Whenever humans come together to do anything, they organize themselves either according to principles of cooperation or principles of coercion. Cooperative ventures run the gamut from neighborhood associations to congregations to civic organizations to markets. Coercive ventures typically fall into one of two groups that are, depending on your perspective, very different or very similar: government and organized crime. While a democratically elected government may look like a cooperative venture, it is only cooperative from the perspective of the majority. To everyone else, it is decidedly coercive.

The important difference between cooperation and coercion is that, when things stop working in a cooperative venture, one is free to leave. That ability to walk away fuels a dynamic process that leads to innovation and improvement. By contrast, the inability to walk away from coercive ventures leads to stagnation and ossification. For contrasting examples, compare FedEx to the US Postal Service, Southwest to Amtrak, private schools to public schools, and obtaining a replacement debit card to obtaining a replacement driver’s license.

Those more trusting of government say that, despite government’s flaws, it’s better to trust important activities to coercion than to the vagaries of cooperation. And so, across the planet, we have governments controlling – in full or in part – health care, education, power generation, communication infrastructure, and all manner of other “essential” activities.

Yet there is something more essential than any of these that we routinely leave to cooperation: language. The ability to exchange thoughts – from the basic and immediate (“I’m hungry.”) to the complex and far-reaching (“We hold these truths to be self-evident, that all [people] are created equal…”) – is more important than health, education, or anything else one might conceive. Because, without the ability to exchange thoughts with each other, we cannot bring any of these other things into being.

As a product of cooperation, language evolves, just like markets. Lingual “entrepreneurs” invent new words, phrases, and grammars. Just like customers in a marketplace, people try out the inventions. Where they sense improvement, people repeat. Where people find words, phrases, and grammars too cumbersome, ambiguous, or unpleasant, they stop repeating. Just like customers’ changing needs and preferences cause them to abandon old products for new, so too do people abandon old ways of speaking for new.

All of this should be alarming. We construct our legal documents, encyclopediae, technical manuals, and all manner of vital documents using words, phrases, and grammars that evolved – and continue to evolve – at the whim of people involved in the cooperative venture of communicating. Governments have little to no control over the matter.

One such major evolution is taking place right now in English – the evolution of the singular “they.” While this pronoun has a rich etymology, what matters to most speakers is how well the word serves our current needs. And here we face a problem. On the one hand, we need gender inclusive pronouns. Using “he” to mean either a male or a person of any gender makes non-males feel as if they are linguistic afterthoughts. “It” is dehumanizing. “He/she” is cumbersome, not entirely inclusive, and not a single word anyway. But on the other hand, evidence suggests that “they” isn’t a good solution either.

In the 1300s, people used “they” as both a singular-neutral and a plural-neutral pronoun. In the mid-1700s, usage evolved and “he” took on the double-duty of a singular-neutral pronoun and a singular-male pronoun, leaving “they” as a strictly plural-neutral pronoun. And that’s the way teachers and editors corrected speaking and writing – until recently. The Oxford English Dictionary has now given its blessing to the singular-neutral “they.”

To be useful, language must conform simultaneously to two contradictory principles. The first is that a language needs to be standard. When a speaker converts a thought into a string of words, those words need to recreate that same thought in the listener’s mind – and that requires that we agree on what the words mean. No language accomplishes this perfectly. But, other things equal, the closer a language comes to this ideal, the more useful it is. The second, and contradictory, principle is that to be useful, a language needs to be able to evolve so as to express new ideas and to express old ideas more efficiently. In short, the most useful language is one that is simultaneously fixed and fluid.

And that brings us to the transformation of “they” from singular and plural, to plural, and now back to singular and plural. Why should any word, other than a noun, distinguish between singular and plural? And, while we’re at it, why do we waste so much time learning noun-verb agreement? I, you, and they “walk,” but he or she “walks.” I “am,” but he or she “is,” while you and they “are.”

If language is a cooperative phenomenon, why didn’t such seemingly useless complexity die out over time? The same forces of profit and loss that weed out inefficient businesses in another cooperative phenomenon – markets – should be at play here. The “profit” is successful communication, and the “loss” is the cost of navigating the rules. It’s tempting to say that complicated noun-pronoun-verb agreement is akin to a market failure – the “communication market” should have weeded this stuff out, but didn’t.

But that’s lazy thinking. If a costly phenomenon persists in a cooperative environment, it’s more likely that there is an as-yet-unseen benefit that outweighs the cost. One possibility is that the complexity evolved as a form of error correction. If I say, “I am watching football tonight,” but you don’t hear the “I”, you can infer that I said “I” because of the verb form, “am.” The noun and verb agreement provides the opportunity for error correction. And if this sounds a bit far-fetched, consider supporting evidence that, if not definitive, is at least interestingly circumstantial: the second-person pronoun, “you.”

We have long had the same problem with “you” that we are now creating with “they.” “You” is both singular and plural. And that double-duty causes problems – particularly when someone is talking to a person within a group. When the boss says, “You need to pick up the pace,” is it you personally who is in trouble, or is it everyone in the room? Sometimes the only way to know is to watch the boss’ eyes. To have to rely on sight to clarify whether the speaker’s “you” is singular or plural is incredibly inefficient and error prone.

We know the double-duty we require of “you” is a problem because pretty much everywhere English is spoken, people have developed colloquial variations to distinguish the singular from the plural. In Pittsburgh, “yinz” is the plural form of you. In the U.S. South, it’s “y’all;” in New Jersey, “youse,” elsewhere it’s “you-uns,” “you guys,” “you lot,” and “ye.”

These variations evolved precisely because language is a cooperative phenomenon in which people have found a need to distinguish between the singular and plural pronouns. If that need exists with the second person, it will also exist with the third. It is time that we had a gender inclusive form of the third-person pronoun. But evidence suggests that asking “they” to pick up the slack by doing double-duty isn’t the right answer.

But, I may be wrong.

And that’s the beauty of cooperation. Ultimately, my opinion doesn’t matter. As we have done for centuries, we will rely on hundreds of millions of English speakers employing trial and error. Each individual will adopt solutions that appear to work and discard those that don’t. Better solutions will naturally spread in usage while poor solutions will wither from disuse. Together, through voluntary cooperation, we’ll find the right answer.

And if we’re willing to trust something as supremely important as our language to cooperation, how much more should we be willing to trust to cooperation our education, health care, wages, and the many other things we think are so important as to require coercion?