Even a crappy amateur of a Philosopher of Science such as myself knows that these words don't relate to any reality at all; that they mean nothing in the scheme of the universe. When we are talking about these words, we are talking about mental constructs - purely human, purely conceptual, used for the way they simplify our thinking about something that resists cognition by pretending to unify and define a certain segment of the unbroken swath of the fabric of reality. That metaphor is important to me - the fabric itself is essentially unbroken, but we look at the crest of a wrinkle in the cloth, which from our perspective looks like a line separating one part of the whole from the rest, and we say "ah, on this side the cloth is different from the other side." You can nitpick about it. You can say "well, the wrinkle itself is a real aspect of the cloth, so we are talking about something real after all." But it happens in science (and in bed-making) all the time: the closer you get to the wrinkle (the better to inspect it), the less and less it looks like a wrinkle as your perspective changes. It might as well be an axiom that the better you understand a subject, the more and more it starts to look like a part of other related subjects - until, in the final estimation, it all comes down to brute physics, and perhaps even math.
It also goes without saying - but is worth reiterating for effect - that this goes very deep. Down to the most infinitesimal aspects of our understanding of physics: everything is energy, made only apparently different by entropy and space. Up to the grandest cosmological viewpoint: Our universe may well be part of a single infinite multiverse. Difference is purely the cognitive short-cut our minds take in order to perceive the world in a timely fashion.
So what. So reality is actually an un-individuated, atomic totality - who cares? We seem to be able to actually learn a semblance of truth about the universe from flawed perception; science delivers repeatable results; God's in his heaven and all's well with the world. Yet I think some fascinating things follow from this line of thinking. More trivially: it reinforces the notion that what we discover through science is not reality per se, but rather a stochastic revelation about a certain perception of reality, given other perceptions. This was already essentially made clear via quantum physics at the latest (but could have perhaps been arrived at earlier had more attention been paid to the philosophy of science). It's trivial because it has little to do with what we care about (i.e. we only care about adequately predicting that which we can perceive to begin with), but it does serve to set us on the right path about what we expect science to tell us and how it will tell it to us.
Less trivially, I think it teaches us to look at our language and its semantics as a potential source of difficulty in doing science. In English and Spanish at the very least (the two languages I know the most about), the very structure of the language works by using "systems thinking" to represent aspects of the reality it is trying to describe; that is, it breaks reality up into chunks that are used in the language as though they were discrete units, separate from the rest of reality. For example, I talk about my ingrown big toenail as though it can be understood separate from my body; it is part of a discrete limb. But this toe is only being discussed because it is causing the rest of my body pain, for it is connected to my body by my leg, and because my body is connected to a world in which various forms of infectious disease live, because it also contains enough living things in it to sustain populations of these diseases, because it has an environment that can sustain these living things, and so on ad nauseam. How does it make sense to call it a toenail, when you can't even understand everything about it unless you consider it as part of a larger system, which itself must literally be infinite in scope before the toenail can be understood completely?
More seriously, think of ants and humans as a biologist would. There are biologists out there who would call an ant hive a superorganism; that is, it is an organism that is composed of smaller pseudo-organisms. Pseudo-organisms, they might argue, because a worker ant or a queen ant by itself does not actually fit the definition of an organism - it cannot maintain homeostasis or reproduce without its compliments in the ant hive. Nobody calls a human a pseudo-organism that is just a cog in a superorganism, though - even though humans as they exist today cannot, for the most part, maintain homeostasis or reproduce without society for help. A tiger could - tigers need no tiger friends. But then even a tiger is unable to reproduce or maintain homeostasis without some help from at least a few unfortunate goats now and again. Or from another angle still - what would the tiger be without his eye cells? What would the tiger's eye cells be without his brain cells or his gastrointestinal cells? Well, the tiger would certainly not be an organism; perhaps a superorganism, though. So? So this is all semantic; so what! How we break it down is arbitrary; where the line between one word's meaning and another's is wholly arbitrary so long as what we mean is conveyed properly!
"So long as what we mean is conveyed properly." That is the kernel of the whole problem. It is very, very difficult to keep track of all the things we mean when we use "systems thinking." We talk about an organism as though that word allows us to think of it differently from a solar system, but both maintain a certain kind of homeostasis, and both grow, reproduce, and die (not necessarily in that order). So what if, in our false conception of separation between the notions of galaxies and game birds, we lose out on a revelation for one that is perfectly applicable, yet not applied because discovered in "another" discipline?
But Adam, I ask myself, why did you belatedly tack on "narrative" and "metaphor" to the list of words that lead you here? Because these are more tools we use to simplify our perceptions for ourselves; a narrative is a pattern that we use to break up chunks of time into discrete units with causal chains running from their beginning to their end, and a metaphor is a way of understanding something by simply saying that it shares properties with another thing, as if it were possible for two things to be actually different from each other; devoid of connections that influence both ends of the metaphor, however raggedy and attenuated, like some literary form of quantum entanglement.
Basically, humans do science in spite of all the most-used tools in their language toolkit, near as I can tell, yet it's not clear that any science could be done without language (at least, no science of any use to us!). That paradox is fascinating, but it took so long to arrive at that any further consideration will have to wait until another post.
Of course, if anyone has gotten so lost on the internet as to wind up here, and they think they can add to the idea, then please do. This is the current limit of what I have consciously perceived.