Following up on a recent post on agency: Today I'm reading Susanne Langer, and I found this passage: "The misinterpretation of signs is the simplest form of mistake... Where we find the simplest form of error, we may expext to find also, as its correlate, the simplest form of knowledge. This is, indeed, the interpretation of signs.It is the most elementary and most tangible form of intellection; the kind of knowledge that we share with animals..."
To be mistaken, I contend, is a sign of agency. The deep-sea fish that bites on one of those head lures the deep-sea predators sport has made a mistake; what he took for a sign of prey was, in fact, a sign of predator. It is the fact that what the fish took as the means to achieve his end was not, in fact, such a means that we deem him mistaken. We never say of an electron, hydrochloric acid, a vein of iron, an ocean current, a volcano, the planet Neptune, or the Andromeda Galaxy that it made a mistake -- we take them all to have no ends at which they can (mis)direct their aim.
Langer posits animals as interpreters of signs; I think, given recent research, she would include plants as well. It took us a long time to recognize the actions of plants, because they act on a time scale and through media different to us. But when a plant detects an insect infestation on a branch, it directs a surge of insecticides to that area; we make sense of this by positing that it occurs with the aim of ridding the plant of the parasites.