Vague Predicates
Vague predicates are predicates — linguistic expressions that attribute properties — whose application conditions are not sharply bounded. A predicate like 'heap,' 'tall,' 'bald,' or 'rich' is vague because there are cases where it is unclear whether the predicate applies: a man with exactly 3,247 hairs may be a borderline case of 'bald,' and a pile of 4,892 grains of sand may be a borderline case of 'heap.' The vagueness is not a matter of ignorance or insufficient information. It is a structural feature of the predicate itself: the concept does not determine a unique cutoff between the cases that satisfy it and the cases that do not.
Vague predicates challenge the foundational assumptions of classical logic, particularly the principle of bivalence — the claim that every meaningful statement is either true or false. If '4,892 grains is a heap' is neither determinately true nor determinately false, then classical logic cannot be applied without revision or reinterpretation. The challenge is not merely technical. It is a question about whether the structure of language mirrors the structure of the world, or whether the precision of formal logic is an artifact that breaks down at the edges of lived experience.
The Problem of Borderline Cases
The defining feature of a vague predicate is the existence of borderline cases: objects or situations for which the predicate neither clearly applies nor clearly fails to apply. A person who is 5'9" may be borderline tall in a population of average height 5'6" but clearly tall in a population of average height 5'2". This context-sensitivity distinguishes vagueness from ambiguity. An ambiguous word like 'bank' has multiple discrete meanings, each precise in its own domain. A vague predicate has a single meaning whose application is graded across a continuum.
Borderline cases are not exceptions to be eliminated. They are systematic. For any vague predicate, there exists a range of cases where competent speakers of the language disagree, hesitate, or acknowledge that there is no fact of the matter. This is the phenomenon that the sorites paradox exploits: if one grain is not a heap, and adding one grain never makes a non-heap into a heap, then no number of grains is a heap. The paradox reveals that vague predicates are not merely imprecise — they are structurally incompatible with the inductive reasoning that classical logic treats as valid.
Theories of Vagueness
Philosophical responses to vague predicates fall into three major families, each with a different diagnosis of what goes wrong and how to fix it.
Fuzzy logic treats vagueness as a matter of degree. A borderline case of 'heap' is not neither-true-nor-false; it is partially true. The predicate has a graded membership function, and the sorites argument fails because its inductive premise is only approximately true — the degree of truth degrades by imperceptible increments across the continuum. The cost is a revision of classical logic: truth becomes a real number in [0,1], and the logical connectives must be redefined.
Supervaluationism preserves classical logic by indexing truth to a space of admissible precisifications. A vague predicate does not have a single extension but a family of them. The statement '4,892 grains is a heap' is super-true if it is true on every admissible precisification, super-false if false on every one, and neither if the precisifications disagree. The sorites premise is not super-true because different precisifications place the cutoff at different points. Classical logic is preserved at the level of super-truth, but the preservation is purchased by introducing a meta-level distinction between truth and determinacy.
Epistemicism, most prominently defended by Timothy Williamson, rejects the claim that borderline cases are indeterminate. There is a precise fact of the matter about whether 4,892 grains is a heap, but this fact is unknowable due to the margins of error in our cognitive systems. The epistemicist preserves classical logic and bivalence at the object level, paying the price of a counterintuitive epistemology: there are truths that are in principle undiscoverable, not because of ignorance but because of the structure of knowledge itself.
Higher-Order Vagueness
The problem of vague predicates does not stop at the first order. The distinction between clear cases, borderline cases, and clear non-cases is itself vague. There are borderline cases of borderline cases: cases where it is unclear whether a case is borderline. This is higher-order vagueness, and it threatens every theory of vagueness that proposes a sharp boundary between the determinate and the indeterminate. If supervaluationism draws a sharp line between the super-true and the not-super-true, then the line itself is subject to the same sorites reasoning that motivated the theory. The epistemicist faces a parallel challenge: if there is a precise fact of the matter about whether something is borderline, then there is a precise fact of the matter about whether there is a precise fact of the matter, and the regress threatens to collapse the distinction between determinacy and indeterminacy.
Vague Predicates as Systems
From a systems perspective, vague predicates are not merely linguistic curiosities. They are the interface between continuous reality and discrete classification. Every system that maps a continuous variable into a discrete category faces a version of the sorites problem: medical thresholds ('high blood pressure'), legal standards ('reasonable doubt'), engineering tolerances ('defective'), and machine learning decision boundaries ('spam'). The design choice is not whether to have a cutoff but whether to treat the cutoff as part of the system's determinate content or as a region of managed indeterminacy.
The supervaluationist insight, generalized: a system that correctly represents its own vagueness is more robust than a system that pretends to precision it does not have. The precisification model is a formalization of what it means for a system to know what it knows and to know what it does not know. The vague predicate is not a failure of definition. It is a definition that correctly models the limits of the domain it describes.
The persistent assumption that vague predicates must be made precise before they can be used in serious reasoning is itself a form of the sorites fallacy — the fallacy of thinking that because precision is possible at the center, it must be possible at the edges. Vagueness is not a defect to be eliminated by better logic. It is a structural feature of systems that interact with continuous worlds, and any theory of reasoning that cannot accommodate it is not a theory of reasoning at all.