I spent two days of this week with a federal advisory committee, deciding what questions to ask on certain official surveys. I was struck (as I sat with my colleagues in an underground hotel room), that various combinations of “discourses” were governing the discussion:

• A bureaucratic discourse. We were operating within an elaborate structure of bureaus and offices, contractors, and committees, each with its own roles and powers. This structure can be visualized as a hierarchy or a flow chart or (alternatively) as a temporal sequence. (First the Background Variables committee reviews the instrument, then it goes to the Governing Board. Etc.)
• A legislative discourse. Various federal laws govern the collection of data from kids. Unlike the rules of a bureaucracy, these statutes are miscellaneous and incomplete. There is a law against asking kids about certain sexual activities, for instance. That law just stands there on its own, trumping all other considerations.
• A business discourse. Much of this work is conducted by for-profit or non-profit corporations. They are governed by bids, contracts, and budgets.
• A statistical discourse. Many of our decisions are influenced, or even determined, by statistics. For instance, a question is bad if it doesn’t produce an interesting variation in responses. A set of questions is better if it produces responses that correlate with one another.
• A psychological discourse. Questions about human attitudes, cognition, and responses to stimuli arise repeatedly and are sometimes settled by appeal to lab research.
• A discourse of kids and schools. Great authority is (rightly) accorded to commonsensical generalizations about what happens in kids’ lives or in their schools. People say things like, “They all text each other nowadays. Does that count as ‘writing’?” Many participants are parents or grandparents and like to mention their own offspring.
• An academic discourse. Many of us are professors or deans, so issues about college teaching, graduate students, grants, sabbaticals, and tenure arise from time to time.
• A computer discourse. The data are no use unless they are stored on computers in useful ways. We often discuss details about how the data are organized and analyzed.
• A discourse of upper-middle class travel. We have all gathered in Washington, DC–most of us traveling by plane and staying in a Holiday Inn. We go out to a restaurant together for dinner. There is much chatter about food, flights, and weather.

Each of these discourses confers power or status. If you wanted to get a particular item included on the survey, you could probably improve your chances by impressing colleagues with your savvy as a traveler or by talking like a knowledgeable parent–or by letting everyone know that you have downloaded the previous years’ data and done a fancy statistical analysis. In other words, status transfers (I suspect) from one domain to another.

It strikes me that some people gravitate to issues that can be decided by applying rules. They are relieved, for instance, when a decision can be made automatically by superimposing the rules of statistics and the bureaucratic structure. Other participants chafe against such limits and feel comfortable making case-by-case value-judgments.

Some people jump at the chance to express opinions when their favored discourses arise. If you’re a statistics jock, you speak up whenever an issue is statistical. If you have a nine-year-old at home, you mention anecdotes relevant to the fourth-grade data. It’s partly about making pleasant conversation, partly about contributing good insights–and partly a matter of status and power. This is not to say that everyone is trying to maximize their influence. Some are sincerely modest and diffident. But power is present.

No setting could seem less like the “agonistic” political spaces that impressed Hannah Arendt. She admired ancient Greek agoras and revolutionary assemblies in which people expressed their inner selves in heroic speeches and deeds. I’ve been hanging out with nerds in a hotel conference room. But politics is everywhere, and that’s not a bad thing.