We sometimes think of words like who or it as pronouns — that is, signals that represent or substitute for other nouns or phrases. But as we've seen, pronouns don't refer to words so much as to partial states that are active in the listener's mind. In order to refer to such an activity, the listener must assign it to some short-term memory- unit — that is, to some pronome. However, communication will fail unless the listener can correctly guess which pronome the speaker wishes to assign to that activity. This can be a problem when there is more than one available choice. For example, consider the pronoun it in the following sentence:
The thief who took the moon moved it to Paris.
How does the listener understand that it must mean the moon? English grammar constrains the choice: it cannot be assigned to the thief because it cannot refer to a person at all. (We apologize to all the small children we have called it in this book.) But grammar alone can't determine the choice, since it could also mean the sun, as it does in this little dialogue:
Good grief; what's happened to the sun? Oh, that! The thief who took the moon moved it to Paris.
The way it works is not so much grammatical as psychological. The expression moved it causes the listener's language-agencies to seek a pronome that represents something that could have been moved. This could be either the sun or the moon. But the previous question, What happened to the sun? has already prepared the listener to expect to hear about an action whose Object pronome represents the sun, just as our earlier question about cotton made the listener anticipate an answer concerning that topic. Furthermore, the new phrase, The thief . . . moved it, fulfills this expectation by activating a Trans- frame whose Actor and Action pronomes already have assignments; this frame requires only an Object to be complete. So the word it is perfectly suited to fill the role of sun in that unassigned Object slot.
What does expectation mean? At each point in a dialogue, both parties are already involved with various concerns and desires. These establish contexts in which each new word, description, or representation, however ambiguous, gets merged with whichever short-term memory best matches it. Why do we make such assignments so quickly, instead of waiting until all the ambiguities are resolved? That is a practical matter. Our language-agencies must dispose of each phrase as soon as possible, so that they can apply their full capacities to deal with what comes afterward. If something in the conversation does not match anything that came before, the listener must activate a new memory-unit. This tends to slow the process down, because it consumes our limited supplies of short-term memory and makes subsequent matching more difficult. If the listener cannot make suitable assignments quickly enough, the conversation will seem incoherent and communication will break down.
Eloquent speakers avoid this by designing each new expression to be easily attached to structures already active in the listener; otherwise the listener is entitled to complain that the language isn't clear. A speaker can also indicate which subjects have not been mentioned yet,
to spare the listener from struggling to make a nonexistent match; we use expressions like by the way to tell the listener not to attach what comes next to any presently active pronome. To do such things, the speaker must anticipate some of what is happening inside the listener's mind. The next section describes a way of doing this — by using the speaker's own mind as a model and assuming that the listener will be similar.