Sobre el concepto de ley natural

Abstract

The purpose of this article is to give an account of the concept of natural law. It is known that the expression “natural law” is used in scientific-philosophical literature in various senses; all of them more or less interrelated. In this paper the concept of natural law is characterized as a pattern of nature’s behaviour, in opposition to natural law as a valid statement in the context of a theory. Therefore, the main question raised is: what feature would a law has that would describe the universe’s real behaviour? As we can see, this is a question that has an ontological component. But we can also see that this is not the most ontological version of the problem of natural laws. Ignoring the ontological problem in a strong sense, an ontological problem in the weak sense is left: still admitting that a law is a simple statement, in other words, a certain linguistic entity, what logical-semantical-grammatical feature must a statement have to deserve the name of a description of the universe’s behaviour? The last question leads us to establish some necessary conditions for calling a statement a natural law. The following, expressed in an informal language, are such conditions :
(a) Veracity: natural laws must be true statement about nature’s behaviour;
(b) Universality: natural laws must be valid statements in general, that is, not restricted to a determinate space-time zone of the universe;
(c) Non-contingency: natural laws must be statements free of every shade of contingency. Of course, unless the meaning of “contingency” is explained, this condition is useless.
The precise purpose of this paper is to establish the logical form of natural law according to the intuitive conditions previously adopted.
(d) Communicability: natural laws must be theoretically communicable statements in the following sense: every intelligent being acquainted with the meaning of our general terms and who dominates the grammar of our language must be in condition to understand the statements we called natural laws.
In virtue of (c) and (d), we can draw a well-known conclusion, namely that natural laws must not contain terms which designate particular objects in essential (not eliminable) form. Actually, the essential occurrence of the name of a particular object seems to make the universe’s behaviour depend on the existence of an object. This can be thought of as an excessive dose of contingency, if we think —by way of an example— that every isolated object is in itself contingent.
On the other hand, in virtue of the condition (b), specifications of time and space cannot figure in essential form in the statement of a natural law.
From conditions (a–(d) the following basic conditions can be derived:
(e) No essential mention should be made of particular objects;
(f) No essential use of numerical specifications of space-time character should be made.
The primitive conditions (a), (b), (c), (d), and the derived ones, (e) and (f), are interesting, but not enough. Two fundamental problems emerge: the problem of the purely qualitative predicates of Hempel-Oppenheim, and the problem of the cosmic accidents of Popper-Kneale-Simpson.
By cosmic accident, we understand something that describes a contingent natural regularity. This is very clearly exposed by Popper’s classical example about the moas in New Zealand. In spite of being a strictly universal statement, “every moa dies before reaching fifty years of age”, its contingent character, put into evidence by a natural law, would not be produced in determinate conditions, objectively different from those that were really given.
We must keep in mind that a cosmic accident is a fact expressable by means of a strictly universal statement, and that it provokes some problems which are tentatively resolved in this paper. The problem raised here is that in order to characterize a cosmic accident we need the previous introduction of the concept of natural law. This seems circular, but as we shall see it is only apparent. Therefore, the approach to the possible solution is based on the concept of “legaloid statement”.
Of what has been said above a reasonable plan to solve the above-mentioned problem arises. It consists of the following: (1) to give a characterization, with the help of the conditions (a–(f), of a previous concept for a law, which is the legaloid statement; (2) to give a characterization, among the legaloids statements, of those which express a cosmic accident, exemplified paradigmatically by the example of the Popperian moas; (3) to define as a natural law every legaloid statement which is not a cosmic accident (and that has no other minor inconvenience).
What is first offered is a first approximation to that scheme, which seemingly translates the intuitive ideas exposed above into formal terms, and which outlines the profound difficulties, of logical-semantic character, that arise in the development of this plan.
A formal and precise definition of legaloid statement and of cosmic accident statement are given and also commented on in detail. Later some commentaries on ontology and truth are made, because the definition of a legaloid statement requires that such a statement be true inasmuch as it refers to reality.
The formal requirement (e) has been discussed abundantly in scientific-philosophical literature for the last century. In connection with the difficulties that arise when we want to define precisely what is meant when we say that a statement does not contain essential references to particular objects, we can use the notion of purely qualitative predicate; a predicate is purely qualitative “if the establishment of its meaning does not require reference to a particular object and to space-time localization”.
Hampel and Oppenheim think the following difficulties to be unavoidable: even if the notion of purely qualitative predicate is referred to a formal language L, “our problem remains open for the primitive [terms] of the language, whose meanings are not determined by internal definitions of language, but by semantic rules of interpretations”. These authors observe that this way of dealing with the problem of characterizing the purely qualitative predicates is to push it into the metalanguage in which the interpretations of terms of L are realized. But if we adequately formalize the metalanguage, the problem is pushed into the meta-metalanguage, and so on.
The answer for this is the following: in the context of this paper (and also in the cited work of Hempel and Oppenheim) and in view of the present state of the philosophical investigations concerning natural laws, the demand of these authors is excessive and lacks relevance, that is, lacks relevance as an objection to a speculative theory of natural laws. It could have some relevance as an objection to a speculative theory of natural laws. It could have some relevance as an objection to a formal theory which puts the accent on problems of decidability; but the argument in this paper consists of establishing that the theory of natural laws is still in an essentially speculative stage, and that it is not in condition to trace problems that are too formal. Therefore, we have to leave the concept of purely qualitative predicate in its present state and vagueness although it still can be of interest to epistemological investigations.
The objections outlined above are important but must be postponed in the first approximation. The fine discussions on ontology and truth, as well as on the definability of predicates, would be more relevant in a scheme acceptable for a theory of natural laws. There are some logical and logico-semantical objections to the scheme proposed in § 3, some of which are relevant to the purposes of this paper. These are connected with what is called, in Quine’s terms, “promiscuous disjunction”, which consists of proposing an alternative between two possibilities that, from an intuitive point of view, have nothing to do with each other.
After examining Popper’s example once again, the author of this paper gives a solution to the problem which is not only provisional but also tentative, an indication of possibilities.
It is possible that the scheme proposed could be refuted by means of trivial examples such as those given against the first approach in § 3. It is the author’s belief, however, that the methodology used in this provisional, tentative solution could be illuminating and fruitful in other more sophisticated approaches to the subject.
The idea on which the new proposal is based has been obtained from the “modus operandi” of natural science (especially physics, and also chemistry and biology): it is meant to explain some organizational levels by means of other more elementary levels. In the course of this new approach, new concepts are defined, such as a totally explainable predicate, which permits the definition of other concepts, and finally the concept of cosmic accident statement, upon which the definition of natural law is based. The plan developed in this paper enables us to avoid enigmatic concepts such as physical necessity, subjunctive conditional, causality, etc.
According to the author, the argument of trivial promiscuity against the first approximation has no validity against the second one. There is no guarantee, however, that the tentative, provisional solution is immune to other more refined kinds of promiscuities, or other perverse counterexamples.

(Arturo Cisneros)