8.11 What is a context?

In the previous section, we gave an informal description of what force operators like ASSERT and INTERR contribute to an utterance. To get a more concrete picture of what illocutionary force operators do with a proposition, it will help to think about why we have conversations in the first place. The “big” question that we are always trying to answer in a conversation is ‘what is the state of affairs in our world?’. In a conversation, you are in a collaborative game with the other discourse participants, trying to figure out which propositions are true, and which ones are false.

We can think of this game as each “player” (discourse participants) updating the context each time they make an utterance. What is a context? You may already have a loose understanding of what a context is, and in fact, we’ve already mentioned the notion when we introduced implicatures earlier in this chapter. Informally, the context of an utterance is the state of affairs in the discourse when the utterance is made: what the topic is, who is a part of the conversation, what things have been said already, etc. More formally, we can characterise the context as a collection of sets that keep track of information in the discourse. This idea is illustrated in Figure 8.5 below.

One of these sets in the context is called the Common Ground (CG). The Common Ground is the set of propositions that all discourse participants in that conversation agree to be true. We can think of the goal of the game of discourse as increasing the common ground: to add as many propositions as possible to this set so that we have as many facts about our world as possible.

Because discourse is a collaborative game, normally, one does not simply get to add propositions to the Common Ground singlehandedly. In other words, you can’t just add a proposition to the Common Ground just because you think it’s true; you have to get the agreement of other discourse participants first. This means that there is another place (set) in the context, separate from the common ground, where this negotiation takes place. This set is called the Question Under Discussion (QUD) set. As the name suggests, the QUD set contains the questions (or topics) that are being discussed in the discourse. The QUD set is a special kind of set that is a stack. Usually, a set is just a collection of things: there is no order to the members in the set. A stack is a special kind of set that has more internal structure than a regular set. Let’s explain this metaphorically first. Think of the QUD stack as a stack of papers, where each paper has a question written on it. Every time you ask a question in a conversation, a “paper” (an issue) gets added to this stack.

The top-most issue in the stack is the issue currently being discussed in the discourse. Sometimes the top-most issue is called the QUD, although it should be noted that the QUD stack itself is also often called “the QUD”. For clarity, we will call the stack the QUD stack in this textbook. Once that top-most issue has been resolved (i.e., someone answered that question), then that issue gets removed from the stack. We of course don’t mean that you literally have a stack of papers during conversations! This is just a metaphor to help you understand this theory of discourse, and to better visualise what we mean when we say that illocutionary meaning “does” something with a proposition.

Sometimes, we can’t come to an agreement in a conversation and the Common Ground doesn’t get increased. Perhaps the QUD is “Is coffee better than tea?” and the discourse participants can’t come to an agreement about it. This means that neither “Coffee is better than tea” nor “Tea is better than coffee” gets added to the CG. But of course, each participant is entitled to their own belief. This suggests that there is another kind of set that is specific to each discourse participant’s beliefs. We can call this kind of set the Discourse Commitment (DC) set of each discourse participant (sometimes also called their Public Belief set). Let’s say that there are two people having a conversation: Aya and Bo. The Discourse Commitment set of Aya is the set of propositions that Aya has publicly committed to as being true. The Discourse Commitment set of Bo is the set of propositions that Bo has publicly committed to as being true. There are as many Discourse Commitment sets as are there are discourse participants. If each Discourse Commitment set is the set of things that each discourse participant believes to be true, then the Common Ground can actually be thought of as the intersection of all of the Discourse Commitment sets in the context (set theory strikes again!). For example, in Aya and Bo’s conversation, there are things that Aya believes to be true, and things that Bo believes to be true: call these sets DC_A and DC_B , respectively. Take the intersection of DC_A and DC_B : that’s the stuff the both of them believe to be true, or the Common Ground.

In summary, the context is the collection of all of these sets that we just mentioned: the Common Ground, the QUD stack, and each discourse participant’s Discourse Commitment set.

Now we have a set of tools to talk about what illocutionary meaning is, and what we mean by you “do” something when you make an utterance. What illocutionary meaning does is take a proposition, and place it somewhere in one or more of these sets in the context: the Common Ground, the Question Under Discussion stack, or a Discourse Commitment set. The relevant question then, is where the ASSERT morpheme places a proposition in the context vs. where the INTERR morpheme places it in the context. This will be addressed in the next two sections.