Knowledge and certainty

[Sound of phone ringing.]

You: Hello?

Voice on other end of line: Hello there! — Bill Oddie here! Did you know there are reports of a rare bird in your area? It’s a yellow-bellied flycatcher. Why don’t you step outside and see if you can catch a glimpse of it?

You: Er… OK, but what should I look for?

Bill: Well, first of all, it’s a bird. As well as that, it has a distinctive yellow belly. And on top of that, it behaves in a rather unusual way: it catches flies!

You: Wait a minute… I think I’ve seen one of those already! Down at the end of my garden there’s a pond, and in that pond lives a green amphibious thing, which has a yellow belly, and it catches flies! When it’s not swimming, it jumps around. Does this yellow-bellied flycatcher thing jump?

Bill: No, no, that sounds like a frog. You’re looking for a bird.

You: Oh… it absolutely has to be a bird?

Bill: Yes, it absolutely must be a bird!

You: So is it what you might call a “necessary bird”?

Bill: Well, I’m not sure what you mean by that, as I don’t believe in natural necessity, but it is “necessarily” a bird in the sense that it wouldn’t count as a yellow-bellied flycatcher unless it was a bird.


The above exchange is straightforward enough, but it exemplifies a misunderstanding that has constipated epistemology — the philosophical study of belief and knowledge — for centuries. Ever since Plato wrote Theaetetus, one of his deepest dialogues, it has generally been agreed that to know something, first, you have to believe it, and second, it has to be true. We might think of those two conditions as the “first two conditions on knowledge”. But there has to be something else — Plato suggested that you have to be able to be able give an “account” of your belief, so you don’t just believe it by accident or for the wrong reasons. A more recent way of putting that is to say that you must be “justified” in your belief.

Even if we accept that “third condition on knowledge”, it should not be understood as a demand for certainty. But over and over again, it has been understood as just that, for much the same reasons as you thought the yellow-bellied flycatcher was a “necessary bird” in the telephone conversation above.

The mistaken idea is that to know something, you “cannot be mistaken” about it. Well, I agree — to count as knowledge, a belief must be true, or in other words it cannot be false. But words like ‘must’ or ‘cannot’ here signify that a criterion is to be met, not that there can be no uncertainty in the mind of the believer, or that there is no conceivable way he might be mistaken. It just means that as a matter of fact, if he happens to know something then he happens not to be mistaken, because if he were mistaken, what he had wouldn’t count as knowledge.

Alas — too often, philosophers have understood words like ‘must’ in this context to mean that we can only know “necessary truths”, whatever they may be. Presumably, they’re things we “cannot be mistaken” about, which as far as I am concerned means nothing at all, because we can be mistaken about absolutely everything. To me, the idea of a “necessary truth” is as wonky as the idea of a “necessary bird”.

“Philosophy is a battle against the bewitchment of our intelligence by means of language”, wrote Wittgenstein, and this is a good example of a battle that the intelligence of very intelligent men — such as Descartes — has repeatedly lost.

The concept of “moral rights” ain’t right

Most philosophy students and academics have become so habituated to thinking about morality in terms of rights that a “denial of moral rights” sounds to them like a rejection of morality itself. If you read what follows, please resist that thought.

To understand why I reject the idea of “human rights”, or any other variant of the concept of moral rights, whatever we call it, let us first be clear that I’m completely in favour of a wide variety of legal rights, robustly enforced by a sound legal system. For example, in the UK a pedestrian has a legal right to cross the road at a “zebra crossing”. That right exists by virtue of a legal rule which says that drivers stop (or ‘must stop’) for pedestrians. The rule is explicitly written down in black ink on page 64 of the Highway Code. This rule describes a particular kind of behavior — namely, stopping for any pedestrian who starts to cross. As a bit of language, it consists of symbols that have to be interpreted. These symbols are quite often interpreted by courts of law (for example, a person wheeling a bike is a pedestrian, but what about someone cycling at walking speed?). The behavior it describes (and by virtue of the context, thereby prescribes) is habitually observed by most drivers, and real legal sanctions are brought to bear against those that don’t. They get points on their licences. Some go to jail.

That big complicated legal apparatus is necessary just to create the simple little legal right that a pedestrian has to cross the road at a zebra crossing. Much more is required for more complicated legal rights — such as the right to utter one’s thoughts freely (assuming we are speaking of a legislation in which such a legal right exists, as it surely does in a few places like the USA).

My main point here is that rights are carved out by rules, in conjunction with all of the machinery required by rules to work properly, including the physical symbols the rules are encoded in, the interpretation required to assign those symbols a (workably determinate) “meaning”, and so on. Where there are no such rules, there are no rights.

We know where to look to find legal rules that create legal rights — they’re printed in black and white in the statute books. But where do we look for the rules that create moral rights? — Nowhere, because they simply don’t exist, nor does any of the other machinery needed to make rules work properly, i.e. the apparatus that make them genuine rules rather than something spelled out via some variation of Kant’s Ouija board.

People used to talk about “God’s Law” and “Natural Law”, but these were obviously part of a fantasy — a fantasy woven by the mistaken idea that morality and the law are “pretty much the same thing”. This is a backward idea. We might expect such confusion from primitive tribal societies who spend their lives imagining ghosts, and unquestioningly accepting the judgements of hereditary “chiefs” about everything. But we should expect much more from advanced Western civilization.

Nowadays, the fact that rights are carved out by rules is routinely overlooked. Why? — The word ‘rights’ is used as a substantive, as if it refers to abstract things, like prime numbers perhaps. Such a thing is thus habitually conceived of as “figure” rather than “ground”, if you follow my metaphor. The more people use the word ‘rights’ like that, and the lazier people get about asking what these abstract things are, the more firmly entrenched our new fantasy becomes. It is one of the many ways in which language bewitches our intelligence.

Even if there were some way of divining abstract moral rules out of thin academic air, a la Kant, and thereby of coming up with fairly determinate moral rights, there would still be countless further problems. Plato had a brush with one of them. (He didn’t have the concept of a moral right, so it was a “brush” rather than a direct hit.) When rights conflict, there has to be some way of deciding between them, in other words of deciding how one right should “trump” another right in this or that set of circumstances. To make that decision, something other than rights must be appealed to. With legal rights, this poses no real problem because we have law courts and a highly structured legislature. With moral rights, we have to refer to some other moral “court of appeal” than rights. But then this other court of appeal is the real place where moral questions are answered.

Plato used that argument to dispose of the divine commands theory, the idea that morality is a matter of following God’s commands. And it also happens to drive a wooden stake through the heart of all such concepts as “moral rights” or “human rights”. These ideas refuse to die among the general public, of course, because most people are thick. But if philosophers can genuinely leave divine commands theory behind, isn’t it about time we did the same with the concept of “moral rights”?

Time to leave the “theater”

Plato thought that most of us do not really see objects outside our own heads. Instead, he thought that all we really see are mere representations of things — mental representations that exist entirely within our own minds. According to a famous passage in Plato’s Republic, these representations are like shadows on the walls of a cave, which remain inside the cave, even though they are cast by things outside the cave.

Other philosophers such as Descartes and Kant had similar ideas. Kant thought that what we really observe are not things as they really are ( = “noumena”) in the world outside our minds, but mere mental representations of those things inside our minds ( = “phenomena”). To Descartes, the mind is like a unitary “eye” of consciousness that looks at a “screen” of conscious experiences like a moviegoer in a cinema. It is anyone’s guess whether what’s happening on the “screen” is anything like the physical world outside the cinema.

Although this idea is very common among philosophers of nearly every tradition — it probably preceded Plato, and certainly continued long after Descartes — it was so clearly and compellingly expressed by Descartes that it is usually associated with him more than anyone else. The contemporary American philosopher Daniel Dennett dubbed it the “Cartesian Theater” view of the mind, and the name seems to have stuck.

The idea is very widespread, but it is deeply and insidiously mistaken. Remarkably, it survived almost completely unquestioned for centuries. Philosophers only began to cast real doubt on the idea in the twentieth century, when the American Pragmatists, the later Wittgenstein, some existentialists, and a few other distinctively “twentieth century” philosophers began to see the mind not as a “centre of consciousness” so much as “what the brain does” — it actively and directly engages with the world, as part of an evolved animal’s inherited physical equipment.

Here is a typical expression of that newer understanding from Donald Davidson (a recent American philosopher of the Pragmatist tradition): “we have unmediated touch with the familiar objects whose antics make our sentences true or false”. In other words, my belief that the cat is on the mat is true if one familiar object (the cat) and another familiar object (the mat) are arranged just as the sentence ‘the cat is on the mat’ says they are arranged. Davidson is expressing nothing more than modest common sense here, but it is remarkable that common sense needs the voice of a great philosopher to be heard.

The most important thing to note here is that it isn’t internal ideas — some sort of conscious experience of the cat or the mat — that are so arranged. It is the actual physical external-to-my-mind cat, and the actual physical external-to-my-mind mat, both of which exist independently of me, outside my head. These are external physical objects rather than internal mental ideas or experiences. Davidson felt it necessary to explicitly say that we have “unmediated touch” with these objects, because the alternative “Cartesian Theater” view is so all-pervasive, and it has been around for ages.

Why did it last so long? Why does it still have such a hold on the popular imagination? The answer is that in effect it assumes the mind is non-physical. Whether we are aware of it or not, we are all steeped in a long religious tradition that understands the mind as a non-physical “soul”. The soul is supposed to survive the death of the body (including the brain). Our religious tradition understands the mind as a sort of “economy” whose currency is not brute physical pushes and pulls, but conscious experiences. For example, it is widely assumed that if a person knows anything, then he must be “justified” in believing it — in other words, he must have a conscious assurance that it is true. Non-conscious physical pushes and pulls are not a valid currency in this economy, because it is assumed that there is something deeply problematic about the mental and the physical coming into contact with each other at all. As supposed “centres of consciousness”, our minds are thought not to be in direct contact with the world. Far from having Davidson’s “unmediated touch” with the physical world of everyday objects, our minds can only reach out with enormous difficulty to a world that is quite alien to us. Some think that the difficulty of bridging the gap is so enormous that the mind “never quite makes it” — so we live in a Teletubby world of our own imaginings. The physical world is an alien planet to our mental selves, the idea seems to suppose. As mental rather than physical beings, our proper place is therefore somewhere “above” the messy physical realities of death, sex, chemicals and parasites. Diseases are assumed not to be caused by physical invaders, but are instead the self-inflicted wounds of mental imbalances. And so on.

The idea that our minds do not engage with the world directly has some terrible consequences. Some of them are mere curiosities, of interest mainly to academic philosophers. For example, Thomas Kuhn (the recent American historian of science) said that scientists who work in different scientific traditions “live in different worlds”. The idea here is that each scientist’s world consists of his own ideas. (But as Davidson remarked, “there is at most one world”, and we all live in it.) Kuhn’s metaphor is striking, of interest to people who wonder about science, but it doesn’t affect ordinary people’s lives much. However, the “Cartesian Theater” idea has other consequences that really do have a bearing on ordinary life.

To see why, consider the difference between perception and volition. With perception, the world affects our minds by causing ideas to form in them, whereas with volition it’s the other way round: our minds affect the world through our actions. Although the “causal flows” of perception and volition go in opposite directions, the two are closely analogous.

When applied to perception, the “Cartesian Theater” idea says that we do not perceive things in the world directly, but instead only perceive things internal to our own mental economies — such as experiences of things rather than physical things themselves. Analogously, when applied to volition, the “Cartesian Theater” idea says that we do not desire that things in the world be arranged in any particular way, but instead desire that our own mental economies come to achieve a particular internal state. The almost universal idea here is that we don’t act in order to arrange things in the world, but instead we act in order to get a particular kind of experience, typically called “pleasure” (as if this were a uniform type of experience).

For example, suppose I want to build a tree-house for my children. Someone who believes in Freud’s “pleasure principle” might say that what I “really” want is pleasure — the pleasure of seeing my children enjoying themselves, or perhaps something less innocent than that. A follower of Nietzsche might say that what I “really” want is a sense of power. Instead of wanting to have some bits of wood arranged in a physical location outside my head, both of these approaches assume that my real goal is something “internal” to my own mind.

But I ask you: Do you want the mere experience of your lover being faithful to you? Or do you want your lover to genuinely be faithful to you? Do you want to have an erotic dream? Or do you want the real thing?

To put the idea another way, the “Cartesian Theater” idea says that when we hunt rabbits, we are really hunting pleasure. It is very common throughout the academic world. For example, economists (right across the spectrum from Marxist to libertarian) tend to assume that rational agents are seeking the most money or profit in all of their transactions, because money delivers pleasure, and at root we are motivated to seek pleasure. Or again, psychologists tend to assume that if we seek goals that involve pain, we must be masochists for whom pain is a form of pleasure. And so on, right across the groves of academe.

Perhaps the worst damage that this terrible “Cartesian Theater” idea does to ordinary lives is in the area of love and sex. For example, a large proportion of people who get married end up getting divorced. The ones who initiate divorce are usually people who have decided that they have “fallen out of love”. They usually check whether they are “still in love” by examining their own “feelings”. But as artists from Shakespeare to Chris Rock remind us, love always involves “negative” emotions, including feelings that are incompatible with the very feeling that is supposed to signal the continued presence of love. No married couple has ever lived together for any stretch of time without some such negative “feelings”. A much better test of love than introspected “feelings” is the extent to which two people’s lives are actually intertwined — how much they routinely do together, especially how often they have sexual intercourse (with each other). If all you do is look at the screen in the “theater”, much of what you see will be illusory.

In my view, the “Cartesian Theater” idea is hopelessly misguided, all the more poisonous and damaging for being so widely held. Apart from a narrow strand of academic philosophy, and a slightly broader strand of common sense, it remains an unquestioned bit of religious dogma. Essentially, it is the idea that the mind is isolated from the physical world because it is “spiritual” rather than physical. It is a hugely mistaken view of what we are, and why we do things. It has corrupted philosophical thought for centuries, and continues to wreak untold havoc in everyday life all over the world.

It is time to get ourselves out of this “Cartesian Theater”. We must walk out into the daylight, blinking and disoriented, and know the place for the first time.

Are there any “alpha men”?

Today’s topic is filed under the category “Mind and evolution”. I have purposely avoided the term “evolutionary psychology”. Although the latter ostensibly aims at understanding the mind from an evolutionary perspective, it has so far done little better than extend the methods of academic psychology a bit beyond its traditional “blank slate” view of the mind. I reject those methods because I regard them as unscientific.

As an example of the sort of sloppy thinking that evolutionary psychologists get up to, consider the popular idea that there are “alpha male” humans. The idea is all wrong, I think, and it’s misinformed by bad logic and bad ideology.

It’s bad logic, because it’s really an “argument by analogy” gone wrong. In an argument by analogy, two things are noted as sharing various characteristics, and then a further characteristic of one of them is extrapolated to the other. Here’s an example of a perfectly good argument by analogy:

A has four legs, and B has four legs
A barks, and B barks
A fetches sticks, and B fetches sticks
A wags its tail
Therefore, B wags its tail as well (probably).

That is a pretty good argument, because tail-wagging is a typical characteristic of dogs. (It’s an inductive argument rather than a valid deductive argument, if anyone’s interested.)

Here’s an example of a bad argument by analogy:

A has four legs, and B has four legs
A barks, and B barks
A fetches sticks, and B fetches sticks
A responds to the name “Hieronymus”
Therefore, B responds to the name “Hieronymus” as well (probably).

This is a bad argument because responding to the name “Hieronymus” is not a typical characteristic of dogs. Arguments by analogy work when the characteristics in question are truly representative of their respective kinds.

The supposedly “scientific” grounds for thinking there are human alpha males follows a similar pattern:

A has a humanoid body (bipedal, hands, face, etc.), and B has a humanoid body
A uses tools, and B uses tools
A has genes x, y, z,…, and B has genes x, y, z,…
A belongs to a species that has alpha males
Therefore, B belongs to a species that has alpha males (probably).

The reason why that’s a bad argument by analogy is that there is no such thing as a “typical ape reproductive strategy”. The reproductive strategy of a given ape species can never be representative of apes in general, because apes have diverse reproductive strategies. For example, gorillas are polygamous, but gibbons are monogamous. Even very closely related species such as chimpanzees and bonobos have strikingly different sexual arrangements and different societies. These reproductive strategies are diverse in the same way as dogs’ names are diverse in the above example.

Some people assume that the genetic closeness of humans and chimpanzees means we are “really” chimpanzees in disguise. Scratch Tarzan and you’ll find Cheeta underneath. But very small genetic differences can lead to very great behavioral and functional differences. There are half a dozen species of Galapagos finch, and they are genetically very similar to each other. But their eating habits are completely diverse. Their similarities in other respects (similar habitat, similar metabolic rate, etc.) drove the diversification of their eating habits. They had to diversify in order to exploit the limited food supply.

Instead of relying on a “sophisticated” argument, evolutionary psychologists should simply observe the living world. The living world of humans, that is, not of chimpanzees. If they did, they would see that human life is full of conflict and competition, but there’s nothing like a single sexual hierarchy. It’s more a feeding frenzy than a pecking order. Women are attracted to rich, clever, tall, charming men, with no single characteristic “trumping” all the others. Henry Kissinger might charm the pants off one woman, yet repel another. One woman’s Shaggy is another woman’s George Formby.

I mentioned above that the human “alpha male” idea was bad ideology as well as bad logic. What I meant was this. Some people find the idea very attractive. Men like it because it suggests that men are “tough”. Women like it because it suggests that men are stupid.

Underneath the surface, there’s a sort of Hitler–Stalin pact between laddish men and feminists. The laddish men embrace the idea that the proper role for men is “slugging it out in the world of competition”, whereas feminists embrace the idea that women remain in charge of the home, where they have always dominated. These supposed “opposites” have remarkably similar agendas, and remarkably similar ideologies, which helps explain why they dislike each other so much.

The ghost in the machine

Most of the world’s population thinks that consciousness continues after death. The “default” condition of Man is to believe that your self or soul will survive the death of your physical body.

To manage that, the self or soul would have to be rather special. It would have to be distinct in some way from the rest of the physical world. A vast amount of bad philosophy begins with the assumption that the mind, self or soul – it doesn’t matter what we call it as long as we understand it as an individual’s center of consciousness and agency – is not part of the physical world at all. The body may be a purely physical machine, but the mind commanding it is a “ghost in the machine”.

The idea we are considering here is very widespread, but it got its most rigorous treatment in Descartes’ dualism – his theory that there are two “substances” (i.e. two kinds of stuff that can exist on their own). One of them is matter, whose distinctive characteristic is that it is extended in space (i.e. it takes up room) and the other is a more mysterious “immaterial substance”, whose distinctive characteristic is that it is conscious.

Now for the bad news. Cartesian dualism is probably mistaken. (Cartesian  = “of Descartes”.) There is just too much perfectly obvious interaction between the physical world and the mind to take it seriously. For example, getting a knock on the head or drinking a bottle of gin (i.e. physical events) affects the mind. A decision to take the bus (i.e. a conscious mental event) affects your physical body (because it causes you to stand at a bus stop). If physical events routinely cause mental events, and vice versa, surely any mental event is just a special sort of physical event?

Almost all present-day philosophers would answer Yes. Most of them have abandoned the “default opinion” of Man. However, this is easier said than done. In trying to slough off the default opinion, they have to try to embrace a rather unnatural new opinion. So it is hardly surprising that many of them have only half-embraced it. Despite openly professing to reject Cartesian dualism, many continue to suffer from its after-effects – the many misguided and misleading assumptions that are part of the dualist package.

I was once invited to give a speech to some academic psychologists. I gave my talk the title “The Worst Hangover of All Time”. This was meant to suggest that although we are trying to “sober up” from the effects of Cartesian dualism, having explicitly rejected it, most of us have yet to completely de-toxify our minds of its noxious assumptions.

For example, it’s reasonable to think that everything that happens is caused to happen. But if so, are all the decisions we make determined by conditions that existed before we were born? How then could we be genuinely free? – This old problem of “free will versus determinism” is widely held to have been solved by philosophers such as Hobbes and Hume, who gave much thought to cause and effect. Our decisions and actions are caused, they said, but being caused is different from being coerced. Only if we are coerced are we unfree. Many people find this solution unsatisfactory. But their dissatisfaction is the result of an unreasonable assumption, a symptom of our Cartesian hangover. We are wrong-footed by a mistaken expectation: we expect genuine freedom to mean not being part of the causal fabric of the universe at all. In other words, we assume that being free means having a non-physical mind. The “ghost in the machine” may be the “pilot”, but we don’t much like the idea of the pilot being “strapped into the pilot’s seat”.

As another example, consider motivation. What motivates us to do anything? The almost universal answer – and the wrong answer, in my opinion – is to say that we do things in order to get pleasure. Suppose I build a tree house, which involves physical discomfort of various kinds. Why do I do it? – The standard answer is that I am going for “deferred” pleasure. As soon as I see my children enjoying themselves in the tree house I built, I will get pleasure, the thing I’m “really” after.

I don’t think so. I think I undergo the physical discomfort because I want to have a tree house. I don’t want a distinctive sort of experience. I want things in the world to be arranged in a particular way – planks of wood nailed to branches of a tree. The reason why we assume we do things for pleasure is we suppose we are disconnected from the physical world. We suppose our minds are “self contained” centers of experience, with their own internal economy of pains and pleasures, rather than an integral part of the physical world. The “ghost in the machine” supposedly has its own “spiritual” ends that exist wholly within its separate realm of consciousness.

As a final symptom of our Cartesian hangover, consider knowledge. It is widely held that to know anything, we must have a true belief that is “justified”. What does it mean for a belief to be “justified”? – Supposedly, it must be based on experience. To know that there is an orange in front of me, say, I must have the experience of an orange-colored disk in my visual field. This assumption leads to trouble, as Descartes himself was famously aware. What if I am merely dreaming that an orange is in front of me, and in reality I am lying asleep in bed?

These considerations lead some people to think that knowledge is impossible. But instead of rejecting the possibility of knowledge, we should reject the supposition that all knowledge is “based” on conscious experience. An insect, which probably has no conscious experience at all, can form a reliable “map” in its head of the world outside its head. That is a rudimentary form of knowledge, and it depends on reliable causal connections between its brain and the physical world outside its brain. Those causal connections are the very thing that a “ghost in the machine” cannot have.

I already mentioned the bad news: death is annihilation. But there’s a little bit of good news too: there probably aren’t any ghosts, either in the machine or anywhere else.

Am I atomic or what?

The Ancient Greeks believed in atoms. Or rather, they believed in the existence of what they called “atoms” – tiny particles of matter that “cannot be cut” or subdivided any further. What we call atoms can be “split” into smaller constituent parts – protons, electrons, neutrons, etc.. But the old idea of un-cut-able a-toms is still intuitively compelling. Suppose we do keep cutting matter into smaller and smaller parts: will we eventually reach rock-bottom, particles that cannot be subdivided any further? If so, they would be the fundamental constituents of matter. Scientists are currently trying to find out if they can isolate the most basic particles of all, using unnatural amounts of energy to do so. We’ll keep our fingers crossed.

A vaguely analogous project was undertaken by Descartes in his Meditations. He was trying to isolate the most basic constituents of knowledge. – Huh? – It seemed to him that everything we know must be “justified” either by being basic and certain itself, or else by being based on something more basic and certain.  So he began to “dig” down into the everyday beliefs he had hitherto considered as knowledge, by actively doubting absolutely everything it was possible to doubt. He would accept nothing except what had to be accepted because it could not be doubted. First off, he couldn’t claim to know anything about the world outside his own mind. He might think he was sitting by the fire in his dressing-gown with a piece of paper in his hand, but he might only be dreaming that he’s sitting by the fire in his dressing gown. And so on…

Descartes thought he had hit rock-bottom when he tried to doubt the existence of his own self. He could doubt the existence of his own body – after all, he might be a ghost – but he couldn’t doubt that he was a thinking, above all experiencing center of consciousness. “What is this ‘I’ that I am?” he asked. His answer: I am a thinking thing. Cogito ergo sum – “I think, therefore I am”.

By the way, what Descartes really meant by “I am a thinking thing” is “I am a conscious, sentient thing.”

Although that may seem undeniable at first, there is more to it than first appears. Descartes wasn’t just claiming that “thinking is going on”, but that his self was thinking – he was assuming the “‘I’ that I am” is an indivisible, “atomic” center of consciousness.

And that certainly can be doubted. Some time after Descartes, Hume toyed with the idea that his self was just a bundle of experiences of various kinds. He never sees his “self” as a single subject having those experiences. That was a real departure. Much later, in the twentieth century, some patients with extremely severe epilepsy underwent an operation which in effect prevented direct communication between the left and right sides of their brains. Consciousness itself was not significantly diminished, apparently, but it certainly was disrupted, and in ways that strongly suggest our “atomic” consciousness is nothing of the sort. The ease with which these patients were able to “multi-task” became rather an inconvenience for them. Various experiments and observations revealed that at times their “left hands didn’t know what their right hands were doing”. And yet most of the time their overall mental integrity remained intact. It shows that when the rest of us “multi-task” in the usual way, our brains rule our bodies more like a “parliament” than a “king”. And we all “multi-task” all the time. (Even men.)

The main lesson I learn from Descartes’ certainty that he was “atomic” is this: we feel subjectively certain about whatever we are familiar with. But a subjective feeling of certainty is not objective certainty, which is anyway impossible. To be a philosopher, a person must try to lose his familiarity with familiar things. He must force himself into exile, as it were, so that he is a stranger in a strange land.

There are many lessons to be learned from Descartes’ Meditations, many of them the very opposite of what Descartes intended. But that is one lesson that he did intend.

Being a pig versus being Socrates

Famously, JS Mill wrote:

It is better to be a human being dissatisfied than a pig satisfied; better to be Socrates dissatisfied than a fool satisfied.

But almost ten years before he wrote those words in his 1863 essay Utilitarianism, Mill wrote the following in an entry in his diary:

Quality as well as quantity of happiness is to be considered; less of a higher kind is preferable to more of a lower. The test of quality is the preference given by those who are acquainted with both. Socrates would rather choose to be Socrates dissatisfied than to be a pig satisfied. The pig probably would not, but then the pig knows only one side of the question: Socrates knows both.

I think these earlier words are historically fascinating, and philosophically profound.

The passage is interesting historically for two reasons. First, it reveals something of Mill’s personality. He was clearly the sort of guy to mull over an idea or analogy before committing it to print. In this, he was like his contemporary Darwin, who took twenty years rather than Mill’s ten. Second – and the reason why it is so philosophically profound – the passage represents both a bridge to, and a break with, the traditional hedonism of the Epicureans.

Epicurus and his followers were hedonists – that is, they judged the value of everything in terms of pleasure. (Not just their own, I hope I don’t have to explain.) For example, a morally right act would be one that increased pleasure for all concerned. Similarly, in the theater, a good play would be one that generated a lot of pleasure in the audience.

For various reasons, the Epicureans had to draw a distinction between “higher” and “lower” pleasures. (Roughly, this is the distinction between comfort and joy.) For example, relief from thirst was a higher pleasure than getting drunk. To a thirsty, non-abstemious person, a cold, strong beer would have hit both spots. And so on.

Mill’s diary entry echoes the Epicurean distinction between “higher” and “lower”, but something absolutely crucial has changed. He has started to talk about happiness rather than pleasure. There’s much, much more to happiness than pleasure.

Pleasure is a type of conscious experience, but happiness is a type of situation to be in. The former has everything to do with the way thingsseem to be arranged. The latter has everything to do with the way things really are arranged. In other words, the latter has a lot to do with luck. (Words like ‘happen’ and ‘happenstance’ have similar Icelandic origins. I know this because I once walked past an Icelandic Lotto outlet with a language expert.)

My father was a boy of eleven when the Second World War began, and an adolescent of seventeen when it ended. He always spoke of those years as the “happiest days of his life”. He was being funny and extravagantly amoral, as always, but he was honestly expressing a truth too: we are happy when we are doing things that we consider worthwhile, however little pleasure we may get as an experiential “reward” for doing them.

Mill’s understanding of happiness is subtle and complicated – as befits a subtle and complicated subject. It has a lot to do with what we count as the “quality” of life. Here’s a reminder of the way Mill judges the quality of the life of Socrates and the life of a satisfied pig:

The test of quality is the preference given by those who are acquainted with both.

There is a theory of value taking shape here. Mill would regard nothing as inherently valuable, i.e. as valuable in itself. Something is valuable only because a conscious agent prefers it to something else. If I’m prepared to sell my gold ring for a glass of water, then a glass of water is more valuable to me than a gold ring.

As far as I know, Mill’s diary entry is the first expression of preference utilitarianism, the form of utilitarianism based on this “relative” understanding of value. Traditional hedonistic utilitarianism counted pleasure as inherently (i.e. “absolutely”) valuable.

Reflections on light

When we look at medium-sized occurrences, such as balls colliding on a pool table, what we always actually see is a large number of particles interacting with each other. That involves the dissipation of motion. The “break” at the start of a game of pool or snooker is the most obvious example of the dissipation of motion, but of course it occurs at the microscopic level as well. A rolling ball knocks aside molecules of air, which absorb some of its momentum. They speed up, and it slows down. When the balls collide with each other, more dissipation occurs. They get distorted slightly, which causes the plastic molecules they are made out of to squeeze together, which makes them push each other apart – always a bit too much, which causes them to retract again, and so on. In other words, they vibrate, which means they heat up slightly, and of course they make a noise when they hit each other.

In short, macroscopic events involving large numbers of particles always involve the dissipation of motion. That’s why no pool ball will ever keep rolling forever, nor can it be perfectly elastic.

Imagine various physical processes in the universe. Some of them are symmetrical with respect to time, and others are not. For example, a perfectly elastic ball that bounces up and down over and over again – to the same height with every bounce – will keep going forever. If you made a movie of that happening and played the movie backwards, it wouldn’t look different from the proper forwards-playing movie. In the real world, of course, no balls are perfectly elastic, because their motion is dissipated, creating noise and heat.

Or again, if you took a movie of the entire solar system, with all of the planets going round the Sun in one direction, and then played it backwards, the only obvious difference is that the planets would go round in the other direction. As long as you ignore the weather and the tides on the planets – processes that really do involve the dissipation of motion – the most basic mechanism of the solar system is time-symmetric.

Now to speculate: suppose that all of the time-asymmetry in the universe is the result of the dissipation of motion. If there is a very close connection between dissipation of motion and time-asymmetry, then processes in which there is no such dissipation of motion might all be time-symmetric.

Most events that occur on a microscopic scale are indeed like that, as long as they only essentially involve very small numbers of particles. Their time-symmetry helps to explain much of their strangeness and unfamiliarity to our eyes – eyes that are accustomed to the dissipation of motion in familiar macroscopic processes, and completely unaccustomed to time-symmetry.

For example, consider the behavior of light particles. Light changes direction when it passes from one medium into another (such as from a vacuum into glass, or from air into water). Now it so happens that the path light takes in travelling between any two points – from A to B, say – is the path that takes the least amount of time to traverse. That is remarkable. – But it can be explained in terms of time-symmetry.

when light passes from one medium to another, its path bends

Consider the path that leads from A to B in the diagram above. It might be the path that a lifeguard A takes as he runs across the sand and into the water to rescue B, a swimmer in difficulty. Or it might be the path that swimmer B takes as he runs out of the water to apprehend thief A, who is rifling through the pockets of the clothes he left on the beach. Both of those paths have a kink in them because of the constraint that they be traversed in minimum time (which they are, roughly, as long as the people involved know they can move faster over sand than through water).

Such a path might also be the one that light follows as it travels from A to B, with the light travelling more slowly in the lower medium than in the upper medium.

How does light “figure out” where it should end up? If we assume that the movement of light is time-symmetric, this is a mistaken question. The departure point has the same “status” with respect to time as the destination point. So there isn’t any question about how light “knew” where to start off from, and there there isn’t any question about how it “knew” where to end up. The only apparent question is how it “knew” which intermediate points to take as it “chose” its path between them.

Recall our assumption of time-symmetry. If light took a path that did not take the minimum time, it would have to stray off that path, to the right or to the left, say. Suppose it strays off the path to the left when traveling from A to B. Then, looking at that path with the time direction reversed, in other words from the future towards the past, it would seem have to strayed off the path to the right when travelling in the other direction, from B to A. That is a violation symmetry. If it strayed off to the right instead, the whole process would lose its time-symmetry, which demands in effect that processes look the same from either time-direction. In other words, if the process is time-symmetric, then straying off to one side when the path is traced from past to future entails straying off in the same direction when the path is traced from future to past. The same applies when we consider its straying off in any direction. Straying to the right, or up, or down, or anywhere off the shortest path is a violation of time-symmetry.

In effect, the path the light “chooses” is like a thread that is drawn as tight as it can be between two end-points. As with all “tightness” of threads, there is no “slack”, with a slight twist: slack in loose cotton thread is entirely spatial, because any slack means its length in space could be shorter; “slack” in a beam of light would be “spatio-temporal” because any such slack here means its duration (i.e. its length in time) as it passes through space could be shorter. The time-symmetry of light particles entails that no such “slack” can exist with light.