I was in a car with a group of children and one of them, full of recently acquired knowledge about hawks, was testing the rest of us. "What is the largest hawk?" "What is the fastest hawk?" I noted my growing feeling of irritation, all out of proportion to the innocence of this boy's pleasure in his new knowledge.

Now I wouldn't mind a request for the information I have about hawks because he would really like to know, or even an inquiry about whether I know much on that subject . Each of those is a genuine desire on the part of the asker for new knowledge--worlds apart from asking, with knowledge in hand, to see how I rate.

I hate to be tested--and oral testing is by far the worst. The many times when I know the answer to the test question don't make up for the stupidity I feel when the space after the question grows longer and I just can't find the answer, or know that what I'm about to say to break that silence is probably wrong.

Maybe there are some people who like being tested--people who make a vocation of learning the right answers to questions, or people who are knowledgeable or skilled enough in one particular area that they welcome the chance to show off. But I would also guess that I'm in the majority, that most of us would exert some effort to avoid being put in that position.

All of which brings me to the subject of testing children. Quite apart from what goes on in school, we parents do an enormous amount of testing, and we start very young. "Who is that sitting across the table?" "What color is this?" "How many fingers do I have up?" "What is eight and four?" "What animal is in that picture?" We do it so much that we don't even notice. Yet I can't imagine that they like to be tested any more than we do.

What if we never, ever asked a question that we already knew the answer to? Some people might worry that the children would never learn, or that we'd never know if they were learning or not. But if you talk in conversation about colors, about people's names and names of things, children will pick them up just as they do the rest of the language. Sooner or later we'll just discover that they know, without ever having to test.

If we're really desperate to know if a child knows, we can always ask: "Do you know what this is called?" That's a legitimate question--one that we don't know the answer to. (Of course, if they respond with an unadorned "yes" or "no", then we should be satisfied because our question has been answered.)

We can also ask if a child would like to be tested--another question to which we may not know the answer. I have a son who actually enjoys answering math questions, and so long as he's the one who decides, the testing can be fun for both of us. (He's also very clear about when he's done.)

Adults rarely test each other in normal conversation. In a way, it's a simple matter of respect, of treating children as we would any other human being. So--what if we never asked a question that we already knew the answer to? What a question!