The development of Euclidean axiomatics. The systems of principles and the foundations of mathematics in editions of the \textit{Elements} in the early modern age (Q343827): Difference between revisions

This is a very much needed richly commented survey of the many editions of Euclid's \textit{Elements} in the early modern age, each with its own version of the axioms and postulates, deviating widely from those in the original text. As such, this paper is a treasure trove of early modern axioms and postulates, most of them in no relation to anything in Euclid's \textit{Elements}. Some of them will turn out much later to be of use in the axiomatic foundations of the late 19th and the 20th century. If we look, to take an example, at the group of axioms that can be fairly described as ``principles of incidence and continuity'', we find (1) ``If a straight line divides a rectilinear angle and is produced to infinity, it will intersect a given straight line which is applied [with its ends] to the straight lines forming the angle. And a straight line drawn from a rectilinear angle to a point inside its base will divide the angle. And a straight line drawn from a point of the base of a triangle to the opposite angle will divide the angle.'' (Richard, 1645) (which contains a statement of the ``crossbar theorem'', see p.\ 116 in [\textit{M. J. Greenberg}, Euclidean and non-Euclidean geometries. Development and history. 4th ed. New York, NY: W. H. Freeman and Company (2008; Zbl 1127.51001)]; (2) ``If a point \(F\) is taken in the base \(BD\) of a triangle \(BCD\) and a straight line \(FG\) is drawn inside the triangle and extended enough, it will intersect one of the other sides \(BC\) or \(CD\) at a point \(G\), or both of them at their common point \(C\) (and \(C\) is then the same point as \(G\)).'' (Roberval, 1675) (which is the statement of the Pasch axiom, if we choose to ignore the ``inside the triangle'' part as an irrelevant verbal flourish); (3) ``An infinite straight line divides the infinite plane in two parts.'' (Kästner, 1758) (a plane separation axiom, which, in case separation is defined, becomes equivalent to the Pasch axiom). If we look at the ``principles of continuity'' we find the so-called ``Line-circle continuity axiom'' in various forms: (1) ``If a segment has one end inside a circle and one end outside it, it intersects the circumference at one point.'' (Kästner, 1758), (2)``An infinite straight line passing through a point inside a circle intersects the circumference at exactly two points.'' (Pascal, 1655), and (3) ``If a circumference has a point on one side of an infinite straight line and the center on the other side, it intersects the straight line at exactly two points.'' (Pascal, 1655). Similarly, we find the ``Circle-circle continuity axiom'' as ``A circumference passing through a point inside a circle and a point outside it intersects the circumference of the circle at exactly two points.'' (Pascal, 1655) or ``If two circumferences both have some points inside one another, they will intersect one another at exactly two points.'' (Pascal, 1655). Among forms of the parallel postulate one finds the so-called ``Principle of Aristotle'' (which was used by Proclus in his commentary on the \textit{Elements} (\textit{In Euclidis}, 371) to prove the parallel postulate), ``If two straight lines forming an angle are infinitely produced from a point, their distance will exceed any finite magnitude.'' (Commandino, 1572), and the one later adopted by Tarski as the parallel postulate for his axiom system, ``A straight line passing through a point that is inside an angle cuts at least one of the sides of the angle.'' (Lorenz, 1773) On page 617, the author meant to say (as communicated to the reviewer) not that ``the acceptance of Aristotle's principle is in fact enough to grant the full axiom of Archimedes (AA1) for geometrical magnitudes'' but rather that ``the acceptance of Aristotle's principle is in fact enough to grant the fact that the sum of the angles in any triangle does not exceed two right angles.''