Language is Bad Metaphor
What do electrons look like?
This question is made up of words, and probably (to you) seems to have meaning. You probably feel that you know what I’m trying to ask, even if you don’t have a clear answer. But, some of the folks who have spent the most time thinking about words would find this question to be meaningless (Wittgenstein), or even a lie (Derridas).
Today, we’re going to talk about words — about whether words have meaning, or (perhaps), whether we should treat words as though they have meaning. Looking for meaning in words, and using words to narrate that search, is kind of like sprinting down a greased hallway or trying to get a piece of egg shell out of a bowl of eggs. It seems easy until you actually try. Then, the harder you try, the less it seems to work.
Even the scientific process, and the useful information about how to effectively interact with the universe around us, runs into constraints and confusions in understanding, articulating, and sharing those insights through the medium of words. Much of the language we use to talk about science creates false confidence in our degree of understanding, because most of the words borrowed by science to communicate are really just bad metaphors. But, then again, almost all language is just bad metaphor.
Let’s start with something extremely on the nose: the spin of electrons.
The thing about spin is that it isn’t spin
It can’t be spin. At least, not what we mean when we say spin, usually — like a ball with that is rotating around an axis. For context: we don’t really know how big electrons are, in fact we’re not totally sure that they aren’t just single points. But, even if electrons were on the upper limit of how big they might be, the amount of angular momentum that would be required for them to produce their magnetic fields would have to be, like, enough that they’re spinning faster than the speed of light. Or they’d have to be larger than the whole atom. Also, like we said before, they may be a literal point in space in which case they lack the necessary dimensions to “spin.” (Don’t @ me about “intrinsic angular momentum” and Larmor processions — I’m making a point here, and this article is hard enough to read as it is.)
But, why would we use a term so outstandingly confusing, evidently incorrect, and dumb? Because electrons have a magnetic field. Sort of. The kind of magnetic field that charged particles should have if moving in a loop. Sort of.
To describe the magnetic field of a charged particle moving in a loop, I’d have to tell you stuff like the direction of the loop, which could be in any direction. Electrons don’t do that, they only produce a field in a completely binary way: Up or down.
What does up and down mean?
It means in the direction you’re measuring, or the other direction. No in betweens, no other directions. So if you take the Stern Gerlach experiment (not gonna go into details, but basically you shoot magnets [or spinning charged particles] through a device and it exerts force up or down on them depending on how tilted upward the north pole of their magnetic field is) a regular bunch of magnets, or spinning charged particles with a magnetic field (which would be oriented in random distribution — which is what you would expect the silver atoms used in an experiment like this to do if electrons were really just spinning [or, properly, looping] charged particles creating a magnetic field) would do this:
But, magical electron “magnets” do this (each magnetic field is perfectly oriented up or down vs the measuring device):.
This is a very obvious example of how language in science is confusing. Not only are electrons not spinning (or looping) particles creating a magnetic field, but the magnetic field they create isn’t even like that of a spinning or looping particle of their mass — in its strength, its potential properties (like direction), or how it interacts with observation. Almost all the language used in science is confusing: energy, charge, work.
We use terms like “particles” and “waves” to describe particles because there are a few partial mathematical analogs in equations (or macro analogs in properties) that can describe the outcomes of interacting with the black box of certain particles. Some of those are analogous to “waves,” some are analogous to “particles,” by which we usually mean tiny little balls.
Why do this? Because we understand balls on a visceral level. We (kind of) understand waves on a visceral level. We can go, oh, so it’s kind of like that.
But the properties of the “that” we are describing are only emergent properties of large groups of the quantum entities we are describing.
An atom cannot act like a ball, because ballness is a concept that describes the macrobehaviour of large numbers of quantum entities. An electron can’t be a wave because waveness is a description of our experiential understanding of emergent properties of large numbers of quantum entities.
When a large crowd in a stadium holds up a bunch of little squares to create an image of their teams banner, a language and experience developed around a world in which you’ve only interacted with those banners doesn’t suit itself to describe the individual people in the crowd.
Our language is designed to capture and express macro experiences and so we end up using bad metaphors for things that operate different on a micro-level.
And we constantly find fundamental difficulties in understanding these things due to a lack of an apt analog. Electrons don’t look like any of their depictions in the below “atoms”.
They don’t, actually, “look” like anything. They, in fact, aren’t “things” in the sense we usually mean the word. A lot of the attributes we understand about these non-things through the language used to teach us about them are not true. They, for example, aren’t in a place. Or, at least, aren’t in a place the way we think of things being in places. They’re in superposition, which we understand as being kind of like being in many places at once on a probabilistically distributed (god damn it) wave function, but that really just isn’t the same thing as being in a place.
What is wrong with language?
Language is a patchwork creation. It was not made by some philosopher god and given to us an an optimized tool to service our needs around communicating and reconstructing relevant information for manipulating our world or creating metaphysical insights: it’s a organically generated set of sounds, and symbols, and symbols-for-sounds, that have affiliated implications that we (think) we mutually agree upon (accurately enough that they have broad usefulness). Language is not only unlikely, based on that origin, to serve a philosopher’s precise desire to communicate any possible thought or feeling he might imagine, but also unlikely to serve everything we encounter and try to reach an understanding about in the physical universe.
Wittgenstein came up in a world whose predominant understanding of language was that necessity and sufficiency fully defined the space of a word’s meaning. A necessary condition is that something needs to be present for a thing to be what it is. An animal must have canine genes to be a dog, that is a necessary condition. A movie being #1 in the charts is a sufficient condition to call it a blockbuster hit, but it is not a necessary condition — as a movie could be #2 in the charts and still be considered a hit. Wittgenstein (with a background in engineering) found this approach to defining the meaning of words….let’s say…insufficiently rigorous. In math, you had a much more well defined, clean, and self contained system than the above definition allows for language.
If I draw a dog, you can still point and say: “that’s a dog.” You can call the stuffed animal my sister had with her for much of her childhood a “dog”(although it usually went by “big puppy”). The exercise of tracing every actual sufficient or necessary criteria for a good use of the word “dog” can become a surprisingly difficult problem surprisingly quickly, and that approach is going to leave most words undefined.
All this language stuff gets far too squishy for Wittgenstein, so he decided to create a more principled definition of the meaning of words — closer to something we have in math. I mean, both language and math are approximations of formal logic, right?
We use “and” and “then” and “so” and “because” and “therefore” — a lot of what we’re doing is making logical statements about the world that we interpret as being cause-and-effect oriented.
Wittgenstein interpreted the reality of the world as being logical, so thoughts or statements like “that object is a a baby, and all babies poop, so that object must poop” reflect a real logical structure that exists in reality as a fact outside of the use of language itself, and so is an appropriate use case of language. It is language that creates a clear thought that reflects an experienced logical structure in reality. This kind of creation and communication of logical structures is what proper language is.
He was so obsessed with figuring this out that he isolated himself in a cabin in Norway and wrote Tractatus Logico Philosphicus — a short but horrifyingly dense list of statements that attempts to describe the proper definition of language, which is constrained to describing existing states of affairs in the world (or states of affairs that could exist in the world). The proper use of language conveys the exact picture you have of something in your head to someone else. If that picture depicts something that does exist it is true; if it depicts something that could exist but doesn’t, it is false; if it depicts something that could not exist — it is meaningless.
Saying an electron has spin, by this way of thinking, isn’t just wrong — it’s meaningless. It can hardly be called language.
Wittgenstein was so proud of this, that he quit philosophy basically thinking he had full-completed the entire enterprise of philosophical thinking. Trying to use language — which is suited and useful only in the construction of logical statements about the physical world — to answer things like “what is good?” or “what is the meaning of life?” is like trying to use a marshmallow to take a photograph. There is no point in talking (or, forming thoughts) about those things.
Thinking about it cannot get you an answer (language is unsuited to the task) and if you arrive by an answer through some other means, it would be impossible to articulate that answer into words: “Where of one can not speak, one must remain silent.”
Wittgenstein might suggest trying something different from philosophical argumentation for those questions, like, I dunno, eating ten grams of mushrooms and climbing a tree in a thunder storm. In language, such questions are meaningless. Framing the question itself in words shows you don’t understand how language works. It would be as meaningless as asking “Who is the brick that made computers less orange?” Asking the question makes you look foolish, and answering it makes you look equally foolish.
This probably comes off as an egregiously constrained and oversimplified definition of the use of language, and Wittgenstein — later in his life — would agree with you. So, he un-quit philosophy and decided to write another book (topic for a different article because this is a long rant already). I guess it seemed like all this talking humans did — talking that wasn’t simply going around and as precisely as possible articulating logical structures that observably exist or could exist in the world around them — must be more than meaningless noise.
So how do we do it? How do we seem to share common meanings of words without all having clear, logically tight, definitions of words in our head that we can rattle off whenever asked? Socrates spent countless hours trying to come up with a necessary and sufficient definition of beauty, and largely failed. We know what words mean but cannot articulate their definitions fully.
I can tell you that the picture, the stuffed animal, and the animal are all dogs, and you will be like “yeah, duh, no insight there,” yet neither of us could articulate a logically complete definition (all sufficiencies and necessities covered) for the word dog.
Perhaps language is a gift created by generations of communities making loose associations, and meaning is derived by commonly understood use? Language requires at least one person to understand what you mean in order to have meaning. This is where Wittgenstein ended up later in life — even this guy ended up in a pretty squishy place as far as creating a logical constraint for where and when words have meaning.
This might not lend itself to as easy and straightforward an interpretation as Aristotelian or Platonic ideas about language: that not only the objects, but also the concepts and values that words refer to are real, stable, things that exist independent of our definition of them (which would mean we can wield these concepts in conversation, accurately reflecting them through precise word choice and syntax to any rational listeners’ minds), but it still provides for the possibility that if we are careful enough, conscious of what words mean not only to yourself but to those who are listening, and work really hard to communicate effectively — we can do some super useful stuff with language. We can mutually agree on real objects or logical constructs that words refer to — allowing the careful use of language to accommodate the cooperation of two (or more) individuals on mutually understood models of reality.
Post-structuralist philosophy on language is a lot less reassuring (and, as I’ll come back to later, a lot less useful).
Philosophers these days tend to have a much more “YOLO” attitude toward word choice, thanks to the postmodern and poststructuralist thinkers.
Derridas thought this whole thing of thinking that words have specific meaning and that, by using them carefully, we can precisely communicate ideas was (at best) hilariously naive.
In the broader branch of semiotics (the study of the meaning of symbols), he would divide the meaning of a word into its “synchronic” and “diachronic” meanings (pretentious, I know). Synchronic meaning looks at the meaning of a word in a specific point in time, while diachronic meaning includes the entire body of a word’s transforming meaning through time. Things like “bruh,” “Karen,” “crypto,” and “gay” have very different synchronic meanings at various points in human history.
What words are is extremely complicated. To understand what a Juul is, you’d have to understand what nicotine is, what a brand is, what a vape is, and what a battery is. To understand those would then also require an understanding of what it means to breathe in, or to charge, which implies an understanding of subjective experience, of lungs, and so on, and so on, turtles all the way down.
All words, in fact, only have meaning (even synchronic meaning) in the context of their relationship to every single other word existing in the language system of every person who is exposed to it — both through its positions of related meaning, and the relation that is its unrelated-ness (a Juul is not a cigarette, a Juul is not used for nuclear fusion). Then, even if you’ve achieved this, you have only created a snapshot of the meaning of a word at a specific time, and you’re missing the majority of the depth of its meaning (…its relationships to every other word) by not including the diachronic meaning of the history of each of its relationships with the rest of the existing words or words that once existed.
At the very least this applies if we are having the broad discussion of what the meaning of a specific word is, in general. A word by itself is defined only by more words, which are in turn defined by more words, etc. — a single word is a bizarre, complex thing. Especially in the poststructuralist worldview that takes no for-granted obvious objective existence for the self or it’s surrounding reality, words seem both profound and silly.
The diachronic meaning loses some of its weight when we are talking about a specific word being used in a specific setting by a specific speaker for a specific audience: but, that comes with its own complications.
These specific contexts, to Derridas, exist within different language games for which the word’s use is entirely context dependent regarding the purpose of its utterance (or writing). The meaning of a word is — in his words — non-present. We have been wrong to assume there is a proximity between words and their meaning. This becomes more obvious in written language, where the artifact containing the “word” can exist outside of the physical presence of a speaker, his intentions, or his surroundings.
To Derridas: according to Calhoun, every statement is a lie. There is too much possible meaning within the evolving synchronic meanings for a word to be used in a way that is “true” — language is too chaotic. We necessarily suppress most of the meanings of any word when we use it. We have to pretend it has a clear stable meaning, like the concepts and values and definitions that Aristotelians thought existed. The separation between speaker/writer, audience, and referenced object (or referenced near-infinite cloud of associated meanings) is too great.
This argument contends that true effective communication in the way Wittgenstein sought is impossible, before even factoring in the fact that much of the time we aren’t trying to do that — we consciously lie, or misdirect, or make statements to make people do things and not because those statements are accurate “pictures” of objectively existing logical structures (or “states of affairs” to use Wittgenstein's term).
This line of thought leads to Derridas’ idea of logocentrism and that there is “nothing outside the text”: we are not individual actors wielding language as a tool to communicate and cooperate in our reaction to an objective reality, but meat sacks that are slaves to the software language of words — meat sacks that propagate, participate in, and manifest language games as an extension of the smushy word-soup of interconnected meanings and interactions that we swim in and use to articulate thoughts and statements. This is, in part, due again to the poststructuralist world view that the objective realness of things is not a given.
By this way of thinking, if you were alone in the universe and had no concept of language, you would have not reason (necessarily) to separate a rock in a stream from the stream itself. There is not objective reality to the rockness of the rock vs the streamness of the stream, no meaning inherent to the distinction in their properties (what does “physical properties” mean? Who cares?). We create all of that through the words. The meaning isn’t real in an objective way so there is no objective meaning for the words to point to. There are only the words. Nothing exists outside the text.
All we can do with “reality,” in this line of thinking, is recombine metaphors using the symbols of words, which don’t necessarily refer to anything real outside of other words we have made.
Well this is a bit constrained — experience is as real as anything else so can’t direct experiences have meaning that exists outside of words? I think so. So would logical positivists, but that’s a story for a different time (they basically think that you can’t do reasoning in a useful way that cannot connect to a direct verifiable experience — an epistemological foundation from which things can flow up).
So what’s the point of all of this?
In a way that I plan to explore in more depth in future articles, I’m a big believer in practical philosophy. This pretty much excludes the conclusions Derridas’ lines of thinking would lead us to, and those that Wittgenstein’s first hot take on language would lead us to. Nihilism isn’t useful, and neither is disregarding the vast majority of human speech as meaningless. This doesn’t mean that the question of language isn’t useful to think about: there are practical, useful things to extract from all of these frameworks for our daily lives.
Challenges to our unexamined use of language, like Diogenes’ prank-lesson, where he responded to Plato defining a “human” as a bi-pedal hairless animal by plucking a chicken and saying “I have created a human,” can teach us how to use language more carefully, in a way that creates better outcomes in the world.
This provides good reason, by way of example, to be cautious about words you use to label yourself — conscious of how those words mean different things to different people, and conscious of the risk that by using a word that has a different meaning to you than to someone else they could consider you a liar while you consider yourself having uncompromising integrity. Are you a feminist? Are you an environmentalist? Are you conservative? Neoliberal? Progressive? Are you a Jets fan? Name the whole starting lineup.
An interesting side-note here is the idea that using logic at all has any ultimate philosophical basis. It pre-supposes types of relations that exist and have meaning the world — primarily causal relations. It is extremely hard to come up with any justification for favoring this way of thinking besides and a-priori preference.
From the perspective of practical philosophy, though, this is easy to answer. It seems everyone can at some level understand the kinds of relations we describe as logical, and having a kind of abstract relation that humans are independently good at understanding can give us the benefit of more effectively thinking together even if it is not directly reflective of reality. At the very least we have evolved a commonly well adapted brain to think of things in terms of cause and effect, even if that model we share isn’t an accurate depiction of how the universe fundamentally works, it allows us to coordinate effectively and seems — thus far — to help us in manipulating the world toward our goals (I will write another article on why we might consider the ability to manipulate the world toward our goals good).
Thinking logically doesn’t have to be true to be effective, and being effective is sufficient reason to do it. Things that preclude either the legitimacy of using logic (Derridas) or the meaning of statements that are anything other than pure logic — even if they are mutually intelligible and useful (early Wittgenstein) aren’t (if adopted in their entirety) useful.
To return to the original example: I was taught that electrons have “spin” in high school without being given the understanding that for the electron to literally be spinning would break the laws of physics. Wittgenstein would consider the statement that an electron has “spin” a meaningless statement in most contexts. Derridas would call it a lie — a worse one than most — in that the vast majority of the meaning of “spin” is suppressed in the intended meaning, and most people who hear the phrase have an especially different understanding of what is meant than what the statement is supposedly attempting to depict.
But, being particular and skeptical about meaning allows us to pick apart what this means in a way that is useful. And, attempting to find the useful meaning requires a framework for understanding language that assumes some useful meaning does (or might) exist. The metaphor is a bad one; all language is bad metaphor. But it has salvageable meaning that reflects a communicated understanding of what — in cause and effect terms that impact consequences of human experience — the universe is like, and how to interact with it. “An electron interacts with magnets in a way that (in certain constrained ways) is similar to it having a magnetic field that is (in some limited ways) similar to what equations describing a charged particle moving in a loop would predict” is closer to the meaning that is intended, but just pretending it’s spinning will work fine for a lot of equations. This metaphor is dangerous, it has positive utility only if the miscommunication ends up creating less bad outcomes than the good outcomes gained by the efficiency of communicating in practical terms how to calculate things.
It isn’t always obvious to us how dangerous a metaphor is. With the example of describing gravity by saying spacetime is like a mattress where heavy objects create depressions and cause things to roll toward it, we are not concerned that people will assume spacetime is made from manufactured fabric stretched over rows of springs with a wooden frame, or that there exists a “memory-foam” spacetime alternative that has different properties, or that I can go to Casper’s website and receive a space-time delivered to my door.
But, even though there exist risks in using metaphor for communication, taking these kinds of risks is necessary, as is having some degree of trust for others to parse them carefully (along with some degree of responsibility for your own struggle to parse them carefully). A similar non-zero risk tolerance is necessary for the application of any practical philosophy — we don’t have to prove beyond any formal logic that other people exist for us to reject solipsism as a bad philosophy. It is useless — in fact it precludes the possibility of usefulness.
I love playing board games, especially Chess and the trading card game Magic the Gathering. You often find yourself needing to make a certain kind of strategic move — one that exposes yourself to a chance of losing sooner, or creates negative short term outcomes, or fails to optimize short-term board state, in order to optimize your chance to win. My favorite example of these kinds of moves are ones where there is only one path to victory, however unlikely. Maybe, you could survive a bit longer, and certainly lose, but if you play cards or make moves in a particular sequence, sacrificing the short term, and then you happen to draw a specific needed card or provide an opportunity for your opponent to make a specific mistake, there exists, however unlikely, a path to victory. If other paths do not include the possibility of victory, there is no point in taking them. Losing now vs losing later is irrelevant if the failure is guaranteed.
Both Wittgenstein's Tractatus and Derridas’ logocentrist theories provide no path to victory. They preclude it, regarding using language to reach higher ideals and achieve something of value. But, portions of their pessimistic views can be abstracted to help us craft our own way, of struggling and working at language to try to create something of value.