The inclination among mathematicians to view numbers and mathematical structures as objective, external features of reality—a philosophical position broadly known as Mathematical Platonism—stems from several distinct aspects of mathematical practice and scientific inquiry.

1. The Phenomenological Feeling of Discovery

Working mathematicians often experience their daily work not as creating something out of nothing (like a novelist crafting a story), but as exploring an existing, external terrain.

• Resistance to Will: A mathematician cannot simply "decide" that a new prime number exists or alter the value of $\pi$. Mathematical objects exhibit rigid, stubborn properties that resist human preference.

• Unexpected Connections: Purely abstract concepts developed for internal mathematical curiosity frequently turn out to have unexpected, deep links to completely unrelated subfields decades later. This inter-connectedness suggests an underlying objective structure rather than an arbitrary collection of human rules.

2. The "Unreasonable Effectiveness" in Physical Sciences

Physicist Eugene Wigner famously coined the phrase "the unreasonable effectiveness of mathematics in the natural sciences."

• Predictive Power: Abstract mathematical frameworks created without any empirical goal routinely end up perfectly describing physical phenomena discovered much later. For instance, Bernhard Riemann’s non-Euclidean geometry (developed in the 1850s) became the essential mathematical framework for Albert Einstein’s theory of General Relativity in 1915.

• The Quine-Putnam Indispensability Argument: Philosophers Willard Van Orman Quine and Hilary Putnam argued that because advanced mathematics is indispensable to our best physical theories (quantum mechanics, general relativity), and because we believe physical entities like electrons or spacetime curvature exist, we ought to be equally committed to the reality of the mathematical entities required to describe them.

3. Convergence Across Independent Minds

If mathematics were a purely social or cultural construct—like language, legal frameworks, or etiquette—one would expect different civilizations to produce fundamentally incompatible mathematical systems.

• Independent human cultures (and hypothetically, intelligent alien life) arrive at identical mathematical truths. The prime factorization of $101$ or the ratio of a circle's circumference to its diameter ($\pi$) remains invariant regardless of the notation, language, or cognitive architecture of the observer.

• This convergence suggests that human notation is the "overlay," but the underlying structural relations being described exist independently.

4. Objective Truth-Values and Incompleteness

In standard discourse, mathematical statements are treated as objectively true or false, independent of human knowledge or consensus.

• Prior to the proof of Fermat’s Last Theorem by Andrew Wiles in 1994, mathematicians did not believe the theorem's truth value was suspended in limbo; they believed it was already objectively true or false, waiting to be verified.

• Gödel’s Incompleteness Theorems: Kurt Gödel demonstrated that in any consistent formal system capable of doing basic arithmetic, there are true mathematical statements that cannot be proven using the rules of that system. This implies that "mathematical truth" is a broader concept than "human provability," suggesting that mathematical facts exist independently of the formal axiomatic systems humans construct to capture them.

Key Counter-Views

While Platonism is widespread among practicing mathematicians, it is not universal. Alternative views include:

• Formalism: Mathematics is a game played with arbitrary symbols according to agreed-upon operational rules (associated with David Hilbert).

• Intuitionism/Constructivism: Mathematical objects are mental constructs, and a statement is only true if it can be explicitly constructed in the human mind (associated with L.E.J. Brouwer).

• Fictionalism/Nominalism: Mathematical statements are useful fictions, extremely effective for modeling the physical world but not literally referencing abstract objects.