My fourth-grade science teacher, Mr. Watt, was the first person I’d ever heard talk about the scientific method. He told us that when a scientific question needed to be answered, the only reliable way to answer it was through firsthand observation carried out in a controlled manner — in other words, an experiment — whose results are recorded in a lab report.

A proper lab report, Mr. Watt told us, had five parts:

1. Question: The question that the experiment is intended to answer
2. Hypothesis: A statement of what the outcome of the experiment was expected to be
3. Method: A description of how the experiment was conducted
4. Results: The data generated by the experiment
5. Conclusion: The answer to the initial question, based on the experiment’s results

This made sense, and seemed quite elegant, except for one annoying thing: the hypothesis. What good did it do to guess at what the results would be before the experiment was conducted? What possible bearing could my prediction have on the experiment’s conclusion? I found that I actually enjoyed writing lab reports, except for the part where I had to arbitrarily decide how I expected the experiment to turn out. That felt completely unscientific.

What brought this to mind, oddly enough, is the presidential debates. By now, everyone realizes that these “debates” aren’t debates at all, but — at best — merely joint press conferences. A real, formal debate has a single question to be decided, a series of well-structured arguments made by each side, and an opportunity for each side to rebut the other’s arguments. So far as I know, there hasn’t been a genuine debate between presidential candidates since Lincoln debated Douglas in 1858.

But thinking about debates reminded me of something I’d always wondered about: Why is the topic of a debate traditionally expressed in the form of a resolution? For example, rather than address the question of “Should there be a nationwide requirement to wear masks for the duration of the COVID-19 pandemic?,” a formal debate would address the statement “Resolved [or “Be it resolved] that there should be a nationwide requirement to wear masks for the duration of the COVID-19 pandemic.” This never made sense to me. Why is the matter considered to be resolved before the debate takes place? And if the resolution were expressed in the negative (“Resolved that there should not be a nationwide requirement…”) how would the debate be any different?

It only just occurred to me that this is the same question I had about the hypothesis in a lab experiment. In each case, why is it considered necessary to predict the outcome in advance?

And in thinking about it, I realized that Mr. Watt had it wrong. (Or, equally likely, I misunderstood what Mr. Watt was telling us. I was only in fourth grade, for heaven’s sake.) A hypothesis isn’t a prediction, or a guess, as to what the outcome of the experiment is likely to be. It’s simply a statement that the experiment is designed to prove or disprove. Similarly, the resolution that initiates a debate isn’t intended to represent the predicted outcome; it’s just a proposition that one side can argue in support of and the other can argue against. The reason this type of formulation is necessary is that unlike a question, to which one can always hedge an answer (“Well, it depends on how you look at it…”), a hypothesis follows an ironclad rule of logic: Either a statement is true or its negation is true; it’s impossible for both to be true at the same time.

As far as I know, this is also why courtroom trials are structured to decide whether a defendant is “guilty” or “not guilty,” rather than “guilty” or “innocent.” In the former case, a jury must decide between two mutually exclusive conditions. In the latter, it would be possible for a jury to decide that the defendant isn’t strictly guilty, but isn’t quite innocent either.

For my friends who are scientists or lawyers, this principle is most likely in the “Duh!” category, but for me,  it was an eye-opening realization. Be it resolved, for me at least, that stating a hypothesis makes sense.