[ Content | Sidebar ]

Steven Pinker talks about indirect speech at Pop!Tech 2007

Written on October 22, 2007

Steven Pinker, who is the Johnstone Family Professor of Psychology at Harvard University, spoke to jam-packed audience on the first day of Pop!Tech. Pinker, known as one of the world’s top researchers on language and the brain, manages to translate his findings in accessible language that is understood by both mainstream media and non-academic audience.

DSC00447

Pinker starts by acknowledging that any complex subject can be approached in several ways. For example, anthropology can do it by documenting universals as well as variations across human societies. Biology can document the process of evolution as well as to help map the developments of genes, brain, and so on. Psychology can help to document subjects' behavior through laboratory studies. Even fiction can help document recurring plots and themes, and thus shed more light on what topics and issues people are passionate about it.

In his new book “The Stuff of Thought” (which all of Pop!Tech participants got in their welcoming bags) Pinker projects a view from the language perspective, which, he believes, can give us a unique insight into thoughts, emotion, and social relations.

So, in more concrete terms, what can we learn form language? One can use verbs as a window into causality; prepositions as a window into the concept of space; tenses as a window into the concept of time. Swearing is a good window into our emotions. And even different ways in which we name babies can provide an insight into the various social networks we belong to.  

In his Pop!Tech presentation Pinker concentrated on one particular subset of language – that of indirect speech, which, as he argued, can offer us a very valuable window-- into various social relationships.

To illustrate his case, Pinker reminds the audience of a scene from the movie “Fargo”, in which Steve Buscemi attempts to indirectly bribe a police officer who pulled him offer. ““I was thinking that may be the best thing would be to take care of this here in Brainerd”. Buscemi veiled his real intentions in this innuendo, expecting his interlocutor to read between the lines.

DSC00453

There are numerous other examples of such innuendos happening in real life. Take fund-raising requests from your alma mater, which are often worded in a way that disguise what they are actually urging you to do (“it's time to demonstrate leadership”). Same with the trite invitations to see etchings at one's place, which are in most cases just a pretext for a sexual relationship.

Why recur to indirect speech when both parties know quite well what is meant?

Practical importance is often the leading factor; the fear of being accused of extortion, bribery, or sexual harassment makes indirect speech the ideal candidate for providing legal cover.

Yet, there exists a more complex three-part solution that explains “this surprisingly naughty problem”, as Pinker calls it. Whenever it comes to activities like bribing, we would often recur to indirect speech because of logic of three kinds: that of plausible deniability, relationship negotiation, and of mutual knowledge.

To better understand what it means, one needs to study game theory. There is one pertinent identification problem in game theory: how to do you figure out the rational course of action when the outcome depends on the actions by another intelligent agent,but you don't know much about the agent's values?  

Suppose that when faced with a police officer, we have just two options: bribe or don't bribe. Consequently, the police officer can also be either honest or dishonest. This give us a matrix of potential outcomes for various types of behavior by both parties. If you are facing an honest police officer, your decision to bribe will be costly: you'll be arrested for bribery. If you don't bribe, you'll just get a traffic ticket. If you meet a dishonest officer, a bribe will let you go without a ticket and no bribe will earn you a traffic ticket. With just two options, things are fairly straight-forward.

However, when we throw a veiled bribe communicated via indirect speech into the mix, things get complicated. A veiled bribe can't get you in prison with an honest cop (he won't have enough evidence to prove if if the bribe is truly veiled) while a dishonest cop will accept it and you'll go free. When presented with these three options – bribe, don't bribe, and offer a veiled bribe—the rational actor will choose the latter.

DSC00450

While the usage of indirect speech in cases that have involve bribery and tinkering with the law is more or less clear, it's still not clear why we often observe people recurring to it in perfectly legal situations. Pinker gives an example from “Pocket full of dough”, an article from Gourmet magazine, in which its author Bruce Feiler, goes to Manhattan restaurants and tries to secure himself and his date a table by bribing the maitre. Despite his fear and outright shame, Feiler got seated every time he tried within just 2-4 minutes.

This brings Pinker to outline a theory of language, which, he says, usually does two tings: to convey content (bribe, command, propose) but also to negotiate a relationship type. We normally use language at both levels. This is why the person attempting to bribe the maitre at a restaurant would use the safest literal form to signal the safest of relationship to the listener, while relying on the listener to interpret the real and much deeper meaning of what has just been said. If taken literally, the language may make not sense; it may be overstated and irrelevant. The listener thinks that if the speaker says that an outcome is good, then he must be requesting. So intended content becomes imperative, only without the presumption of dominance.

What other kinds of relationships do people negotiate? There are three major kinds and each of them have a distinct evolutionary basis. Those three are: dominance, communality, and reciprocity. Dominance relationships are usually of “don't mess with me” kind and appear in dominance hierarchies. Communality relationships are usually of “share and share alike” type , and appear in kin selection and mutualism (kin, spouse, close friends). What's curious is that behavior that is acceptable in one relationship type is anomalous in another.

When relationships are ambiguous, divergent understanding can be costly (this is when “awkwardness” occurs). Consider a friendship with one's boss: should it be seen through the prism of dominance or that of friendship? Or consider selling a car to a friend: communality competes with reciprocity. Same with sexual harassment (dominance vs sex) or dating (friendship or sex-- this it the kind of a conflict that provides raw materials for romantic comedies).

The emotional costs associated with awkwardness can duplicate payoff matrix of legal identification problems. Consider the already examined issue of bribing a maitre at a restaurant-- we get the two conflicting types (dominance vs reciprocity). A simple matrix with two options—bribe and don't bribe and just two possible types of maitres, honest and dishonest, gives us results similar to those from the car pulling example. If those are the only options available, a rational actor would choose a long wait over awkwardness that would result if an honest maitre turns down the bribe.

Indirect speech, as one may have suspected, changes the picture. When the speaker knows the listeners' values (i.e. all maitres take bribes), any deniability is not plausible. So when you offer an ambiguous bribe, the two outcomes you are faced with is a short wait or a long wait, but no awkwardness, and the rational actor would always choose to bribe.

DSC00463

Sexual innuendos usually invoke the logic of mutual knowledge. Pinker says it's important to distinguish between individual and mutual knowledge. Individual knowledge is of the type A knows x and B knows x. Mutual knowledge (that's how economists and logicians prefer to call it) is more subtle: usually it means that A knows x and B knows x, and A knows that B knows x and B knows that A knows x and A knows that B knows that A knows x and so forth.

One of the interesting illustrations of this is in how freedom of assembly often leads to political revolutions (when you go to protest against something, you feel that good that your neighbors are present because they are also disgruntled and you know that neighbors probably know that you know that they know, etc – this helps to mobilize the movement).

Same happened in the Emperor's New Clothes. When the boy stated the obvious, nobody was blind to what was happening. But the moment the boy cried, everybody knew that everybody else knew that everybody already knew and so forth. In other words, says Pinker, language creates mutual knowledge.

While innuendos merely provide individual knowledge, direct speech creates mutual knowledge. Relationships, in turn, are maintained or nullified by a mutual knowledge. If you approach a woman or a man of your dreams and directly offer to have sex together, this eliminates any uncertainty that there might be and creates mutual knowledge. By using indirect speech, one can maintain a healthy relationship (“the fiction of friendship”, as Pinker calls it) with the person even if he/she turns down the veiled innuendo.

Summing up, Pinker repeats that people often have to convey messages while unsure of their relationships. Indirect speech can minimize the risks in legal contexts with tangible costs (e.g. Bribes, threats). The same thing can happen in everyday life, because relationship mismatches have an emotional cost. Pinker emphasizes that indirect speech prevents individual knowledge from becoming mutual knowledge; mutual knowledge is the basis of our relationships. Humans think a lot about what others think about them, and their relationships are ratified by their mutual knowledge. As a result, to preserve their relationships (while transacting the business of their lives) humans often engage in hypocrisy and taboo.

Filed in: Conferences, Ideas.

No Comments

Write comment - Trackback - RSS Comments

Write comment