The paper revisits a classic result from 1978 by the mathematician Yu. F. Borisov, who showed that if you start from a small set of reasonable physical assumptions plus Einstein's relativity principle, only two consistent pictures of motion emerge: the relativistic one (where transformations between observers follow the rules of special relativity) and the classical Newtonian one (where they follow Galilean relativity). The authors recast Borisov's argument in a rigorous logical framework and then ask a subtle question: what happens if you do not assume that physical quantities behave like ordinary real numbers, but instead allow them to come from a more exotic mathematical structure called a non-Archimedean ordered field? Such fields contain numbers that are infinitely large or infinitely small compared to ordinary numbers, which is unusual but mathematically legitimate.
When they make this change, the clean two-option result breaks down in interesting ways. The authors show that over non-Archimedean fields, there is a third type of consistent model in which the transformations between observers are Euclidean rotations and reflections rather than anything recognizably physical. More strikingly, instead of just a handful of models, you get infinitely many, arranged in an endless descending chain, something that simply cannot happen when quantities are ordinary real numbers. This richness comes directly from the strange arithmetic properties of non-Archimedean fields.
The practical upshot is about what it takes to rule out the unwanted, physically meaningless Euclidean models. Over real numbers, two simple and natural assumptions do the job: that time flows in one direction and that observers cannot be in two places at once. Over non-Archimedean fields, however, these same assumptions are not strong enough, and the Euclidean models survive no matter what. This suggests that the familiar uniqueness of special and classical relativity as the only options is more fragile than it looks, depending quietly on deep properties of the number system underlying the physics.