The negligent student of biology will be forgiven for thinking individual survival is the only naturally prescribed goal. The anomaly of constitutionally communal species aside—ants, bees, and the like—the replicating unit is the single animal, compelled to interact with its kin only by the necessity of procreation and, for some, the expediency of collaborative hunting or defence. Even then, others seem merely instrumental, adopted and discarded as individual goals dictate. Examples of sacrifice, say, in the animal kingdom are seen as either aberrations or illusory anthropomorphizing. Our biological selfishness is inescapable, we are told, for it is itself the product of another: the selfishness of genes, competing in evolution's lawless market.
What is it, then, that makes us “kind”, “considerate”, “sympathetic”, “humane”? The traditional view is that these characteristics develop in spite of biology not because of it, the outcome of social forces operating outside its domain. The resultant dualism is palliated by talk of “emergence”, the notion of a humanity built on biology but somehow, mysteriously, emancipated from it, like a house-reared slave risen to overcome his master. A few obvious issues of causality aside, this idea makes the apex of biology—mankind—its denial, its final loss of control. You have to be very naive to believe biology could be that stupid.
So finding a biological substrate of empathy would be a “disruptive” advance, to use the contemporary jargon. We have known for some time that destroying someone's prefrontal cortex can make him callous, but only amongst enough other defects of character to make the relation too non-specific. What we need is a clear “neural signature” of empathy, one that identifies the fundamental biological mechanism of its operation. This is what Rizzolatti and his colleagues imagined they had found when they came across a population of neurons in a monkey’s premotor cortex with a peculiar pattern of activity. The neurons fired when the monkey performed a certain movement but also when it viewed another doing the same thing. These “mirror neurons”, as they came to be known, seemed to provide the identity link between my actions and yours, and by extension my intentions and yours. Understanding others is then easily explained, in Rizzolatti's words:
“The proposed mechanism is rather simple. Each time an individual sees an action done by another individual, neurons that represent that action are activated in the observer’s pre-motor cortex. This automatically induced, motor representation of the observed action corresponds to that which is spontaneously generated during active action and whose outcome is known to the acting individual. Thus, the mirror-neuron system transforms visual information into knowledge.”
The appeal of so beautifully simple an idea is hard to deny. And very few have, especially those in the humanities hungry for neuroscience at least vaguely relatable to the human condition. So more has been written about these neurons in the non-neuroscientific press than perhaps any other. But if there is one rule about the brain, it is that any simple idea of its operation is likely to be wrong. Look carefully and you will see all the magic here is illusory - not Jesus, but his profile on a piece of toast.
One cannot come to know what one is doing by inspecting the activity of a “motor” neuron in one's brain alone. Yes, such activity may become a neural label of some kind, but only by becoming associated with the sensorily perceived consequences of an action—visual, auditory, proprioceptive, haptic, etc—on which conscious dissociation from other actions necessarily depends. To understand this, consider what would happen if the neuron got its action wrong and became associated with (say) a kick rather than a handwave. Without sensory feedback there is no way - information theoretically no way - it could correct its error. And since the association cannot be specified in the genome—there is nowhere near enough room for it there—it must arise by trial and error in the first place.
The identity of an action is thus determined in part by seeing ourselves perform it: were it not so, the easily elicited rubber hand illusion—mistakenly believing a strategically placed, visualised artificial limb to be your own—would be impossible. Of course, one does not need sensory feedback on every occasion, only on enough occasions for the association to be established. When someone observes himself performing an action, he is learning that this is what his action looks like, an association in need of constant updating because it is liable to change over time. Now a neuron engaged in this process is bound to respond to the image of an action even when someone else is performing it because its identity is precisely what it is contributing to learning. The “mirror neuron” does not “know” your action is the same as mine, it simply cannot tell them apart.
"Mirror neurons are no more mediators of empathy than mediators of selfishness: such terms simply cannot apply here."
Hang on, you might say: its business is establishing the commonality, another part of the brain establishes the differences. But the whole point is that neither is knowable without the other: activity of the kind that mirror neurons exhibit is needed to know what we are doing ourselves. It will be present equally in misanthropes and philanthropes, sociopaths and sociophiles, even if our lives were conducted in perfect solitude. These neurons are no more mediators of empathy than mediators of selfishness: such terms simply cannot apply here.
But I want to call them empathic, you say. Then consider a few other absurdities. Are the congenitally blind incapable of empathy because their mirror neurons are definitionally silent? Must our empathic power be in proportion to the visibility of a passion? Since mirror neuron activity is synchronic, is empathy impossible when the action and passion are diachronic? How do we distinguish empathic mirror neurons from envious ones, whose activity, after all, fits the mirror profile much better? My empathy for your thirst is not a desire to sink the contents of your glass. For your marriage, not to seduce your wife. For your redundancy, not to get your job. And so on.
In all fairness to neuroscience, this is the neural version of an error it took philosophy two thousand years to notice. The idea that you “know” what you are doing by inspecting the activity of a set of neurons in your brain mirrors the idea that you “know” what you are thinking by inspecting the contents of your mind. In each case such “knowing” could never work because there cannot be—information theoretically cannot be—a purely internal standard of correctness. And here, as there, the problem is not that such knowledge could not be communally shared but that it is impossible to acquire it at all for oneself. Private neural signalling here replaces a private language: the rest of the conceptual framework is more or less the same, and equally incoherent.
So next time you are feeling beguiled by talk of mirror neurons, sober up with a shot of Wittgenstein. Empathy is, of course, no less biological than any savagery; if we find it hard to explain biologically it is only because we have too narrow a conception of the biological. “I am not so hard up for categories”, Wittgenstein said: it is time we realised our conceptual penury is self-inflicted.