We have still to consider the extension of the methods of number theory to infinite ordinals—or to transfinite numbers as they may also, as usual, be called.

The means for establishing number theory are, as we know, recursive definition, complete induction, and the “principle of the least number.” The last of these applies to arbitrary ordinals as well as to finite ordinals, since every nonempty class of ordinals has a lowest element. Hence immediately results also the following generalization of complete induction, called transfinite induction: If A is a class of ordinals such that (1) ΟηA, and (2) αηA → α′ηA, and (3) for every limiting number l, (x)(xεl → xηA) → lηA, then every ordinal belongs to A.

In this paper we shall be concerned with questions regarding decision problems for certain classes of sentences (without quantifiers) of various kinds of algebra. In the first section we shall establish a result of a general nature, which enables one, in a number of cases, to reduce a decision problem of the type considered to a somewhat simpler problem. In the second section we formulate a rather broad sufficient condition for the existence of a decision method for sentences without quantifiers; and in the last section we show that this condition holds of lattices.

Two more or less standard methods exist for the systematic, logical construction of classical mathematics, the so-called theory of types, due in the main to Russell, and the Zermelo axiomatic set theory. In systems based upon either of these, the connective of membership, “ε”, plays a fundamental role. Usually although not always it figures as a primitive or undefined symbol.

Following the familiar simplification of Russell's theory, let us mean by a logical type in the strict sense any one of the following: (i) the totality consisting exclusively of individuals, (ii) the totality consisting exclusively of classes whose members are exclusively individuals, (iii) the totality consisting exclusively of classes whose members are exclusively classes whose members in turn are exclusively individuals, and so on. Any entity from (ii) is said to be one type higher than any entity from (i), any entity from (iii), one type higher than any entity from (ii), and so on. In systems based upon this simplified theory of types, the only significant atomic formulae involving “ε” are those asserting the membership of an entity in an entity one type higher. Thus any expression of the form “(x∈y)” is meaningless except where “y” denotes an entity of just one type higher than the type of the entity denoted by “x” It is by means of general type restrictions of this kind that the Russell and other paradoxes are avoided.

Peano's five axioms for the sequence of natural numbers run as follows: (1) 1 is a number, (2) Any number α has a unique sequent a′ (3) Different numbers have different sequents, (4) 1 is not a sequent, and (5) The principle of mathematical induction.

Now, in a system of axioms, it seems desirable, in the first place, to avoid symbols for particular individuals and symbols for functions, and, secondly, to secure the property of complete independence. In fact, Peano's system can be replaced by an equivalent one which fulfills both requirements. In such a system, the symbol “1” and the function “sequent” will not appear, although they may be introduced afresh through definitions. Besides, the principle of mathematical induction will no more play the rôle of an axiom, because, under the second requirement, it can hardly be stated in a simple and elegant form.

The motives for seeking economy in the basis of a system are much the same as the motives for constructing the system itself. A given idea A need be left as primitive in a system only so long as we have discovered between A and the other primitives no relationship intimate enough to permit defining A in terms of them; hence the more the set of primitives can be reduced without becoming inadequate, the more comprehensively will the system exhibit the network of interrelationships that comprise its subject-matter. Of course we are often concerned less with an explicit effort to reduce our basis than with particular problems as to how to define certain ideas from others. But such special problems of derivation, such problems of rendering certain ideas eliminable in favor of others, are merely instances of the general problem of economy. Thus it is quite wrong to think of the search for economy as a sort of game, inspired by an abnormal love of superficial neatness. Some economies may be relatively unimportant, but the inevitable result of regarding all economy as trivial would be a willingness to accept all ideas as primitive at the outset, making a system both unnecessary and impossible.

In this paper it will be shown that the three primitive symbols used by Lewis are independent in each one of his systems S1–S6.

Lemma I. “·” is not definable in terms of “∼” and “◊” in any one of the systems S1–S6.

Proof. This is obvious, since no binary operation can be defined by unary operations alone.

Lemma II. “∼” is not definable in terms of “˙” and “◊” in any one of the systems S1–S5.

The concept of confirmation occupies a central position in the methodology of empirical science. For it is the distinctive characteristic of an empirical hypothesis to be amenable, at least in principle, to a test based on suitable observations or experiments; the empirical data obtained in a test—or, as we shall prefer to say, the observation sentences describing those data—may then either confirm or disconfirm the given hypothesis, or they may be neutral with respect to it. To say that certain observation sentences confirm or disconfirm a hypothesis, does not, of course, generally mean that those observation sentences suffice strictly to prove or to refute the hypothesis in question, but rather that they constitute favorable, or unfavorable, evidence for it; and the term “neutral” is to indicate that the observation sentences are either entirely irrelevant to the hypothesis, or at least insufficient to strengthen or weaken it.