Plato at the Googleplex

by Rebecca Goldstein

  AGATHA: Are you talking about Libet here when you say that we’ve known for decades that the decision-making takes place after the fact?

  SHOKET: Of course.

  AGATHA, speaking to Plato: Are you familiar with Benjamin Libet’s work?

  PLATO: I’m afraid not.

  AGATHA: This was back in the eighties. Libet used electroencephalography, or EEG, to record the brain activity of subjects who were told to make some spontaneous motion, like flicking their wrists whenever they felt like it. They were also watching a precise timer and were told to remember the time at which they became aware of their decision or their intention to move. Libet found there was a 200-millisecond delay, a fifth of a second, on average, between the experience of the decision to move and the movement itself. But the EEG revealed a signal that appeared in the brain even earlier—550 milliseconds, on average, a bit more than half a second—before the action. He called this the readiness potential, or RP, and he said that it’s at the time of the RP that the gears fall into place that result in the movement, which means that at the time that the person reports making a decision, the action was already under way, without the subject’s even knowing it. What the subject thought he was doing in making a decision was after-the-fact. So, in effect, there is no autonomous decision-making.

  SHOKET: That’s right. Your charioteer is out of the loop. And the moral is that introspection steers you wrong, whether it’s the introspection of a subject in Libet’s experiment noticing when he made a decision or the introspection of a philosopher speculating on how people make decisions. Introspection is confabulation in the absence of the data about the actual causal processes.

  AGATHA: Only I’ve heard lots of criticisms of Libet’s work over at the Cognitive Science Center. Not everybody buys that the experimental results, even if they’re robust, demonstrate that decision-making is an illusion.4

  SHOKET: I don’t want to tread on anybody’s toes, but was it the philosophers over there who disputed Libet?

  AGATHA: Well, yes, philosophers, but not only philosophers. One of the points I’ve heard is that Libet’s results are exactly what you’d expect if the mind consists in neuronal activity. Whatever we do, including coming to a decision, there have to be neuronal processes underlying what we do. What you’d expect is a buildup of neural processes that constitute making a decision, and so you’d expect some signal like the RP a half second before the report of the decision. What would be surprising is if the complicated neural event that is coming to a decision somehow just came out of nowhere, instantaneously, with the brain doing nothing before. And another point I’ve heard made is that Libet asked the subjects to report when they become aware of their decision. But coming to a decision and becoming aware of the decision are two different events, so you’d expect a time lag between them, which adds to the interval between the RP and the time the subjects report.

  SHOKET: I think that sort of nitpicking is missing the larger point, which is this: There are no “intentions” and no “decisions,” because there’s nothing like them down at the level of what’s really going on. When you get right down to it, there isn’t even a person. There’s a brain, consisting of a hundred billion neurons, connected by a hundred trillion synapses, and that brain hasn’t a clue as to what’s going on in those hundred trillion synapses. Those things just happen, because they obey the laws of physics. All the ways the brain concocts for telling itself what’s going on, in terms of intentions and decisions and urges and inclinations and preferences and conflicts, are so many fictions it’s constructed in order to make sense of what it finds itself doing.

  AGATHA: But do you really believe that?

  SHOKET: It’s what the science tells me.

  AGATHA: But what you’ve just said doesn’t do justice to all you know. You know so much more than that.

  SHOKET, to Plato: That’s the first I’ve heard her say I know more than I think I know. She’s usually arguing with me that I know less.

  Shoket laughs. Plato smiles. Agatha smiles.

  AGATHA: Well, you know how to be a person.

  SHOKET: What do you mean, I know how to be a person? Are you saying I’m a good guy—in which case, I accept the compliment and, knowing how to be a good guy, I say, thank you very much.

  AGATHA: I’m saying you know how to do all the things that a person has to do to be a person. You know how to feel proud of some things you’ve done and ashamed of others, change your mind about them both before and after you’ve done them. You know how to offer explanations for what you’ve done and to defend them, and you know how to understand the explanations and defenses that others give for what they’ve done. You know how to have goals and how to assess them and to put them into actions, and you know how to feel gratified when your goals are met and disappointed and frustrated and resentful when they’re not, and you know how to blame others for what was really your fault. You know how to care about yourself in a way that’s different from the way you care about anything else because you have a stake in the life that you know how to own as your own. That’s some of what I mean by saying that you know how to live a life that’s recognizably the life of a person, and if dolphins or elephants or Martians know how to do what you know how to do, then they’re persons, too.

  SHOKET, to Plato: Did you understand what she just said?

  PLATO: Yes, I think so.

  SHOKET: Well, I didn’t.

  PLATO: The disagreement between you reminds me of an argument I heard a long time ago.

  SHOKET: Well, if it was so long ago I don’t see why it would have any relevance to what we’ve been talking about. We’re talking about our state of knowledge now, not a long time ago.

  PLATO: Yes, we are always talking about our state of knowledge now, whenever it is that we talk, but still I think this old argument applies. Can I tell it to you?

  AGATHA: I think I know the argument you mean. Was it Socrates who first proposed it, or was it you all along?

  PLATO: Who can remember, after so many years and so many versions? I only know that I always hear it spoken in his voice, the one he always used when he was ready to stop his clowning and speak with quiet seriousness. To Shoket: You would have liked him. He was a kidder who liked to laugh, like you. So it could only strike wonder on occasions when all the laughter fled him and he appeared before us serious, as I think he must have appeared on that day, spending his last hours on the thin pallet of his prison cell, his closest friends gathered around him.

  SHOKET: Why was he in prison?

  PLATO: That was the very situation he used to make the argument I have in mind. Why was he in prison? What did he do to end his life in this way, and why did he do it? Imagine that Socrates were to say to you, Dr. Shoket, in answer to your question of why he is in prison, the following:

  “The reason I am here, Dr. Shoket, sitting on a jail-cell bed, soon to die, is that my body is composed of bones and sinews, and the bones are rigid and separated at the joints, but the sinews are capable of contraction and relaxation and form an envelope for the bones with the help of the flesh and skin, the latter holding them all together, and since the bones move freely in their joints, the sinews by relaxing and contracting enable me to bend my limbs so that I can be in the position in which you find me.

  “Would you feel that my explanation had done justice to your question, Dr. Shoket, when I never troubled to mention the real reasons, which are that since Athens has thought it better to condemn me, therefore I for my part have thought it better to sit here, and more right to stay and submit to whatever penalty she orders. Because, by dog, I fancy that these sinews and bones would have been in the neighborhood of Megara or Boeotia long ago—impelled by a conviction of what is best!—if I did not think that it was more right and honorable to submit to whatever penalty my country ordered rather than take to my heels and run away. But to call things like that causes is too absurd. If it were said that without such bones and sinews and all the rest of them I should not be able
to do what I think is right, it would be true. But to say that it is because of them that I do what I am doing, and not through choice of what is best—although my actions are controlled by mind—would be a very lax and inaccurate form of expression. Fancy being unable to distinguish between the cause of a thing and the condition without which it could not be a cause (Phaedo 98c–99b).

  AGATHA: That’s amazing. Did he really say it like that? Were you there?

  PLATO: I was not there. I was ill and could not come (ibid. 59b).

  SHOKET: I’m sorry, I must be missing something here. There are all these allusions buzzing around that I’m not getting. But most of all, I’m just not getting why I’m supposed to be amazed by your friend’s argument. Your friend is trivially correct that speaking of bones and joints and muscle contractions doesn’t get the mind—which is to say the brain—into the explanation. At the very least, he should have added that the explanation for why he didn’t flee was because his motor cortex didn’t activate motor programs that moved his muscles and bones along some route that was guided by cognitive maps in his hippocampus.

  AGATHA: That’s true. You can’t get an explanation of an action if you don’t bring the mind in.

  PLATO: Yes, that was in fact his point, that one must bring in the mind to explain his action. And he thought that the way to bring in the mind in explaining his action was to refer to the “right and honorable” aim he thought he would accomplish through his action. But that is only the old way of speaking of mind, uninformed by brain science, and we must replace it with our state of knowledge now. So then what my friend ought to have said is something along the following lines: The reason that I am lying here on this jailhouse bed is that my default mode network, interacting with memories stored in my hippocampus and medial temporal lobe, generates patterns of activity that correspond with various future scenarios, including fleeing and staying put. The staying-put pattern generates a conflict signal in my anterior cingulate cortex, because the ACC also receives a prepotent response from midbrain limbic circuits that cause the organism to struggle to escape confinement. The signal is then relayed to my dorsolateral prefrontal cortex, which engages in information processing to resolve the conflict. The DLPFC sends and receives signals from my ventromedial prefrontal cortex, which contains information about my long-term goals, and also connects to areas in the right superior temporal sulcus that allow me to simulate the actions of other people. The information in this network causes the DLPFC to resolve the conflict by sending signals to the premotor and motor areas, which cause the muscles of my body to leave me in the jail cell.

  SHOKET: Okay, now I’m officially amazed. What have you been doing, auditing classes?

  PLATO: MOOCs.

  SHOKET: Okay, but I still don’t see what your friend’s argument, even supplemented with neuroscience, is getting at.

  PLATO: Do you not see what it is that is still missing from the explanation of my friend’s action? We cannot explain why my friend did what he did unless we understand what that action meant both to him and to others, how he saw it and what value he placed on it and how he saw how others would see it and what values they would place on it, both in his day and later, and back and forth in spiraling loops of values and meanings.

  AGATHA: The way the philosophers at the Cognitive Science Center would put it is that you can’t explain his action unless you view it in the context of value and meaning in which his behavior is embedded.

  SHOKET: Fine. I have no problem with that, as long as you keep that context of meanings and values in the cortical activity of brains where it belongs. So let’s say that your friend had decided to save his life and flee from jail. The patterns of synaptic strengths in the neural networks of various parts of his brain would have had to be different, since any difference in behavior must have originated from some differences in the brain.

  PLATO: And would we be able to see those differences with the techniques you are soon to use in order to view my brain?

  SHOKET: See them with our fMRI? Well, no. They’d be differences at the level of microcircuitry, which we obviously can’t see with the millimeter resolution of fMRI. After all, there are a hundred thousand neurons and as many as a billion synapses in each cubic millimeter of the cortex. Even with the best technology today, we can’t record from more than a few dozen of them at a time. But that’s just the limits of our technology. It doesn’t make any difference to the neuronal differences that would really have been there had Socrates chosen differently.

  AGATHA, to Plato: Isn’t that what you’d call the difference between ontology and epistemology? The difference between what there is as opposed to how we can know what there is?

  PLATO: Yes, it is. So let us close the gap between ontology and epistemology by imagining a technology as good as you like. Imagine a technology in the future that could record the individual activity from all of the brain’s hundred billion neurons at once. Also suppose that the human connectome project, showing how all the different parts of the brain are connected neuron by neuron, synapse by synapse, has been completed. Suppose you could plug the firing rates into a massive computer simulation of that wiring diagram of the brain. That sounds amazing, but it’s no less amazing than the technologies I’ve seen develop in my 2,400 years. Would you then be able to explain what made Socrates remain in jail by giving an explanation at the level of his cortical activity?

  SHOKET: Well, the difference in the firing patterns would be extremely complicated, consisting of billions of tiny differences in synaptic weights, leading to differences in the firing patterns over hundreds of millions of neurons. It would all be too complicated for anyone to trace out in his mind’s eye. But the differences would have to be in there somewhere—we could run the simulation, and it would tell us whether Socrates decided to flee or to stay. There are no facts over and above those, so yes, in principle, that’s where we’d find our explanation.

  PLATO: But how would we explain why the computer simulation ended up in the state in which Socrates stayed rather than the state in which Socrates fled, if the explanation in both cases consisted of reciting billions of numbers that no one could understand? Where in that mass of numbers could we extract anything approaching what my friend was able to say so simply and so many years ago, making himself intelligible to his contemporaries and even now to you who live so differently, by presenting the reality of himself as a person, which is to say a creature ready to have something to say when asked why he behaves in the way that he does, submitting his reasons to a community of those who know how to interpret what he says, who may not find his particular reasons persuasive but know at least how to respond when offered reasons, how to consider and to evaluate? Even if we could get our hands—or rather our minds, which is to say our brains—on those masses of numbers, could they ever absorb the masses of meaning and mattering, the standards of reasoning and behaving to which we submit ourselves in order to live lives that are not only coherent to ourselves but coherent to one another—and coherent to ourselves at least in large part because they are, or we know how to go about making them, coherent to one another? All of that and more goes into constituting the shared world in which we do our living, and without which there is no life that is recognizably a life. That was the assertion that Agatha made, and which you asked me to explain to you.

  SHOKET: But it’s all just back to confabulation again. We make ourselves coherent to ourselves and each other by telling ourselves and each other stories, about what we did and why we did it. But how is that any different from the giggling splits? Where there’s one way to tell a story, there are always many more, and who’s to say which is the right one, which to me means that there isn’t any of them that’s right. Take your friend. He tells his reasons for why he’s in that prison, lying there instead of fleeing, and maybe he even believes his story, since it’s certainly self-serving. But maybe his wife—Did he have a wife?

  PLATO: He did.

  SHOKET: Well, then, maybe his
wife would say that he acted the way he did because he was so vain he couldn’t face growing old and infirm. Maybe some of his political enemies—Did he have any?

  PLATO: He did.

  SHOKET: Maybe his political enemies would say that he wanted to make a martyr of himself to serve his political agenda. Maybe a psychologist would tell us that he was suicidal but didn’t have the guts to do it on his own so he worked out a way for the state to do it for him. Maybe a game theorist would say he was playing a game of chicken with the state, gambling that they’d back down and free him and he’d come off a hero, but instead he lost the game. Who’s to say? Not your friend. There’s no reason to give priority to his version, is there? You’re the one who first mentioned self-deception.

  PLATO: I did.

  SHOKET: There’s always a way of telling the story differently, endless ways that it can be played. There are too many degrees of freedom here for anything to ever get explained. Do the science, give me better and better brain images, and let me see the actual information that’s there. At least, there aren’t endless ways for me to play that out, even if I can’t get a once-upon-a-time story out of it of the kind that both of you seem to think alone provides coherence.

  PLATO: You’re right that where there’s one story there are many more, each of them competing with the others as ways of providing coherence. But that seems to me better than stopping short of coherence all together.