Formal Translation in Aquinas

In every translation I've read of Aquinas' discussion of love, I find a completely worthless translation of the two categories of love he discusses. If you translate them with a formal-equivalence model, you get "the love of desire" and "the love of friendship". What he means by those is that the love of desire is when you love someone or something for the benefit you get from it or them, and the love of friendship is when you love someone in a way that takes what they desire as becoming among your own desires, and you desire it for its own sake and not just to get something out of them.

To an English speaker, the expressions "the love of desire" and "the love of friendship" suggest no such thing. They sound more like the thing you love is desire for the first, and the thing you love is friendship for the second. A much better translation would be "desire-love" and "friendship-love". Those preserve the connection with desire and friendship rather than paraphrasing them, but they change the form of the grammatical construction in order to remove the different sense that the form carries in English.

A formal-equivalence translation has this danger. It preserves the form as a higher priority than the basic meaning of the expression in its context, and you get this kind of misleading nonsense that someone teaching the material then has to explain. Isn't it better just to translate the expression in a way that conveys its meaning? If this can be done without altering the basic linguistic units, as my translation above does, then that's ideal. The problem with most dynamic-equivalence or thought-for-thought translations is that they don't do that. They might translate this as something like "self-seeking love" and "unconditional love". Such a translation would make no sense of Aquinas' attempt to explain why love having to do with desire is self-seeking and why love having to do with friendship is unconditional. It doesn't translate what's said but adds to it based on the background knowledge about how Aquinas is using the terms. It's probably rare that you can find the happy medium that I've come to with this case, where you avoid both extremes, but that seems to me to be the goal.