I have never heard anyone ask whether two men can be friends. The cross-gender version of the question gets asked constantly, in movies, at dinner tables, by people who have clearly been holding it for a while and want to set it down. The strange thing is that nobody asking it seems to notice the asymmetry. Nobody says, huh, why is this the version of the question that exists. It is treated as if it were the obvious one to ask, when in fact it is the only one that gets asked, which means something is making it ask-able in a way the others are not.
So that is where I want to start. Not with the answer. With the asymmetry. Because the asymmetry is doing something the question is hiding.
Here is my first attempt at making sense of it. Maybe the question gets asked because the answer is genuinely uncertain, and the same-gender versions do not get asked because their answers are obvious. Two men can clearly be friends. Two women can clearly be friends. Whether a man and a woman can is up for grabs. Fine. But this only pushes the question back one step. Why is the cross-gender case up for grabs while the same-gender cases are settled? There has to be something structural about the cross-gender case that the others lack, and if I can name that something, I will have my answer.
The obvious candidate is romantic or sexual possibility. Two men in the default case are not seen as candidates for each other romantically, so the friendship question has nothing to compete with. Same for two women. But a man and a woman in the default case are seen as candidates, and the question of friendship has to share the space with the question of romance, and the sharing is what makes the friendship question feel unsettled. Okay. That seems right as a first pass. But it does not yet tell me whether cross-gender friendship is possible. It only tells me why people keep asking about it.
Let me try to define what I mean by friendship and see if cross-gender cases can satisfy the definition.
Here is a working definition I would accept.
Apply this to a cross-gender case. Two people who have known each other for years, like each other, do not act on whatever flickers might exist between them, and care about the friendship as a thing in itself. Do they meet the definition? On a first read, yes. They meet every clause. So the answer to "can men and women be friends" appears to be yes.
But there is an asterisk and I cannot ignore it. Most people who have spent time inside a cross-gender friendship know what the asterisk is. There is a thing being managed, an awareness that has to be kept in its lane, a low-level negotiation that the same-gender friendships do not seem to require. The friendship satisfies the definition but it satisfies it with effort, and the effort is not nothing. So now I am stuck. The definition says yes. The asterisk says yes, but. The question is what to do with the asterisk.
Second attempt. Maybe the asterisk proves that cross-gender friendships are a different kind of relationship and they need a different name. Call them whatever you want, but stop calling them friendships, because the word is doing too much work and pretending there is no difference. This has appeal. It is honest about the asterisk instead of waving it away. And it lets people on both sides of the question stop arguing, because they would agree the relationship exists, they would just not agree on what to call it.
But then I notice something that bothers me. My friendship with John (not a real name of someone I know), who I have known for more than ten years, also has asterisks. We used to compete for the same things. There was a stretch where one of us was doing much better than the other and it strained things. We come from different backgrounds and there are subjects we have learned to step around. None of those asterisks make me think John and I are not friends. They are textures inside the friendship. So why does the cross-gender asterisk feel different? Why does that one make me reach for a new category, when the John asterisks do not?
Maybe it does not feel different and I am being inconsistent. Maybe friendship is just a category that contains a lot of internal variation, and the cross-gender asterisk is one variation among many, no more or less category-breaking than the others. If that is true, then the answer to the original question is yes, with the same kind of yes that applies to any friendship that has things being managed under the surface, which is most of them.
I want this to be the answer. It is the cleanest one. But I do not quite trust it, so let me try the third attempt, which is to see if I can save the categorical distinction.
Here is the move. Look closer at "valued for its own sake." That clause is doing more work than it might seem. It is not saying you have to value the individual for their own sake. It is saying you have to value the friendship for its own sake, the friendship as a thing. And maybe what the cross-gender asterisk does, structurally, is make it harder to value the friendship for its own sake, because at least one party is at least sometimes valuing it as a stand-in for a different relationship that they actually want, or might want, or are trying not to want. The friendship becomes a placeholder.
This is the strongest version of the categorical claim and I have to take it seriously. If it is right, then cross-gender friendship is not just friendship-with-an-asterisk. It is a relationship with a structural pressure toward instrumentality that other friendships do not have, and the instrumentality violates the definition in a way the John asterisks do not.
But here is where I get stuck again. The placeholder dynamic can happen in cross-gender friendships and probably does, often. But it does not have to. I can imagine, and have known, cross-gender friendships where both people looked at whatever romantic possibility existed, decided the friendship as it stood was the thing they actually wanted, not as a substitute, and then proceeded to value it for what it was. Those friendships exist. They are rarer than people pretend, but they happen. And once I admit they happen, the categorical claim collapses into a probabilistic one. Most cross-gender friendships are placeholders. Some are not. Which means the asterisk is real and common but not category-defining.
So here I am at the fourth attempt and I notice I have been doing something for the last ten minutes that I should probably look at. I have been trying very hard to make the categorical claim work. I tried the definition. I tried the placeholder argument. I tried narrowing the clause about valuing for its own sake. Each attempt almost works and then slips. The slipping is a kind of data. What is it data of?
Here is what I think it is data of. The question "can men and women be friends" is not actually a question about men and women in general. Nobody asks it in the abstract. People ask it when they are inside a specific friendship, or watching one, and they are looking for a general answer that will tell them what to do about the specific case. The general answer is what they want. The general answer does not exist, and the reason it does not exist is that the question is the wrong question. The right question is the one the asker is avoiding. What is going on in this specific friendship, and why am I reaching for the abstract version instead of asking the personal one?
That is where the gap is. Not in the proposition. In the asker. The same-gender questions do not get asked because nobody is hiding from anything when they look at their same-gender friendships, so the implicitness holds. The cross-gender question gets asked because somebody, somewhere, is hiding from something, and the abstract question is the screen.
So the answer to "can men and women be friends" is yes, in the abstract. In your specific case I do not know, and the fact that you are asking me the abstract version tells me you already know more than you are letting on about the specific one.
I do not think I am going to do any better than that today.