chosen; it seems, rather, to be the unchosen condition of the choice. One might object at once that freedom is above all a moral question, and that it arises and becomes intelligible in the face of queries of this kind. And yet here too the matter is not clear at first glance. Suppose I am faced with the choice of stealing money that, I think, will make me happy, or else abstaining in order to do my duty (another problematic little word); well then, is the intention, let us say, not to steal the result of a free act, or do I simply find myself faced with the fact of having formed, or not, the intention to “not-steal” (or its opposite)? What I mean is: is it ever possible to choose the content of an intention if one does not already possess the general intention of which the concrete act is a specification? But how, in short, is it possible to choose the content of an intention without returning to the case of the ice cream, to the well-known libertas indifferentiae? How is it possible to choose the content of an intention, given that every choice, if it is not directed and guided by some intention, is precisely wholly blind? How, then, is it possible to be a moral subject if moral intent is not, to say it again, already at work? Is one therefore, in fact, so it seems to follow, moral or not, without this being-so being the result of a free choice? What, then, becomes of freedom? Or rather, as Bontadini intimated, what does freedom properly consist in, if one does not wish to reduce it to mere chance? Bertrand Russell used to say that those discourses are philosophical which call into question things that only a madman could doubt, only to arrive at conclusions that only a madman could accept. It may well be so; but since they have paid me my whole life to produce them, I carry on with the little game.

Clearly I shall not tackle head-on the discourse on the nature of freedom. I have already done so in a book years ago, with, in hindsight, meagre results. I can, however, pause briefly to consider determinism, which for many, the incompatibilists, is a necessary condition for there to be no freedom. On what conditions can one be a determinist? Obviously, in the theory the concept of cause is dominant. Every action of mine, every intention that guides it, would be caused, whatever the antecedent. The fact is that one often fails to reflect on the nature of causality. There are two dominant conceptions of the concept. According to the first, intuitive, commonsensical, and shared by most men of science, that of cause is a real relation between single events in succession. If you give me a kick on the shin, the kick produces, effects, creates (all synonyms of this concept of cause) the painful bruise. In the case where the two events recur under the same conditions, I shall think that causality goes on operating, that is, that the bruise always follows the kick. Let us call this causality ontological causality. There is, however, a second conception of causality, the Humean or nomological one (it seems Pyrrho, the most famous of the ancient skeptics, already held it). Between what we call cause and what we call effect, the nomological conception maintains, there is no relation other than that of succession and, possibly, of contiguity (though this last could be dispensed with), plus the regularity with which the two events recur, a regularity that precisely permits the hypothesis of laws, whence the label “nomological”. Nature, in short, is in fact made up of events that regularly recur. This is all, and beyond it “ask no more”; do not, therefore, seek any further, primitive causal nexus.

For many, this second conception is too weak to speak of causality. And yet, I believe, I think, it is the only tenable one. Let us reflect. If it is true that a certain relation subsists, then the terms of the relation must necessarily subsist. To say that Franco is beside Giovanna, but that Giovanna does not exist, is nonsense. But if, when the cause exists, the effect does not yet exist, and when the effect exists the cause exists no longer, then the causal relation goes up in smoke. The only way to save ontological causality is to maintain that cause and effect are simultaneous, not in succession. But, if that is so, it follows that: a) if the two are simultaneous, it is difficult to observe and to state which is the cause and which the effect; and above all, what matters most, b) events in succession, the regularities studied by science, can never be connected according to ontological causality. The only causality observed and studied is therefore that of regularity, the Humean or nomological causality precisely (the discussion then grows more complex, but this is the starting point).

Well then, if things stand thus, on what conditions can human life be regarded as deterministically established? Human beings are, besides being individually different from one another, above all historical beings; behaviours change, and historical events, if not in general terms, do not recur. It follows, since only the nomological sense holds, that human history and my personal history do not display that regularity of successions which would allow them to be subsumed under causal laws. Certainly I can go on being a determinist, asserting that every pair of successive events in my existence would have recurred had the conditions been the same; but it is clear that the empirical evidence for an assertion of this kind is entirely absent, and that nomological causality, in historical events, is almost wholly absent.

That my life is therefore deterministically established thus rests on a metaphysical inference devoid of empirical basis, that regularity which, in causal processes, is the only thing that counts. Have we thereby gained at least one necessary condition, certainly not a sufficient one, for being able to call ourselves free? To the reader, given the brief considerations set out at the outset, the arduous verdict.

Luigi Cimmino teaches Epistemology and Ethical Paradigms at the University of Perugia.