René Descartes believed animals were automata — flesh-machines with no inner life, responding to stimuli like a clock responds to winding. Ivan Pavlov, whose conditioned reflex experiments began in earnest around 1897, produced dogs that anticipated food before any food was present. The machine had learned to want. Descartes’ philosophical corpse has been drooling on the floor of experimental psychology ever since.
This is not a minor historical curiosity. The question of what animals know — and whether “knowing” is the right word at all — sits at the intersection of philosophy of mind, neuroscience, and ethics. If Descartes was right, then animal suffering is not suffering in any morally relevant sense. If Pavlov’s data means what Merleau-Ponty thought it meant, then something like meaning-making happens below the threshold of language, in the dark, electrochemical space where reflex becomes anticipation.
Descartes in the Discourse on Method, Part V
The argument is laid out with characteristic precision in the Discourse on Method (1637). Descartes needs to explain why the human body, which is also a kind of machine, produces behavior that differs categorically from animal behavior. His answer is mind — the res cogitans, the thinking substance that he identifies with the soul. Animals have no such substance. They are, in his phrase, “automata naturalia” — nature’s mechanisms, operating through the arrangement of their organs the same way a clock operates through the arrangement of its springs and gears.
The test Descartes proposes is linguistic. A machine, however sophisticated, cannot use language to respond to the sense of what is said to it rather than to the mechanical stimulus of the sounds. A parrot can say “I want that” when taught to do so, but it cannot form novel combinations of words to respond to novel meanings. This linguistic plasticity is, for Descartes, the reliable marker of the presence of mind.
He is not entirely wrong about the linguistic point, which is part of what makes this so interesting. The argument is philosophically careful. What he gets catastrophically wrong is the implicit model of what a body-without-mind can do.
What the Conditioned Reflex Actually Shows
Pavlov’s Nobel Prize came in 1904 for his work on digestive secretions, but the conditioned reflex research, published in its canonical form as Conditioned Reflexes (Oxford University Press, 1927), is what haunts the philosophy of mind. The experimental setup is famous: a dog is exposed to a neutral stimulus — a bell, a metronome, a light — simultaneously with the presentation of food. After repeated pairings, the neutral stimulus alone produces salivation. The dog salivates to the bell.
For Descartes, this should be impossible in any philosophically interesting sense. A machine responds to what is mechanically present. But the conditioned dog is responding to what the bell means — to its acquired relationship with food — not to the bell’s acoustic properties. The bell’s acoustic properties have not changed. What has changed is the dog’s internal representation of the relationship between that sound and a subsequent event.
This is not mechanism. Or rather, it is mechanism of such a specific kind that it forces us to expand our concept of what mechanism can accomplish. The body has stored a relation between two events that have no natural connection. In storing that relation, it has done something that looks, from the outside, very much like primitive inference: when this, then that. To produce salivation in the presence of the bell, the nervous system must represent the bell not merely as a sound but as a sign.
Descartes’ animal-machine cannot have signs. Signs require something that interprets them. The conditioned dog interprets the bell. Whether we call that interpretation “meaning-making,” “proto-cognition,” or simply “associative learning” is partly a terminological decision — but it is a decision that carries enormous philosophical weight.
Merleau-Ponty’s Judo Move
The most underappreciated act of intellectual judo in the history of philosophy of mind occurs in Maurice Merleau-Ponty’s The Structure of Behavior (1942; English translation, Beacon Press, 1963). Merleau-Ponty was attempting to dismantle behaviorism — the doctrine that psychology should concern itself only with observable stimulus-response relationships and not with inner states. His target was not really Pavlov but the American behaviorists, particularly Watson, who had turned Pavlov’s careful experimental findings into a sweeping philosophical program: all behavior is conditioned reflex, all inner states are eliminable, consciousness is an epiphenomenon at best.
Merleau-Ponty’s move is to use Pavlov’s own data against this program. He points out that the conditioned reflex is not a simple mechanical link between stimulus and response. The animal’s behavior varies with context in ways that cannot be explained by the conditioning history alone. A dog conditioned to salivate to a bell in one room may not salivate to the same bell in a different room. The relevant unit is not the stimulus but the situation — the whole perceptual field in which the stimulus appears. This means that what is being conditioned is not a mechanical response but a structured relation between an organism and its environment.
This is the beginning of what Merleau-Ponty will develop, in Phenomenology of Perception (1945), into the concept of the body-schema — the idea that the body is not a neutral instrument operated by a mind but is itself a form of intentionality, a being-toward-the-world that has its own intelligence. The dog’s salivation is not evidence against the dog having inner states; it is evidence that inner states are embodied before they are cognitive.
The philosophical significance of this is hard to overstate. Merleau-Ponty is not rejecting Descartes’ dualism by simply asserting that animals have minds. He is rejecting the terms of the debate. The question is not whether there is a mind distinct from the body that the body serves. The question is whether the body-world relationship itself constitutes a form of orientation — a primitive intentionality — that precedes and makes possible the explicit cognition that Descartes identified with mind.
The Cartesian Residue
It would be convenient to conclude that Descartes was simply wrong and that Pavlov proved it. The situation is more interesting. Descartes was wrong about the specific claim — that animals cannot store relations between unrelated events and act on those stored relations — but he was not wrong that there is a difference between the kind of cognition that produces salivation to a bell and the kind that produces the Discourse on Method.
What Pavlov’s data establishes is that the line Descartes drew — at the boundary of language, at the threshold of mind — is not where he thought it was. The body can do more than mechanism, in Descartes’ sense, predicts. Whether what the body can do constitutes thinking is a question that remains genuinely open.
Merleau-Ponty’s answer — that embodied intentionality is a mode of knowing prior to and not reducible to explicit cognition — is compelling, but it raises its own difficulties. If the dog’s anticipatory salivation is a form of knowing, what follows for how we treat dogs? The ethical implications were not Merleau-Ponty’s primary concern, but they are ours. A being that knows, even in this primitive, pre-linguistic sense, is a being to whom something can happen — a being that can be harmed, not merely damaged.
Descartes’ animal-machine doctrine was not a casual prejudice. It was a philosophical position in service of a broader metaphysical project — the strict separation of matter and mind, the identification of the self with the thinking substance alone. That project is now widely regarded as untenable. But its shadow persists wherever we act as though the question of animal minds has been settled, or wherever we treat the body — animal or human — as mere mechanism waiting to be set in motion by a sufficiently elevated mental event.
The dogs in Pavlov’s laboratory salivated to bells they had learned to interpret. What they had, before the learning, was a body with the capacity to learn. What Descartes couldn’t account for was that capacity itself — the openness to relation, the ability to be changed by experience in ways that persist and produce anticipation. That openness is not mechanism. And it is not nothing.
You Might Also Like: The Problem of Induction: David Hume, Black Swans, and the Limits of Predictive Statistical Models | The Demarcation Problem: Karl Popper, Falsifiability, and the Boundary Between Science and Pseudoscience | The Extended Phenotype: How Your Genes Build Structures Beyond Your Body | The Conscious Mind by David Chalmers — The Last Wall Science Hasn’t Scaled
Sources
Descartes, René. Discourse on Method and Meditations, trans. Laurence Lafleur. Bobbs-Merrill, 1960. | Pavlov, Ivan. Conditioned Reflexes, trans. G.V. Anrep. Oxford University Press, 1927. | Merleau-Ponty, Maurice. The Structure of Behavior, trans. Alden Fisher. Beacon Press, 1963. | Comfort, Nathaniel. The Science of Human Perfection. Yale University Press, 2012. | Nobel Prize — Ivan Pavlov
