Often, philosophers of love want to distinguish between romantic love and other kinds of love. Some appeal to sex in some way to make this distinction. Schopenhauer, for example, reduces romantic love to sexual desire alone; in fact, to opposite-sex sexual desire alone. He says, for example, that 'all love, however ethereally it may bear itself, is rooted in the sexual impulse alone, nay, it absolutely is only a more definitely determined ... sexual impulse. ... It is merely a question of every Hans finding his Grethe'.
Others appeal to exclusivity as a distinctive feature of romantic love. The idea behind this move is that what’s distinctive about romantic love is that one only feels it towards one person (or perhaps slightly more plausibly, one person at a time). Nozick, for example, defends what is sometimes called a 'union view' of love, holding that it is a desire to form a union (or 'we') with a beloved. In The Examined Life, he appeals to exclusivity to help distinguish this from other kinds of love which might also involve desires for union of some kind. Nozick writes: 'I believe that the romantic desire is to form a we with [one] particular person and with no other' (emphases in the original).
There is (or at least: has been) a culturally prominent stereotype of romantic love on which it consists in a sexualized relationship between exactly one woman and exactly one man. Nozick and Schopenhauer, no doubt influenced by the cultural sway of this stereotype, end up claiming that certain features of the stereotype are in fact defining characteristics of love.
This means the metaphysical theses (theses about what love is) advanced by Nozick and Schopenhauer cannot accommodate the possibility of love which does not conform to stereotype in various ways. Schopenhauer's reductionism makes same-sex romantic love and asexual romantic love metaphysically impossible, and seems to misclassify friends-with-benefits as romantic lovers. Nozick's union view makes polyamorous romantic love metaphysically impossible, and seems to misclassify desire for exclusive friendship as a case of romantic love. And so on.
Methodologically speaking, this kind of shortcoming suggests that a more rigorous metaphysics of love will benefit from closer attention to the existence and variety of non-stereotypical kinds of love and related phenomena ('queer love' for short). In fact, the pre-existing range of varieties of queer love is able to serve metaphysicians as a very handy source of ready-made counterexamples to oversimplistic metaphysical views.
This sort of methodological point is not peculiar to the metaphysics of love. Metaphysicians in general are not well-advised to restrict themselves to considering only the most commonly-acknowledged or most self-presenting cases of the phenomena they are investigating. And this fact is evidently appreciated by metaphysicians in many areas. In fact, metaphysicians often turn attention to extremely strange scenarios. For example, metaphysicians studying mereological composition (that is, the relation between parts and wholes) routinely ask what might seem outlandish questions about the possibility of the entire universe being a single thing with no parts, or the possibility of matter being infinitely divisible even if it turns out that space is quantized. Similarly, metaphysicians of time ask whether it is possible that there is no time, and spend a lot of time thinking about bizarre time-travel scenarios. Science-fiction-inspired thought experiments of various kinds are often part of a metaphysician’s toolkit.
And this isn't (just) whimsy; it's methodologically advisable. Many metaphysicians are interested in limning the contours of a phenomenon, and this kind of work can’t be done by thinking only about what are already the most well-known instances of that phenomenon. Metaphysics often has to seek out and focus attention on cases that might sound queer, if it is to discover how far and in what directions the phenomena it studies extend.
In the case of love, of course, queerness isn't science fiction, or even particularly unusual. In this respect its methodological value is even more straightforward and accessible to metaphysicians: no efforts of imagination are required to learn about a wide range of possibilities that go beyond the kinds of stereotypes that guide the likes of Schopenhauer and Nozick. There's testimony and experience around that do lots of the required work. Imagination and science-fiction-like scenarios might also present interesting cases that may be important to consider, but metaphysicians of love stand to learn much of relevance to their projects just by paying attention the range of actual cases.
[This post has been updated in response to helpful feedback from Rachel McKinnon.]