"It's all about that semantic space map."

[Originally posted 7/14/2022 on Substack.]

One of the coolest illustrations of societal consensus is this cartography research project from a few years ago called Bostonography. Boston residents were asked to draw the boundaries of various neighborhoods and their responses were amalgamated onto a single map, with color-coding to designate areas with high or low consensus.

Some neighborhoods, like the tony Beacon Hill, have extremely high consensus and a cohesive shape. In part this is aided by its surroundings — with a major avenue to the North, a river to the West, and a park to the South, the only meaningful area for disagreement appears to be over a small parcel on the East, close to the downtown core. In contrast, few can apparently agree on where Chinatown begins and ends.^[1]

Other neighborhoods, like Dorchester, cover such a massive area that the boundaries fade into the horizon.

Neighborhood boundaries are a semantic Schelling point. Even though Boston residents each have their own subjective interpretation of boundaries, convergence towards a shared understanding remains possible. There may be disagreements at the margins, but Boston residents generally agree on where what is. And because of this shared understanding, neighborhood names will be useful methods of coordination. Telling someone that you bought a house in Beacon Hill, or that you work in Allston, or that you’ve never been to Bay Village, etc. all are used to quickly communicate useful information to someone, without the need to pull out the sextant each time.

Another useful method of coordination is agreeing on a unit of measurement. History has provided us with a cavalcade of now-obsolete units of measurement. Vernacular units of measurement based on some dead guy’s foot or whatever were likely good enough for most purposes, especially if you’re dealing with the same locals for decades (everyone eventually knows that Yosef has the biggest cubits), but once you step outside of your community you’re bound to run into some problems. Coordination across commerce and enterprise cannot meaningfully happen without establishing a shared understanding of measurements.

Tackling this problem on a global scale was the aim behind the metric system. In creating the kilogram standard, the enlightened French Revolution wanted it to be feasible for anyone with enough scientific know-how to reproduce it by deriving it from common natural phenomena, no matter where they were located. No reliance on a mummified foot necessary.

If you take the distance between either orbital pole and the Equator and divide it by ten million: voilà, you have a meter! If you use that meter to make a cube, fill it with near-freezing water, weigh that volume, then divide it by a thousand: voilà, you have a kilogram! But carting around melting ice across the Republic was never going to be practical, so the French created a more durable reference point cast from platinum and secured it inside a government vault. A cylinder about the size of a golf ball, the Kilogramme des Archives became the definition of a kilogram from which other copies were made.

But this didn’t last. The platinum cylinder had to be replaced by a platinum-iridium alloy version, supposedly because the alloy was far more stable of a material. But that also didn’t last, because scientists eventually realized that despite their best efforts, the new alloy cylinder was (very slowly) absorbing microscopic particles from the surrounding air and (very slowly) gaining weight (which meant, by definition, everything in metric denomination was losing weight). They eventually gave up on physical reference objects entirely in 2019 and now base the kilogram on esoteric mathematical relationships used in quantum mechanics that normal people don’t understand.

Here’s the point though: it does not matter what the Real Definition™ of the kilogram is! The history of the kilogram is a cute bit of trivia, and how it was derived is an interesting exercise in reproducibility, but the perennial goal here is communication and coordination above all else. A kilogram can be defined however the fuck you want because all that matters in the end is that people have a shared understanding of its meaning.

That’s what’s weird about words: they’re thoroughly meaningless in a vacuum. Words are an arbitrary string of arbitrary phonemes, and meaning can only attach and anchor a word if someone (anyone!) acknowledges the link. Language is clearly most useful when it is broadly adopted, but this also means that the evolution of any major language is going to be an organic and decentralized process outside of any individual’s direct control.

Sure, anyone can make up a word and internally assign a meaning to it, but if no one else accepts the meaning it will remain gibberish. Sometimes these efforts are successful, especially in slang, like that time someone made fetch happen. These types of efforts are far more likely to succeed when an institution throws its formidable weight behind the endeavor. This is what the French do with the Académie Française, the government body with final authority over all aspects of the French language. It’s run by forty members known as The Immortals, whose official uniform is a sword and intricate robes that cost $50,000 to make. It’s as pompous as you can imagine:

The mission of the Académie is primarily motivated by national pride (AKA hating English). For example, as the word email became more widespread, the Immortals endorsed the clever patriotic alternative by concatenating the French words courrier (mail) and electronique to come up with courriel. The government tried to get everyone to use that pronunciation gallop, but outside of bureaucrats who had no choice, it hasn’t caught on. They tried a similar gambit to combat the inescapable popularity of le weekend, but the nasally fin-de-semaine substitute hasn’t worked either.

Despite Herculean efforts and fancy embroidery, the best the Académie can do is meekly pantomime aspirational control over the communication decisions of millions of francophones across the world. People will use whatever words they want, and those words will mean whatever people want them to mean. The only necessary ingredient here is consensus, and pulling that out of thin air is mostly a fool’s errand.

Consider Bostonography again, but instead of a city, picture a map of idea space. If you could somehow map out semantic space and ask participants to draw boundaries of any given word as a neighborhood on this idea map, you’re very likely to get solid consensus for most common words. Some words are relatively unambiguous and precise and thus will generate a high degree of consensus on a small area of this idea space map, while others are more like a vague cloud. Think Beacon Hill versus Dorchester above.

For the same reasons a word might have vague and fuzzy boundaries, it necessarily will also overlap with and be adjacent to related concepts. For example a word like insurrection generally has a negative connotation and will likely bring to mind similarly disapproved of concepts such as disorder, rebellion, violence, etc. If you were to map all those word neighborhoods, they’re likely to occupy the same general province of idea space. Even with disagreements over specifics, all this largely works as intended because the entire point of language is to serve as a tool of communication and coordination. Words are what we use to point people to the right part of the idea map. In other words, the same shared understanding.

Debates over definition boundaries can be a fun conversational frivolity (Are Pop-Tarts a sandwich? Are Algerians Latino? Is Old Town Road country music? Etc.) but they’re most often deployed with other goals in mind. Arguments over definitions are often disguised queries for something else entirely. In Yudkowsky’s example, a factory worker is tasked with sorting blue furry egg-shaped objects (called “bleggs”) from smooth red cubic objects (“rubes”) on an assembly line. This job goes fine until the worker encounters a purple egg and has no idea how to sort it. The worker and his supervisor get distracted by the debate over definitions (Can bleggs be purple? Can rubes be furry?) until the true purpose of the sorting job gets revealed: bleggs contain vanadium ore, and rubes contain palladium ore, both of which need to be industrially processed differently. The ore processing plants do not give a fuck what color or shape their supply chain materials are so long as they accomplish the purpose they were built for. The question “Is this a rube or a blegg?” therefore is used as a good enough (and presumably cheaper) way of solving the ultimate (and presumably more complicated) logistical question of “Does this need to be processed by the palladium or the vanadium plant?”. But without a shared understanding from both parties for why the distinction matters, no answer will communicate any useful information to whoever is asking. It’ll remain a useless question.

Similarly, if a tree falls in the forest and no one hears it, does it make a sound? Another useless question. Both yes and no can be correct answers, depending entirely on whether your definition of sound is “acoustic vibration” or “auditory sensation”. Without first establishing why the distinction between the two definitions matters, no useful response is possible. Here too it’s possible to get infinitely distracted over which definition is the “correct” one, but the purpose of language ultimately is communication, and the more productive avenue would be to acknowledge that it’s helpful for distinct concepts to have distinct words. You can even make up new words (alberzle and bargulum for example) to avoid future confusion.

Sometimes ambiguity is intentional. Was January 6th an insurrection? Again, that depends entirely on why the distinction matters. The reason this might matter to a federal prosecutor would be maybe to determine whether they should file criminal charges under 18 U.S.C. §2383. But outside of that niche analysis, it’s reasonable to be suspicious of questions like this because of the serious negative connotations that insurrection inevitably conjures up. The question is very likely intended as a cover to ask “Was January 6th a Very Bad Thing™?”. Whoever asking the question then would have an interest against establishing a shared understanding for why the distinction matters. Because the goal with dishonest questions like is to score points, not gain information.

I am not the first to notice this deceptive practice. Parrhesia wrote about the word games used in these types of discussions, most blatantly with the word racism. In terms of heavyweight champions in the semantic space, racism is a sought-after bruiser. It’s a strong word with serious negative connotations, and everyone is eager to use it to bludgeon their opponents. Asking whether something is or isn’t racist is the mother of all disguised queries. The goal here isn’t to gain information, it’s to find out who to bludgeon.

And so let’s finally ask the question that brought us here: What is a woman?

It’s a fascinating piece of contemporary commentary that this question is now seen as a weaponized statement in the culture war. The inquiry has become a raucous crowd favorite among the “conservative”^[2] side of the trans identity debate, because it’s presented as a facially simple question which nevertheless is highly effective at confusing the hell out of progressives and needlessly handing wins to sentient bags of douche. 

No doubt many of the progressive efforts in this arena have been thoroughly embarrassing. To their credit, trans-inclusive feminist philosophers have tried to earnestly grapple with the definitional quagmire. But in the end, they introduced a process euphemistically-titled “ameliorative inquiry” only to offer a revisionary effort indistinguishable from just asserting “the dictionary is wrong”:

An ameliorative inquiry into the concept of woman invites feminists to consider what concept of woman would be most useful in combatting gender injustice. This opens the way for a revisionary analysis that can be tailored to avoid exclusion and marginalization. (emphasis added)

But the opposing side shouldn’t get a pass here. The conservative retort to the vexing question is smugly heralded as “adult human female” and while it is miles more coherent as a cognizable concept, it isn’t without problems. As Aella pointed out, secluding yourself inside the “adult human female” bunker will leave you vulnerable to many flanking attacks (e.g. Is someone who gets their uterus removed still a woman under this definition?) because the “but what about” response train is endless and will inevitably find a way to run you over. Lily Alexandre, a trans YouTuber, pointed out in their (thoughtful and highly-recommended) video essay on the subject that the very basic blocks of the “adult human female” definition leave us with many unanswered questions. What is an adult? Is it anyone over 16 years-old? 18? 21? 26? More poignantly, what is a human? This is not so facile a question when you realize how endemic campaigns to dehumanize outgroups^[3] have been to mankind’s violent conflicts.

Which one of these is the correct definition? This is a meaningless question, because there is no such thing as a right answer. You can point to the dictionary, consult linguists, and research etymology all you want but none of that shit will matter. Disagreements over definitions is not a factual problem with factual solutions, it’s a coordination problem. The flaw undergirding both sides of these definition debates is a distracted aversion to the much more important question:

Why does it matter?

I admit a well of deep confusion within this topic. I do not understand why these linguistic slap fights have such emotive zeal burning behind them and I do not understand why there is such a distinct lack of curiosity behind answering that first question.

So what is a woman then? It’s literally whatever you want it to be, because that’s how language works. But to the extent you want your language to remain a useful method of communication, it’s helpful for other people to agree with your definitions.

Reading about “ameliorative inquiry” fills my head with a cacophony of WHY WHY WHY WHY because I do not understand why anyone bothers. Trans-inclusive feminists argue that a rejiggering of the definition of “woman” is necessary for “feminism” to be “inclusive”.^[4] But if the “solution” they come up with requires changing the definition of a word anyways…why bother keeping the word at all? So far the suggestions they’ve come up with have been marred with serious and intractable deficiencies, from both under and over-inclusion.^[5] Instead of forever debating whether sound is “acoustic vibration” or “auditory sensation”, why not just make up a new word?

I like precision in my language. Whether I’m speaking or writing, I want some reasonable assurance that whatever message I am communicating is conveyed clearly to my audience. I want both myself and my audience to simultaneously point to the same spot on the Boston map. Words are my arsenal in this jihad. Yet despite my adoration for these tools, if a word gets dull and starts failing at its basic task of communicating a shared understanding, it is useless and I have no qualms with tossing it in the figurative garbage. I may keep some around for their poetic or aesthetic value, but everything is subordinate to the mission of clear communication.

Multiple meanings in words are not always a problem, and can often be goddamn poetic. The issue here is getting close to the equivalent of a Bostonian and a Seattleite vociferously arguing about the boundaries of Beacon Hill without even realizing they’re referencing maps of entirely different cities. Regardless of how you personally feel about what the definition of “woman” should be, the fact that this has apparently turned into a hot debate is abundant evidence that we’re starting to deal with a dull knife here. I want to throw “woman” in the garbage. Because of its high potential ambiguity in some contexts, my solution personally has been to increasingly try to avoid using the word “woman” because I can’t always reasonably certify that my message will be received the way I intend it to be.^[6]

That’s why as much as it is a source of derision and mockery (much of it earned) I actually appreciate the lexical intent behind “birthing person”. The common justification trotted out (that it’s necessary to include the theoretically-possible transman who somehow can get pregnant and apparently suffers no dysphoria from carrying a fetus to term) is completely daft. But it’s still true that for many different reasons, not every “adult human female” can get pregnant. So while it may be aesthetically horrific (I can’t help but picture spawning vats in a laboratory) “birthing person” is undeniably more precise within the context of human pregnancy discussions. Credit where credit is due.

Yet this fidelity towards precision is not necessarily reciprocated. For example when Them wrote about a survey on trans dating preferences, the author tried to feign surprise that the vast majority of lesbians who were trans-inclusive in their dating were willing to date transmen. This would only be a shock if you “think in words” to borrow Parrhesia’s phrase and ignore factual reality where whether someone finds you attractive is not likely to have anything to do with your internal gender identity.^[7] Any efforts to cut down on this apparent ambiguity (for the record, I think people are just pretending to be dumb) such as with “biological women” or “super straight”, quickly get derided as hate speech. The answer might be obvious, but nevertheless it remains worthwhile to explicitly decouple the blurred debate: Are you objecting to the neighborhood name or the neighborhood itself?

The mixed results outlined above lead me to conclude that a principled adherence to precision is not really the motivating principle behind the language revisionist efforts.

In the few years since I wrote about non-binary identity, I am not any closer to understanding what the term means. If anyone identifies to me as non-binary, it’s the equivalent of someone telling me the brand of shampoo they use: What am I supposed to do with this information?

The debate over the definition of woman has diluted its meaning to such a degree that it’s close to joining the same ranks. If anyone identifies to me as a woman, the same question and more: What am I supposed to do with this information? What new information has this communicated? Why should I care? Why does it matter?

I have some theories, but they’re both uncharitable and unsatisfying, because they don’t fully explain the phenomena. For example, in most industries, the title of vice president is only used by a handful of senior executives within a company. But in the financial industry, they hand out this title like Tic Tacs. To anyone who isn’t wise to this practice, interacting with someone with the vice president title likely misled them to assume they were dealing with someone with much higher authority and importance than the reality reflects. Similarly, I gather that most people assume that anyone who goes by the title Doctor to be some sort of medical physician, and the title-holders who aren’t physicians don’t seem very eager to correct the misconception. Realtors also have a financial incentive to play fast and loose with neighborhood boundaries. If their property listing is located inside an unsavory neighborhood, they might have an incentive to list it under an adjacent neighborhood with a better reputation. If words mean whatever we want them to mean anyways, each of the above can try and claim plausible deniability.

Those examples all share deception as an element. Perhaps the theory here is there is an expectation that the word woman will (intentionally or not) dredge up in people’s minds everything else tangentially associated with the concept. It’s the one star used to offer you the entire constellation. Or the barnacle that gives you an entire container ship. When a word has accumulated such a formidable coral reef within our collective consciousness, it might be too tempting to just walk away from it.

But why put in so much effort to occupy the neighborhood? Even if you accept my uncharitable theory, what exactly do people hope to gain? I can only sort of maybe identify scenarios where constellation dredging is sort of maybe relevant/beneficial. Maybe if I told someone I intended to buy them a clothing gift and they told me that they identify as a woman, I would be expected to read between the lines and assume this to mean “I am hoping that you select clothing that is generally identified as feminine within our shared cultural understanding”. But it would be better for both of us if she just told me this directly, without inviting potentially insulting or erroneous assumptions. To the extent that woman is a cluster of traits, I struggle to contemplate a scenario where communicating the cluster is a more efficient or more thoughtful method of communication than just communicating the specific pertinent trait. Just tell me what you want me to know directly. Use other words if need be.

My advice is to flee the area. The woman neighborhood will be subject to artillery bombardment for a long time to come. It isn’t worth it and there are some open listings available in nearby neighborhoods. In a few years, you might even forget why attacking/defending it was so important. If you identify as someone with an intense attachment to a specific definition of the word woman (no matter in which direction), my question to you is: Why does this matter to you? How does it help me not get lost in Boston? I’m listening.

1. ^{^}

   Survey responders were not relegated to answering only about their own neighborhood, so this lack of consensus most likely reflects how underrepresented Chinatown residents were in the city-wide sample. This is perhaps a cautionary tale about outsiders weighing in on local issues they’re unfamiliar with.

2. ^{^}

   This essay is ultimately about the precision of language, and so I fully own up to the irony that I am using less than precise terminology here.

3. ^{^}

   I am funny, I know.

4. ^{^}

   I’m using quotes to indicate I have no idea what these words mean in this context.

5. ^{^}

   See Tomas Bogardus Why the Trans Inclusion Problem cannot be Solved (2022) for a thorough catalog of the issues with the “Ameliorative Inquiry” approach.

6. ^{^}

   This should be a concern for trans individuals too — if someone insists that I see them as a woman, how do they even know that my understanding of the term is in accord with theirs? My internal interpretation of the semantic label “woman” could be “wool sweater” for all they know.

7. ^{^}

   In a previous essay I used the example of a woman-identifying masculine-presenting Jason Statham.