An evolutionary psychologist with a focus on language, Steven Pinker is the author of several bestselling books, including The Language Instinct, How the Mind Works, Words and Rules, and The Blank Slate. No stranger to controversy, in The Blank Slate Pinker challenged the view that all people are born equal, instead arguing that genetics shapes much of personality and predisposes people towards processing information certain ways. He teaches at Harvard and is an active researcher as well as a popular public lecturer.
Pinker's latest book is The Stuff of Thought: Language as a Window into Human Nature, which Wired calls "a fascinating look at how language provides a window into the deepest functioning of the human brain." One rainy afternoon, Dr. Pinker stopped by to discuss causality, the concept of concepts, how to swear in several languages, and the way irregular verbs can lead to romance.
Doug Brown: To start off, I have a general question about a point of ignorance. I come from a field biology background, where you do research by going out and looking at animals doing things, and sometimes by doing experimental manipulations. How does a linguist do research and make contributions to the literature?
Steven Pinker: Well, I'm a psychologist, not a linguist, so my stock in trade is bringing people into the lab and flashing words and pictures at them on a computer screen, and asking them questions and seeing how fast they respond. I also will go out to day care centers with a bag of ducks and bunnies and teach kids new words or ask them to act out sentences. I have also done neuroimaging work where, in collaboration with others, we stick people's heads in a brain scanner and see what happens to their brains as they're converting a verb into the past tense, or a noun into the plural, or repeating it.
Most recently, together with a student, I've been doing intracranial recordings. Epilepsy patients have a set of electrodes implanted in their brain in order to record activity to see where seizure originates, and they sit around waiting to have a seizure. When they do, the computer has recorded their brain activity for the period beforehand, and therefore the surgeon can detect where the seizure originates and extirpate just that little spot of the brain without removing healthy tissue. While they're sitting there waiting to have a seizure, they're bored; they have all this data coming out of their brain, and they're happy to participate in our language tests. So we can get the kind of recordings that ordinarily you can only get from animals. A college student comes in for beer money and you can have him look at pictures on a screen, but you can't ask him to let us drill a hole in his skull and sink an electrode into his brain. (Laughter) This is the next best thing.
There's also the main method of real linguists in linguistics departments, which is simply to use themselves as subjects to craft sentences or words and just set down their own reactions on the assumption that their reactions are likely to be the same as any English speaker's. I do a fair amount of that as well, and throughout the book, in cases where it's obvious to any English speaker that this sounds weird, this sounds natural, you can mine a lot of implications out of that.
Doug: In terms of language acquistion, is there a critical period in the early-mid teens past which the brain has a harder time learning foreign languages?
Pinker: Yes, without an accent, and without the fine points of grammar and conjugation instinctively mastered. In fact, there's a decline even earlier, but it seems to kind of fall off a cliff in early adolescence.
Doug: Does it seem to be primarily the rules that are harder to acquire, or the words as well?
Pinker: I suspect the rules more so. We learn new words all our lives. Not just new vocabulary items like modem, reboot, defragmenter, and so on. But also anytime you meet a new person, you hear, "I'd like you to meet Marvin Klempfish," and if it sticks in memory, you've learned a new word.
Doug: While some of your previous books, such as How the Mind Works and The Blank Slate, were broad explorations of human behavior, The Stuff of Thought is more grounded in language.
Pinker: It covers the same territory, more or less, but uses language as the main source of evidence. Just as How the Mind Works was divided into sections on cognition, perceiving, and thinking, a section on emotion, and a section on social relationships, that's also true of The Stuff of Thought. Here, though, the section on cognition is a discussion of nouns, verbs, prepositions, and tense markers as entry points into people's concepts of matter, causality, and space and time, respectively. That's the cognition part. The part on emotion is an exploration of swearing, and the part on social relationships is an exploration of innuendo and euphemism in indirect speech.
Doug: What was the genesis for the book, and did it change shape along the way?
Pinker: It did change as I was working it out and writing it. Originally I thought of concentrating on the inventory of the basic thoughts out of which all complex thoughts are composed. That is, could you come up with an alphabet of thought, so to speak, which provides the raw materials for all our complex thoughts, where everything we can think is either an assembly of simpler concepts or a metaphorical abstraction of concrete concepts? So when we say that the economy rose and fell, we're taking a physical source, that is where an object is in space, and we're extending it to a much more abstract domain. A lot of that ended up in one chapter, chapter five, called "The Metaphor Metaphor," about the metaphorical basis of thought, and a little bit in the final wrap-up chapter. But I realized that I had a lot more to say than just that, and it became clear to me as the book was taking shape that it was language as a window into human nature more generally.
Doug: In the first part of the book you examine some alternative models of the relationship between language and thought. I was struck by a parallel between Extreme Nativism and Intelligent Design, in that both appear to rest upon faulty assumptions of irreducible complexity. Would you say this is a fair comparison, or no?
Pinker: It's interesting; I hadn't thought of that parallel, but it's not unreasonable. The logic is irreducible complexity, that is true — that is, things that have no parts that could not have been assembled. In the case of Intelligent Design, it's organisms that could not have been assembled over the course of evolution. In the case of Extreme Nativism, it's concepts that couldn't have been assembled during the course of childhood learning.
This is the position, just to flesh it out, that rather than say the verb kill, meaning cause to become not alive or the verb paint, meaning cause paint to be on — this would be the idea that to paint is an irreducible concept; it doesn't have any parts. Likewise the verb kill is an irreducible concept. If it is irreducible, then it has to be innate, according to this line of thinking — one that I do not agree with. Neither that it's innate, because that would be evolutionarily unlikely, nor that it's irreducible. I spent a good chunk of the early part of the book showing how word meanings are built out of parts, and you can dissect a word meaning and see simpler elements in it like cause, change, goal, manner, act, and a few others.
Doug: As someone known as an evolutionary psychologist, do you find yourself being pulled into Intelligent Design debates?
Pinker: Yes. I usually try to hand over the assignment to someone who's more prepared for that style of debate. But I have occasionally weighed in. For example, two years ago, Time magazine wanted contributions from four viewpoints — around the time that President Bush offered the opinion that he thought Intelligent Design should be taught in schools. They had a fundamentalist Christian minister; they had a Christian spokesperson who believed that Christianity was compatible with evolution; they had a biologist who was a devout believer, namely, Francis Collins — the head of the Human Genome Project; and then they had an undiluted, died-in-the-wool mechanistic, scientific world-view proponent — namely, me. Each of us got to write about two paragraphs, so it was kind of a soundbite debate on the page.
Doug: How do you respond to critics of evolutionary psychology who say evolutionary psychologists are simply making up just-so stories to justify their preconceived notions about human nature? What are the elements that make evolutionary psychology science?
Pinker: You test hypotheses to see whether they are true or not. You derive predictions from theories, which are generally derived both from the mechanics of natural selection and knowledge of the kind of environment in which we evolved. For example, an environment without money, without police force, without medicine, without written language. Via engineering analysis we determine what kind of system a priori is capable of attaining a particular goal in a particular environment — that is, what are the requirements of stereo vision, what are the requirements of maximizing gene transmission in a polygynous versus a monogamous mating system. All of those predictions are derived before you actually go out and look at a human being. But then, once you have an optimality analysis — from computer vision, genetics, mechanical engineering, optics, whatever the system you are looking at is — you see how well the engineering specs match the observed characteristics of the human being. The closer the match, the greater confidence you have that the adaptive hypothesis is true.
Lots of adaptive hypotheses don't pass muster. To give an example, here's a hypothesis that I didn't think was particularly plausible in the first place, but it is testable: namely why are there a substantial percentage of men who are homosexual, given that you might expect natural selection to have weeded them out. Any genes for homosexuality would not perpetuate themselves if they are unlikely to seek heterosexual matings. One hypothesis was that gay men make up for it by devoting so much attention to their nieces and nephews that they increase their fitness. The increase in fitness from their care of their nieces and nephews who share a quarter of their genes makes up for the fact that they aren't transmitting genes to their own offspring. This is something that you see in many species of animals, helpers in the nest.
Doug: Kin selection.
Pinker: Yes. So I thought this was a little fishy, but it's testable — you can see if gay men spend more time, resources, and care on their nieces and nephews than straight men. The answer is no. If anything, it's the other way around. So that's a falsified hypothesis, and it's a sign that evolutionary hypotheses can be subject to the acid test of science, namely falsifiability.
Doug: In The Stuff of Thought, another model you discuss is Linguistic Determinism. You demonstrate fairly well that it puts the cart before the horse in suggesting language shapes thought. But one instance in which it seems that language might influence thought is languages with gendered nouns. If a child growing up in Germany is told cats are feminine and dogs are masculine, how does that affect them?
Pinker: I think that's actually a good example of the limitations of Linguistic Determinism. It would be hard to believe that German speaking children literally think of cats as more feminine; there might be a tiny effect in terms of overall how masculine or feminine you perceive them, but not so that it would affect anything you do in daily life. In fact, gender in most languages with grammatical gender isn't so much a marker of biological sex as a phonological marker. Often you do a better job of predicting a noun's gender based on what vowels or consonants it ends in than whether it has anything to do with a male or a female. The reason we even call them masculine or feminine is that in the pronouns, that is, the he and she, one of them goes with the male, and one goes with the female.
Doug: They could just as easily be called A, B, C...
Pinker: In fact, that is literally what gender is in the broad sweep of the world's languages. Gender means kind, as in genus in biology, or generic, or genre. It just so happens that in many European languages the only two or three genders align with biological sex in the pronouns. But in many languages you can have thirteen genders. It doesn't mean you have transsexuals, cross-dressers, hermaphrodites, and so on. (Laughter) It might mean they separate the nouns into things like animate/inanimate, human, long and skinny, flat and flexible, and animal. They have different ways of dividing up the world of things, and those are also genders, but it's basically the same grammatical phenomenon.
Doug: In the Chapter "Cleaving the Air," you discuss the topic of causality. Are you arguing causality is a construct of how we look at the world, or just that we need to be careful in choosing our words when discussing events and their antecedents?
Pinker: It's very hard to pin down exactly what causality is. There have been some philosophers of science who have said that we should just dispense with the concept altogether. Science gives us basically a bunch of differential equations that describe how, given the state of the world at time 1, we can predict what the state of the world will be at time 2, but X causing Y is not a real part of it. Bertrand Russell famously said the concept of causality persists in science for the same reason the institution of the monarchy survives in Britain: because people mistakenly think it does no harm. (Laughter)
I think that's too extreme, but it is true that it is a very slippery concept. One example that I give in the book is that when James Garfield was shot by an assassin, the bullet didn't hit any major organs. But he was subjected to the harebrained medical practices of his day — like doctors who didn't wash their hands from the previous patient and then probed his wound with unwashed and ungloved hands — and he eventually died of infection and starvation. The assassin at his trial said, "The doctors killed him, I just shot him." (Laughter) He raises an interesting philosophical point: who caused his death? They kind of both did, in different ways. If it wasn't for either of them he would have been alive. Which one was the cause? Well, there's no real answer to that question. I think the human mind seeks that kind of answer, but it doesn't mean that our best science will satisfy us.
Another example: you take Vioxx, and you die of a heart attack. Now, we know that the vast majority of people who take Vioxx don't die of heart attacks, and a lot of people die of heart attacks who don't take Vioxx. Let's say though, that Vioxx does increase the probability from X percent to Y percent. How can you answer the question of whether John Smith was killed by the Vioxx, or whether he would have died anyway? All we know about Vioxx is that it ups the probability. How do we apply a probabilistic statement to an individual? That too is a case where the human mind craves a yes or no answer. And a scientist will say, "Don't look at me, I can't answer that question."
Doug: In the "Seven Words" chapter, which was great fun for a verbal libertarian like myself, you mentioned in passing that the brain's right hemisphere handles conversation fillers like um. A friend of mine is currently taking Japanese language and etiquette courses in preparing to be a translator, and one thing taught was how to say um in Japanese.
Doug: Do all languages have their own word for um?
Pinker: I suspect so. I don't know about all languages, but I do know that um is an English word. If you speak another language, you have to learn its translation. Growing up in Quebec, a bilingual society, I learned French in school. I remember the first time I had a French girlfriend; when I spoke in French to her, it drove her crazy that I would say um. She would say "You sound so English when you say um! Say u, not um!" (Laughter) The interesting thing is that these response cries, as Erving Goffman called them — these interjections like aha, oops, brr, yuck, YES! — even though they don't have parts of speech and you can't slot them into a sentence, they are words in a particular language.
The relevance to that chapter is that cathartic swearing, when you hit your thumb with a hammer and you say "oh shit" or "oh fuck," has to be learned, too. I think it originates from the rage circuit that you see in many mammals; that is, a vocal angry outburst when suddenly injured or confined. But once it's patched into the language system, there is also some language-specific learning you have to do. And again, if you translate your cathartic swear words from one language to another, the results can be quite comical. In Quebecois French, if you want to yell at someone, you say "damn tabernacle," or "damn chalice." (Laughter) In Portuguese, you say "whore." In French, you say a word that is literally "cunt," con, although nowadays most French speakers have lost that connection. But even though for us cunt is an obscene word. It's not the word we say when we stub our toe. You have to learn which word to use; it's not just a reflex.
Doug: In Colin McGinn's review of The Stuff of Thought in the New York Review of Books, he posed this question which I'll give you the opportunity to address: "Where exactly might a concept end and its interpretation begin? Is our concept of something identical to our conception of it — the things we believe about it?"
Pinker: There's a chapter on the logic of names that deals with that issue. Most semanticists now would say that our conception of something is not the same thing as our concept of it; that our concept of it inevitably involves some connection to the real world. A classic example would be that if you look up a name in the dictionary — William Shakespeare is the example that I give — it says "English playwright, lived in the 17th century, wrote Romeo and Juliet and Hamlet, etc." Is that what the name William Shakespeare means, and is that what the concept William Shakespeare is? That sounds plausible, but it turns out not to be true. If we were to learn that William Shakespeare didn't write any of the plays attributed to him — let's say that we learned he didn't even live in Stratford, that there was a clerical error and he really lived in Warwick. He would still be William Shakespeare, and we wouldn't posthumously dub the real author of Shakespeare's plays William Shakespeare. We would just say we were mistaken about what we believed about William Shakespeare.
So what is the concept of William Shakespeare, the meaning of the word William Shakespeare? Basically, when Mr. and Mrs. Shakespeare christened their son William, and the name stuck, and then everyone who knew him, and then who knew someone else, who knew someone else, and passed it down to us — that unbroken chain of transmission of the name from the moment of first dubbing is what gives William Shakespeare its meaning. There's a sense in which to have a concept necessarily means to be connected to the world through this chain of transmission of a name going back to the moment of first dubbing.
Some philosophers who want to answer that question more fully, more satisfactorily, say there are two senses of a concept or a meaning: there's a narrow sense and a wide sense. The narrow sense is what's just in the head; the wide sense is what's in the head and its connection to the world, à la William Shakespeare actually referring to that guy. So a full answer to the question "What is a concept?" is: it depends on what you mean by concept, whether you mean a narrow concept or a wide concept.
Doug: What are some of the main ideas you hope readers take away from The Stuff of Thought?
Pinker: That the world doesn't just stamp itself into the brain, but that the brain always actively construes a situation in one of a number of ways. That our construal of the world is often very different from what our best science teaches us is the nature of the world, and that education consists of overcoming our innate way of understanding the world. That is also true of our social relationships; our instinctive social relationships often are at cross purposes to the requirements of a fully functioning democracy. We have to unlearn our nepotism, our cronyism, and our dominance, which are widespread in the animal kingdom, but are no way to run a society. And most generally, that there is such a thing as human nature, both in our thoughts and in our emotions, and that understanding what it is and knowing how we can transcend it gives us our best hope towards knowledge and wisdom.
Doug: Everyone else seems to be writing books on atheism these days...
Pinker: (Laughter) That's right.
Doug: Do you see yourself writing one, or do you think it's pretty much been said by Dawkins, et al?
Pinker: Yeah, it's pretty much been said. I'm often lumped with them because I have a view of the human mind that doesn't need a soul endowed by god, and that also grounds morality in particular in naturalistic origins, which I articulated in How the Mind Works and The Blank Slate. Those are essentially secular humanist books, but I don't see myself devoting a whole book to that, especially as it has been done.
Doug: Where do you currently stand on free will?
Pinker: It depends on what you mean by free will. I believe that there is a distinction between behavior that can be caused two different ways. In the extreme, if someone hits my knee with a hammer and it jerks, that's one way that behavior is caused. If I ponder what to have for dinner while examining a menu, that's caused in a different way. The second is much more complex. It involves the frontal lobes in the brain. It can be influenced by other people and the norms of society that I've assimilated throughout my lifetime. That second decision making process is what we call free will. A difficult to predict, highly complex process of the higher parts of the brain, one that can be influenced by our standards of holding people responsible, and therefore responsibility still has a role to play in affecting people's choices. What I don't believe is that that second category of behavior requires a soul or any immaterial entity that won't ultimately be explained in terms of the dynamics of neural functioning.
Doug: What are some questions in science that you would like to see answered?
Pinker: How does the brain represent a proposition; that is, a belief that can be true or false? We have some inkling about how the brain might represent the concept of, say, a dog, the concept of a person even, but the difference between "the man bit the dog" and "the dog bit the man" — the same concepts, but they're in different logical relationships. Or "a dog bit a man" and "all dogs bite men," where it's a logical distinction. We have no idea what goes on in the brain, that stores information in that sophisticated form, even though it's the lifeblood of human conversation and social arrangements.
To the extent that any of our emotions or skills are innate, how do they get coded in the genes and wired in the brain during fetal development? That would be the second scientific puzzle. Third would be why are people different, other than the fact that they are genetically distinct and grow up in different cultures. So you have two identical twins, they grow up in the same culture with the same parents, and they have the same genomes, but they are not indistinguishable. They are correlated, but they can have individual quirks. Where do they come from, if neither genes nor environment? I suspect chance — stochastic processes in gene expression and neural growth — but it would be nice to know whether that's the case. Or whether it's chance events in the early years of life, or perhaps chance events throughout one's lifetime that multiply or have some chaotic effect that sends you along one life path versus another. I think we don't have the conceptual vocabulary to understand what makes us what we are because we think in terms of these two categories, of genes and upbringing.
Doug: Do you set time aside to write, or write simply as the muse (or data) strikes?
Pinker: When I begin work on a book, I try to set aside time to write. Not just hours in a day, but weeks and months during which I have fewer other responsibilities. Times at which I'm not teaching, where I can more or less question myself, and really concentrate.
Doug: What are you currently teaching at Harvard?
Pinker: I teach a large lecture course called "The Human Mind," which is kind of a biologically oriented introduction to psychology, but not primarily for psychology majors. It's for other majors like government, history, economics, biology majors. Last year I taught a course called "Morality and Taboo" jointly with Alan Dershowitz of the law school on the concept of taboo, of ideas that it's considered immoral to think. What is the psychological basis for taboo, and what are its implications for freedom of speech, for religion, for morality, for legal issues? And then I teach smaller graduate courses.
Doug: Did some of the themes of taboo in the "Seven Words" chapter form during that course?
Pinker: That's right. Those are the clearest taboos, where just the word itself is a taboo. Shit is a taboo, but feces is not. That's when the taboo is all focused on a single word. Then there are more general taboos in terms of belief systems such as in religious societies, the possibility that, say, Mohammed was not a divine prophet. In politically correct society, the possibility that men and women aren't biologically indistinguishable. I think taboos continue to shape intellectual life, even secular intellectual life; there are some topics where we feel we just shouldn't go there. In fact, for a volume called What Is Your Dangerous Idea?, I wrote the introductory essay asking the question, "Are there areas where scientists shouldn't go?"
Doug: What have you been reading lately?
Pinker: Over the summer my partner and I — my partner is also an author, Rebecca Goldstein, who wrote The Mind-Body Problem, Betraying Spinoza, and a number of other books — and neither of us had ever read Moby Dick. It's one of those big gaps…
Doug: Neither have I.
Pinker: Oh, I highly recommend it; it's great! I think I was turned off by it by the fact it was assigned in high school, and anything a high school teacher assigns has got to be boring. (Laughter) But I was wrong! We read it to each other over the course of the summer, a couple of chapters a night. Especially for a biologist, it's got a huge amount on — is it cetology?
Doug: Cetaceans, yes.
Pinker: An enormous amount on cetology, and an enormous amount on the technology and economics of whaling in the 19th century, as well as this gripping drama about this mad monomaniac. Highly recommended.
I also read Ian McEwan's On Chesil Beach, partly because I'm going to be appearing onstage with Ian McEwan the week after next in England. On Chesil Beach hinges on the fact that a newly married couple in the early 1960s could not talk about sex. So why did language break down in that circumstance? I think it's part of the general taboo we that have on sexual topics, that there's certain things we just don't like talking about, or we have to force ourselves to talk about. And I'm reading my stepdaughter's debut novel, Yael Goldstein's Overture, about a mother-daughter conflict in the case of two very talented artists. She and I are going to appear onstage together as well, and I'm going to explore her use of indirect language in the context of her characters. I like exploring connections between fiction and psychology, and this is a great opportunity.
Doug: Are there particular books you have read more than once?
Pinker: 1984 rewarded a second reading. Isaac Bashevis Singer's novel Enemies, a Love Story, which I used to conclude my book The Blank Slate as a way in which fiction explores the complexities of human nature. The Evolution of Human Sexuality by Donald Symons, which may be one of the very first books in evolutionary psychology, but applied in this case to sexuality. A marvelously rich book, and I learned a lot in the second reading. And I might read Moby Dick again. It was a fabulous book.
Doug: Mark Twain defined a classic as "something you want to have read, but don't want to read."
Pinker: That's what I thought Moby Dick was, and I was wrong.
Doug: Are there any classics left on your list?
Pinker: Oh, tons. Next summer Rebecca and I are going to read The Pickwick Papers together, because I've never read a novel by Charles Dickens, another big hole. Again, it's the Mark Twain phobia — if it's called a classic it can't be fun. But I'm sure it will be. We're also going to read something by Saul Bellow, either Herzog or Mr. Sammler's Planet, because I've never read Saul Bellow. Actually, in terms of books I've read twice, I've read Rebecca's first novel twice, The Mind-Body Problem, and enjoyed it both times.
Doug: Had you read it before you met her?
Pinker: Yes. In fact, ours was a literary romance; each had read the other before meeting one another. We met through the fact that in one of my books I cited her use of an irregular verb, in Words and Rules, as an example of how today's great writers, when given a choice, prefer an irregular verb over a regular one. She used "stridden" instead of "strided." She picked the book up, saw her name in it, and knew that I had therefore read her novels, and that's what started off our relationship. It's either a literary romance, or two nerds finding each other, but we met over an irregular verb. (Laughter)
Doug: Are there any books you think everyone should read?
Doug: Do you know yet what your next project is?
Pinker: Yes, I'm going to be writing a book called The Better Angels of Our Nature: The Decline of Violence and Its Implications, exploring the fact that contrary to popular belief, violence has steadily gone down in human history, and that we're probably living in the most peaceable era in the history of our species. Even with Iraq and Darfur and Afghanistan, genocides have gone down, totalitarian dictatorships have fallen, invasions have gone down, political assassinations have gone down, torture has gone down, slavery — at least if you think quantitatively. It doesn't mean that they are absent; a lot of people have trouble making that distinction. So what's happened? Why have we gotten nicer? That's the mystery of the next book.
Steven Pinker spoke to Doug Brown on September 28, 2007, just before his reading at the Bagdad Theater.
Books mentioned in this post