The paper wrestles with a fundamental puzzle arising from Gödel's incompleteness theorems. Those theorems show that for any sufficiently powerful mathematical system like Peano Arithmetic (the standard axioms for whole numbers), there exist statements that can neither be proved nor disproved within that system. This raises a philosophical question: if you can consistently add either a statement or its negation to your axioms, is there any principled reason to prefer one extension over the other? Some philosophers, called arithmetical pluralists, say no, both extensions are equally legitimate. The author pushes back against this view.
The author's central proposal is what he calls "graded epistemic arithmetical non-pluralism." Rather than a simple yes-or-no stance, he argues that our justification for choosing between competing extensions of arithmetic comes in degrees, and those degrees track how deeply the statements in question connect to the broader landscape of set theory and large cardinal axioms. Large cardinals are extremely powerful infinity-related assumptions that mathematicians use to settle questions that ordinary axioms cannot resolve. The author defends, for example, our right to prefer the system that includes the consistency of Peano Arithmetic itself, and he argues that a rival non-pluralist position proposed by philosopher Peter Koellner fails to handle certain concretely meaningful unprovable statements discovered by mathematician Harvey Friedman.
The upshot is that deciding which arithmetical extensions are more justified is not merely a local, number-theory-internal matter. It pulls us into deep questions about the justification of large cardinal axioms and what is sometimes called Gödel's programme, the broader project of finding and justifying new axioms powerful enough to settle otherwise undecidable questions. The paper thus connects a seemingly narrow debate about arithmetic to some of the most ambitious and contested foundational questions in mathematics.