Experiment #37 · Thought experiment

Swampman

Does causal history matter for mental content?

Donald Davidson · 1987 · Philosophy of mind, content

First published: D. Davidson, "Knowing One's Own Mind", *Proc. APA* 60 (1987): 441–458.

A lightning strike in a swamp produces a perfect molecule-for-molecule duplicate of Davidson, just as Davidson himself is incinerated nearby. Does Swampman have thoughts?

Davidson walks through a swamp; lightning strikes both him and a nearby tree. By coincidence, the molecules of the tree are rearranged into a perfect duplicate of Davidson, who walks off and lives Davidson's life. Davidson's claim: Swampman has no thoughts, no language, no genuine mental content — because content requires causal-historical connections to the world that Swampman's newborn structure cannot have. The case is the sharpest test of teleosemantic and externalist theories of mental content: if content requires history, then a duplicate without that history is mind-empty despite identical behaviour. Internalists demur.

Formulation

Stipulate: subject S₁ (original Davidson) destroyed at t; subject S₂ (Swampman) created at t, atom-for-atom identical. S₂ walks off, indistinguishable in behaviour and inner experience. Question: does S₂ have thoughts with content?

Compare this experiment

Dimensions Engaged

◎

Observer

Bears on Observer · Knowledge Extent: is what one knows fixed only by current state, or by the history that produced the state?

⧉

Information

A pressure-test for Information · Ontological Status: is informational content intrinsic or relational?

⧖

Time

Bears on Time · Direction: does the past constitutively enter present mental life, or is the past merely the route by which present states arrived?

Responses — How Schools Engage

Denies / rejects the premise 2

Phenomenology

Swampman has experiences from the moment he steps out of the swamp — intentional, embodied, lived. To deny them on grounds of missing history is to confuse a phenomenological fact with a metaphysical condition.

Structuralism

Content supervenes on structural-functional organisation; Swampman has the same structure, hence the same content. History is causally relevant but not constitutive.

Reframes the question 2

Naturalism

Many naturalists side with the externalist conclusion: content is grounded in evolutionary-historical function. Without history, Swampman has structures that look like contentful states but are not.

Pragmatism

Whether Swampman "has content" depends on how the community treats him — whether his utterances earn a place in the linguistic practice. The metaphysical question is less determinate than the practical one.

Holds it inconclusive 1

Analytic Metaphysics / Logical Atomism

A live battleground: teleosemanticists (Millikan, Papineau) follow Davidson; internalists (Fodor, Searle) hold Swampman has thoughts. The case reveals which theory of content one is implicitly using.