Mid-Pour
On Noticing, Wanting, and Other Things That Happen Before You Do
My first book, philosophy meeting science. It is not published yet, but you can already open it and read part of Chapter Two.
- Status
- Forthcoming
- Genre
- Philosophy / Science
- Sample
- Chapter Two, in full
A glitch is only a glitch until someone decides to paint on it.
A mockingbird on an ordinary afternoon. A jam tasting that goes quietly wrong. A cup of coffee in the moment the cream is still curling.
Mid-Pour follows one small observation all the way down: something in us moves before we do. We notice before we decide to look, want before we have reasons, and find things meaningful before we can say why. From the edge of our senses to the physics of why anything settles at all, it asks what it means to be a shape in a long fall, and why that might be wonderful rather than bleak.
Robin Khanduri is a technologist who builds software, sound and images. Mid-Pour is his first book.
The pull comes first.
Chapter Two opens with this diagram. Move below the line: the pull goes where you go, at once. The reason arrives about half a second later, and settles as if it had led.
A pull forms first, somewhere beneath awareness. Reasons arrive later and make a case for it. We experience the case as the cause.
Chapter Two.
The Floor Below Choice.
On preferences that arrive before their reasons, and the part of us that writes the explanation afterwards. The whole chapter, set in the book’s own type. Turn the pages, or read it as text.
Chapter Two
The Floor Below Choice
the reasonthe pullIf you ever want to watch human freedom in action, go to a technology conference.
Stand near the coffee and look around. You will see, within a few minutes, a remarkable degree of agreement. The same quilted vests. The same muted sneakers. The same laptop stickers, the same water bottles, the same carefully casual haircut, the same phrasing of the same three opinions about the future. Ask any individual why they are dressed the way they are and you will get a sincere answer about practicality. The vest is warm but leaves the arms free. The sneakers are comfortable. And every one of these several hundred independent thinkers will have arrived, through their own reasoning, at an identical outfit.
It would be unfair to single out the technology industry, because the pattern is everywhere. Walk into a gathering of academics, or punks, or finance people, or evangelicals, or climbers, or musicians, and you will find the same convergence in a different costume. People who belong to a group come to look like the group, dress like the group, and eventually sound like the group, sometimes down to the rhythm of their sentences. Even people who consider themselves above all this tend to dress in the recognisable uniform of people who consider themselves above all this.
Ask anyone in any of these groups why they like what they like and they will tell you. “I like how it looks.” “It’s comfortable.” “It’s just me.” They are not lying. The liking is real. But the answer does not explain what it claims to explain. It does not say why this particular person came to like precisely the thing that everyone around them also likes. The preference was there first. The explanation arrived afterwards and wrapped itself neatly around it.
That, stated plainly, is the idea this chapter is about. Much of what we experience as choosing is not choosing in the way it feels. A pull forms first, somewhere beneath awareness. Reasons arrive later and make a case for it. We experience the case as the cause. The order is backwards.
It is a large claim, and a claim that large deserves evidence. Fortunately, psychologists have spent decades collecting it, often in ways that are very funny if they are not happening to you.
Start with the simplest mechanism, which is exposure.
In the 1960s, readers of student newspapers at the University of Michigan and Michigan State started noticing small boxes on the front page, like advertisements, each containing a single strange word. The words were Turkish, or Turkish-sounding, and meant nothing to the readers. Some appeared many times over the weeks, others only once or twice. Anyone who asked the editors was told that the advertiser wished to remain anonymous.
The advertiser was a psychologist named Robert Zajonc. When the run was over, his team sent questionnaires to students asking whether each word seemed to mean something good or something bad. The words that had appeared most often were rated more positively. Nobody knew what they meant. Nobody had any reason to prefer them. They had simply been seen more.
Zajonc called this the mere exposure effect, and it has since been found with faces, shapes, paintings, sounds, characters from unfamiliar alphabets and a great deal else. Familiarity, it turns out, breeds not contempt but fondness. And one detail makes it especially relevant here: the effect tends to be strongest when people are not aware they were exposed at all. The preference forms most easily precisely where you cannot see it forming.
Now apply that to the conference. Nobody sat down and decided to love quilted vests. They were simply surrounded by them, month after month, in the company of people they admired or wanted to resemble, until one day a vest in a shop window looked obviously right. The reason, comfort, practicality, was perfectly true. It just was not the reason.
Exposure explains where some preferences come from. The next experiment is more unsettling, because it catches the explanation being made up on the spot.
Two researchers at Lund University in Sweden, Petter Johansson and Lars Hall, borrowed a technique from stage magicians, and published the result in 2005. They sat across a table from volunteers and showed them pairs of photographs of faces, asking them each time to point to the one they found more attractive. Then the experimenter slid the chosen photograph across and asked the volunteer to explain why they had picked it.
The trick was that on some trials, using sleight of hand with black-backed cards on a black table, the experimenter handed over the other photograph. The one they had not chosen.
Most people did not notice. No more than about a quarter of the swapped trials were caught, and the rest looked at a face they had just rejected and explained, with complete confidence, why they preferred it. The details were the best part. One participant reportedly said he liked the woman because she was smiling, though it was the face he had actually chosen that had the smile. Another reportedly said she liked the earrings. The face she had originally picked had not been wearing any. When the researchers analysed the explanations, the reasons given for the swapped faces were delivered with the same confidence, detail and emotion as the reasons given for faces people had really chosen. And afterwards, when asked, participants tended to insist they would have noticed a switch.
It would be comforting to think this only works on something as subjective as faces. So the same team set up a tasting stall in a supermarket. Shoppers sampled two jams and picked a favourite, then tasted it again and explained what they liked about it. Except that the jars were rigged, and the second taste was the jam they had rejected. Overall, about one in seven noticed at the time, one in five counting those who realised afterwards, about a third counting every kind of reported change. Even when the difference was enormous, cinnamon-apple swapped for bitter grapefruit, only about one in five shoppers noticed on the spot, and fewer than half noticed at all. The rest described the virtues of a jam they had just turned down.
The name the researchers gave this is choice blindness. We are not only unaware of where our preferences come from. We are often unaware of what we chose a moment ago, and we are fully prepared to supply reasons for it anyway.
The clearest picture of how those reasons get produced comes from a stranger place: patients whose brains had been, in a very literal sense, divided.
For a time, some people with severe epilepsy were treated by cutting the corpus callosum, the thick bundle of fibres connecting the two halves of the brain. It reduced their seizures and left them, in everyday life, surprisingly normal. But it also meant that the two hemispheres could no longer share information directly, and in the 1960s the neuroscientist Michael Gazzaniga, working in the lab of Roger Sperry, realised this offered an extraordinary opportunity. By flashing an image to only one side of a patient’s field of vision, he could deliver it to only one hemisphere. And in most people, only the left hemisphere can talk.
In one often-told experiment, a patient’s left hemisphere was shown a chicken’s foot, and his right hemisphere was shown a snowy scene. He was then asked to choose, from a row of pictures, the ones that went with what he had seen. His right hand, controlled by the left hemisphere, picked a chicken. His left hand, controlled by the right hemisphere, picked a shovel. So far everything made sense.
Then he was asked why he had chosen the shovel.
The talking left hemisphere had never seen the snow. It had no idea why the left hand had picked a shovel. But it did not say “I don’t know.” Without hesitation, it produced an answer: the shovel was for cleaning out the chicken shed.
Gazzaniga came to call this part of the brain the interpreter. Its job, as he describes it, is to build a coherent story out of whatever the rest of the brain is doing, and it does that job whether or not it has access to the real reasons. It is not lying. It genuinely believes the story. It simply was not in the room where the decision was made, and it would rather improvise than admit it.
I have come to think of the interpreter as a press secretary. Decisions are made in closed rooms, by committees it has never met. Then it walks out to the podium, faces the cameras, and delivers a confident, articulate account of the reasoning behind them. It is very good at this. It is so good that most of the time even you believe the press briefing. The split-brain patients are simply the rare case where someone slipped a photograph under the door, and we got to watch the secretary work without notes.
The split-brain patients are an extreme case. The most famous experiment on the same question in ordinary brains is worth mentioning for honesty’s sake, because it turned out to be less clear than its reputation.
The neuroscientist Benjamin Libet asked volunteers in 1983 to flex a finger whenever they felt like it, and to note the moment they became aware of deciding to do so. Recordings of their brain activity showed a slow build-up of electrical activity, called the readiness potential, starting several hundred milliseconds before the moment people reported deciding. For decades this was cited as proof that the brain decides before you do and your awareness is merely informed afterward.
More recent work has complicated that. In 2012 Aaron Schurger and his colleagues proposed that the readiness potential is not the signature of a hidden decision at all, but the average of the brain’s ordinary background fluctuations, the natural ebb and flow of neural noise. When you are told to move whenever you feel like it, you tend to move at the moment that noise happens to drift across some threshold. This reading has gained ground.
It is a good correction, and I do not want to lean on Libet more than he can bear. But it is worth noticing that the new explanation does not exactly restore the feeling of a sovereign self pulling a lever. It replaces “your brain decided first” with “you acted when the random background murmur of your brain happened to peak.” Which is, if anything, a stranger answer. The volunteers acted when the moment struck them. The phrase is apt. Something struck. The person arrives, if only just, afterwards.
So if preferences are not chosen in the way they feel chosen, where do they come from?
The honest answer is: from a layer of you that is not available for inspection.
Down there, beneath awareness, is more or less everything that has ever happened to you, mixed with everything you were born with. Your genes. The tone of voice your parents used. Whatever was on television in the afternoons when you were six. A particular kid at school who was cool in a way you could not have put into words. The smell of some room you no longer remember. Thousands of small exposures, each one trivial, together amounting to a vast set of leanings. None of it is stored as a story you can go back and read. It lives as a set of weightings, the reason one thing feels right and another feels faintly off before a single conscious thought has formed.
You can find the edge of this layer quite easily. Ask someone why they like something, and they will give a reason. Ask why they like that, and they will usually manage one more. Ask a third time and most people arrive, with a slightly puzzled look, at the same place: “I just do.”
It is tempting to treat “I just do” as a lazy answer, a failure to think harder. I suspect it is the most accurate thing said in the whole exchange. It marks the precise boundary of conscious access. Beyond that line the machinery is still running and still producing the liking, but in a format the conscious mind was never built to read. Anyone who writes software will recognise the problem. It is a program trying to inspect the kernel it is running on, using the very system it wants to examine, which was never designed to be viewed from the inside.
The obvious next step is to try anyway. If the pull happens below the floor of awareness, what happens if you turn awareness downward and try to watch the floor?
I have tried this more than once, and it is less dramatic than it sounds, and also stranger. You sit quietly, and instead of attending to a thought you try to attend to whatever is producing thoughts. You try to catch a want at the instant it appears, before the press secretary has written anything down.
What happens, at least for me, is not clarity. It is a gradual loosening. The ordinary sense that “I am the one doing this” becomes faint and uncertain. I tried this once while typing this book out. My fingers went on moving across the keys, and for a few seconds they felt as though they belonged to someone sitting just beside me. Not a stranger. Someone adjacent. They were obviously my fingers, but the feeling of their being mine had come slightly unstuck, like a label starting to peel at one corner.
For a long time I thought of this as a purely philosophical experience. Then I learned about the senses neuroscientists have added to Aristotle’s list, and it looked quite different. Remember the sense of agency, the feeling that you are initiating your own movements, and the sense of ownership, the feeling that your limbs are yours. These are not facts you deduce. They are perceptions, produced by the brain in the same way it produces the colour red or the pitch of a note. And like any other perception, they can be turned up or down. After certain strokes, for instance, some people report feeling as though their own arm is being moved by someone else.
So the peeling label is not a mystical glimpse behind the curtain. It is something more ordinary and, I think, more interesting. The feeling of “mine” is itself one of the things the machinery produces. When you turn your attention hard enough onto it, you can watch that particular signal flicker, the way a word flickers into nonsense when you say it forty times. The self is not revealed to be absent. It is revealed to be, among other things, a sensation.
This points to a problem at the centre of the whole enterprise, and it is a problem I do not think anyone has solved.
To understand anything, there has to be someone who does the understanding. A subject. An “I.” So if you want to examine the machinery underneath you, the “I” is the instrument you need. But the “I” is also one of the things the machinery produces. It is not standing outside the system, looking in. It is part of the output.
This means you cannot simply take it out and use it to look, because the moment you try, the thing doing the looking begins to come apart. It is like trying to photograph a camera using only that camera. You can angle mirrors and get clever with reflections, but you will never catch it directly.
Buddhist contemplative traditions arrived at this problem many centuries ago, and their answer, broadly, was to follow it all the way down. Look long enough, they said, and the subject dissolves, and what remains is awareness with nobody in particular being aware. That may well be true. It is certainly one of the more elegant conclusions human beings have ever reached. It does come with a small practical difficulty, though. At the moment of discovery, there is no longer anyone around to write it up. The finding, as it were, eats the finder.
All of this is remarkably difficult to explain to another person, and I think the reason is built into language.
Consider how we describe what people do. “She left because she was bored.” “I bought it because it was on sale.” Every sentence of this kind has the same skeleton: a person at the front, making a decision, for a reason. The grammar assumes an author before the explanation has even started. So when you try to describe something that happens beneath the author, the sentence itself resists you. Your listener’s whole model of human behaviour begins with the agent, and to get them anywhere near this idea you would first have to take that model apart. Very few people are willing to sit through the demolition before hearing the point, and I cannot say I blame them.
There is also a second, rather comic difficulty. If preferences really arrive before reasons, then this idea cannot win anyone over on the strength of its reasons. It will only take hold in someone who is, for their own obscure reasons, already drawn to it. You cannot argue a person into it. You can only offer an invitation and see if it lands. In my experience the invitation that works best is a single, simple question: have you ever found yourself liking something before you had any reason to?
Most people say yes. What they do with that yes is up to them.
It would be easy to end a chapter like this feeling slightly robbed. If our preferences are assembled out of sight, our reasons are often improvised after the fact, and even the feeling of being the one in charge is a signal that can flicker. It can seem as though there is not much left for us to be.
I do not think that is the right conclusion, and the evidence itself suggests why.
Look again at the mere exposure effect. If preferences are shaped by what we are repeatedly exposed to, then the one lever we do seem to hold is the choice of what to expose ourselves to. We may not be able to decide, in the moment, to like something. But we can decide what rooms to stand in, what to read, whose company to keep, and over time the layer beneath the floor will quietly reorganise itself around what it finds there. It is a slow form of authorship, more like gardening than giving orders, but it is authorship all the same.
There is an obvious objection to this, and Chapter Four takes it up. Who decides which rooms to stand in? The wish to stand in a different room is also a pull, and it too arrives before its reasons. So the claim here is a modest one. I am not saying the steering comes from outside the machinery. I am saying the machinery can be steered, slowly, by what it is shown, and that the steering is made of the same material as everything else.
And look at choice blindness from the other side. The people who noticed the switch were paying attention. Attention is not unlimited, and it is not always ours to command, but it can be trained, and many contemplative traditions have some method for doing so. The press secretary will always be there, but it is possible to become a more sceptical member of the press.
It is a great deal of work to be the sole author of every wish you have ever had. It is much lighter to be something more like a reader: someone who notices what the machinery offers, raises an eyebrow at the official explanation, and chooses, as far as possible, what to read next.
And the machinery, it turns out, has been running for far longer than any of us. The pull that made me stop for a mockingbird, and made you pick up a book like this one, did not start with us. If you follow it down far enough, beneath the self and beneath the species, it runs into a much older question about why anything in the universe is pulled toward anything at all.
That is where the floor goes next.
Chapter Two
The Floor Below Choice
the reasonthe pullIf you ever want to watch human freedom in action, go to a technology conference.
Stand near the coffee and look around. You will see, within a few minutes, a remarkable degree of agreement. The same quilted vests. The same muted sneakers. The same laptop stickers, the same water bottles, the same carefully casual haircut, the same phrasing of the same three opinions about the future. Ask any individual why they are dressed the way they are and you will get a sincere answer about practicality. The vest is warm but leaves the arms free. The sneakers are comfortable. And every one of these several hundred independent thinkers will have arrived, through their own reasoning, at an identical outfit.
It would be unfair to single out the technology industry, because the pattern is everywhere. Walk into a gathering of academics, or punks, or finance people, or evangelicals, or climbers, or musicians, and you will find the same convergence in a different costume. People who belong to a group come to look like the group, dress like the group, and eventually sound like the group, sometimes down to the rhythm of their sentences. Even people who consider themselves above all this tend to dress in the recognisable uniform of people who consider themselves above all this.
Ask anyone in any of these groups why they like what they like and they will tell you. “I like how it looks.” “It’s comfortable.” “It’s just me.” They are not lying. The liking is real. But the answer does not explain what it claims to explain. It does not say why this particular person came to like precisely the thing that everyone around them also likes. The preference was there first. The explanation arrived afterwards and wrapped itself neatly around it.
That, stated plainly, is the idea this chapter is about. Much of what we experience as choosing is not choosing in the way it feels. A pull forms first, somewhere beneath awareness. Reasons arrive later and make a case for it. We experience the case as the cause. The order is backwards.
It is a large claim, and a claim that large deserves evidence. Fortunately, psychologists have spent decades collecting it, often in ways that are very funny if they are not happening to you.
Start with the simplest mechanism, which is exposure.
In the 1960s, readers of student newspapers at the University of Michigan and Michigan State started noticing small boxes on the front page, like advertisements, each containing a single strange word. The words were Turkish, or Turkish-sounding, and meant nothing to the readers. Some appeared many times over the weeks, others only once or twice. Anyone who asked the editors was told that the advertiser wished to remain anonymous.
The advertiser was a psychologist named Robert Zajonc. When the run was over, his team sent questionnaires to students asking whether each word seemed to mean something good or something bad. The words that had appeared most often were rated more positively. Nobody knew what they meant. Nobody had any reason to prefer them. They had simply been seen more.
Zajonc called this the mere exposure effect, and it has since been found with faces, shapes, paintings, sounds, characters from unfamiliar alphabets and a great deal else. Familiarity, it turns out, breeds not contempt but fondness. And one detail makes it especially relevant here: the effect tends to be strongest when people are not aware they were exposed at all. The preference forms most easily precisely where you cannot see it forming.
Now apply that to the conference. Nobody sat down and decided to love quilted vests. They were simply surrounded by them, month after month, in the company of people they admired or wanted to resemble, until one day a vest in a shop window looked obviously right. The reason, comfort, practicality, was perfectly true. It just was not the reason.
Exposure explains where some preferences come from. The next experiment is more unsettling, because it catches the explanation being made up on the spot.
Two researchers at Lund University in Sweden, Petter Johansson and Lars Hall, borrowed a technique from stage magicians, and published the result in 2005. They sat across a table from volunteers and showed them pairs of photographs of faces, asking them each time to point to the one they found more attractive. Then the experimenter slid the chosen photograph across and asked the volunteer to explain why they had picked it.
The trick was that on some trials, using sleight of hand with black-backed cards on a black table, the experimenter handed over the other photograph. The one they had not chosen.
Most people did not notice. No more than about a quarter of the swapped trials were caught, and the rest looked at a face they had just rejected and explained, with complete confidence, why they preferred it. The details were the best part. One participant reportedly said he liked the woman because she was smiling, though it was the face he had actually chosen that had the smile. Another reportedly said she liked the earrings. The face she had originally picked had not been wearing any. When the researchers analysed the explanations, the reasons given for the swapped faces were delivered with the same confidence, detail and emotion as the reasons given for faces people had really chosen. And afterwards, when asked, participants tended to insist they would have noticed a switch.
It would be comforting to think this only works on something as subjective as faces. So the same team set up a tasting stall in a supermarket. Shoppers sampled two jams and picked a favourite, then tasted it again and explained what they liked about it. Except that the jars were rigged, and the second taste was the jam they had rejected. Overall, about one in seven noticed at the time, one in five counting those who realised afterwards, about a third counting every kind of reported change. Even when the difference was enormous, cinnamon-apple swapped for bitter grapefruit, only about one in five shoppers noticed on the spot, and fewer than half noticed at all. The rest described the virtues of a jam they had just turned down.
The name the researchers gave this is choice blindness. We are not only unaware of where our preferences come from. We are often unaware of what we chose a moment ago, and we are fully prepared to supply reasons for it anyway.
The clearest picture of how those reasons get produced comes from a stranger place: patients whose brains had been, in a very literal sense, divided.
For a time, some people with severe epilepsy were treated by cutting the corpus callosum, the thick bundle of fibres connecting the two halves of the brain. It reduced their seizures and left them, in everyday life, surprisingly normal. But it also meant that the two hemispheres could no longer share information directly, and in the 1960s the neuroscientist Michael Gazzaniga, working in the lab of Roger Sperry, realised this offered an extraordinary opportunity. By flashing an image to only one side of a patient’s field of vision, he could deliver it to only one hemisphere. And in most people, only the left hemisphere can talk.
In one often-told experiment, a patient’s left hemisphere was shown a chicken’s foot, and his right hemisphere was shown a snowy scene. He was then asked to choose, from a row of pictures, the ones that went with what he had seen. His right hand, controlled by the left hemisphere, picked a chicken. His left hand, controlled by the right hemisphere, picked a shovel. So far everything made sense.
Then he was asked why he had chosen the shovel.
The talking left hemisphere had never seen the snow. It had no idea why the left hand had picked a shovel. But it did not say “I don’t know.” Without hesitation, it produced an answer: the shovel was for cleaning out the chicken shed.
Gazzaniga came to call this part of the brain the interpreter. Its job, as he describes it, is to build a coherent story out of whatever the rest of the brain is doing, and it does that job whether or not it has access to the real reasons. It is not lying. It genuinely believes the story. It simply was not in the room where the decision was made, and it would rather improvise than admit it.
I have come to think of the interpreter as a press secretary. Decisions are made in closed rooms, by committees it has never met. Then it walks out to the podium, faces the cameras, and delivers a confident, articulate account of the reasoning behind them. It is very good at this. It is so good that most of the time even you believe the press briefing. The split-brain patients are simply the rare case where someone slipped a photograph under the door, and we got to watch the secretary work without notes.
The split-brain patients are an extreme case. The most famous experiment on the same question in ordinary brains is worth mentioning for honesty’s sake, because it turned out to be less clear than its reputation.
The neuroscientist Benjamin Libet asked volunteers in 1983 to flex a finger whenever they felt like it, and to note the moment they became aware of deciding to do so. Recordings of their brain activity showed a slow build-up of electrical activity, called the readiness potential, starting several hundred milliseconds before the moment people reported deciding. For decades this was cited as proof that the brain decides before you do and your awareness is merely informed afterward.
More recent work has complicated that. In 2012 Aaron Schurger and his colleagues proposed that the readiness potential is not the signature of a hidden decision at all, but the average of the brain’s ordinary background fluctuations, the natural ebb and flow of neural noise. When you are told to move whenever you feel like it, you tend to move at the moment that noise happens to drift across some threshold. This reading has gained ground.
It is a good correction, and I do not want to lean on Libet more than he can bear. But it is worth noticing that the new explanation does not exactly restore the feeling of a sovereign self pulling a lever. It replaces “your brain decided first” with “you acted when the random background murmur of your brain happened to peak.” Which is, if anything, a stranger answer. The volunteers acted when the moment struck them. The phrase is apt. Something struck. The person arrives, if only just, afterwards.
So if preferences are not chosen in the way they feel chosen, where do they come from?
The honest answer is: from a layer of you that is not available for inspection.
Down there, beneath awareness, is more or less everything that has ever happened to you, mixed with everything you were born with. Your genes. The tone of voice your parents used. Whatever was on television in the afternoons when you were six. A particular kid at school who was cool in a way you could not have put into words. The smell of some room you no longer remember. Thousands of small exposures, each one trivial, together amounting to a vast set of leanings. None of it is stored as a story you can go back and read. It lives as a set of weightings, the reason one thing feels right and another feels faintly off before a single conscious thought has formed.
You can find the edge of this layer quite easily. Ask someone why they like something, and they will give a reason. Ask why they like that, and they will usually manage one more. Ask a third time and most people arrive, with a slightly puzzled look, at the same place: “I just do.”
It is tempting to treat “I just do” as a lazy answer, a failure to think harder. I suspect it is the most accurate thing said in the whole exchange. It marks the precise boundary of conscious access. Beyond that line the machinery is still running and still producing the liking, but in a format the conscious mind was never built to read. Anyone who writes software will recognise the problem. It is a program trying to inspect the kernel it is running on, using the very system it wants to examine, which was never designed to be viewed from the inside.
The obvious next step is to try anyway. If the pull happens below the floor of awareness, what happens if you turn awareness downward and try to watch the floor?
I have tried this more than once, and it is less dramatic than it sounds, and also stranger. You sit quietly, and instead of attending to a thought you try to attend to whatever is producing thoughts. You try to catch a want at the instant it appears, before the press secretary has written anything down.
What happens, at least for me, is not clarity. It is a gradual loosening. The ordinary sense that “I am the one doing this” becomes faint and uncertain. I tried this once while typing this book out. My fingers went on moving across the keys, and for a few seconds they felt as though they belonged to someone sitting just beside me. Not a stranger. Someone adjacent. They were obviously my fingers, but the feeling of their being mine had come slightly unstuck, like a label starting to peel at one corner.
For a long time I thought of this as a purely philosophical experience. Then I learned about the senses neuroscientists have added to Aristotle’s list, and it looked quite different. Remember the sense of agency, the feeling that you are initiating your own movements, and the sense of ownership, the feeling that your limbs are yours. These are not facts you deduce. They are perceptions, produced by the brain in the same way it produces the colour red or the pitch of a note. And like any other perception, they can be turned up or down. After certain strokes, for instance, some people report feeling as though their own arm is being moved by someone else.
So the peeling label is not a mystical glimpse behind the curtain. It is something more ordinary and, I think, more interesting. The feeling of “mine” is itself one of the things the machinery produces. When you turn your attention hard enough onto it, you can watch that particular signal flicker, the way a word flickers into nonsense when you say it forty times. The self is not revealed to be absent. It is revealed to be, among other things, a sensation.
This points to a problem at the centre of the whole enterprise, and it is a problem I do not think anyone has solved.
To understand anything, there has to be someone who does the understanding. A subject. An “I.” So if you want to examine the machinery underneath you, the “I” is the instrument you need. But the “I” is also one of the things the machinery produces. It is not standing outside the system, looking in. It is part of the output.
This means you cannot simply take it out and use it to look, because the moment you try, the thing doing the looking begins to come apart. It is like trying to photograph a camera using only that camera. You can angle mirrors and get clever with reflections, but you will never catch it directly.
Buddhist contemplative traditions arrived at this problem many centuries ago, and their answer, broadly, was to follow it all the way down. Look long enough, they said, and the subject dissolves, and what remains is awareness with nobody in particular being aware. That may well be true. It is certainly one of the more elegant conclusions human beings have ever reached. It does come with a small practical difficulty, though. At the moment of discovery, there is no longer anyone around to write it up. The finding, as it were, eats the finder.
All of this is remarkably difficult to explain to another person, and I think the reason is built into language.
Consider how we describe what people do. “She left because she was bored.” “I bought it because it was on sale.” Every sentence of this kind has the same skeleton: a person at the front, making a decision, for a reason. The grammar assumes an author before the explanation has even started. So when you try to describe something that happens beneath the author, the sentence itself resists you. Your listener’s whole model of human behaviour begins with the agent, and to get them anywhere near this idea you would first have to take that model apart. Very few people are willing to sit through the demolition before hearing the point, and I cannot say I blame them.
There is also a second, rather comic difficulty. If preferences really arrive before reasons, then this idea cannot win anyone over on the strength of its reasons. It will only take hold in someone who is, for their own obscure reasons, already drawn to it. You cannot argue a person into it. You can only offer an invitation and see if it lands. In my experience the invitation that works best is a single, simple question: have you ever found yourself liking something before you had any reason to?
Most people say yes. What they do with that yes is up to them.
It would be easy to end a chapter like this feeling slightly robbed. If our preferences are assembled out of sight, our reasons are often improvised after the fact, and even the feeling of being the one in charge is a signal that can flicker. It can seem as though there is not much left for us to be.
I do not think that is the right conclusion, and the evidence itself suggests why.
Look again at the mere exposure effect. If preferences are shaped by what we are repeatedly exposed to, then the one lever we do seem to hold is the choice of what to expose ourselves to. We may not be able to decide, in the moment, to like something. But we can decide what rooms to stand in, what to read, whose company to keep, and over time the layer beneath the floor will quietly reorganise itself around what it finds there. It is a slow form of authorship, more like gardening than giving orders, but it is authorship all the same.
There is an obvious objection to this, and Chapter Four takes it up. Who decides which rooms to stand in? The wish to stand in a different room is also a pull, and it too arrives before its reasons. So the claim here is a modest one. I am not saying the steering comes from outside the machinery. I am saying the machinery can be steered, slowly, by what it is shown, and that the steering is made of the same material as everything else.
And look at choice blindness from the other side. The people who noticed the switch were paying attention. Attention is not unlimited, and it is not always ours to command, but it can be trained, and many contemplative traditions have some method for doing so. The press secretary will always be there, but it is possible to become a more sceptical member of the press.
It is a great deal of work to be the sole author of every wish you have ever had. It is much lighter to be something more like a reader: someone who notices what the machinery offers, raises an eyebrow at the official explanation, and chooses, as far as possible, what to read next.
And the machinery, it turns out, has been running for far longer than any of us. The pull that made me stop for a mockingbird, and made you pick up a book like this one, did not start with us. If you follow it down far enough, beneath the self and beneath the species, it runs into a much older question about why anything in the universe is pulled toward anything at all.
That is where the floor goes next.
Click a page or use ← →
Something in us moves before we do.
Mid-Pour is not published yet. If you want to know when it is, send a note and I will write back when there is news.