Mechanical logic in three-dimensional space (Q2855950)

The author continues the old tradition of developing logic as a mechanical calculus. In contrast to the commonly applied algebraic approach, his solution is based rather on arithmetic. In Part I he introduces the notion of \(n\)-dimensional logical space, which is analogous to the space of arithmetic where each number occupies a well-defined position. Similarly, in logical space, propositions are identified by means of finite sequences consisting of 1 and 0. The number of dimensions refers to the number of different variables involved. Part II is concerned with a detailed development of a theory of syllogism-like inferences in three-dimensional space based on the operations of product, sum, subtraction and division. Finally, in Part III, some generalizations and possible applications are sketched. In particular, it is suggested (after Peirce) that the three-dimensional space is sufficient since all inferences containing more variables than 3 and more premises than 2 may be reduced to simple inferences dealt with in Part II. Also some comments on reducing relations to unary predicates are given and possible applications of the presented machinery in artificial intelligence and quantum calculations. NEWLINENEWLINENEWLINENEWLINEAlthough the general idea is interesting, and in the text one can find several remarks of heuristic value, the approach offered in the book can hardly be treated as a serious alternative to modern development in logic. The author's proposal could be considered either as a practical tool for making inferences by humans or as a basis for automated deduction; unfortunatelly it fails on both fields. Despite the claims of the author, all the machinery involved in the transformation of inferences into arithmetical operations is really hard to follow. The book contains numerous lenghty tables containing the results of combinatorial encoding formulas and inferences into sequences of numbers. I admire the author's effort in doing all this but I really doubt if it can be applied as a ``simple'' method of proving. Of course the notion of ``simplicity'' is vague but I'm strongly convinced that natural deduction or tableau methods offer much simpler solutions for the analysed examples.NEWLINENEWLINEThe arithmetic engine for mechanical inferences offered by the author cannot be also treated as a serious proposal for automated deduction. In fact, the book contains no reference to mainstream research in this field. There is no comparison with resolution or tableau methods, no consideration on efficiency of the proposed method. The problem of reducing other types of inferences to those which can be performed in three-dimensional space is not developed in a rigorous fashion. In fact, even the author seems to be not fully convinced of the possibility of such a reduction since he claims that a theory of spaces with more dimensions should also be developed. Even historical remarks are far from being complete; for example, neither Jevons nor Schroeder are mentioned. All these deficiencies make this book rather a mere curiosity than serious scientific achievement.