We submitted the paper, then another thing happened. Steve and I were asked to give a talk at the MIT Center for Cognitive Science seminar series. And so we were set to give this talk and expound the position of this paper, and then I found out that the commentators would be Stephen Jay Gould and Noam Chomsky.
I was absolutely terrified. Besides his obvious status in the field—he’s like the Descartes of our time, people will look back a thousand years from now and will know his name—Chomsky is utterly merciless in debate, and I didn’t really want the experience of getting my ass kicked by him.
And there were other people who were very unhappy about the article. One colleague of Steve’s at MIT was extremely upset. He was very much of a Chomskyan and was really mad. He thought we were being hugely naive about evolution and wrote a long letter that was quite angry, accusing us of all sorts of things.
And I think a lot of people were really upset in part just because we disagreed with Chomsky. A lot of that anger was directed at Steve. I might have been thought of as a poor graduate student who was seduced into it, an Oliver Twist to Steve’s Fagin. But Steve was at MIT, Chomsky was at MIT, and I think some people felt it was betrayal. You’d expect it from Phil Lieberman at Brown or Elizabeth Bates at the University of California, San Diego, they were always disagreeing with Chomsky, but Steve was at the center of things.
That evening, something happened. I think Chomsky’s back went out and he couldn’t do it. I felt transcendent relief. Chomsky was later replaced by Massimo Piatelli-Palmarini.
Still, going up against Stephen Jay Gould in debate was no small feat for any academic, let alone a student:
We met before for an informal dinner, but I was too nervous to eat. When we got there, the auditorium was packed, and they sure as hell weren’t there to see me.
The room was crowded for Gould, but a lot of people wanted to hear what Steve had to say. Leda Cosmides was there. And there were a lot of major figures like Ray Jackendoff and Daniel Dennett. On a previous Tuesday night thing, Steve and Alan Prince had had a major battle with Jay McClelland over computational models of language. It was an astonishing intellectual event, and the graduate students were talking about Steve’s presentation many months later.
So Steve and I split our talk. Then Gould began his talk with, of all things, a mildly offensive joke. He said something like, “I just got back from a flight from Japan, and I’m exhausted—I got jet rag!” People hissed.
Pinker likewise remembered the auditorium for the Tuesday night colloquium overflowing with people. “The crowd was far bigger than any previous audience at the series, and a partition had been taken down to double the room size. They were all there to hear Gould.8
“But what Gould said,” observed Pinker, “was surprisingly feeble. It was clear that he hadn’t prepared at all. He said something like, ‘Well, language can’t be an adaptation for communication because it isn’t always used for communication. For example, when I came here tonight from the airport, people asked me how I was, but they didn’t really mean it.’”
The main thrust of Pinker and Bloom’s argument was that it was obvious from the design of language that it had evolved: “If someone told you that John uses X as a paperweight, you would certainly be hard-pressed to guess what X is because all sorts of things make good paperweights. But if someone told you that John uses X to display television broadcasts, it would be a very good bet that X is a television set or is similar in structure to one, and that it was designed for that purpose. The reason is that it would be vanishingly unlikely for something that was not designed as a television set to display television programs; the engineering demands are simply too complex.”
No matter how you look at it, they said, whether you consider organs that evolved to serve a specific purpose or something that started off as one thing, like a heat exchange, and then evolved to fulfill another purpose, a wing; there was no a priori reason that language could not have evolved stepwise like many other products of evolution.
It made as much sense to say that language evolved as a spandrel as it did to say that the eye could be some kind of architectural side effect of another kind of evolutionary change, said Pinker and Bloom. The reason you have all these very specific parts of the eye that perform particular jobs is because they evolved to do those jobs. Their jobs were their very reason for existence. The same is true for language. The rules of syntax and intonation and words matured over time into the system we have today because they were progressively refined by use and the forge of survival and reproduction—not because the brain got big and complicated for some other reason, and all of a sudden we discovered we could now manipulate symbols as well.
In addition to arguing for evolution from the design of language, Pinker and Bloom said there were many reasons why language could not be a Gouldian spandrel. Language was just too complicated. Spandrels are usually quite simple features. Even if a spandrel ends up being modified by evolutionary change and used in complicated ways, spandrels are typically “one-part or repetitive shapes or processes that correspond to simple physical or geometric laws, such as chins, hexagonal honeycombs, large heads on large bodies, and spiral markings.”9
They reiterated that one of Gould’s main problems with language evolution was that its supporters tended to rely on “just-so” stories, like the Rudyard Kipling tales that told how the leopard got its spots, to explain some critical developments. (Chomsky calls them “fairy tales.”) In academia this is considered a term of abuse, and it essentially means you are making things up. The fear of being accused of fabrication was one reason that people stayed away from the issue of language evolution, said Pinker. But he and Bloom laid out many reasons why the evolution of language was a legitimate area of study. There are other compelling clues to the ways that language evolved; for example, our vocal tracts are shaped to produce speech, just as our hearing is specialized to register it.
Finally, they said, the argument against the evolution of language seemed to be based on nothing at all but the force of incredulity.
“Dan Dennett was there that night,” said Pinker, “and more than once since then he has told me that he thought the debate was won by us. And yet as everyone was leaving, he was shocked to hear many people saying that Gould had clearly won.” (Dennett was so incensed by this that he was inspired to write Darwin’s Dangerous Idea, a bestselling book about the theory of natural selection.)
“After the talk,” said Pinker, “we sent the paper to Behavioral and Brain Sciences. It had one round of reviews. We made the changes, and then it was accepted. We wrote it in 1989 and it got published in 1990, which for academia is fast.”
Behavioral and Brain Sciences has an unusual format. When it publishes a paper, it includes comments from many other academics, to which the authors then get a chance to respond. Compared with just reading a single paper from a team of researchers (and then possibly reading a response to it two years later in another journal), it’s a rich way to gauge the complexities of a subject and to understand what is at stake. The effect is of a dialogue.
The back-and-forth on the subject of the natural selection of language ran seventy pages: Pinker and Bloom’s original paper was twenty pages, thirty-seven pages of comments from thirty-one different sources followed, and Pinker and Bloom responded in thirteen additional pages.
Many commentators were delighted by the paper. Jim Hurford, a linguist at the University of Edinburgh, who had been interested in the area of language evolution for some years, was thrilled. “I felt freed,” he recalled, and aptly titled his reply “Liberation!” “Pinker and Bloom’s target article is deeply satisfying,” he wrote. “They correctly diagnose the consensus in linguistics and cognitive science, nurtured by the writings of Chomsky and Gould, that ‘language may not be the product of natural selection.’ Pinker and Bloom confront this stifling consensus head on.”
The overwhelming impact of Pinker and Bloom’s contribution stemmed not so much from the s
pecific ideas about adaptation they proposed as from the stand they took against the idea that language evolution was an uninteresting or intractable subject. Working out the details of how language might have evolved remained a monumental task, but with their paper it was as if a door had been flung open. From that point on, more and more researchers felt that studying the origin and evolution of language was a legitimate academic inquiry. After a hundred years or so of uncomfortable silence, it had become intelligent, respectable, and interesting to wonder aloud how on earth we had come to be a species with words.
Influence isn’t easy to define in academia. It may be obvious that a person or his ideas are powerful, but it can be hard to prove beyond simply pointing out that everyone seems to accept them. A more specific, if incomplete, measure of influence is counting how many times a scholar’s papers are mentioned by other scholars in their own work. Yet another measure is the prestige of the journal in which the scholar publishes. (The influence of a journal is determined by how many times anyone cites papers it has published.) For instance, Language, the biggest journal in linguistics, has an Impact Factor (a measure of how often it is cited) of only 3. Behavioral and Brain Sciences has a score of 15.6, making it a powerhouse. In the case of Pinker and Bloom, although it’s not possible to determine the relative contribution of all these factors, it’s clear that together they had an impact. Before their paper, relatively few books and papers were published on the topic. Since then, many books and more than one thousand papers have been published on language evolution.
Why did the paper have such an impact? There’s no guarantee that a clever, fascinating, and quite possibly correct academic article is going to be read. The products of science, like works of art, require intense focus and a lot of time to create, and then, typically, all but a few are ignored.
In part, the paper had the effect it did because of Pinker’s stature. “Finally,” Jim Hurford explained, “someone prominent, someone sort of in the Chomsky camp, someone generativist, was interested in language evolution.”
Pinker agreed, “I think people liked it possibly because I was coming from so close to the politburo headquarters—being at MIT, where Chomsky was, and also just down the street from Gould at Harvard. Many people saw the paper as coming from someone who had no ideological ax to grind against Chomsky. I’m often seen as Chomskyan, even though I disagree with him on many things. But I’m close enough that that statement was all the more attention-getting.”
Ironically, others were angered by the piece for the same reason—that Pinker was seen as an influential Chomskyan and yet he was disagreeing with Chomsky.
“Overall,” said Pinker, “some people were grumpy. I think they were disgruntled because we contradicted the official line. And there was one exceptionally long and sarcastic letter written in response to our paper that was withdrawn from publication.”
The language evolution paper also turned out to be a turning point in Pinker’s development as an academic, for it got him started in evolutionary psychology. About a year after it was published, he started to think about writing a book for nonspecialists. In 1994 he published The Language Instinct, a prizewinning account of language as a biological instinct that hit the bestseller lists.
Bloom also did well. The Behavioral and Brain Sciences piece, only the second paper he had published, drew a lot of attention, and at the time he was on the job market—a fortunate coincidence for any graduate student. Today Bloom is a professor at Yale and a successful author, as well as a coeditor of Behavioral and Brain Sciences.
After Pinker and Bloom, more and more people stopped asking, “Did language evolve?” and instead wondered, “How did language evolve?” Instead of being treated as an indivisible mystery, the problem of language evolution began to fracture into many good and answerable questions, like “What does gesture have to do with human language?” “How did categorical perception evolve?” “What’s the relationship between music and language?”
In addition to its political impact, Pinker and Bloom’s paper had the effect it did because they were writing about an idea whose time had come. Indeed, it was remarkable how many of the commentaries on their paper began with a remark like “Oh, how nice to see that Pinker and Bloom are now saying what I’ve been saying for twenty years. How nice that they agree with me.”
It’s true—while no one had previously enjoyed the attention that Pinker and Bloom got, a number of researchers had been toiling for years on the mystery of language and adaptation. By encouraging scholars to move beyond the Chomsky-Gould consensus, Pinker and Bloom not only inspired them to ask questions anew but created an opportunity for scholars to seek out earlier research on the topic and find out what had already been discovered beyond the borders of mainstream linguistic respectability.
4. Philip Lieberman
Because the light of evolution is not instantaneous or blinding, it is difficult to visualize the immensely slow and gradual change that is brought about by mutation and natural selection. When you consider a protozoan cell or an amphibian, on the one hand, and dolphins or, say, commuters, on the other, there is no intuitive way to make sense of the line that runs from one form of life to the next.
The popular cartoon of evolution, where an ape slowly unbends, straightens up, starts walking, and mutates into some form of modern-day human, is probably the easiest way to think about it. But as Stephen Jay Gould insisted, this caricature is misleading. Evolution does not follow the course of a single line. The tree of life bristles with stems, boughs, and branches. Most lines from one form to another are densely surrounded by branches leading to different species or to dead ends.
When it comes to the idea of language as an adaptation, the challenge of grasping evolution is further compounded by our inability to imagine ourselves without language. Language not only fills our lives, but we do our imagining, to a large extent, with language. Every now and then, we get a glimmer of what it might be like to exist without words. Sometimes there is a moment on waking when we are conscious but not self-conscious and our thoughts aren’t shaped by language. We are looking up at the ceiling or across the room, and the ceiling or the objects in the room are just there, as we are there. We’re awake but not much more. Is this what it’s like to be pre-linguistic?
In addition to the natural obstacles to imagining how language, or anything, evolved, the way language was defined by generative linguistics made its evolution seem even more incomprehensible. Although Chomsky forswore explicit discussion of the language evolution question, many scholars thought the answer was implicit in his model of language. Indeed, Chomsky spoke often of innateness, and when you invoke innateness, it’s hard not to make a few assumptions about genetics and evolution.
As a result, it seemed to many linguists and other cognitive scientists that the only way an innate universal grammar could exist, the only way humans could be born with a language organ, was if it was genetically endowed. The implication was that the language organ was specified in the genome, and generally it was assumed that there was a gene or genes specifically for language.
At the same time, Chomsky saw language as a perfect, formal system. So it appeared that a gene for this mathematical entity must have appeared out of nowhere with no precursors in other animals. This contributed to the widespread view that language evolution was impossible and language’s very existence was miraculous.
Although Pinker and Bloom helped considerably to challenge that belief, some researchers had been resistant to this idea even earlier—Philip Lieberman, for example. Although Lieberman was once a student of Chomsky’s, there is no interaction between them now. Both men are famously combative, and they have taken opposite positions on the subject of the evolution of language. In the 1980s and 1990s, while Chomsky expressed no interest in its study, Lieberman was examining skulls, listening to apes, and testing brains, all in search of clues to language’s origins. Lieberman argues that not only should you study language evolution, but you can’
t even begin to understand language if you don’t start with evolution. His research is grounded in the basic tenets of messy biology. When you look at the problem through his eyes, it becomes harder to see language evolution as either mystical or impossible. Instead, it looks merely insanely complicated.
Lieberman was born to a family of idealists and fix-it types. Both his parents had gone to the Soviet Union in the 1930s to save the world, but after a few of their Russian friends disappeared in purges, they left. Still, Lieberman grew up with books like The Commissar of the Gold Express lying about the house in Brooklyn. (He suspects his mother remained a sympathizer.) Lieberman’s father, who learned his plumbing skills in the Soviet Union, ended up building highly classified plants for the atomic bomb project.
Lieberman himself completed a B.S. and an M.S. in electrical engineering at MIT in 1958, but after working on a few real-world projects for General Electric, he was bored with transistors and breadboards and decided to take a linguistics class. It was a low-key arrangement. There were only three other students, and the teacher handed out purple-ink ditto-machine copies of his notes on syntactic structures and transformations. The idea of transforming one syntactic structure into another by preordained steps had been around in linguistics, but in this class it was taken a step further. Transformations weren’t just notational devices, said the teacher, but actual operations of the mind. It was the first linguistics class that Noam Chomsky taught.
Lieberman, who was twenty-two years old, found the class exciting, for he liked language and was intrigued by the idea of using it to understand the mind. Despite his enjoyment, however, his path soon diverged from Chomsky’s. One day soon after his shift to linguistics, he wandered through the department—it was housed in a wooden building on campus where the first laser had been built—and was drawn by funny noises coming from a room off the hallway. He had heard the DAVO, one of the first speech synthesizers, and the engineer-turned-linguist became interested in how speech actually works.
The First Word: The Search for the Origins of Language Page 8