The \textit{casus irreducibilis} in Cardano's \textit{Ars magna} and \textit{De regula aliza} (Q2347246)

In his \textit{Ars magna}, Cardano provided a well-structured procedure for solving cubic equations. He classified all ``interesting'' cases into thirteen families of cubic equations that cannot be solved in elementary ways (that is, by directly extracting the cubic roots or by reducing the order of the equation). However, Cardano did not succeed in grounding the whole construction. The reasons for this are discussed in detail in the paper which is primarily concerned with the analysis of the solution methods for the families of cubic algebraic equations, \( x^{3}+a_{1}x=a_{0} \) and \( x^{3}=a_{1}x+a_{0}, \) known as depressed equations lacking a second degree term. There is a significant difference between the solution methods for the above two families since in the case of the first equation the cubic formula never contains square roots of negative numbers. A particular attention in the paper is paid to the so-called \textit{casus irreducibilis}, the case where square roots of negative numbers occur in the solutions; this can happen for the second equation. The author demonstrates how the difficulty entailed by the \textit{casus irreducibilis} seemingly in only one family of equations in Cardano's text is in reality pertinent to most families of equations. Although Cardano suggested several temporary ``patches'' for dealing with this problem in a number of particular cases, he did not resolve this issue fundamental for finding general solution techniques. \textit{De regula aliza} is referred to as the treatise that should have permanently corrected the problem. Due to its many inconsistencies, it is not as well known as the \textit{Ars magna}. Nowadays, we know that imaginary numbers could not be avoided in the solution formula when a cubic equation has three different solutions. This, unfortunately, was not known to Cardano who did not complete the analysis of the problem. Nevertheless, the author concludes the paper by arguing that \textit{De regula aliza} deserves to be better known as the work demonstrating the ways Cardano used to deal with equations.