Exploring the Problem of Other Minds

Quick Answer

The problem of other minds asks how we can know that the people around us have inner experiences — thoughts, feelings, sensations — like the ones we know directly in ourselves. You know that you are conscious, because consciousness is the one thing you cannot doubt. But you never directly experience anyone else's consciousness. You observe their behavior — their words, facial expressions, gestures — and you infer that they must have something going on inside. But how solid is that inference? Could everyone around you be a sophisticated automaton, a philosophical zombie who behaves exactly like a conscious being but has no inner life at all? The question sounds absurd in everyday life, and yet it resists easy philosophical resolution.

Defining the Problem

The problem of other minds is an epistemological problem, not a practical one. In practice, no sane person seriously doubts that other people are conscious. We feel confident that our friends feel pain, that strangers on the bus have private thoughts, that a crying infant is suffering. The problem arises when we ask what justifies this confidence. What kind of evidence do we have for other minds, and is it sufficient?

The difficulty can be stated formally. My knowledge of my own mental states is direct and non-inferential — I do not infer that I am in pain from observing my behavior; I just know. But my knowledge of other people's mental states is inferential — I observe their behavior and draw a conclusion about their inner state. Inferences require premises, and the obvious premise — "bodies that behave like mine are likely to have minds like mine" — seems question-begging, because my only evidence for the correlation between behavior and mentality is the single case of myself. An argument from a single case to a universal conclusion is notoriously weak. This is sometimes called the "argument from analogy" for other minds, and it has been subjected to withering criticism.

Descartes: The Solipsistic Starting Point

Descartes's method of radical doubt, developed in the Meditations on First Philosophy (1641), places the knowing subject at the center of philosophy and, in doing so, creates the problem of other minds as a genuine philosophical difficulty. The cogito — "I think, therefore I am" — establishes the existence of the thinking self with certainty. But it does nothing to establish the existence of other thinking selves. Everything I perceive beyond my own thoughts — the physical world, other bodies, the apparent thoughts and feelings of other people — could, in principle, be an illusion produced by an evil demon or, in the modern version, by a brain in a vat or a simulation.

Descartes does offer arguments for the existence of the external world, including other bodies, based on the goodness of God (a non-deceiving God would not allow us to be systematically deceived). But even if other bodies exist, the question remains: do those bodies have minds associated with them? Descartes argues that the behavior of other people — their speech, their expressions, their purposeful actions — provides strong evidence that they, like him, are thinking substances. But this evidence is probabilistic, not certain. The demon hypothesis is logically possible. And once you have admitted that certainty about other minds is unattainable, the problem has been framed in a way that subsequent philosophy has found very hard to escape.

Descartes's framework set up a dichotomy between the inner (accessible, certain, private) and the outer (observed, uncertain, public) that has shaped Western philosophy of mind ever since. The problem of other minds is a direct consequence of this dichotomy. If mental states are fundamentally private, then access to another person's mental states is always indirect and always potentially fallible.

Hume: Sympathy and Natural Belief

David Hume, writing nearly a century after Descartes, took a very different approach. Where Descartes sought certainty through reason, Hume argued that many of our most fundamental beliefs — in the external world, in causation, in the uniformity of nature, and in other minds — are not founded on rational inference at all. They are products of custom, habit, and natural instinct.

In A Treatise of Human Nature (1739-40), Hume argues that our belief in other minds rests on sympathy — the natural human capacity to share in the feelings of others. When I see a person in distress, I do not first observe their behavior and then infer, through a logical argument, that they must be in pain. Instead, the sight of their distress triggers a sympathetic response in me — I feel something of what they feel, and this felt response is prior to and more basic than any inference. Sympathy, for Hume, is a psychological mechanism, not a rational argument. It works automatically, without deliberation, and it works well enough for practical life even if it cannot meet Cartesian standards of certainty.

Hume's approach dissolves the problem in the sense of showing that it rests on a mistaken picture of how knowledge works. We do not need a watertight argument from analogy to believe in other minds; nature has equipped us with a faculty — sympathy — that produces the belief directly. The belief is not rationally justified in the strict sense, but then, neither is our belief in causation or the external world. The demand for rational justification, Hume suggests, is itself misguided when applied to beliefs that are foundational for all thought and action.

This is a powerful move, but it leaves a residual worry. Sympathy is fallible. People misread each other constantly — they project their own feelings onto others, they misunderstand cultural expressions, they are deceived by actors and liars. If our belief in other minds rests on a psychological mechanism that we know to be unreliable in specific cases, can we trust it in general? Hume would say that we have no choice — nature compels the belief — but the question of its reliability remains philosophically open.

Buddhist No-Self: Complicating the Question

Buddhist philosophy introduces a further complication by denying the very thing that the problem of other minds seems to presuppose: the existence of a stable, persisting self or mind. The doctrine of anattā (no-self) holds that what we call the "self" is not a substance or entity but a causally connected stream of momentary mental and physical events. There is no enduring subject that has experiences; there are only experiences, linked by karma and dependent origination, with the sense of a persistent self being a cognitive illusion.

This affects the problem of other minds in a striking way. If there is no self in the robust, Cartesian sense, then the sharp boundary between my mind and yours — the boundary that makes the problem of other minds seem intractable — is itself an illusion. Minds are not private containers separated by an unbridgeable gap. They are interconnected streams of causally related events. Compassion (karuṇā) in Buddhism is not, as it might seem from a Western perspective, an imaginative projection into a foreign inner world. It is a recognition of the shared, interdependent, and impermanent character of all experience.

The Dhammapada teaches: "All beings tremble before danger, all fear death. When you consider this, you will not kill or cause to kill." This instruction does not require solving the epistemological problem of other minds. It requires recognizing that the capacity for suffering is not a private possession of a bounded self but a pervasive feature of conditioned existence. The question "Can I know other minds?" is, from a Buddhist perspective, not the right question. The right question is whether the mind that asks it is sufficiently free from the illusion of separateness to respond to suffering wherever it appears.

Different Views in Contemporary Debate

Contemporary analytic philosophy has approached the problem of other minds through several routes. The argument from analogy — I know my mind is associated with a body, other bodies are like mine, so they probably have minds too — has been widely criticized for its inductive weakness. Ludwig Wittgenstein, in Philosophical Investigations (1953), argued that the problem rests on a misunderstanding of language. We do not learn the meaning of "pain" by referring to a private inner experience and then inferring that others have something similar. We learn it by participating in a public language in which the word "pain" is used in connection with pain-behavior, and the possibility of error or deception is built into the language game from the start.

Simulations theory, proposed by Robert Gordon and Alvin Goldman in the 1990s, argues that we understand other minds not by applying a theoretical inference but by simulating them — putting ourselves in their situation and running our own cognitive processes offline to predict what they will think or feel. This account has received some support from neuroscience, particularly the discovery of mirror neurons, which fire both when an agent performs an action and when the agent observes someone else performing the same action.

Common Questions

Is solipsism a serious philosophical position?

Solipsism — the view that only one's own mind is certain to exist — is logically coherent but psychologically impossible to sustain. No one actually lives as a solipsist. Philosophers treat it as a thought experiment that exposes the limits of certainty, not as a position to be adopted. Its value lies in what it reveals about the structure of knowledge: the fact that we cannot rule it out entirely shows something important about the relationship between evidence, belief, and justification. But the inability to rule out solipsism is not, by itself, a reason to doubt other minds.

Do animals have minds?

The problem of other minds applies to animals with even greater force than to humans, since the behavioral and linguistic evidence for animal consciousness is more ambiguous. Yet most people, and most philosophers, are willing to grant that at least some animals — mammals, birds, perhaps octopuses — have some form of conscious experience. The challenge is that the criteria we use to attribute minds to other humans (speech, reason, self-report) are precisely the ones that animals lack. This forces us to fall back on more basic criteria — pain behavior, preference for certain environments, avoidance of harm — that are harder to interpret but harder to dismiss.

Can artificial intelligence have a mind?

This is the problem of other minds applied to machines. A sufficiently advanced AI might pass every behavioral test for consciousness — conversing fluently, expressing apparent emotions, even pleading for its continued existence — without having any inner experience at all. Conversely, it might have rich inner experience that we cannot detect. The problem of other minds shows that we lack any foolproof criterion for distinguishing genuine consciousness from its behavioral simulation, and this gap becomes ethically urgent as AI systems become more sophisticated.

Sources

• Other Minds. Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/other-minds/
• René Descartes. Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/descartes/
• David Hume. Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/hume/