The Tonelli-Shanks algorithm is a well-known procedure for computing square roots in modular arithmetic, meaning it finds a number whose square equals a given value when you divide by a prime number and look at the remainder. The algorithm works by breaking the prime minus one into two parts: its largest odd divisor and a power of two. It then computes what this paper calls the "Tonelli-Shanks power," an intermediate quantity built from the largest odd divisor, and multiplies it by a correction term to get the final square root. This process is fundamental in computational number theory and cryptography.
The paper works within a mathematical framework called the ring of integers modulo infinitely large primes, sometimes written as a nonstandard arithmetic setting. This is a way of doing number theory "at infinity," treating an idealized prime that is larger than any specific finite prime. The central question is whether certain quantities arising from the Tonelli-Shanks algorithm, when viewed in this infinite setting, are algebraic or transcendental over the rational numbers. A number is transcendental if it cannot be a root of any polynomial with rational coefficients, the way pi and e are transcendental in ordinary analysis.
The main result is that several key quantities in this setting turn out to be transcendental. Specifically, the largest odd divisor of the prime minus one, the largest power of two dividing the prime minus one, the Tonelli-Shanks power for any integer other than negative one, zero, or one, and the correction term in the algorithm are all transcendental when lifted into this nonstandard ring. This is surprising because these quantities come from purely arithmetic, combinatorial operations on integers, yet they escape the world of polynomial equations over the rationals. The result gives a deeper structural picture of how classical algorithms behave in nonstandard or asymptotic number-theoretic settings.