A new approach to the Dyson rank conjectures (Q6098228): Difference between revisions

A partition \(\pi\) of a positive integer \(n\) is a finite weakly decreasing sequence of positive integers \(\pi_1\geq\pi_2\geq\cdots\geq\pi_r\) such that \(\sum_{i=1}^r\pi_i=n\). The \(\pi_i\) are called the parts of partition \(\pi\). Let \(p(n)\) denote the number of partitions of \(n\) with the convention that \(p(0)=1\). \textit{S. Ramanujan} [Cambr. Phil. Soc. Proc. 19, 207--210 (1919; JFM 47.0885.01)] discovered and later proved the following three beautiful congruences for the partition function \(p(n)\), namely, \begin{align*} p(5n+4) &\equiv0\pmod{5},\tag{RAMA-cong-1}\\ p(7n+5) &\equiv0\pmod{7},\tag{RAMA-cong-2}\\ p(11n+6) &\equiv0\pmod{11}.\tag{RAMA-cong-3} \end{align*} \textit{F. Dyson} [Eureka, 8, 10--15 (1944)] went on to remark that although there at least four different proofs of (RAMA-cong-1) and (RAMA-cong-2), it would be more satisfying to have a direct proof of (RAMA-cong-1). For this purpose, he supposed whether there was some natural way of dividing the partitions of \(5n+4\) into five equally numerous classes. He defined the rank of a partition as the largest part minus the number of parts, and conjectured that the residue of the rank modulo \(5\) does the job of dividing the partitions of \(5n+4\) into five equal classes. He also conjectured that the residue of the rank modulo \(7\) in a similar way divides the partitions of \(7n+5\) into seven equal classes thus explaining (RAMA-cong-2). More specifically, let \(N(m,n)\) denote the number of partitions of \(n\) with rank \(m\), and let \(N(m,t,n)\) denote the number of partitions of \(n\) with rank congruent to \(m\) modulo \(t\). Dyson conjectured that for any \(n\geq0\), \begin{align*} N(0,5,5n+4)=N(1,5,5n+4)=\cdots=N(4,5,5n+4) &=\dfrac{1}{5}p(5n+4),\tag{Dyson-conj-1}\\ N(0,7,7n+5)=N(1,7,7n+5)=\cdots=N(6,7,7n+5) &=\dfrac{1}{7}p(7n+5).\tag{Dyson-conj-2} \end{align*} The corresponding conjecture with modulus \(11\) is definitely false. The Dyson rank conjectures (Dyson-conj-1) and (Dyson-conj-2) were first proved by \textit{A. O. L. Atkin} and \textit{P. Swinnerton-Dyer} [Proc. Lond. Math. Soc. (3) 4, 84--106 (1954; Zbl 0055.03805)] in 1954. Atkin and Swinnerton-Dyer's proof involved proving many theta-function and generalized Lambert (or Appell-Lerch) series identities using basically elliptic function techniques. Their method also involved finding identities for the generating functions of \(N(k,5,5n+r)\) for all the residues \(r=0,1,2,3,4\). In 2017, \textit{D. Hickerson} and \textit{E. Mortenson} [Math. Ann. 367, No. 1-2, 373--395 (2017; Zbl 1367.11077)] used the theory of Appell-Lerch series to obtain results for the Dyson rank function including the Dyson rank conjectures. In 2019, the author [Trans. Am. Math. Soc. 371, No. 1, 199--248 (2019; Zbl 1469.11404)] how the theory of harmonic Maass forms can be utilized to prove the Dyson rank conjectures and much more. In this paper under review, we describe a new and more elementary method for proving Dyson's rank conjectures. The main ingredients in the proofs are some identities for Hecke-Rogers series. Based on these identities, the author obtained (Dyson-conj-2) for partitions of \(5n+4\) without having to deal with the partitions of \(5n+r\) for the other residues \(r=0,1,2,3\).