How to Do Things with Contexts

"Is There Anything in the Oven?"

Children are a never-ending well of interpretive puzzles and delights. A few days ago, I was standing in the kitchen with my four-year-old when he asked me, "Is there anything in the oven?" I said no. Now that was not literally true. Inside there were heating elements, as well as two oven racks, though one could say that those were just part of the oven. But I also knew that hanging from the bottom oven rack there was an oven thermometer. That was a "thing." And it was in the oven. But I said no. And not just that, I knew that in the oven there were two cast-iron skillets. Those are certainly "things." On no theory would the cast-iron skillets be part of the oven, and they were in the oven. And yet I said no when asked "Is there anything in the oven?"

Did I answer truthfully? Of course. The reason, as you have no doubted intuited, gentle reader, is that my four-year-old son was asking whether there was any food in the oven.

How did I know? This is not because "food" is one of the senses in the dictionary for "anything." In the linguistic community of our family or neighborhood or city there is no special usage by which "anything" could be a special term for "food." Nor was there any semantic ambiguity that would make me decide to turn to background or context. To the contrary, his question was about as free from semantic ambiguity as one could get. The referent for "oven" was clear. What part of "anything" didn't I understand?

But I knew that it was 5 pm and said four-year-old was eager for supper. In other words, I knew the setting. I knew–to put it in legal language–what the mischief was to this four-year-old.

Now imagine a different setting. I've told my four-year-old that we're going to clean the oven. And I've told him that the first step is to clear everything out of it. And he asks: "Is there anything in the oven?" Now I answer differently: "Yes, there are two cast-iron skillets, a couple oven racks, and a thermometer. We have to take those out before we can clean it."

In this second setting, his question was verbally identical. The semantics were the same: there was no distinctive meaning of "anything" in one question or the other, no semantic seam that would allow you to distinguish the two cases. Instead, the question was operating in a different context, against a different background. Which made it a different question. And, critically, I needed to know the context and background before I could make sense of the question.

For the longer, more elaborate version of this point, you can see The Mischief Rule. And as I argue there, the mischief is not just a device for resolving ambiguity, but it is part of the background and context in which a court discerns that there are multiple possible meanings for statutory text.