The Language of Silence: Implication and the Role of Conversation

16

The Language of Silence

IMPLICATION AND THE ROLE OF CONVERSATION^*

Peter M. Tiersma

Assuming that people can communicate by their silence, the next question is, how do they accomplish this feat? Normally, silence is an absence of speech, and communicates nothing, or at most allows for the drawing of certain inferences.¹ What, then, are the mechanisms by which saying nothing can become a means of communicating something?

***

Finally, an act or silence may obtain meaning by implication. In examples of convention, agreement and declaration, the meaning of silence is fixed in advance. Any political science student knows that inaction by the president for the specified period means that a bill will just become law. Similarly, certain colonists knew what the lamps in the church tower meant prior to Revere’s historic ride.

Implication, on the other hand, gives meaning to an act on the spot and on a much more ad hoc basis. For example, if you ask me what I would like to do this evening, I could respond verbally by saying that I would like to play a game of chess. In the alternative, I could respond by conspicuously picking up the chess board and holding it up, while making eye contact to indicate my intent to communicate.²

Now suppose that I say nothing in response to your question. Rather, I silently leave to set up the chess board in another room. Here I have not intentionally communicated anything. It is rude as well, since while I obliquely suggest that I want to play chess, I have failed to answer your question directly; I have left you to deduce or infer my intent.³

Another example of implication is if, in response to your question about what we should do this evening, I mimic the driving of an automobile, pretending to turn an imaginary steering wheel and changing gears while imitating the sounds of a roaring motor and screeching tires. Pretending to drive a car has no conventional meaning. Still, I have answered your question.⁴ The reason is perhaps a more general interpretive convention that conspicuously mimicking a physical action can refer to, or symbolize, that action. Thus, ritualistically mimicking the driving of a car is not just the act of pretending to drive a car, but may stand for or represent the act of driving itself, especially when there are other indications, like eye contact, which show that the actor intends to engage in intentional communication.

In comparison to actions, silence provides fewer clues of intent to communicate. With people being silent the vast majority of their lives, how can we isolate the relatively rare occasions when their silence is meaningful by implication? Or to phrase it differently, when—if ever—can silence be meaningful if we can identify no convention, declaration, or agreement that gives meaning to the silence?

One fairly obvious requirement is that to be meaningful by implication, the silence must be part of a conversation or discourse.⁵ The silence that most of us engage in during our daily existence, therefore, does not communicate if we are asleep or alone.

In light of the relevance of these phenomena, it is useful to outline in broad strokes some of the work that has been done recently on conversation or discourse. Traditional linguistic theory, which even today is heavily influenced by the work of Noam Chomsky, has almost exclusively concentrated on analyzing linguistic units no larger than the sentence.⁶ Only more recently have linguists and other social scientists begun to systematically analyze larger units of speech, such as conversations.⁷

Modern research on discourse analysis has shown that conversations have a number of identifiable characteristics.⁸ One focus of discourse analysis is what is called the turn.⁹ A turn lasts as long as one speaker holds the floor.¹⁰ A turn may be a complete sentence, of course, but it may also be more than one sentence, or an elliptical (partial or incomplete) sentence consisting of just one or two words.¹¹ Ideally, speakers’ turns do not overlap, but rather follow each other sequentially, for obvious reasons.¹²

Turns, which are in many ways the building blocks of a conversation, are almost always organized into larger units.¹³ The next larger unit is generally called an exchange.¹⁴ It consists of at least two turns, one directly following the other. Consider the following short exchange:

Thea: “Did you go to the festival in Oak Park last weekend?”

Anne: “No.”

Note that Thea’s turn begins the exchange, and is therefore called the initiation.¹⁵ Schoolmarms and legal writing teachers will be happy to observe that it is a complete sentence. Although this is not an essential trait of an initiation, it is quite common. Of course, there is a great deal of background information or shared knowledge (e.g., which festival the conversants are discussing; where Oak Park is located) that need not be explicitly mentioned. But someone who initiates an exchange must give enough information so other participants understand what is being said. Consequently, the initiation usually cannot be too elliptical or cryptic.

The second turn is Anne’s response.¹⁶ By definition, a response must follow an initiation.¹⁷ Note further that because the initiation provides most of the essential information, the response can be very elliptical (i.e., it can omit most material that was already provided in the initiation). If Anne spoke in full sentences, she would have to say, “No, I did not go to the festival in Oak Park last weekend.” But because of the information given by Thea in the initiation, all Anne needs to say is “no.” The missing material can be “recovered” or filled in from the initiation.

Another attribute of the structure of such an exchange is that the initiation frequently predicts or expects a response.¹⁸ This necessarily entails that the initiation cannot end the exchange; something else must follow. In the above example, the initiation does not predict whether the response will be positive or negative. It predicts only that a response of some kind will be forthcoming.¹⁹ For Anne to fail to respond would indicate that something went awry: that she had not heard Thea’s comment, did not understand it, or maybe did not want to talk to her.

A final relevant characteristic of exchanges is that the initiation not only predicts that there will be a response, but it greatly limits the range of appropriate responses. The potential topics of any initiation are vast; someone can always change the subject and raise a new topic. But assuming that conversational partners are relatively cooperative, an initiation in the form of a greeting (“Hi”) expects or predicts that the response will be another greeting. A yes/no question (“Did you eat breakfast?”) anticipates either a “yes” or a “no” in reply.²⁰ In fact, the range of expected responses can be reduced to one particular response. As every trial lawyer knows, the leading question “You were at the scene of the crime, weren’t you?” clearly anticipates only one response: “Yes.”²¹

If we posit that ordinarily silence is meaningful through implication only if it is part of a structured conversation, we come a long way in explaining when and how silence can communicate.²² Most notably, it greatly narrows the occurrences of silence that can be considered meaningful. As one commentator has noted, participants in a conversation “can, by remaining silent, answer a question or agree to a request.”²³ On the other hand, silence that is not part of a conversation is far less likely to involve intentional communication, although it may obviously support various inferences.²⁴

It also appears that for silence to communicate, it must not only be part of a conversational exchange, but it must virtually always be a response.²⁵ In other words, silence or inaction cannot initiate an exchange. This is consistent with the observation that silence has no conventional meaning in isolation. Something that is meaningless in isolation cannot sufficiently convey the specific information required to initiate a conversational exchange.

Thus, absent convention, agreement, or declaration, silence normally is meaningful only as a response in an exchange.²⁶ This observation produces at least two significant benefits. One is that, as noted, it serves to define the context in which silence can be meaningful by implication. Not only must silence be part of a conversation, but it must occur as a response to a relatively complete initiation. The second benefit is that it shows how silence can communicate a relatively specific message absent convention, declaration, or agreement to give it meaning. In isolation, it is hard to say that silence has any articulable meaning. Yet recall that responses are often elliptical and derive much of their meaning from the initiation. Because a silent response secures meaning from the turn that it follows, the fact that no convention, declaration, or agreement exists to give meaning to the silence is not fatal.

Of course, these principles assume that the parties are indeed engaged in conversation. More specifically, the person initiating the exchange must have good reason to expect a response in the first place. Between strangers, for example, one person might be free to simply ignore a question, thereby refusing to engage in conversation. The fact that the person has remained silent in response to a question is meaningless in this situation, at least if the questioner has no right to expect or demand a response.

This principle is nicely illustrated by an example of Muslim marriage, [where the prospective bride’s silence conventionally means consent only if she is asked by a close relative; it indicates rejection when asked by a stranger.] Presumably, where the family has arranged the marriage, the expectation is that she will consent. Thus, the family can expect a positive response. The woman does not need to say anything, unless she wishes to upset that expectation. But if a nonrelative asks, she is under no obligation to say anything. Her silence would therefore indicate a nonresponse, especially because it is probably impolite or unusual for a nonrelative to seek her consent. And practically speaking, a nonresponse is equivalent to a rejection.

A final issue is that although the conversational approach tremendously limits the range of possible meanings of silence, it does not always suffice. We have made a great deal of progress if we can interpret silence to mean either “yes” or “no.” But the distinction between “yes” and “no” is fundamental. How can we select one of these two possibilities?

I propose that silence acquires meaning here on the basis of the participants’ expectations. Sometimes these expectations can arise linguistically, often through the use of what are called tag questions.²⁷ For instance, the question “It’s all right if I use your bathroom, isn’t it?” anticipates a positive response. If you hear me and say nothing, I can safely conclude that you will let me use the bathroom, at least if I have the right to expect a response. In contrast, the question “You don’t mind if I use your bathroom, do you?” expects a negative answer. These expectations can arise not only by linguistic structure, but also through real-world experience. For instance, someone may on many occasions have asked his friend Mary the neutral question, “May I use your bathroom?” and always have received a positive answer. Based on expectations of previous experience, he can construe Mary’s silence as a positive answer.

Another legal area where silence may be significant is evidence. A major part of evidence law is devoted to the hearsay rule, which excludes testimony regarding what other people said outside of court. Thus, if you hear John say that “the light was red when the car entered the intersection,” your testimony regarding what John said is inadmissible under the hearsay rule.²⁸ The rule has numerous exceptions, however. One such exception relates to admissions.²⁹ If you hear John, who is now the defendant in a lawsuit, admit that the light was red when he entered the intersection and rammed into the plaintiff, John’s statement is an admission. Thus, your testimony regarding John’s statement is admissible.

What interests us here is what is called a tacit