Above is our cat Oliver, in a photograph made by my wife Loret. Oliver is four years old. He takes to visitors, though he can be skittish, and they take to him. He stares, and his eyes seem to interrogate you. His lankiness makes him seem almost human at times; as he reaches for objects it is easy to think of his front legs as arms.

He is also smart and bonded with me in particular. Several times a day he calls for attention, and he appears most at ease when he can join me on the couch. He will address me in the kitchen, either vocally or with a tilt of his head, and if I say, Yes, Oliver, let’s go to the couch (I emphasize the word as a training technique) he runs as fast as he can to the hallowed piece of furniture and stands guard, watching me until I stretch out. Then he lies down, his body relaxes, and his features soften into the closest things cats come to a smile. If he hears a noise, though—his littermate Leo playing with a catnip mouse, perhaps, or the letter carrier pushing mail through the slot—he will jump off and run from the room without the slightest hesitation or sign of regret.

I think it likely that cats entertain thoughts, but I am certain that they do not entertain second thoughts. This is one of the disconcerting things about them, and about our relationships with many non-human animals. When I am caught up in the emotional connection I seem to have with Oliver I relate to him as one feeling being to another, and the sensations, though different in degree from those I experience with my wife, are not so different in kind that I would hesitate to call them love. In that openness there is no distinction between feline and human. But the sudden rupture that plunges him into hunting mode reminds me that he is profoundly alien, not just in his habits but in the lack of any self-awareness that carries over from one activity to another. The grace and purity of feline gestures, like the perfection of Kleist’s marionettes, come from their full absorption in whatever they may be doing. Their acts are not clouded by consciousness of the past or by worries over the future:

For there is nothing sweeter than his peace when at rest.

For there is nothing brisker than his life when in motion.

What goes on “inside” Oliver’s mind? What is it like to be our cat? Those two moments in our everyday interactions seem to map out two opposing possibilities. When he nuzzles me and settles between my legs it is hard not to see Oliver as a being imbued with a rich emotional life and the awareness that comes with those feelings. William James saw emotions as intuitions of bodily states, and as Mark Solms points out, they would be useless unless they were experienced. But when Oliver springs into action he seems to be identical with his acts, living in a pure responsiveness without reflexive awareness.

These happen to be the only two possibilities that Descartes could imagine. He argued that humans were ensouled, self-aware beings capable of both thought and the passions, but “beasts” were merely animate bodies. They had a large repertory of possible actions and underwent the same physical states that we experience as emotions, but they were unaware of those states and acted mechanically. They had no feelings.

Descartes’ alternatives have been heavily criticized, generally with justice, yet we seem to be unable to think beyond them. It is clear that other animals are not fleshy automatons or philosophical zombies, but those who reject that alternative tend to embrace the other one, seeing them as “highly emotional people of very limited intelligence.” (The phrase originates with Oskar Heinroth.) This is common in popular science writing; there’s a steady stream of articles claiming that crows or pigs or parrots are self-conscious, “just like humans.” But there are serious scholarly articles that make similar claims—that coyotes and others have a concept of death, for example.[i]

We are still trapped in the Cartesian scheme: there is either a body propelled by blind instinct or something that knows itself to be willing, seeing, and feeling. But though Oliver is obviously an agent—he interrupts me to get my attention, jumps up to watch the birds outside the bathroom window, and tells me when he wants a nice shared nap—the relevant question is not whether he has actual agency but whether he knows himself to be an agent. I can see that he acts purposefully, but is he aware of those purposes? Can he reflect on what he is doing or on his desires and fears? It is not obvious that he can, and that doubt points towards what one could call two real but less-often-recognized “hard problems:” what self-consciousness consists of and how it emerges from consciousness per se.

Descartes’ on-or-off view of consciousness has given way to more nuanced schemata, such as one that distinguishes perception, access consciousness “in which mental representations may be poised for use in rational control of action or speech,” phenomenal or subjective consciousness, and second-order or self-consciousness. The representationalist bias in this analysis is questionable, though, and the boundaries among the “higher” states are rather porous; the definition of even access consciousness presumes some kind of subjectivity which can exercise control over conduct. What is more, the entire schema conceals an unjustified assumption which come closest to the surface in its definition of self-consciousness.

This is interpreted as thinking about our own thoughts, as “an organism’s capacity for second-order representation of the organism’s own mental states.” The definition seems rather straightforward, even self-evident. We in the West (as opposed to many in the Indic tradition, for instance) assume that agency and perception are properties of an autonomous subject, coextensive with the brain or body, and that self-consciousness arises when we turn the light of awareness on events within that subject instead of on events without. If one can solve the canonical hard problem and demonstrate the neural basis of awareness, it seems, nothing more needs to be done to show how self-awareness comes into being. Once a being’s brain has sufficient neural complexity to engage in reflection it will encounter an already-existing ownness, those mental states which it can represent to itself as the contents of the self.

But that assumption cannot be justified. As Dieter Henrich pointed out decades ago, reflexivity cannot account for self-knowledge.[ii] For one thing, we could not identify ourselves as selves in reflective mental operations if we did not already have a concept of a self. This is only one of his three “paradoxes of self-reference,” but it is enough to cast the common sense model of the self into doubt, and the elision of consciousness with self-consciousness along with it. Even if the self were there to be found, reflexive awareness would not recognize it.

This error affects much contemporary writing about self-consciousness. Anil Seth, for example, writes that “every emotional experience is rooted in top-down perceptual best-guessing about the state of the body (and about the causes of this state),” and his parenthetical gives the game away. Self-guessing the state of the body cannot generate a distinction between experience and its causes. It encompasses both, but it gives us no grounds for differentiating them. Seth assumes what he needs to account for.

What does account for self-consciousness? No philosopher in the European tradition has grappled more earnestly with that problem than Fichte, whose “original insight” prompted Henrich’s essay. Fichte is often taken to be the philosopher for whom the I is the center of all thinking and experience, but he was more concerned with problematizing and deconstructing the experience of selfhood. And while he had little or no interest in the experience of other animals, Fichte’s insights turn out to be extraordinarily helpful in distinguishing between consciousness and self-consciousness and in imagining what it is like to be a cat.

There is no substantial self for Fichte, no entity which is there for the finding. There is only the infinitely repeated act of positing a self and a not-self, the world. Kant had argued that we are objects to ourselves, known not directly but only through appearances. Fichte asked how that inner split could have come about, and his answer was that we make it ourselves. The self exists when we parse our experience so as to constitute self and world, forming a self as a subject amid objects, as an active agent surrounded by passive materiality. It is nothing more than that structuring.

Positing does not take place in a vacuum, however. It arises within a reciprocal web of activity that is woven from the responses of each organism to every other organism. Even what we construe as internal biological events are demanded by and coupled with the organism’s environment, physical and social. Positing the self lays claim to independent agency, but that claim is made within a complex of distributed agency in which intention and action are out in the open. These ideas originate with Kant, if not earlier, but they are consistent with much contemporary cognitive science; serious scholars can write about a “cognitive” biosphere woven from “a dense web of feedback loops and interactions that emerges as a higher-level phenomenon from the activity of zillions of lower-level players, like microbes.” The “ownness” which is supposedly discovered in introspection thus looks like a back-projection from self-consciousness. In reality, perception and action flow uninterrupted from one body to another. As Susan Hurley wrote, there is nothing “specially oomphy” about the skin; beings are “transparent to causality.”[iii]

If the self is not an entity but a way of framing experience, though, there is every reason to think that there can be experience without that frame. How Oliver, Leo, and other animals act is consistent with their perceiving and experiencing everything that we do, specificities of perceptual organs aside, with one fundamental difference: they do not posit themselves. They do not parse what they perceive into introspection and extroception, and thus they do not separate themselves from their experience, which is alive with sensation and pregnant with intentionality and agency. “Do not look for anything beyond the phenomena,” Goethe said. “They themselves are the theory.” Oliver does not need that advice.

In experience unsundered into self and other there is no separation between feeling and perception. Our cats’ lives are lit up with physical well-being and dimmed with pain, and the experience of cruelty would not leave them wounded souls in an objective world; it would poison everything. This is not unimaginable to us. Tennyson famously evoked grief in which the percipient vanishes:

He is not here; but far away

The noise of life begins again,

And ghastly thro’ the drizzling rain

On the bald street breaks the blank day.

And one could also cite the stupendously wrong but brilliant Julian Jaynes, who thought he could imagine what inner life might be like without a self.

This is a way out of the Cartesian false choice. Other animals are neither machines nor self-conscious mini-mes. They are fully sensitive to perception and emotion, but they shift from pleasure seeking to aversion or from fear to security like sailors riding the wind, doing spontaneously whatever falls to hand. Recognizing this possibility offers insight into a number of philosophical problems.



[i] Other animals appear to have stable and revisable classifications of objects and patterns of activity, but these do not entail self-consciousness. It is otherwise with genuinely propositional concepts, which are explicitly interposed between subject and experience. See, e.g., Albert Newen and Andreas Bartels, “Animal Minds and the Possession of Concepts”, Philosophical Psychology, 20:3, 283 – 308.

[ii] Dieter Henrich, “Fichte’s Original Insight” in Contemporary German Philosophy, Vol. 1 (Univ. Park, PA: Penn State University Press, 1982), 15-53

[iii] Susan Hurley, Consciousness in Action (Cambridge, MA: Harvard University Press, 1998), p. 336.