Placeholder Image

Subtitles section Play video

  • >> MERRILL: I'm Douglas Merrill. I'm a VP of Engineering here at Google, and as side

  • note I have a PhD in Cognitive Science. In my dissertation, I spend about a chapter and

  • a half fairly but superlatively sighting you and saying why I think you're wrong, so. For

  • the record every time Steven and I have argued he is being right, and I'm sure it was the

  • case this time as well. Steven is the Johnstone Family Professor of Psychology at Harvard,

  • I believe. Is that roughly correct? He recently devolved. He was at MIT for many years, but

  • that's okay just to make it shorter. I asked Steven what he wanted me to say if anything

  • in particular and he wants me to definitely call out two things. One, well, and he wanted

  • me to call out one thing, which is that he was listed by Time Magazine as one of the

  • most 100 Most Influential People of All Time, which I find fairly creepy. But Steven wanted

  • me to mention that he appeared on Colbert and didn't suck. And with that it's a great,

  • great, great honor to introduce one of the fathers of the field of actually understanding

  • how human mind works, Steven Pinker. >> PINKER: Thank you so much. It's a real

  • pleasure and honor to be here. This old wood cut of the story of the blind man and the

  • elephant is a reminder that any complex subject can be studied in many ways. And that is certainly

  • true for a subject as complex as human nature. Anthropology can study universal patterns

  • of the belief and behavior across the world's societies as well as the ways in which they

  • defer. Biology can document how the process of evolution selected the genes that helped

  • to wire the brain. Psychology, my own field, can get people to disclose their foibles in

  • laboratory studies, and even fiction can illuminate human nature by showing the universal themes

  • and plots that obsess people in their myths and stories. This afternoon, I'm going to

  • give you the view from language: what kind of insight we can gain into thought, emotion

  • and social relations from words and how we use them. I'll talk about grammar as a window

  • into thought, swearing as a window into emotion, and innuendo as a window into social relationships.

  • And in each case, I'll start with a puzzle in language show how it reveals a much deeper

  • feature of the human mind using specific examples from English, the language of which all of

  • us are familiar. But examples that have close counterparts in many languages and that follow

  • an overall logic that can be found in all languages. So let's begin with language as

  • a window into thought. And the puzzle I will start off with comes from a delightful book

  • by Richard Lederer called "Crazy English" which has the following passage: "You have

  • to marvel at the unique lunacy of a language where a house can burn up as it burns down,

  • and in which you fill in a form by filling it out. Why is it called 'after dark' when

  • it is really after light? Things that we claim are underwater and underground are surrounded

  • by, not under the water and ground." So the first puzzle is why languages talk about the

  • physical world in such crazy ways. And the answer I'm going to suggest is that there

  • is a theory of physics embedded in our language: A concept of space in our prepositions, a

  • concept of matter in our nouns, a concept of space in our propositions, a concept of

  • a matter in our nouns, a concept of time in our tenses, and a concept of causality in

  • our verbs. That understanding the intuitive physics in language helps to explain not just

  • the quirks of language itself but the mental models that humans use to make sense of their

  • lives. So, let's start off with space. How do we locate an object relative to a place,

  • a reference location or coordinate frame? Well, you can imagine an ideal hypothetical

  • system of prepositions where every proposition was composed of six syllables: one each for

  • distance in the up-down, left-right and front-back direction, and then one each for the angle

  • of pitch, roll and yaw. Needless to say, no language uses this system. Instead, location

  • is digitized. Languages make distinctions like near versus far, on versus off, in versus

  • out, on versus under. Which is why Groucho could say, "If I could held you any closer,

  • I'd be on the other side of you." Also scale is relative. You can use the same spatial

  • term across to refer to an ant walking across a hand or a bus driving across the country.

  • And the interpretation of the word "there" will defer in a sentence like put it there,

  • depending on whether the person uttering it is a crane operator or a brain surgeon. Also

  • shape is schematic. In reality, all objects are 3-dimentional arrangements of matter.

  • But language idealizes them as essentially 1-deminsional, 2-dimensional, or 3-dimensional.

  • So we've got a line which courses a 1-dimentional, but also a road which is conceived up as 1-dimentional

  • with a little width flashing it out, and a beam which is also conceived as 1-dimentional

  • but with a finite thickness flashing it out. In contrast, we've got a surface which is

  • 2-dimentional or a slab also construed as 2-dimentional with some finite thickness.

  • This idealized geometry governs are used of prepositions. So, for example the preposition

  • "along" requires an essentially 1-dimentional object. You can say the ant walked along the

  • line or along the road or along the beam, but not the ant walked along the plate or

  • along the ball which sounds a little anomalous. It governs the way we apply nouns to shape.

  • So we don't refer to a wire as a cylinder as a long, skinny cylinder, nor a CD is a

  • cylinder, a short fat one even though geometrically speaking that's what they are. But because

  • we ignore certain dimensions as insubstantial and idealize the shape as one of the remaining

  • dimensions. And I think it goes into our overall sense of shape, what we conceive of as similar

  • to what else, as when a child says, "I don't want a little crayon box. I want the box that

  • looks like audience." That is not the eight box of eight crayon box of Crayola, but the

  • sixty-four Crayon box where the crayons are arranged in pitched rows like the balcony

  • of an auditorium. A fourth quirk is that the boundaries of object are treated like objects

  • themselves. And this is something you may have heard of, heard from Ray Jackendoff,

  • my colleague who I understood stands--spoke here recently. We have words like "edge" which

  • refer to the 1-D boundary of a 2-D surface. And so, we could say the ant walked along

  • the edge of the plate, even if we can't see the ant walked along the plate. Or, and word

  • like "end" which is the boundary either of a 1-D ribbon or a 2-D beam, as long it's essentially

  • 1-D. And you could even cut the end off a ribbon, which geometrically speaking ought

  • to be impossible but we conceive of the end as if it was an object itself. That explains

  • the mystery of why we say, "underwater" and "underground" when the thing is surrounded

  • by water or ground. It's because the word "water" or "ground" can refer to the 2-D surface

  • of the 3-D volume, not just the 3-D volume itself, and you can be under that surface.

  • So why is the language of space so crazy? Well, I think the main reason is that preposition

  • divide up space into regions with different causal consequences. And the clearest illustration

  • of that comes from a story that I clipped out at the Boston Globe a few years ago: Woman

  • rescued from frozen pond dies. A woman who fell through thin ice Sunday and was under

  • water for 90 minutes died yesterday. The Lincoln Fire Department said a miscommunication between

  • the caller who reported the accident and the dispatcher significantly delayed her rescue.

  • The rescue workers believed that a woman had fallen on the ice, not through it, and that

  • left the rescuers combing the woods to find the scene of the accident. So that digital

  • distinction between "on" and "through" in this case was literally a matter of life and

  • death even though it involved just a couple of feet in analog space. Let me turn to substance

  • in language. Language distinguishes stuff from things. Indeed, language taxonomises

  • matter into four categories. There are countable things as an apple; masses as in much apple

  • sauce; plurals as in many apples; and collections as in a dozen apples. These aren't so much

  • different kinds of matter as different frames or attitudes in looking at matter which is

  • why we can look at the same mass of little rocks and think of it either as pebbles, a

  • collection of individuals or as gravel, an amorphous stuff, and why we have the cliche

  • about the person who can't see the forest for the trees. In Crazy English, Lederer asks,

  • "Why does a man with hair on his head have more hair than a man with hairs on his head?"

  • Why is the language of substance so crazy? Well, words for matter allow people to agree

  • on how to package and quantify the continuous material world. In an obvious context in which

  • we see that is at the supermarket where chunks of matter have to be transacted and they can

  • be priced per item, which is what a count noun does; by weight, which is what a mass

  • noun does; or by the dozen, which is what a collective noun does. And in fact, that

  • same mindset that we apply to packaging matter in the physical world, we also apply to abstract

  • concepts. So just as we have the distinction between pebbles and gravel, we have a distinction

  • between many opinions as if they were discrete object and much advice as if it was an amorphous

  • mass. We do the same thing to happenings in time. We package the flow of experience in

  • the same way that we package the continuum of matter. For example, let's say I would

  • ask you, how many events took place in the morning of 9/11 in New York City? One answer

  • is there was one event, because a single plan was executed. You can demarcate events by

  • the realization of a plan. Another answer is two, because two buildings were destroyed.

  • You can demarcate time by salient physical events. This might seem like the height of

  • pointless semantic nitpicking or hairsplitting, but in fact it is a question with consequences

  • because the lease holder for the World Trade Center had an insurance policy that entitled

  • them to 3.5 billion dollars per destructive event. If 9/11 comprised one event, he stood

  • to gain three and half billion. If it comprised two events, he stood to gain seven billion.

  • And in a number of court cases tied up for many years, the lawyers debated this issue

  • in semantics. So if anyone says, "How much is a semantic distinction worth?" The answer

  • is $3.5 billion. Well, this brings me to the language of time. And this illustration reminds

  • us the time in many ways is conceived like space, and happenings are conceived like matter

  • as if there's a kind of "time-stuff" that could be chopped into the equivalent of objects,

  • except we call them events. We see this in the many spatial metaphors for time like "the

  • deadline is coming," or "we're approaching the deadline." We see it in kind of errors

  • that children make like, "Can I have any reading behind the dinner?" That is, after the dinner

  • as if events were stretched out in front of us. And we see it in the semantics of verb

  • tense. Now, verb tense, in many ways follows a semantics that is parallel to the semantics

  • of space and matter in the case of prepositions and nouns. First, time is digitized, and second

  • time is relative. That is, no language has tenses for precise intervals of time like

  • an hour, nor for locations in time like November 7, 2007. Instead, location in time is trichotomized

  • in English into three regions to find relative to the moment of speaking. An event can be

  • located in the specious present, an interval of about three seconds in which we don't make

  • temporal distinctions. It's the basic unit of nouns. This is the--specious present is

  • a term from William James, and it refers to an interval of time that embraces a deliberate

  • action like a handshake, a quick decision like how long you alight on a channel while

  • channel surfing and decide whether to click again, to the decay of unrehearsed short-term

  • memory to a line of poetry, and to a musical motif like the opening notes of Beethoven's

  • Fifth which we don't perceive as just one note, one note, one note but rather as a coherent

  • motif. Then there's the past stretching backwards indefinitely. So, every event from four seconds

  • ago back to the big bang is treated as identical by the English language, which is why Groucho

  • could say, "I've had a wonderful evening, but this wasn't it." And then, there's the future until eternity,

  • that is, everything from four seconds from now until the heat death of the universe all

  • lumped together. There are not only locations in time, what I just referred to, but shapes

  • in time what linguist call "aspect," that is, how a happening begins, unfolds, and ends.

  • Shape in time, like shape in space is treated schematically. We conceive of some happenings

  • as amorphously spread out in time without any crisp beginning or end, such as, the verb,

  • "to shake." We conceive of other events as momentaneous or a punctate, such as to swat

  • a fly. And then, still other events like to cross the street have no crisp beginning,

  • but are terminated until some goal has been achieved. In this case, you get to the other

  • side. Now the stretches of time that are defined by verbs can also be mentally packaged. In

  • the same way, that we can take a noun "beer" which just refers to the stuff generically

  • and then package it in a unit by use of the word "one" as in one beer, turned a mass noun

  • into a count noun. You can take an amorphous stretch of time like "shake it" and with the

  • use of particle like "out" turn it into an accomplishment that ends at a defined boundary

  • as in "shake it up," that is, shake it until up to completion. Likewise, we can take "wring

  • it" which is indefinite in terms of when it ends, and give an endpoint with the particle