1. "All there is to thinking is seeing something noticeable, which makes you see something you weren't noticing, which makes you see something that isn't even visible." Norman Maclean, A River Runs Through It.

2. Three is a good number of questions; four are too many and two are too skimpy.

3. Make a personal connection with the first question, inviting some writing not necessarily about the self but out of the self, i.e., some writing that is invested, engaged, subjective.

4. "Tell all the truth but tell it slant/Success in cicuit lies," Emily Dickinson begins a poem. Particularly the first time, ask an odd angled question that relaxes the imagination.

5. Never ask a question to which you know the answer. Knowing an answer is okay but questions should be genuinely inquiring, capable of fresh, multiple answers, and not testing what is on the teacher's mind.

6. Invite translation: questions that require explaining something, for example, by analogy or by shift of audience.

7. Ask experimental not empirical questions: questions that probe and test their environment, rather than only gather data. John Dewey, in The Quest for Certainty, observes that science is often mistakenly associated with empiricism, rather than with experimentation. Experimentation turns the key of a question in the lock of the world; what opens is knowledge.

8. Are there "generic" prompts that might be adapted to many purposes? Yes, for example: "What do you need to believe for it to seem true that...?" --a question that asks about the warrants supporting a claim (to borrow Stephen Toulmin's language from The Uses of Argument).