Announcer
The following program features simulated voices generated for educational and philosophical exploration.
Leonard Jones
Good afternoon. I'm Leonard Jones.
Jessica Moss
And I'm Jessica Moss. Welcome to Simulectics Radio.
Leonard Jones
Yesterday we discussed the epistemology of disagreement and how to respond when equally informed peers reach opposite conclusions. Today we turn to a puzzle that has troubled philosophers since antiquity—the sorites paradox, which concerns vague predicates and the absence of sharp boundaries. How can removing a single grain from a heap make it not a heap?
Jessica Moss
This seems like a trivial puzzle about language, but it raises profound questions about the structure of reality, the limits of thought, and whether the world contains genuine indeterminacy.
Leonard Jones
Our guest is Dr. Timothy Williamson, Wykeham Professor of Logic at Oxford University, whose work on vagueness, knowledge, and the philosophy of language has fundamentally shaped contemporary analytic philosophy. Welcome, Professor Williamson.
Dr. Timothy Williamson
Thank you. The sorites paradox presents a genuine puzzle about how vague language relates to precise reality, and different solutions have radically different implications for metaphysics and epistemology.
Jessica Moss
Let's start with the basic paradox. A collection of ten thousand grains of sand is a heap. Removing one grain doesn't make it not a heap. But if we keep applying this principle, we eventually conclude that a single grain is a heap, which is absurd.
Dr. Timothy Williamson
Right. The paradox arises from three seemingly compelling premises: first, that ten thousand grains constitute a heap; second, that if n grains constitute a heap, then n-1 grains also constitute a heap; and third, that one grain doesn't constitute a heap. These premises are jointly inconsistent, so at least one must be rejected.
Leonard Jones
Let me be precise about the logical structure. The second premise is a universal generalization over a tolerance principle—that small changes don't affect whether the predicate applies. But iterated application of this principle leads to contradiction. So which premise should we reject?
Dr. Timothy Williamson
Most philosophers reject the second premise—the tolerance principle. They accept that there's some precise number n such that n grains constitute a heap but n-1 grains don't. The question is what to say about this boundary.
Jessica Moss
But this seems deeply counterintuitive. How could there be a precise boundary when the concept 'heap' seems inherently vague? What's the difference between 5,347 grains and 5,346 grains?
Dr. Timothy Williamson
That's the epistemicist response I defend. There is a precise boundary, but we can't know where it is. Vagueness is fundamentally epistemic—it concerns our ignorance of the precise facts, not indeterminacy in reality itself.
Leonard Jones
This requires unpacking carefully. You're claiming that 'heap' has a precise extension—there's a fact about which collections count as heaps—but this fact is unknowable. What explains our inability to know the boundary?
Dr. Timothy Williamson
The explanation involves a margin for error principle. To know that n grains constitute a heap, you'd need to be in a position to discriminate this case from nearby cases where n-1 or n+1 grains are present. But vague predicates are precisely those where we can't make such fine discriminations. The inexact nature of our evidence makes knowledge of the boundary impossible.
Jessica Moss
But where does this precise boundary come from? It seems mysterious that 'heap' could have a determinate extension when speakers have no precise intentions about borderline cases.
Dr. Timothy Williamson
The boundary emerges from the complex interaction of our usage practices, contextual factors, and semantic constraints. Just as the meaning of 'water' turned out to be Hâ‚‚O even though speakers didn't intend to refer specifically to that chemical structure, the meaning of 'heap' can have precise extension even though we can't specify it.
Leonard Jones
This analogy seems problematic. With 'water,' empirical investigation revealed the underlying nature. But what investigation could reveal whether 5,347 grains constitute a heap? The boundary seems arbitrary rather than discoverable.
Dr. Timothy Williamson
The boundary isn't arbitrary—it's determined by the totality of our usage and the semantic constraints governing vague predicates. But you're right that it's not empirically discoverable in the way chemical composition is. The unknowability is more fundamental—it flows from the structure of vagueness itself.
Jessica Moss
What are the alternatives to epistemicism? How else might we respond to the sorites?
Dr. Timothy Williamson
The main alternatives are semantic approaches that posit genuine indeterminacy. Supervaluationism says that borderline cases are those where different precisifications of the vague predicate give different verdicts. Degree theories say vague predicates come in degrees—something can be a heap to degree 0.7.
Leonard Jones
Let's examine supervaluationism more carefully. On this view, 'heap' has multiple admissible precisifications—different ways of drawing a precise boundary that are all compatible with our usage. A statement is true if true on all precisifications, false if false on all, and neither true nor false if true on some and false on others.
Dr. Timothy Williamson
Right. So for borderline cases—collections where some precisifications classify them as heaps and others don't—supervaluationism says 'that's a heap' is neither true nor false. This respects the intuition that there's no fact of the matter about borderline cases.
Jessica Moss
But doesn't this create logical problems? Classical logic assumes every statement is either true or false. If we abandon bivalence, don't we need to revise logic itself?
Dr. Timothy Williamson
Supervaluationists claim to preserve classical logic by defining validity in terms of truth preservation across all precisifications. But I think this is unstable. The logic they defend isn't really classical because it validates inferences that classical logic doesn't—for example, from 'either n grains is a heap or it isn't' we can't infer that one of the disjuncts is true.
Leonard Jones
There's also a question about what grounds the admissible precisifications. Why are certain ways of drawing boundaries legitimate while others aren't?
Dr. Timothy Williamson
Exactly. Supervaluationists need some account of what makes a precisification admissible, and the usual answer is that admissible precisifications preserve our usage patterns and respect the penumbral connections between vague predicates. But this requires substantive semantic theory that may be no less mysterious than epistemicism.
Jessica Moss
What about degree theories? These seem more intuitive—we naturally talk about things being 'more or less' heaps, 'somewhat' bald, 'fairly' tall.
Dr. Timothy Williamson
Degree theories face their own sorites. If 'heap' comes in degrees, presumably removing one grain reduces the degree by some tiny amount. But what prevents us from iterating this to reach absurdity? We need a tolerance principle for degrees, which recreates the original problem.
Leonard Jones
Let me make sure I understand this objection. If ten thousand grains has heap-degree 1.0 and one grain has heap-degree 0.0, then by continuity there must be some number where heap-degree drops discontinuously, or the degree function must be everywhere discontinuous in a way that seems unprincipled.
Dr. Timothy Williamson
Precisely. Degree theorists might accept discontinuity, but then we're back to the epistemicist picture—precise boundaries that we can't know. The degrees don't eliminate sharp cutoffs; they just relocate them.
Jessica Moss
Does epistemicism extend to all forms of vagueness? What about semantic indecision cases—situations where speakers genuinely haven't decided how to apply a term?
Dr. Timothy Williamson
I distinguish between vagueness proper and semantic indecision. If we simply haven't decided whether to call something a planet, that's different from borderline cases of established vague predicates like 'heap' or 'bald.' But even semantic indecision creates precise extensions determined by our partial decisions and semantic constraints.
Leonard Jones
What about higher-order vagueness—vagueness about where the boundary between determinate cases and borderline cases lies? If I can't know which number marks the heap boundary, can I know which numbers mark the boundaries of the borderline region?
Dr. Timothy Williamson
Higher-order vagueness is a serious challenge for non-epistemic theories. Supervaluationism struggles to give a satisfactory account of borderline borderline cases. Epistemicism handles it straightforwardly—there are precise boundaries for the borderline region, but we can't know them either. The margin for error extends to our knowledge of our own ignorance.
Jessica Moss
Let's consider the metaphysical implications. Does vagueness exist in the world, or only in our language and thought?
Dr. Timothy Williamson
I defend the view that vagueness is exclusively linguistic and epistemic. The world consists of precise facts, and vagueness arises from our inability to know or specify these facts exactly. This is a controversial position—many philosophers believe in ontological vagueness, indeterminacy in reality itself.
Leonard Jones
The quantum measurement problem might provide examples of genuine worldly indeterminacy. Before measurement, quantum systems seem to exist in indefinite states. Could this be metaphysical vagueness?
Dr. Timothy Williamson
That's an interesting case. I'm skeptical that quantum indeterminacy is vagueness in the relevant sense. Quantum superposition might involve determinate facts about wavefunctions rather than indeterminacy about classical properties. But if there is genuine metaphysical indeterminacy, it would challenge epistemicism's assumption that the world is fully precise.
Jessica Moss
What about vagueness in ethics and aesthetics? When we disagree about whether an action is morally permissible or whether a painting is beautiful, is this ordinary vagueness or something different?
Dr. Timothy Williamson
Ethical and aesthetic predicates are vague in similar ways to ordinary descriptive predicates. There are borderline cases of moral permissibility and beauty. Whether the underlying facts are objective or response-dependent is a separate question from whether the predicates are vague. Epistemicism can accommodate either metaethical picture.
Leonard Jones
But moral disagreement seems different from disagreement about heaps. When people disagree about abortion or euthanasia, they're not simply confused about where a precise boundary lies. They have substantive disagreements about values and principles.
Dr. Timothy Williamson
Agreed. Not all moral disagreement involves vagueness. Some involves genuine normative conflict about what matters morally. But there are also borderline cases where the question isn't which principles are correct but whether a particular case falls under a vague moral category.
Jessica Moss
How should we handle vagueness in legal contexts? Laws are often vague—what counts as 'reasonable care' or 'excessive force'? Should legislators try to eliminate vagueness through precise definitions?
Dr. Timothy Williamson
Some vagueness in law is unavoidable and perhaps desirable. Legislators can't foresee all cases, and excessive precision might lead to perverse outcomes. But vagueness creates problems for rule of law—if citizens can't know whether their conduct violates the law, this undermines legal guidance and fair notice.
Leonard Jones
There's a parallel to epistemicism here. Even if the law has a precise extension, citizens might be unable to know exactly where the boundary lies. This creates a margin for error in legal reasoning similar to the margin for error in heap cases.
Dr. Timothy Williamson
That's right. Legal vagueness and semantic vagueness share this epistemic structure. Judges must make decisions in borderline cases, effectively precisifying vague legal terms through their rulings, even though the underlying legal facts may already be precise.
Jessica Moss
What about the phenomenology of vagueness? When I consider whether 5,000 grains constitute a heap, I don't feel like I'm ignorant of a precise fact. I feel like there's no fact of the matter. Can epistemicism accommodate this phenomenology?
Dr. Timothy Williamson
The phenomenology of vagueness is underdetermining. We might feel there's no fact of the matter, but this could simply reflect our inability to access the fact rather than its absence. Consider color perception—there's a precise wavelength where red transitions to orange, but we can't perceive it as a sharp boundary. The phenomenology reflects our epistemic situation, not metaphysical indeterminacy.
Leonard Jones
Let's return to the tolerance principle. Some philosophers argue that vague predicates are precisely those that satisfy tolerance—that competent speakers are disposed to accept small differences as irrelevant. Does this usage-based account threaten epistemicism?
Dr. Timothy Williamson
I think the tolerance principle is false, but explaining why we find it compelling is important. We're disposed to accept it in individual cases because we can't discriminate small differences. But we shouldn't be disposed to accept its universal generalization, which leads to contradiction. The appearance of tolerance reflects our epistemic limitations, not semantic facts.
Jessica Moss
How does epistemicism relate to contextualism about knowledge? Some philosophers argue that whether someone knows depends on context—perhaps whether 'heap' applies also depends on context?
Dr. Timothy Williamson
There's definitely context-sensitivity for vague terms. What counts as a heap might vary between different contexts—perhaps we need more grains in contexts where precision matters. But this context-sensitivity is compatible with epistemicism. In any given context, 'heap' has a precise extension that we can't know.
Leonard Jones
What about the relationship between vagueness and the problem of the many? When I point at a cloud and say 'that cloud,' there are many cloud-shaped regions of water vapor with slightly different boundaries. Which one am I referring to?
Dr. Timothy Williamson
The problem of the many is closely related to vagueness. Epistemicism says there's a precise cloud I'm referring to, even though I can't specify its exact boundaries. The alternatives—semantic indeterminacy or multiplicity of reference—face difficulties explaining how reference succeeds despite this apparent underdetermination.
Jessica Moss
As we approach the end of our time, what do you see as the most important open questions about vagueness?
Dr. Timothy Williamson
I think the key questions concern the relationship between vagueness and broader issues in metaphysics and epistemology. Does quantum mechanics provide cases of genuine metaphysical vagueness? Can we develop a fully adequate epistemology of vague propositions that explains patterns of knowledge and ignorance? How does vagueness interact with other forms of semantic context-sensitivity?
Leonard Jones
And presumably the debate between epistemic and semantic approaches will continue. Neither position is fully satisfactory—both face serious objections.
Dr. Timothy Williamson
I think epistemicism is the least unsatisfactory position, but you're right that challenges remain. The key is to develop the view in ways that address objections while maintaining its explanatory advantages over rival theories.
Jessica Moss
Professor Williamson, thank you for this careful examination of vagueness and its philosophical implications.
Dr. Timothy Williamson
Thank you. These questions about the relationship between language, thought, and reality remain central to philosophy.
Leonard Jones
We'll return tomorrow with more philosophical inquiry.
Jessica Moss
Good afternoon.