One way to decide whether you should include an idea in a paper you’re writing is to ask yourself how willing you are to discuss it. You should, of course, also be able to discuss it; that is, you should be aware of your reasons for believing what you believe and have some sense of how someone else might think otherwise. But this issue won’t arise if you have already decided not to engage with someone who disagrees with you. My advice is to leave such ideas out of your scholarly writing.

You’ll never be able to do that completely, but it is worth trying. It is a companion rule to Oliver Smithies’ suggestion to “never write something you don’t understand”. You will find, I suspect, that your unwillingness to discuss something is often grounded in not quite understanding it. You may be quite certain that it is true, but on closer inspection you realize that this is just because so many people you respect are saying it. Worse, you may remember that the only reason you think what you think is that it sounded good in the “hot take” you read a few months ago. Better go back to the source and do a little fact checking. Once you’ve got your facts in order, you may still believe it–after all, it may still be true–but now you also know why. And this, you will sometimes find, has the added benefit of giving you the will to talk to other people about it. It’s at this point that it should go into your paper.

Now, since you originally didn’t find them worth discussing, these ideas often go unnoticed in your writing. The issue will only really come up when you run into difficulties with a sentence, or when one of your readers does. If the paper is already published, it’s of course too late, and you will just have to step up and admit that you hadn’t thought that point through well enough. (There should be no shame in this.) But you might also notice it during the writing and revision process, when you are struggling with how to express a particular thought. That’s when you should take a step back and ask yourself whether you’re actually willing to discuss it. How would you react if someone tells you you are wrong. And if your answer is that you’d just stop talking to that person then I strongly advise you just to delete it. The implicit subtext of all scholarly writing is (or, rather, should be) “Here are few things I’m happy to discuss.”

Please note that I’m not saying you should not hold beliefs you don’t want to discuss. Nor even that your published work be entirely insulated from those beliefs. It can certainly be useful to put some distance between, say, your scientific investigations and your religious convictions, so that your results can stand or fall independent of your faith, but, through a long and varied life, most of your beliefs will come into contact with each other at some point and, when they do, they will demand a modicum of consistency. So you will sometimes find that the discussion you are willing to have leads (sometimes surprisingly quickly) to one you don’t want to have. That possibility should not prevent you from writing things down either.

I am suggesting mainly that you let ideas you don’t want to discuss go without saying. If you hold beliefs that you really think can be held without question, then you should give your reader the credit of presumably holding them too. You should write some sentences about ideas you do want to discuss in such a way that they sometimes, and sometimes shamelessly, presume things you don’t want to discuss. A reader who would question those things should get the sense, even before raising the question, that doing so will not begin a conversation, but rather end it. You are, in a sense, granting that if you are wrong about this then your point will fall. But your interlocutor will have to wait until you discover your error for yourself before a retraction is forthcoming. That’s a limiting case, however. Your paper should not consist solely of such sentences. Your readers should not be only those that take your hints about what you don’t want to discuss.

Scholarly writing, I always say, is the art of writing down what you know for the purpose of discussing it with other knowledgeable people. It presumes a will to discuss things. This will, I would add, is grounded in a commitment to what Foucault called “the law of coherence”:

a procedural obligation, almost a moral constraint of research: not to multiply contradictions uselessly; not to be taken in by small differences; not to give too much weight to changes, disavowals, returns to the past, and polemics; not to suppose that men’s discourse is perpetually undermined from within by the contradiction of their desires, the influences that they have been subjected to, or the conditions in which they live. (Archaeology of Knowledge, p. 149)

Of course, he talked about this law with some irony. He was not proposing that we do or should at all times obey it, only that it is presumed in discourse and, especially, in our history of ideas. I think this obligation is worth considering from time to time in your own procedures. When you feel undermined by your desires, the influences that you have been subjected to, or the conditions under which you live, ask yourself whether or not you are open to discussing them. If not, leave the idea out of your writing for now. Come back to it when you’re feeling stronger.