What Is the Thermostat Argument?

Direct Answer

The thermostat argument is not a single argument but a family of arguments that use the thermostat — a simple device that detects the ambient temperature and switches a furnace on and off to hold the temperature near a set point — as a test case for theories of mind. The thermostat is the minimal candidate for mentality: it has an internal state that tracks a feature of the world (temperature), it has a goal-like structure (maintaining a set point), and it produces behavior (turning the furnace on or off) that is appropriate to that state. The question is: does the thermostat have beliefs about temperature? Does it have intentional states? Is it, in any sense, conscious?

The argument runs in two directions. In the first direction, the thermostat is used by liberal theorists to support their views. The physicalist or functionalist argues: if a thermostat's internal state counts as a belief that it is cold (it reliably indicates cold, it plays the right causal role), then beliefs can be realized in very simple physical systems, and there is no deep mystery about mentality — it is a natural phenomenon continuous with the mechanics of thermostats. In the second direction, the thermostat is used by critics to attack such theories: if a theory implies that a thermostat has beliefs, or some degree of consciousness, the theory is absurd; the thermostat is "all dark inside," and any account of mind that cannot distinguish the thermostat from a person has failed.

The two directions mirror the two uses of the thermostat in the literature. Daniel Dennett, in Consciousness Explained and elsewhere, uses the thermostat as a friendly example: a thermostat really does "believe" it is cold in the minimal, mechanistic sense, and higher forms of mentality are built by adding layers of sophistication to thermostatic foundations. John Searle and others use the thermostat as a reductio: if a theory cannot tell the difference between a thermostat and a mind, the theory is wrong. The thermostat is thus the perfect boundary object: it sits exactly at the threshold where commonsense refuses to attribute mind and philosophical theories start to waver.

Historical Context

The thermostat entered the philosophy of mind through the mid-twentieth-century debates about the nature of mental states. Hilary Putnam, in his classic 1962 paper "Brains and Behavior," used the example of a "super-super-spartan" and the famous "pain machine" debates to probe behaviorism; the thermostat became a standard example in discussions of functionalism, where any system that instantiates the right input-output structure counts as having the corresponding mental state. The thermostat's claim to mentality was discussed by Dennett in his account of the intentional stance and the "grades" of intentional systems: a thermostat can be treated as having beliefs for predictive purposes (it "believes" the room is cold, "wants" to warm it), even though it has no consciousness.

The thermostat also features in panpsychist and "consciousness everywhere" debates. If consciousness is a fundamental feature of reality (panpsychism) or if it comes in degrees tied to causal structure (as in integrated information theory), then the thermostat may have a tiny but real degree of consciousness or proto-consciousness — a conclusion that strikes many as absurd and others as inevitable. David Chalmers, Galen Strawson, and other philosophers have used the thermostat to test the plausibility of such views: if the theory says the thermostat is conscious, is that a reason to reject the theory, or a reason to take the thermostat's experience seriously?

Key Arguments & Debates

The liberal use of the thermostat argument runs: the thermostat has internal states that track the environment, guide behavior, and stand in the right causal relations; if we refuse to call these beliefs, we must explain what is missing; and any criterion we give (flexibility, learning, consciousness) either applies to some thermostats (e.g., adaptive thermostats, learning thermostats) or reintroduces the mystery. The conservative use runs: a thermostat has no beliefs because it has no understanding — its "beliefs" are not about anything for it; it has no point of view; there is nothing it is like to be a thermostat; and a theory that implies otherwise has confused simulation with reality.

The philosophical debates the thermostat generates are three. First, the intentionality debate: what is the minimum for aboutness? Teleosemanticists argue that a state is a belief only if it has the function of indicating, fixed by history (evolution or design); a thermostat's states may fail this test, or pass it, depending on how the function is specified. Second, the consciousness debate: the thermostat is the standard test for theories that make consciousness a matter of function or information integration. If integrated information theory says the thermostat has some degree of consciousness, does that refute the theory or stretch the concept? Third, the "graded mentality" debate: is mentality all-or-nothing, or does it come in degrees, from thermostats through insects to humans? Dennett's answer is a graded, functional account; Searle's is a sharp, biological one; panpsychists make even the thermostat's states experience-like.

Contemporary Relevance

The thermostat argument is directly relevant to AI and the ethics of machines. If a thermostat can have beliefs in the minimal sense, then current AI systems — which are vastly more sophisticated — presumably have beliefs in a richer sense, and the question of what else they need for genuine mentality becomes urgent. Conversely, if the thermostat's "beliefs" are a mere façon de parler, then the same skepticism applies to claims that large language models have beliefs or understanding, however sophisticated their behavior. The thermostat thus frames the central question of AI philosophy: what is the minimum threshold for genuine mentality, and where do machines stand relative to it? The argument also structures debates about panpsychism and the hard problem: if the thermostat is not conscious, what exactly is the difference between it and a brain, and can that difference be stated in physical or functional terms? As long as these questions are open, the thermostat will remain the simplest possible probe of our theories of mind.

Related Concepts

• Panpsychism: the view that consciousness is fundamental and everywhere, which the thermostat tests.
• Intentionality: the aboutness the thermostat allegedly has or lacks.
• The Chinese Nation thought experiment: a related test of functionalism's liberal consequences.
• The hard problem vs easy problems: whether a thermostat's functions amount to experience.
• The character of consciousness: Chalmers's extended treatment of such cases.

Further Learning

• Internet Encyclopedia of Philosophy: Panpsychism
• Stanford Encyclopedia of Philosophy: Consciousness
• Hilary Putnam, "Brains and Behavior," in Analytical Philosophy (1962).
• Daniel Dennett, Kinds of Minds: Toward an Understanding of Consciousness (1996).