Mathematicians have long known that the classical gamma function, which extends the factorial to non-integer inputs, is uniquely pinned down by a theorem called Bohr-Mollerup: among all smooth solutions to the basic factorial recurrence, only one is convex in the right way. Around 2000, Manjul Bhargava introduced a sweeping generalization of the factorial to arbitrary subsets of integers, replacing the usual set of whole numbers with other collections, such as the set of prime numbers. This paper asks: can we do for the "prime factorial" what Bohr-Mollerup did for the ordinary factorial? That is, can we find a canonical, well-behaved function that interpolates Bhargava's prime-based factorial to all complex numbers?
The authors construct such a function, called the prime cyclotomic gamma function, using several technical tools including what they call factorial calculi, filtered prime-layer completions, and cyclotomic quadrature. The resulting function is entire (meaning it has no singularities anywhere in the complex plane), never equals zero, and satisfies a clean reflection symmetry: the product of its values at any point z and at 1 minus z always equals 1. They also derive a precise asymptotic formula for how the prime factorial grows, analogous to Stirling's famous approximation for the ordinary factorial. This growth formula involves a specific constant built from the prime numbers in a natural way.
Beyond these structural results, the paper addresses several open questions Bhargava posed in his original work, specifically questions 31 and 33 about how to normalize and understand these generalized factorials. The authors also study the statistical behavior of the error term in their Stirling-type formula, proving that assuming a plausible technical estimate about character sums over primes, the squared error terms accumulate at a rate governed by a recognizable value of the Riemann zeta function. They further show that certain classical integral representations familiar from the ordinary gamma function do not carry over to this prime setting, motivating a conjectured replacement.