Sometimes we talk about things that are hard to talk about. Direct language, such as how I'm talking to you now, is such a small slice of our total experience. Language can be (ab)used through poetry, fiction, or song to convey feelings that the words themselves can't approach directly. And we can use analogy to "point at" things: this song isn't warm, but feels warm. Analogies are critically important to conveying internal state because they can find commonalities between things that on their surface are unrelated.
Good poetry often points at something — a concept, feeling, state of mind, an emotional landscape. It's limited by language, so it can't always exactly define the space it wants to. Instead, it uses language to dance around a conceptual mountain, using analogy and metaphor to paint a picture that may not be possible to paint directly.
Music does this. There's emotion conveyed through music that doesn't map cleanly on to words. And perhaps maps differently for different people, but still maps all the same.
Sometimes single analogies are too unspecific. Or perhaps misleading. Or incomplete. We run into this a lot in meditation instruction, because the mental states a practitioner may experience are incredibly difficult to put into words. This is probably why so many meditative states and methods keep their Pali, Hindu, or Tibetan names, because they're just markers to something that has to be experienced to really know what it's like. Teachers use many analogies to try to triangulate a concept, where none are "correct", but hopefully some are directionally useful.
Recently, in a meditation class, we had been talking about a specific meditative state for a while — one that is hard to put into words, and happens fairly spontaneously when doing a certain practice. A participant offered an interpretative dance to illustrate what a specific meditative state feels like. Immediately, I felt I had a much better understanding of the state. The discussion, full of analogies, and a very-non-verbal example triangulated a concept.
Some concepts are as if they're on the top of a mountain, and there's no great straight paths up. Talking about these concepts is hard, because we have no good maps or coordinates for exactly where we want to refer to.
Here's a useful tool: circumnavigate the mountain with many examples and arguments—which by themselves are each incomplete—but together "bound" the space. Then, slowly tighten the bounds, as if you climb the conceptual mountain from many sides at once. What's common between the examples starts to surface more and more clearly, as the superfluous details fall away.
Surfaces and Essences is one of my favorite books because Douglas Hofstadter and Emmanuel Sander do exactly this. They want to make a big claim: all cognition is analogy making. They can't show the reader this directly because this concept can't be pointed at, and they want to dismantle beliefs you likely have but don't know you have. Disproving it would take one counter example, but "proving" it requires conceptually surrounding this big idea from many angles, and slowly tightening them in. So they approach this mountain through physics, linguistics, mathematics, common sense, and stories. The book isn't about physics, but uses it to explain something else.
When using convergence arguments with a conversational partner, it's an interactive and adventurous exercise. When I'm doing it, I readily admit I can't point directly at the perspective I'm offering. So, together, we explore around the conceptual mountain. The other person's examples and arguments can serve as compasses — do they point towards or away from where I want to explore? Keeping a light grasp on what's interesting to follow is best, because surprises crop up often. Even more important is an attitude of exploration and wonder. Sometimes, together, we can find a direct path to the concept. Usually, it tightens up my own understanding.