Wordsmith.org: the magic of words

Wordsmith Chat

About | Media | Search | Contact  


Today's Word

Yesterday's Word



A Chat With Steven Pinker

Steven Pinker's picture
Date:Feb 8, 2001
Topic:Words and Rules: The Ingredients of Language
Duration:One hour

A professor of psychology in the Brain and Cognitive Sciences department at MIT, Steven Pinker is the eminent author of many books on mind and language, including The Language Instinct: How the Mind Creates Language, Words and Rules: The Ingredients of Language, and How the Mind Works. He writes frequently on language for the New York Times, Time, Slate, and other magazines. He is on the Usage Panel of the American Heritage Dictionary. His book The Language Instinct won the Public Interest Award from the Linguistics Society of America and was named one of the ten best books of the year by the New York Times Book Review.

To join the chat, please enter your name and country below and press "Connect" button. Please join a few minutes before the start of the chat.

Transcript of the Chat

Anu Garg (Moderator)
Welcome to the fifth online chat at Wordsmith! Our today's guest is Steven Pinker, MIT Professor and author of several books about language and mind.

Welcome, Steve!

Steven Pinker (Guest Speaker)
Glad to be here; thanks for having me.

Hello Mr. Pinker- I was wondering what your opinion was on the subject of Facilitated Communication for people with Autism?

Steven Pinker (Guest Speaker)
My understanding is that it has never been shown to be valid -- it appears to depend completely on the wishful thinking of parents, who unconsciously insert their own content into the messages supposedly from the impaired child.

Anu Garg (Moderator)
Why do people like certain words? Is there a scientific explanation behind it? For example, our speaker for the last chat mentioned `serendipity' as her favorite word.

Steven Pinker (Guest Speaker)
People often like words that display "sound symbolism" -- the sound, or the articulatory movements in enunciating it, call to mind what the word represents, such as "mellifluous or cantankerous." There are also sound patterns that people prefer all over the world, such as short words coming before long words, or soft sounds coming before hard sounds: hence razzle-dazzle (not dazzle-razzle), and kit and kaboodle, not kaboodle and kit.

John USA
What grammar (syntactical theory) do you favor for representing language competency of a speaker? Would it be the same if you were programming a computer to mimic a speaker?

Steven Pinker (Guest Speaker)
Most of my research has been based on the theory called "Lexical Functional Grammar," devised by the linguist Joan Bresnan at Stanford University. However I use a variety of theories, whatever explains the phenomena best. My views are perhaps closest to those of Ray Jackendoff, whose recent book The Architecture of the Language Faculty (MIT Press), is one of the most cogent discussions of language that I have seen in years.

What do you think the hardest language to learn is, and why?

Steven Pinker (Guest Speaker)
It depends on whether you're a child or an adult. If you're a child, then all languages are equally easy (at least the spoken vernacular -- the scholarly written versions of course are different). Spoken vernaculars have to stay about the same difficulty, because if they got too complicated, children would simplify them, and the language would change. Now, if you're an adult, the situation is different, because you're no longer able to pick up the patterns of the language instinctively, and have to think about everything. In that case, aspects of language with lots of irregularity or complex declensional patterns, like the Slavic languages (Russian, Polish, etc.) can be quite hard. It also depends on what aspect of language we're talking about -- adults generally learn to read and write a foreign language, not just speak it, and that varies radically across languages, because writing systems are artificial inventions, not systems that evolve naturally through children. English spelling is harder than Spanish

Hi Steven, I've enjoyed several of your books. In How the Mind Works, you have a wonderful example of how we fill in meanings, based not on the words and sentences we hear, but on our memories and the patterns we store. The example used three sentences. Janie heard the joingling of the icecream truck. She ran upstairs to get her piggy bank. She shook it till some money came out. Anyone who lives in N America knows more or less how old Janie is, the fact that coins came out rather than bills, and what she wanted the money for. Can you talk more about this phenom and perhaps give other examples?

Steven Pinker (Guest Speaker)
This often falls under the rubric of "pragmatics" or "discourse processing." Another example -- I forget who devised it -- is that when we read the instructions on the shampoo bottle, "Lather. Rinse. Repeat," we don't spend our lives in the shower; we understand that it means "repeat just once." Interestingly, Bill Clinton's cleverly deceptive language in the Paula Jones depositions, depend heavily on this process -- he exploited listeners' tacit understanding to get them to interpret his statements in one way, even though he did not technically lie. I wrote about this in a NY Times Op-ed, available from my web site.

ioana USA
Hi Mr. Pinker- I was wondering if raising a child trilingualy can have a negative effect on the child. Do you have any opinions/experience with this?

Steven Pinker (Guest Speaker)
There is an urban legend that growing up with many languages is harmful -- apparently some pediatricians are still telling this to parents. But it's false -- children have no trouble picking up multiple languages. They confuse the vocabularies only occasionally, and the only negative seems to be that their vocabulary in each language grows more slowly than it would if the child was speaking that language alone (that's because there are only so many hours in the day, and every minute spent learning French words is a minute not spent learning English words). But still, the total vocabulary of all the languages is far greater, and there are no psychological problems.

Do you see much value in using E Prime as an exercise in teaching writing clarity?

Steven Pinker (Guest Speaker)
I am not familiar with that system, unfortunately, so cannot answer the question.

Truncated from my answer about ease of learning languages was the small point that learning to read and write a language like Chinese is particularly difficult because there are no rules for going from sounds to letters, so every word has to be memorized, and readers and writers have no way to decode a character for a new or unfamiliar word. [Again, sorry I am answering someone else's question, not yours.]

I am currently reading "How The Mind Works" and I am wondering if you are aware of any related Artificial Intelligence research that takes the approach of using the structure of language, and the way in which the mind extracts "knowledge" from words to develop an internal representation of knowledge?

Steven Pinker (Guest Speaker)
I am not an expert in AI, but I think there are efforts at the labs at Microsoft, for example, that use information in language databases to build up an understanding of the world. The biggest project is Cyc in Austin, Texas, developed by Doug Lenat, which is being fed with hundreds of thousands of facts of common sense for use in language and reasoning systems. There is a slightly different effort at the Media Lab at MIT, associated with Prof. Marvin Minsky, that has a web site at which thousands of people dump in their own factoids about anything whatsoever, as a way of building up a huge database of common sense understanding (I can't remember the URL or the person whose idea it is, but it is probably accessible via Prof. Minsky's own web site).

guin (canada)
How is it that languages that share their root, but which are separated by distance and culture etc. seem to go through similar developments even after diverging from that root? I'm studying the linguistic developments of both English and German at university and am very intrigued about the similarity of many developments - especially those that are not synchronic. It would seem for all its apparent randomness, a language does develop in part according to pattern. Would you agree or disagree with this?

Steven Pinker (Guest Speaker)
I think languages develop by a mixture of pattern and chance. The patterns may come from psychology -- the needs of speakers to communicate without too much effort, while still remaining intelligible to their listeners, and the needs of children to make sense of the language as they learn it, probably reflecting the structure of their own language and conceptual systems. Chance comes in because fads, influential speakers, neighboring languages providing loan words, dialect mixtures, and countless other processes will differ from one place to another.

clayton usa
Tallyrand thought that the purpose of language was to disguise man's thoughts. Do you agree.

Steven Pinker (Guest Speaker)
If language *completely* disguised one's thoughts, it would become useless, like confederate money, and no one would listen. In most cases conversational partners cooperate by being largely truthful; if you thought your partner was lying habitually, you would no longer listen. The skill of being a good liar is to weave an occasional lie into a largely truthful matrix, so people won't simply write you off as not worth paying attention to.

Except for a story which my dad told me about visiting France and hearing kids there speaking their own version of French, flipping words and meanings around, I have never encountered as much slang in any language except for English. I was wondering if you have any opinions about why English speakers have such a propensity for developing their own words for just about everything? Also, if I'm wrong about the lack of slang in other languages, please let me know.

Steven Pinker (Guest Speaker)
As far as I know, all languages have slang. Indeed lots of standard words started out life as slang (I have a bunch of examples in The Language Instinct and Words and Rules), such as "mob," "fun," and "bully." Even today, slang words such as "to flame" and "to diss" are coming into the mainstream. NO doubt that happens in all languages, because words are not designed by committees, but have to be with a creative speaker who first coins the word.

mattruben usa
Mr. Pinker, I really enjoy your work. This may be asking a bit much, but would you enlighten us with your thoughts about the future of language, given recent advances in communication? Should we expect increased borrowing of words and speech across languages?

Steven Pinker (Guest Speaker)
Yes, I imagine there will be increased borrowings, especially from English, as globalization and electronic media expand their influence. All in all I am optimistic about the future of language because it springs from a constant, human nature, including the faculty of language. For an editorial on the future of English, see the piece I did for the NY Times on the eve of the millennium, available through my web site.

In The Language Instinct, you agreed with Noam Chomsky's assertion that "virtually every sentence that a person utters or understands is a brand-new combination of words, appearing for the very first time in the history of the universe." Are you sure that "virtually every" is warranted in light of the number of ritual expressions ("Thank you.") and common, simple sentences ("I'll call you.")?

Steven Pinker (Guest Speaker)
Well, "virtually" is a fuzzy term. You are certainly right that we use lots of canned expressions, and with short interactions (e.g., a person passing a doorman) most utterances would be formulaic, but as a conversation progresses, and one goes beyond the pleasantries, I would imagine that all the sentences are new. So it would depend on how well the people knew each other, how long the conversation lasts, and other factors.

Do you think the conceptual system is shaped partially by the language acquisition process, or do both rely on a more fundamental mainspring of the way our minds work?

Steven Pinker (Guest Speaker)
I think our concepts are shaped partly by language acquisition, both because a language will force you to pay attention to certain aspects of the world in order to use words the way other people do, and because language is an important conduit by which we learn about the world through communication with other people. But I think the conceptual system also must be able to come to conclusions based on the five senses, on abstract reasoning, on introspection, and other language-independent routes (otherwise where would the first concepts and thoughts come from -- a language can only spread one person's ideas to another, and the ideas had to get into the population of speakers somehow to begin with.

Nicholas (MN, USA)
I enjoyed your article about the future of English that you mentioned (NY Times). I am curious now about your thoughts on the future of the various Englishes. Around the world there are many varieties, each with their unique standard and non-standard dialects. Might these varieties converge through increased communication in the ever-shrinking world, or is English destined to split up into distinct "languages," as did Latin and others? I don't suspect that the current dialects, however changed, can persist as mutually intelligible yet different varieties of the same language. I like your hair.

Steven Pinker (Guest Speaker)
Many thanks (on the hair, that is)! As for Englishes, predictions are hard, especially about the future (as Yogi Berra is reputed to have said). The reason is that there are two opposing forces. People like to borrow cool words they hear from other dialects, but people also like to maintain a distinctive way of speaking to differentiate themselves from those they feel superior to: the hoi-polloi, the rabble, the dweebs, the old f**ts, and do on. Which of those two forces is stronger will determine whether dialects grow more similar or more different. Multiply by all the ways that people divide themselves into groups and you have a chaotic system in which it is hard to make specific predictions.

mattruben usa
Where are you currently going with your research and what can we expect of your next book?

Steven Pinker (Guest Speaker)
My laboratory research is on two topics: language development in twins (do identical twins develop language in closer lockstep than fraternal twins, suggesting a possible genetic clock), and language in the brain (using neuroimaging techniques like MEG and fMRI, can we distinguish words from rules in the brain)? Currently I am working on a new book called The Blank Slate: The Denial of Human Nature in Modern Intellectual Life. It will be published by Viking/Penguin in the fall of 2002 -- but I have to finish writing it first.

I really want to know why language has been changed? Just because of the creativity of language?

Steven Pinker (Guest Speaker)
I have a chapter in Words and Rules called "Broken Telephone," after the children's game (also called "Chinese Whispers") in which one child whispers a word to a second, who whispers it to a third, and so on, until the end result is comically different from the original. That is a metaphor for how languages change: a given generation of children acquires the language imperfectly, and their children introduce still other changes, until Shakespearean English turns into modern English. Also, people introduce new words to express new ideas, try to sound hip or cool or sophisticated, mis-hear other people's speech and pass on the malapropisms, compensate for a poverty in one part of the language by introducing complexity in another part, and so on. As these small changes accumulate, the language is transformed, just as in the biological evolution of organisms.

Nicholas (MN, USA)
What is the main property of human language that distinguishes it from non-human communication systems? Is it generativity, spontaneous use, employment of syntax...?

Steven Pinker (Guest Speaker)
The main property is compositionality: the use of productive rules of grammar to assemble words into combinations, in which the meaning of the combination can be computed by the meanings of the individual words and the way they are arranged.

Keith in Ottawa
Is the process whereby children can (or are compelled to) synthesize a language or a creole from a pidgin understood at all beyond the fact that it is an instinct? It's the process itself that I'm interested in: given the absence of a grammar, are there any particular tendencies that express themselves? I recall from "The Language Instinct" that there are certain syntactic forms that never occur and others that are almost ubiquitous. When a language is being formed in a near vacuum however, is there anything that is predictable?

Steven Pinker (Guest Speaker)
The linguist Derek Bickerton has claimed that children fall back on a "biogrammar" that specifies the creole language almost in full -- that there is a universal list of constructions, rules that generate them, tenses, voices, and so on. His position is probably too strong, but if you turn the "bioprogram" into probabilistic tendencies, you certainly get a picture of the biases and constructs that children inject into the language they collectively invent.

Following up on the slang thread, do you feel dictionaries should take a descriptive or prescriptive approach? France took a very prescriptive stand towards new jargon.

Steven Pinker (Guest Speaker)
I think dictionaries should inform the user what people's expectations are, and let them use their judgment. That is, flag a meaning or usage when it is generally considered uncouth or incorrect, or considered incorrect only by a few remaining pedants, and so on. After all, a dictionary should provide information, not be a policeman. The "usage notes" in the American Heritage Dictionary are a good model -- for a careful writer and speaker, they provide information on how others are likely to react to a usage, and you can do what you want with that information.

Anu Garg (Moderator)
That was our last question for today. Thanks to all who participated, and asked questions, even if we couldn't field your questions due to limited time.

Steven Pinker (Guest Speaker)
Thanks, Anu, for having me, and thanks to all of you for your stimulating questions.

Nicholas (MN, USA)
Thank you for chatting with us today, Dr. Pinker, and thanks, Anu, for arranging this opportunity! I've enjoyed it.

Thank you very much!

clayton usa
How wonderful of you and AWAD to provide this opportunity to "talk" with you.

Nicholas (MN, USA)
Thank you to both of you, Dr. Pinker and Anu!

ioana USA
Danke schoen, Herr Pinker.

Thank you very much!


Thanks from Wales, too!

Anu Garg (Moderator)
Our next guest is David Crystal, author of The Cambridge Encyclopedia of the English Language, and a number of books on the English language. This chat takes place on Feb 26, 2001. For more details, please see wordsmith.org/chat/dc.html. We hope to see you all there.

Chat Events:  


Chat Transcripts:  

Barbara Wallraff  
Atlantic Monthly  

Joseph Pickett  
American Heritage  
Sreenath Sreenivasan  
Columbia University  

Lisa Simeone  
National Public Radio  

Richard Lederer  
Language & Humor  

David Crystal  
Encyclopedia of English  

Wendalyn Nichols  
Random House  

Robert & Jean H.  
Dante's Inferno  

Joseph Bruchac  

John Simpson  
Oxford English Dictionary  

Subscriber Services
Awards | Stats | Links | Privacy Policy
Contribute | Advertise

© 1994-2024 Wordsmith