View source for Predicativism

'''Predicativism''' is a philosophy of mathematics that rejects '''impredicative definitions''' — definitions that quantify over a totality which includes the object being defined. An impredicative definition is one that, in order to specify an object, refers to a collection of which that object is a member. The Predicativist holds that such definitions are illegitimate because they involve a kind of circularity: the object is defined in terms of a whole that presupposes the object itself. This is not merely a technical restriction. It is a principled rejection of self-referential construction in the foundations of mathematics.

The classic example is the definition of the least upper bound of a set of real numbers. To define the least upper bound of a set ''S'', one quantifies over all upper bounds of ''S'' — a collection that includes the least upper bound itself. For the Predicativist, this is viciously circular. The bound is defined by reference to a totality that could not be specified without already having the bound. The classical mathematician accepts this as harmless. The Predicativist sees it as a structural error: a system that defines itself by reference to itself has not genuinely defined anything.

== Historical Origins ==

Predicativism emerged from the same foundational crisis that produced [[Intuitionism]] and the revisions of set theory. [[Henri Poincaré]] was the first to identify the problem clearly, in 1906. He argued that impredicative definitions involve a vicious circle because they rely on a set that could not exist prior to the definition of its members. [[Bertrand Russell]] adopted Poincaré's diagnosis and made it the basis of his '''ramified theory of types''', which stratified the universe of sets into a hierarchy of levels. A set at level ''n'' can only be defined by quantifying over sets at levels below ''n'', never over sets at its own level or above.

Russell's ramified type theory was a direct response to the paradoxes that had destroyed naive set theory. The idea was simple: if self-reference is the source of the paradoxes, then eliminate self-reference by stratification. The price was high. The ramified theory made it impossible to prove even elementary theorems of classical mathematics without adding ad hoc '''axioms of reducibility''', which undermined the very point of the stratification. The Predicativist program, in its most rigorous form, seemed to require either unacceptable restrictions on mathematics or principles that smuggled the impredicativity back in.

== Predicative Mathematics ==

Despite these difficulties, a significant body of '''predicative mathematics''' has been developed. In the 1960s, [[Solomon Feferman]] and [[Georg Kreisel]] showed that a large portion of classical analysis — the theory of real numbers, continuous functions, and much of calculus — can be developed on a purely predicative foundation. The cutoff is surprisingly high: predicative mathematics can prove everything that can be proved in a system called '''''ACA''₀''', which is far stronger than primitive recursive arithmetic but weaker than full second-order arithmetic. This means that the mathematics needed for most scientific applications is predicatively acceptable.

What predicative mathematics cannot do is prove the existence of certain sets whose classical definitions are impredicative. The least upper bound theorem for arbitrary sets of reals fails in its classical form. The well-ordering of the real numbers cannot be proved. And certain constructions in descriptive set theory and measure theory are unavailable. The Predicativist accepts these losses as the price of coherence. The classical mathematician sees them as unnecessary restrictions on a perfectly sound practice.

== Systems-Theoretic Interpretation ==

From a systems-theoretic perspective, Predicativism is a theory about the '''bootstrap problem''' in self-referential systems. A system that defines its own components by reference to the whole system has not genuinely bootstrapped itself. It has presupposed what it was supposed to construct. The Predicativist's stratification is a requirement that the system build itself level by level, each level depending only on levels already constructed. This is exactly the principle of '''bottom-up construction''' in complex systems: a system must be built from components that do not presuppose the system they are part of.

This connects Predicativism to [[Autopoiesis|autopoiesis]], the theory of self-producing systems in biology. An autopoietic system produces its own components, but it does not produce them from nothing. It produces them from pre-existing components in a continuous process. The system's organization is self-maintaining, but it is not self-creating in the sense of defining itself into existence from a void. The Predicativist's complaint about impredicative definitions is analogous: a system cannot define its own organizing principle by reference to a totality that includes the principle itself. The principle must emerge from a process that does not presuppose it.

Predicativism also connects to [[computational complexity theory|computational complexity]]. A predicative definition is one that can be computed by a process with a well-defined stage structure. An impredicative definition is one that requires a global fixed-point computation — solving for an object that is defined in terms of a system that includes the object itself. Fixed-point computations are powerful but computationally expensive, and they are not always guaranteed to converge. The Predicativist's restriction is a complexity constraint: only those objects that can be constructed by a finite, staged process are mathematically legitimate.

_The Predicativist does not reject the infinite. The Predicativist rejects the pretense that the infinite can be surveyed from a standpoint that includes itself. The classical mathematician treats the set of all sets as a completed totality and then defines subsets of it by reference to the whole. The Predicativist sees this as the systems-theoretic equivalent of a snake eating its own tail — not a paradox, but a failure of construction. A system that defines itself by reference to itself has not defined itself. It has declared itself. And declaration is not construction._

[[Category:Mathematics]]
[[Category:Philosophy]]
[[Category:Foundations]]