It is not unreasonable, at least in ordinary usage, to say “plants grow roots to get water.” Note, however, that your original phrasing was "a plant wants to gr

It is not unreasonable, at least in ordinary usage, to say “plants grow roots to get water.” Note, however, that your original phrasing was "a plant wants to grow its roots to find water." In the context of your article, the use of "wants" here invokes, tacitly, the concept of agency. Furthermore, this is done as the coda to a section in which you discuss cockatoos learning to open garbage cans from one another. There's no problem, in the case of humans dumpster-diving, saying that they want to find food (or evidence or secrets, depending on the case.) It is a bit of a stretch, but not much, to suppose that the cockatoo behavior is driven by wanting to find food (especially, if, as you suggest, they are acting on a mental model of garbage cans as good food sources.) In the context you had set up, introducing "wants" into the quoted phrase question-beggingly prompts the unprepared reader to tacitly accept the not-explicitly-stated notion that plants want to get water just like cockatoos want to get food. Perhaps you are thinking that there is a little bit of agency in what plants do, similarly to how some people will say (and have said to me - I'm not making this up) that a building with a thermostat is a little bit conscious. The problem with this sort of claim is not that it is provably wrong, but that it is practically useless, as the differences, between how a building regulates its temperature or a plant gets water on the one hand, and, on the other, how humans - or cockatoos - act on the basis of learning from one another, are so great that trying to discuss and analyze one in terms of the other leads to confusion rather than knowledge. If things reach this point, it has become a discussion of English-language semantics rather than anything about the natural world.