When Newton met Diophantus: a study of rational-derived polynomials and their extension to quadratic fields. (Q1976804): Difference between revisions

The problem is considered of classifying all univariate polynomials defined over a domain \(k\) whose derivatives (up to a certain degree) have their roots also in \(k\). Usually, \(k\) is the field \({\mathbb Q}\) of rationals or its ring \({\mathbb Z}\) of integers. The domain \(k ={\mathbb Q}(\sqrt{m})\) of a quadratic field or its ring \({\mathcal O}\) of integers is also treated. The problem is of interest in algebraic geometry also since it is intimately related to that of finding rational points on elliptic or hyperelliptic curves. For instance, the Jacobian \(J({\mathbb Q})\) of the hyperelliptic curve \[ C: z^2 = 9w^6+195w^4+975w^2+1125 \] is isogenous to a degenerate algebraic surface, namely the product of two elliptic curves. It finally turns out to be isomorphic to the product of the Kleinian four group and two copies of \({\mathbb Z}\): \[ J({\mathbb Q}) \cong {\mathbb Z}/2{\mathbb Z} \times {\mathbb Z}/2{\mathbb Z} \times {\mathbb Z}^2. \] In particular, it has rank 2. (Additonally, the elliptic curves \[ E_m: mz^2 = w(w-6)(w+18) \] for various \(m \in {\mathbb Z}\), \(m > 0\), have rank 1 or 2 over the quadratic fields \({\mathbb Q}(\sqrt{m})\).) However, the classification problem is not completely solved. For let \(p_{(m_1,m_2,\ldots,m_r)} (x) \in {\mathbb Z}[x]\) be a polynomial of degree \(n\) with \(r\) distinct roots, where the \(i\)-th root has multiplicity \(m_i\) and \(n = m_1+m_2+ \cdots + m_r\). Then the problem is unsolved for the quartic polynomial \(p_{(1,1,1,1)}(x)\) and the quintic polynomials \(p_{(1,1,1,1,1)}(x)\), \(p_{(2,1,1,1)}(x)\) and \(p_{(3,1,1)}(x)\). But it is solved modulo the truth of two conjectures. These conjectures take care of the higher cases too. (We mention that for genus one curves Tate's Haverford College Notes [see \textit{J. H. Silverman} and \textit{J. Tate}, Rational points on elliptic curves. New York: Springer (1992; Zbl 0752.14034)] and for genus two curves \textit{J. W. S. Cassels}'s and \textit{E. V. Flynn}'s book [Prolegomena to a middlebrow arithmetic of curves of genus 2. Cambridge Univ. Press (1996; Zbl 0857.14018)] is cited.)