Will our protocol-based future protect liberty?
By Jim Harper
I hope it’s not a historical accident and conceit that language — the way people talk and write — is a bastion of human freedom. By its nature, language is deeply resistant to control. Its ongoing creation and revision are largely decentralized, spontaneous, and opaque, even to language’s user-controllers. We don’t think we’re in charge as we enforce or remake how we speak and write.
Will the languages we use to organize our society in the future be similarly protective? I’m speaking of the protocols that transport an increasing share of our communications and that will administer larger and larger swathes of our activity. If we are free in our use of language but not free to do as we please with it, our horizon is rather dark.
I think technological determinism is correct. Language gave humans our power. The printing press reallocated it. Advances in communications and record keeping shifted things again, creating the administrative state and the multinational corporation.
In a highly stylized sense, governments and corporations today vie for power, the spoils being all of us. But we have language, “systems of sometimes breathtaking sophistication [that] arise through what appear to be the mundane and commonplace traits of everyday speech,” writes Guy Deutscher in “The Unfolding of Language: An Evolutionary Tour of Mankind’s Greatest Invention” (Holt Paperbacks, 2006). Deutscher is talking about emergent order, a phenomenon Neil Chilson assesses as both a policy and personal dynamic in his new book, “Getting Out Of Control: Emergent Leadership in a Complex World” (New Degree, 2021).
Part of why language helps keep power distributed is the fact that it permits hidden meaning and subtext. Small communities have literal code words. Vaclav Havel’s plays were inscrutable enough to confound authority but sufficiently pregnant with meaning for Czechs seeking change. Havel was also direct, and he wielded poems. I hope and trust he has many analogs in present-day China.
Happily, there is no changing the dynamics of language itself. But we communicate over distance, manage our affairs, and organize as citizens and consumers within protocols that are more centrally designed. They are blunt instruments compared to human languages.
Transmission Control Protocol/Internet Protocol, the basic protocol set of the internet, is supposed to transmit content neutrally, but does it? There is greater risk of skew in higher-level protocols, such as those translating text, audio, video, or other content for transmission via the transport layer. The practices of internet service providers are the subject of a shopworn policy debate.
Biases are obvious in things such as the data layer. Have you ever noticed that campaign finance data is readily available, while government spending data is not? That’s society’s subconscious (IMO, errant) decision that democracy should be improved one way and not another.
As protocols do more and take on administrative tasks such as value transmission — yes, I’m talking about blockchain and cryptocurrency — no blog post is complete without them these days — it will be essential to have broad understanding of their biases. And I mean broad. Protocol literacy will soon be a part of financial literacy, economic literacy, and even social literacy. At least 5 percent of the population should be able to understand the mechanics of protocols the way perhaps 5 percent of us really work on our reading and writing. (Did you notice the switch between Latinate and Germanic words in the first paragraph? The iambs? I’ve also been reading Farnsworth.)
I find it unlikely that people will readily modify digital protocols the way spoken-language communities do when their needs are unmet. Changes to protocols are much too clunky, and network effects provide much of protocols’ value. The goal in amending a protocol would be to bring everyone else along. That means huge coordination problems if some small group wants to convince everyone else that a “corner case” — some functionality left out of a protocol design — should be moved to the center.
Given protocols’ rigidity, it would be wise to have a lot of protocols to work with. People should have good exit options because voice will not do.
There is an argument in the cryptocurrency community about whether there should be one true protocol, Bitcoin, or a diversity of coins and tokens, some of which may differ from others only by degree. The Bitcoin maximalists versus — well, the maxis call the others a lot of names.
As a thought experiment, I’ve conceived here of our computer protocols as part of a stack of technologies continuous with spoken and written language. All of them we use to make sense of, communicate about, and act in the world. If we are going to protect freedom, I conclude that the coming world of protocols should be polyglot.