phil-181: Philosophy and the Science of Human Nature
Lecture 9 - Virtue and Habit I [ February 8, 2011 ]
Chapter 1. Norms, Laws, and Habits [00:00:00]
Professor Tamar Gendler: So today's lecture is about the claim that we've been exploring in the context of Aristotle's ethics, that the way to cultivate virtue is by cultivating in ourselves certain sorts of habits. And I want to start by showing you a picture of a T-shirt, which was popular at MIT in the mid-'90s when my husband was a graduate student there. And what it said on the T-shirt was, "Gravity: It's not just a good idea, it's the law."
Now, if you guys were MIT students, you would be rolling on the floor. [laughter] As it is, social beings that you are, you're rolling on the floor at the thought that other beings are rolling on the floor at the thought of this. [laughter] But what's interesting about this T-shirt is that it brings out a distinction between two kinds of laws. Because gravity isn't a law that tells you how you ought to behave. It's a law that tells you how things do behave.
And philosophers make a distinction between two kinds of laws. There are, on the one hand, normative laws, oughts, things that tell you how you how you should do things, things that express, as the name indicates, norms. Those are things like, “look both ways before crossing the street.” That's something you ought to do. It wouldn't be funny have a T-shirt that said, "Look both ways before crossing the street: It's not just a good idea, it's the law." Right? It could be the law, and the T-shirt, for lots of reasons, wouldn't be funny.
It wouldn't be funny to have a T-shirt that said, "Don't eat in the library: It's not just a good idea, it's the law." Because it could be a law in the library that you not eat. And it wouldn't be funny to have a T-shirt that says, "65 miles per hour: It's not just a good idea, it's the law," because in fact, it’s a normative law.
So normative laws express summative judgments about the way things ought to be. They are laws in the sense that you find over at the Yale Law School.
But in addition, there are laws of a very different kind, spread around the rest of campus. Spread around, in fact, everything that you ever do. And those are descriptive laws. They tell you the way things actually are.
So it would be sort of funny to have a T-shirt, "If a car hits you, you will die. It's not just a bad thing, it's the law." Because "If a car hits you, you will die" is a description of the way the world is. It's a fact about the world that is a law in the sense that it is in a position to allow you to make predictions about the future on the basis of the past. It tells you about law-like relations between things in the past and things in the future.
Likewise, it's a law of biology plus chemistry plus physics, roughly, that “crumbs cause book decay.” That's a description of a fact about the world to which the normative law, “don't eat in the library”, might correspond. But they are nonetheless very different claims.
And finally, whereas it's a speed limit on your car that you drive 65 miles an hour, it's a speed limit on everything that it not go faster than 186,000 miles per second. The speed of light expresses a descriptive law about how fast you can go, whereas the speed limit “65 miles an hour” expresses a normative law about how fast you may go.
Now, why did I start out this lecture on Aristotle and habit with a bunch of remarks about the difference between normative and descriptive laws? The reason is this. Habits are tools for turning oughts into ises. Habits are ways of taking normative commitments that we have about the way we want things to be, and making use of the fact that we are psychological, biological, chemical, physical beings in whom patterns of descriptive law-like relations can be created by repeating the same activities over and over.
So when Aristotle says, "We learn a craft by producing the same product that we must produce when we've learned it. We become builders by building, harpists by playing the harp, in the same way we become just by doing just actions, temperate by doing temperate actions, brave by doing brave actions,” he is explaining to us the utility of recognizing the connection between normative laws on the one hand--ways we think we want to be--and descriptive laws on the other--ways that we find ourselves naturally becoming.
Remember, when he describes for us how it is that we cultivate virtues, he contrasts that to two cases where merely descriptive laws apply. He said, it's not like trying to train a rock to stay in the air, because that's a case where a descriptive law applies. Nor is it like a case of watching a plant unfold when given water and light, because that, again, is a case where merely descriptive laws apply.
What's interesting about the middle realm on which almost the entirety of the Nicomachean Ethics focuses is that it's the domain where the principle that I've just articulated holds. It's the domain where it's possible for us to think about the ways we want things to be, to act as if things were already that way, and in a self-fulfilling manner, to have things become that way.
So Aristotle's basic insight is that if you want to become something, act as if that is what you already were. If you want to become instinctively brave, act as the brave one does. And then it will become natural to you.
If you want to become a piano player, train your fingers to act as the piano player's fingers do. You learn the craft of piano playing by producing the product that you must produce when you have learned to piano play. That is, you learn to become a piano player by practicing the piano.
So patterns of behavior that are initially under conscious control can, through a process of repeated practice, become automatized. So initially, when you learned how to drive a car, you had to think very carefully about what to do with each of your feet and each of your hands. Any of you who went to dancing school in junior high know that when you learn how to dance, you start off by counting: “One and Two and…” I won't dance on stage, because this is going forever on the Internet. [laughter] But those of you who learned how to dance in middle school know that behavior that was initially under conscious control became automatized, such that those of you who were trained to do waltz and tango will now, upon hearing the music of waltz and tango, have a kind of motor routine activated in your feet.
The fact that this happens inevitably to biological beings like ourselves gives us a tool for turning normative commitments into descriptive laws. So it has become the case for me that though it began as a norm, “look both ways before crossing the street,” it's now a description of me: that I look both ways before crossing the street.
And any of you who has ever been to England or Australia--countries in which people drive on the left, rather than the right--know how incredibly difficult it is to overcome that ingrained habit. In fact, when I'm in England, I look both directions about thirty times, because I'm so disoriented by the fact that my routine doesn't fit the situation in which I find myself.
If you grew up driving a standard shift car or an automatic car and switch to the other, it's incredibly difficult to make the changes. When we become habituated to a certain pattern of behavior, something that was initially a normative rule for us becomes a descriptive one.
In fact, one of the main goals of parenting is to instill in one's children instinctive responses that accord with one's reflective commitments. I want it to be the case that when handed an item, without reflection, automatically, my children say, "Thank you."
Now, the same capacity that allows us to turn normative commitments that we reflectively endorse into habitual practices can, of course, be deployed in the reinforcement of habits which we wish to get rid of. So for many of us, it is the case that upon opening one's computer, there is an immediate compulsion to open one's Internet browser, and an immediate compulsion to check one's Facebook page.
Now, we talked already in the very first class about one of the ways of dealing with this, which is to eliminate the connection that takes you from the computer to the Facebook page. If you turn off your Internet browser, then no matter how instinctive the action is, you won't be able to respond to the compulsion. If you have an instinctive tendency to go down to your refrigerator at midnight and drink the full fat chocolate milk that's there in the fridge, if you take the milk away, then you don't have to change the habit.
So one of the strategies for self-regulation involves limiting access to the response that you wish to get rid of. But sometimes either the response that we want to get rid of isn't something that we want to eliminate entirely. Right? I don't want to get rid of all the food in my kitchen. Some of you don't want to get rid of Internet access simpliciter on your computer.
Chapter 2. Aristotle on Habituation [00:12:31]
So the question that we're going to consider in lecture today is what additional strategies are available for breaking the link between unwanted habits, or the link between cues and the unwanted habits to which they give rise.
So let me ask you to take out your clickers. These are the only times we're going to use the clickers today, but I promise in later lectures we'll use them in less contrived ways. So I want to ask you. Which is true of you? That you have no habitual behaviors that you would like to change? Everything about you is perfect? You're like the figure in Alan Kazdin's opening chapter who wants to change everybody in the world, but who needs to change nothing in him- or herself? Or are you somebody who has at least one habitual behavior that you would like to change? And we'll keep polling open for another six, five, four, three, two, we've got 110 responses, and let's see how it comes out.
OK. So 8% of you are perfect. I'm thrilled. I've always wanted to have a class full of perfect people. At least I have, I guess it's that back left-hand corner. [laughter] But 92% of you have at least one habit that you would like to change.
For the 92% of you who have that habit that you'd like to change, a second question. So for any of you who has ever gotten rid of a habit and thought to do so, which is true of you? Were you able to change that habitual behavior just by saying to yourself, “You know what? I'm not going to check Internet in class anymore.” “You know what? I'm actually going to practice my violin every morning at ten.” “You know what? I'm not going to leave my bed unmade in the morning.” So how many were able to change your habitual behavior just by talking yourself out of it, and how many of you were not able to change that habitual behavior by talking yourself out of it? And again, we'll cover this for another ten. OK. And let's see how the numbers come out. Four, three, two, one. OK.
Oh my goodness! 35% of you can talk yourself out of habitual behaviors. But the other 65% of you are embodied human beings of the sort whom I took myself to be lecturing in this class. OK.
It is very often the case, though obviously not always the case, that simply trying to talk yourself out of an unwanted behavior is extraordinarily difficult. And the chapter that we read from Jonathan Haidt, and in fact, the readings that we've been doing all semester long, explain why this is so.
Part of the reason that at least 65% of you fall into the category of people who are unable to talk yourself out of an unwanted behavior is that we aren’t just composed of reason. Plato gave us the metaphor of reason, spirit, and appetite. Haidt gives us the metaphor of the rider and the elephant. Every single one of the authors that we've read so far has talked about the ways in which we are fundamentally processing information at a rational level that may or may not reach down to the other aspects of what Plato calls our soul.
So Aristotle, in those incredible closing five pages of the Nicomachean Ethics that I had you read for today, the very last bits of the last book, writes as follows. He says, "If arguments were sufficient in themselves to make people decent, the rewards they would command would justifiably have been many and large. But," he points out, "in a large majority of circumstances," simply saying to people, "Hey, you know what? You probably ought to pay your taxes by sometime mid-spring," doesn't cause them to act in keeping with what it is that they want, or you want, them to do.
Arguments alone, appeals to rationality alone, work in exceptional cases. But they don't work all the time.
Now, because I've been glossing over it in most of the material that we've been reading from the ancient authors, I want to point out that the intervening texts, the texts that I'm skipping over in the rest of 1179, is actually an expression of something that runs through every single one of the ancient Greek texts that we've been reading, which is an expression of a certain kind of cultural elitism about the difference between the well-born and the many.
But I want to, right now, set aside what is, I think a legitimate ground for challenging some of what Aristotle's saying, and focus instead on what I think is true about what he goes on to remark, which is, that it is impossible to alter by argument what has long been absorbed as the result of one's habits.
And Aristotle goes on then to discuss something that we'll talk about in about five weeks. Namely, given this fact about human beings, that early experience shapes subsequent behavior, and that putting regulations in place can shape behavior in ways that is pro-social, it appears that there are implications for how societies ought to be structured. And when we get to the unit on political philosophy, we'll look yet again
at these closing pages of the Nicomachean Ethics .
What I want to point out to you now is that there it is an extraordinary connection between what Aristotle is saying in this ancient text--here's a beautiful fifteenth century edition of the Ethics in Latin, here's the translation that we're using--and it seems to me, and this is the point that I want to make in the remainder of today's lecture, that the fundamental insight of Aristotle's ethics is what lies behind the literature in a certain kind of therapeutic practice--in particular, cognitive behavioral therapy--and I want to show you how this plays out in a particular kind of self-help book. In particular, a parenting guide.
The claim I want to make is roughly that Aristotle's Nicomachean Ethics is the greatest parenting guide ever written. So those of you looking for a baby present for your newborn niece or nephew, look no further! [laughter]
So what I want to do is to contextualize cognitive behavioral therapy by starting off with a discussion of the animal literature of which it is a part. Behavioral therapy exploits the fact that we are evolved beings, continuous with non-human animals, in the kinds of control over behavior which are available to us. So with apologies to those of you who have taken already Psych 110, a brief introduction to Pavlovian conditioning.
Oh! Sorry. With no apologies to anybody, some more quotes from Aristotle. Just as Aristotle says, "We learn a craft by producing the same product that we must produce when we've learned it. We become builders by building, harpists by playing the harp," when Aristotle says that, what he is pointing out is the importance of structuring your experience in such a way that you have the opportunity to perform the behavior that you wish to cultivate. Give yourself the opportunity to do the things that you want have become habitual, and those things will become habitual.
Chapter 3. Classical and Operant Conditioning [00:21:37]
All right. Here's the dog. So as I assume most of you know, but it's worth reviewing, there is a technique of regulating animal behavior, which has of course been known for thousands of years. You hear report of the training of domestic animals in the ancient Greek tradition, and in fact, Descartes has a long discussion of how you can train dogs.
(Descartes had idiosyncratic views about animals. He thought that they were machines. But with regard to this hypothesis, that mistaken assumption doesn't make a difference.)
So the way classical conditioning works, is that before the conditioning takes place, you identify an item--say, food--that produces in the being that you wish to condition a response of salivation, in this case, or excitement. So the dog sees the food, and goes, “wow!”
So the food, in this case, is what psychologists call an unconditioned stimulus. It's something that, without intervention on our part, produces is in the being whom you're trying to condition, an unconditioned response. You show the food, and without any intervention on our part, the dog salivates.
By contrast, there are lots of things in the world--bells, for example--which produce in the dog no reaction at all. You ring a bell. Dog goes, “Big whoop. Who cares? It's a bell.” So in that case, you present what is called a neutral stimulus. and what you get from the dog is what is called no response .
Now here comes the conditioning. You take your dog, and you present him with, simultaneously, the bell and the food. And the presence of the food is sufficient to produce in the dog the "Oh, wow". Right? The dog says, “Food! Salivate! Excellent! Yum! Something to eat!”
And so what you've done is, you've taken paired stimuli, the bell and the food, and produced, as a result of the introduction of the paired stimuli, an unconditioned response that is, at this point still, a response to the food. But animals are associative beings. And eventually, this unconditioned response, the "Oh, wow" that results as a matter of hard-wiring from the introduction of the food, comes to be paired with the neutral stimulus. And when that happens, when the bell is in a position to produce in the dog the response of salivation, one has thereby conditioned the stimulus to produce in the animal a conditioned response.
Now, it's not just with animals that this sort of thing is possible. In the opening pages of the Alan Kazdin textbook that you read, we heard the story of poor Baby Albert who used to think white rabbits were no big deal, until the neutral stimulus of a white rabbit got paired with the unconditioned stimulus of a loud noise that produced terror in him. And when that happened, Albert came to associate what had previously been neutral, the white rabbit, with a feeling of fear.
Human beings are capable of coming to have associations that have valence, either positive or negative, associations that produce in them positive or negative emotional responses. And they are capable of coming to have those associations with entities that previously had no value to them.
So with that fact in mind, psychologists began thinking more generally about the relation between objects and behaviors, on the one hand, and outcomes that those objects or behaviors are associated with, on the other. So whereas classical conditioning is concerned with the passive consumption of some item in the world--food is presented to you, a rabbit is presented to you, and a certain association arises--operant conditioning is concerned with the relation between behaviors that you yourself perform and the outcomes to which those behaviors typically give rise.
So we have, in our behavioral repertoire, a bunch of things we do that are desired behaviors. It's a desired behavior that you do your reading, that you go to the gym, that you're polite to your roommates, that you call your parents at least once a week. And we also have undesired behaviors. Things that we wish we didn't do. The bad habits that 92% of those of us in this room have.
And what psychologists noticed is that we have resources available to us using the mechanisms that conditioning provides for increasing differentially various types of behavior. In particular, if we are attentive to the consequences that behaviors typically give rise to, and if we intervene to control those consequences, we can increase the incidence of certain kinds of behavior, and decrease the incidence of other kinds of behavior.
So if there's a desired behavior that you wish to have more of, you can pair with that behavior as a consequence some sort of reinforcer. You can pair with it a positive reinforcer. I can reinforce your efforts on writing a paper by providing you with the positive reinforcer of a letter on your transcript that has signaling value to future employers. So I can provide you with positive reinforcement for desired behavior. Or one can reinforce a desired behavior through what is technically known as negative reinforcement. The removal of an aversive stimulus. If there's noise in the room, you are negatively reinforced in your tendency to put in earplugs if the putting-in of the earplugs reduces the aversive consequence of being in a noisy room.
So we can increase behaviors that we want to increase by associating with them consequences that serve to increase the value, subjectively, that that behavior has to the individual, as the result of pairing something that the individual likes already with the behavior.
We can, in addition, decrease the likelihood of an undesired, or in fact of a desired, behavior by associating that behavior with some consequence that is a negative consequence. So we can take away something that you like if you act in a way that we don't want. And we will, in the final two lectures before, or just after March break, talk about punishment and how it works.
So behaviors can be associated with positive consequences, or they can be associated with negative consequences. And sometimes, that happens naturally. So when I eat chocolate cake, that is associated with the positive consequence of the flavor that it provides me. It is positively reinforced, because the behavior, eating the cake, gives rise to a natural consequence, the taste of the cake. But it may not be so salient to me that the behavior also gives rise to a consequence which I don't evaluate so positively. It increases the amount of arterial blockage on the way to my heart.
Making salient to people what the natural negative consequences of their behaviors are is one of the ways one can make use of this paradigm. So we can take behaviors that occur, that we want to have more of, and associate them with positive consequences. We can take behaviors that occur that we want to have more of and associate them with negative consequences. We can help make people aware of the positive or negative consequences that these behaviors already have. And finally, we may discover that unbeknownst to ourselves, a certain kind of reinforcer is, in fact, preserving a behavior for us that we wish to get rid of.
So a parent may discover that the reason a child is continuing to whine is because whining, an undesired behavior, is in fact associated with a reinforcer, parental attention. And one might come to recognize that the removal of that reinforcer will reduce the behavior.
So in this regard, operant conditioning takes our capacity for association and uses it in familiar ways to increase desired behaviors and decrease undesired behaviors. But, of course, behaviors don't arise out of nowhere. And the third thing that operant conditioning attends to are the antecedents that gave rise to behaviors.
So whether a behavior is likely to occur depends first of all on the setting events. On whether the reinforcer is likely to increase the probability of the behavior occurring. You will be more likely to drink chocolate milk late at night from your refrigerator if you are hungry. So if you drink a glass of water an hour beforehand, then the reinforcement that drinking the chocolate milk had won't be so great for you. You can set circumstances so as to make the reinforcers more or less effective.
You can also provide yourself or others with prompts. I want to help you do the reading, so I provide you with reading guides in which I direct you to the parts of the text which I think are most important, and I thereby reduce the barriers to entry that the readings might provide. I help prompt the behavior by making the task more manageable.
And finally, we can give rise to behavior differentially by the presence or absence of what are called discriminative stimuli. that indicate, in technical terminology, the availability of a reinforcer. So in the pages of Kazdin that we read today, Alan Kazdin points out that, if the behavior that we're interested in is the answering of one's telephone, that the ringing of the phone is an antecedent that makes the behavior more likely to give rise to the reinforcer. The reinforcer is, being able to talk to somebody at the other end.
So you're standing in front of your telephone. If the phone rings, there's reason to pick it up. If the phone doesn't, there isn't.
So what the framework of operant conditioning provides is a way of cashing out the Aristotelian suggestion that what we want to do is to structure our lives in such a way that the behaviors we want to have become part of our repertoire, become habitual. And one of the most effective ways to do that is to structure experience so that the thing we don't want to do is incompatible with an alternative behavior that we put in its place.
So in the context of, for example, parenting, Alan Kazdin says, “instead of thinking of your child's behavior in terms of what you don't want”--right? I don't want to have him whine; I don't want the siblings to be fighting; I don't want my children to be staying up past midnight--“Start thinking in terms of the behavior that you do want, and reward that opposite.”
So instead of saying to yourself, “I don't want to spend the evening talking to my roommate and ending up having to stay up all night,” say to yourself, “what is the behavior that I wish to cultivate and reinforce? What is the alternative that will preclude my acting in the ways that I don't want, and encourage me to act in the ways that I do?”
In so doing, you open up for yourself the most profound and lasting form of human self control. The form of human self-control that leverages the difference between what reason commits you to and what spirit and appetite may be pulling you towards. You come to associate a behavior that was previously unreinforced for you with positive consequences. “When you get rid of a behavior by rewarding its opposite. says Kazdin, the effects are stronger, “and last longer than if you punish the undesired behavior. The best way to build the behavior you want is through reinforced practice.”
Now, “the best way to build the behavior you want is through reinforced practice” is something that presumably you've heard before. In fact, you've heard it before three times in today's lecture. "We learn a craft by producing the same product that we must produce when we have learned it. We become builders by building. We become harpists by playing the harp."
So what we try to do in Aristotle's challenge, and in the contemporary analogue of it, is to structure our lives in such a way that we can take normative commitments--things that are good ideas--and turn them into descriptions of behaviors that come naturally to us. We take them from being good ideas into being laws.
And this provides, I think, new insight on the passage with which I closed the previous lecture, as well. Aristotle writes that "Abstaining from pleasure makes us become temperate, and that once we've become temperate, we are most capable of abstaining from pleasures. It is similar with bravery. Habituation in standing firm in frightening situations makes us become brave, and once we have become brave, we are more capable of standing firm."
So this, I think, is an undeniable aspect of human nature. But it isn't the full story. And the two essays that we're reading for next class bring out a complication to the story that I've told today, and they ask us stop to think about whether this entirely describes what it is that's required for bringing about the kind of change which we hope to achieve. And I look forward to talking to you about those texts on Thursday.
We have a couple of minutes for questions, if somebody--by my watch, we have three more minutes. I'll give you positive reinforcement if you raise your hand, then a query that you have will be answered! All right.