Sex, brains, robots and Buddhism: looking for free will
New Scientist vol 178 issue 2394 – 10 May 2003, page 46
How much free will do you think you have? Does understanding how cause and effect work in the brain undermine the very idea of it? What does it mean to talk about a “sex drive” as if it were out of our control? Are both robots and humans doomed to have only the illusion of free will? And why talk about free will at all if we only exist and act in relation to others? These were key issues for our panel at the Royal Society for Arts during a seminar held as an introduction to New Scientist‘s two-part series on human nature, starting next week
The panel members were Chris Frith, professor of cognitive neurology at London’s Institute of Neurology; Shere Hite, author of the Hite Reports and professor of gender and society at Nihon University in Japan; Owen Holland, senior lecturer in computer science at the University of Essex; and Geshe Tashi Tsering, a Tibetan Buddhist monk and head of the Jamyang Buddhist Centre in London. Keeping the proceedings focused was Simon Blackburn, professor of philosophy at the University of Cambridge and author of Think
When you spontaneously lift your finger, you are aware, first, of the urge to lift it and then, shortly afterwards, of lifting it. But the brain activity that goes with this simple action (as the famous work by Benjamin Libet shows), produces a time line for the physical event that is different from that of the mental event. On average, there is a change in brain activity almost a second before your finger “spontaneously” lifts, whereas the awareness of the intention to lift the finger happens considerably later. This has been replicated by my colleague, Patrick Haggard.
So does this observation eliminate the possibility of free will? I don’t think so. For me, the moment of freedom occurs much earlier. This delay in our awareness is to give us a sense of being in control of our movements.
There’s another aspect of the original Libet experiment. The awareness of initiating the act also happens at a different time from the physical event – earlier in this case, so you are aware of initiating the act about 80 milliseconds before your finger moves. The consequence is the intention and the action are pulled closer together in mental time compared to what happens in physiological time. So voluntary actions and their effects are experienced as being closer together in mental time than they are, while the opposite is true for involuntary movements. Haggard calls this “intentional binding”.
But what if, rather than being told to lift your finger at a certain time, an experimenter lets you choose from four responses? Is this really a free-selection task? When Libet tells you to lift your finger whenever you feel the urge, you’re well aware he would be cross if you never had the urge. So you’re selecting from a specific sub-category of responses. Your interpretation is to try to produce non-obvious responses so the experimenter can’t predict what will happen next. Patients with damage to the prefrontal cortex show “utilisation behaviour”: they can’t resist making the obvious response. When the French neurologist François Lhermitte took one such patient round his flat, he showed him the bedroom and the patient undressed and climbed into bed!
Now, once we’ve chosen from this restricted repertoire of responses, the selection of the individual response can be determined by the environment. We may decide to behave like a scientist, a citizen, or a hero, but having chosen, we can be driven by the environment. That is why I suggest that “free will” occurs before the selection of a particular action.
Is the decision to behave in any particular mode somehow predetermined? I can’t answer that absolutely, but it seems clear that someone with frontal lobe damage or an animal with a less developed prefrontal cortex will have less free will because their behaviour is driven by the environment in a much more direct way. Studies of the brain certainly don’t eliminate free will, but may specify a bit more what we’re trying to talk about.
In the discipline of intellectual history, which was my first field of study, we try to separate out the ideas that we learn and decide are our own (even if they’re not, but really ideas that society wants us to internalise) from those that we are truly expressing.
For me, the idea of the male “sex drive” is a classic example. It is discussed so often in the media as if it is a scientific precept – but I’m not so certain. It is hard to say if there is such a thing or if it is a product of ideology.
The common, clichéd view of men’s sexuality is that men’s bodies contain a powerful mechanism called “sex drive” connected to “male hormones”, and that sex drive makes men want to “penetrate” and “impregnate women with their seed”. Using the phrase “sex drive” seems to imply that sexual activity leading to reproduction is a biological imperative.
It may be that men really have a choice about how they want to express their sexuality. In fact, “sex drive” may be a deus ex machina, a concept designed to prove everything else, “since this, therefore that, obviously evident”… But there is no proof whatsoever for its existence. The only “evidence” is circumstantial.
If the “male sex drive”, the erect penis being biologically programmed to be attracted by the vagina of a reproductively aged female, exists “in nature”, then it should be possible to prove its existence scientifically. Hormones are usually cited as a “proof” for the existence of the male sex drive. But it must still be demonstrated: first, how do hormones cause desire for orgasm, other than that they and orgasm coincidentally exist, and secondly, how exactly do hormones cause men to “naturally” focus on “penetration of the vagina” of a woman, or “the reproductive act”?
The cause of a desire to orgasm is not scientifically known. “Hormonal urges for orgasm” and focus on coitus are separate issues. Men have fluctuating sexual hormones that may or may not coincide with desire for orgasm. It is an age-old question: just how mechanical is male desire?
Most men in my research feel their desire is largely inspired by a particular desired individual, or by particular images or a fantasy – it does not come automatically or “mechanically”, without their effort. Or if they get an erection at night during sleep, it doesn’t automatically make them feel they desire something or someone.
These questions science must answer. To simplistically assert: “It’s a question of men’s sex drive” doesn’t answer any questions. Using the concept of “sex drive” as a jack-in-the-box may be distorting both male and female sexuality. It puts everyone under pressure to have this mechanical idea of what is happening in the body.
Of course, if it can be proven that there is some mechanism we could call “sex drive”, I’d like to know about it. But so far we really don’t know.
Robots are usually seen as the paradigm case of things which obviously don’t and can’t have free will. Industrial robots certainly are automata, but I want to turn this negative view around by talking about another kind of robot, the biologically inspired robot designed to imitate humans and animals or to have an architecture containing abstractions based on our understanding of humans or animals.
The first of this breed was devised in 1948 by Grey Walter, a pioneering neuroscientist. The way Walter described their design is critical to our understanding of what robots are: “One of the elements of animal behaviour and human psychology, which the [robot] tortoise is designed to illustrate, is the uncertainty, randomness, free will or independence so strikingly absent in most well-designed machines.”
He discovered something else, too. People treat small robots physically as if they were small animals, and, importantly, they attribute to them intention, decision-making, moods. Fifty years on, the ease with which small robots encourage people to project animal qualities onto them has been exploited by large corporations – like Sony with its dog, Aibo.
And what about robots that mimic humans? The best known is Cindy Breazeal’s Kismet at MIT. This is a cartoon-like collection of plastic and metal in the form of a robotic head, with facial expressions, a voice of sorts, and head movements, all linked to a sophisticated internal model of emotion. Kismet is almost a caricature of a humanoid robot – but “she” produces interactions with humans that look and feel natural.
On the Web, there’s film showing a researcher who had worked in the lab for seven years, taking part in an exercise to “scold” Kismet – bad robot, no, inappropriate, bad! Kismet’s ears go down, her eyes go down, head goes down… The researcher turns and looks at the camera, absolutely horrified at the idea of hurting the robot’s feelings! But don’t call this an illusion – some people maintain it’s the real thing, that’s all there is. Go to any bar at 1 am, where it’s too noisy to hear, and you’ll see Kismet talking to Kismet…
Now whatever consciousness is, it does appear to be some kind of user illusion created by the brain for dealing with itself. And while a lot of what consciousness seems to be telling us is wrong, it’s still very useful. In The Illusion of Conscious Will, Daniel Wegner extends this by arguing that conscious [free] will is an illusion that allows us to track and identify the “authorship” of our actions, and to behave consistently. These illusions are engineering solutions from nature, to solve engineering problems which we don’t yet understand. But when our biologically inspired robots get sophisticated enough, we’ll copy those tricks. So robots won’t have free will but they will have the illusion of it, the same illusion that we have.
As a roboticist, I’ll settle for that. As a human being, I’m not so sure.
Geshe Tashi Tsering
In Buddhism, free will is not discussed. One of the main points in Buddhism – known as dependent arising, or dependent origination – is that everything and every event, including human existence, comes into existence dependent on others. My existence and your existence is dependent on others. So free will is not really discussed.
As a human being, my existence is a combination of material matter and consciousness. When this combination is put together, I can say I exist, I function. But if I search within this combination of matter and consciousness, can I find anything that I can describe as a “me” or an “I”? No, I cannot find it. I am a process, a combination of mind and matter. If we go beyond that, and try to find something within it that we can call a “me” or “I”, we cannot find it. Our feelings, such as of happiness of sorrow, also come into existence as a result of causes and conditions.
For me, free will seems very much connected with the concept of the god or the soul. But Buddhism doesn’t believe in that. There isn’t within ourselves something we can call a soul. Simply, everything comes into existence through causes and conditions. Given the right causes and conditions, things will come into existence, events will happen. But nothing exists independently, or inherently. Everything, including our identity, is dependent on others.
Exercising the prerogative of the chair, I would say I agree with Shere Hite that talk of “drives” is often doubtful. It leads people to think that drives aim at their own extinction, which is a mistake. Sexual desire aims at sex, not at absence of desire.
Then I must say I agree with Geshe Tashi that the “interventionist” idea of free will – that the real “me” can stick its finger in and change the direction of the body-brain mechanism – is untenable. It postulates a self as a kind of extra-physical “extra”. And I was intrigued that Chris talked about neurophysiologists searching for the moment of freedom. That strikes me as reflecting the Cartesian or interventionist view.
Most philosophers think of free will more in terms of “reason responsiveness”. What it means to regard you as free, as opposed to being in the grip of some delirium or schizophrenia is that I expect you to respond to reason in a certain way: if you don’t, then I don’t regard you as free and I think of you as a phenomenon to be managed or avoided like bad weather. This is the compatibilist idea of free will. There isn’t a moment of freedom, there’s just the ongoing fact about a person that they are or are not responsive to reason and so responsible for what they do.
University of Birmingham
It hasn’t been said clearly enough that there are different interpretations of the phrase free will. I’ll mention four: there are probably a lot more. There’s the theological notion, that it’s God’s excuse for the nastiness in the world – he didn’t do it, it was down to our free will. Another notion is the romantic one of people wanting to feel they’re initiators rather than products of things, and this often comes out in poetry and novels. Another is a legal and social idea and has to do with the conditions under which you can be punished, blamed or absolved of something you’ve done.
Then there’s another notion waiting to be invented, which is the scientific version. Once upon a time, there were very simple organisms which unlike other physical objects had a store of chemical energy inside them which they could use to resist some external forces and to make selections between them. Gradually evolution found more and more sophisticated ways of building mechanisms that could enable these organisms to take information and use it later. To be like a human being you have to have a very large collection of different capabilities that were produced at different times in our evolution. Owen’s robots have a long way to go but there’s no reason why they shouldn’t get to the same state as us. Then they’ll have what we have: the ability in some sense to control our own destiny instead of being subject to physical forces acting on us.
University of Sheffield
It seems in much of this discussion there exists the notion of a subconscious or a pre-conscious or a not-so-very conscious, not necessarily in the Freudian sense. As a forensic psychiatrist I’m often trying to make a judgement about how responsible someone was for their action, and it’s evident that a rational choice model is inadequate, because the last person who knew what they were doing was the person doing it. One is then in the area of something pre-conscious or unwilled. This kind of thinking is certainly not amenable to the scientific interpretation we might like.