Bitcoin is a crowd. Crowds can be understood.
I’m currently reading a fascinating book called Crowds and Power by Elias Canetti, recommendation courtesy of @HipCityReg who also wrote some good stuff on it a while back. Since Bitcoin’s having a fun week (and not just all of the is-it-real-or-fake price action - lots of real progress, today most notably from Microsoft), it seems like a great time to do some exploratory rumination on the crowd dynamics of Bitcoin through the lens of this deep and thought-provoking text. So that means we get a Bitcoin issue this week!
The first page of Crowds and Power opens so coldly it gave me chills:
There is nothing that man fears more than the touch of the unknown. He wants to see what is reaching towards him, and to be able to recognize or at least classify it. Man always tends to avoid physical contact with anything strange. In the dark, the fear of an unexpected touch can mount to panic. … All the distances which men create round themselves are dictated by this fear. They shut themselves in houses which no-one may enter, and only there feel some measure of security. … The promptness with which apology is offered for an unintentional contact, the tension with which it is awaited, our violent and sometimes even physical reaction when it is not forthcoming, the antipathy and hatred we feel for the offender, even when we cannot be certain who it is - the whole knot of shifting and intensely sensitive reactions to an alien touch - proves that we are dealing here with a human propensity as deep-seated as it is alert and insidious; something which never leaves a man when he has once established the boundaries of his personality.
It is only in a crowd that man can become free of this fear of being touched.
Crowds and Power opens this way for a reason. Canetti establishes a thoughtful definition of what exactly is a crowd: a crowd is a phenomenon where many different people have discharged and flung aside the distance and differentiation that normally surrounds each of their individual selves. A crowd emerges out of a collection of individuals when differences are cast away; the crowd precipitates back into individuals when differences and differentiation return once more.
Crowds are usually temporary phenomena. What kills the crowd is when the individual members of the crowd begin once again thinking of themselves, their awareness of their own personal boundaries, and their own objectives. They may leave peacefully, they may flee, or they may turn against the crowd itself. The number one thing we need to know in order to understand the crowd is: What differentiation has been cast aside in the creation of this crowd, and in what ways might the return of differentiation destroy the crowd?
Some crowds exist only briefly; others exist rather stably for centuries. Crowds need a reason to exist; it’s not unusual for the crowd to owe its existence to some sort of anger or provocation that is related to the differentiation that gets cast aside as the crowd forms. Crowds can be violent, and this violence is often directed towards any visible symbol of differentiation and distance in its way: that’s why it comes so easily to crowds to indiscriminately smash windows, doors, or barriers of any kind that can be seen as such a symbol.
So long as the crowd is growing, it is strong. A crowd that grows, and that has a purposeful direction in which to go, is almost invincible. A crowd that is under threat from the outside is also strong; threats from the government or law enforcement typically provoke the crowd more, not less, as they legitimize the crowd and induce solidarity in outsiders who may take up the crowd’s cause.
The true threat to a crowd is not outside force. It’s subversion from the inside: the return of individual motivation and individual boundaries among its members. In its most intense form, this is a panic: when the crowd rapidly precipitates back into a dense mass of individuals who are now suddenly and desperately threats to each other; uninvited presences in one other’s personal space; obstacles in the way of the exit. The fear of being touched returns, and we seek to re-differentiate ourselves at the crowd’s expense. What threatens the crowd is differentiation itself. This is why the feeling of being persecuted is an almost universal feature of crowds: it reflects the awareness of this dual threat to the crowd, both from the outside and from within.
How does this help us understand Bitcoin?
Unlike the debt or equity shares of an organization, or even most shitcoin cryptocurrencies for that matter - all of which require some sort of intrinsic usefulness in order to be worth anything - Bitcoin is a reflexive representation of collectively perceived, freely transferrable value. Nothing more, nothing less. All of the technical genius behind Bitcoin’s design and architecture is a necessary starting point; it is table stakes to create an internet currency with which we can start fresh.
The default price for a Bitcoin is zero dollars. But it’s not zero dollars; the price of Bitcoin is some positive number, because there exists a crowd of people who are all alike in that they all believe in the value of Bitcoin. If we want to know what the crowd thinks about Bitcoin, and what the crowd will think about Bitcoin in the future, then we ought to try and understand the crowd and what it wants, whether the crowd will remain a crowd, and for how long.
The origin story of Bitcoin matters a great deal. It was created by the anonymous person (or group of people) Satoshi Nakamoto, who had a specific grudge, and created it at a specific moment in time: in the wake of the great financial crisis. We can think of this grudge as follows: The global financial system that exists today is differentiated and unfair. Some people enjoy its benefits and protections more so than others. Some people can be effectively censored by others. Power and personal responsibility are distributed unfairly within the system. This differentiation is wrong. The collapse of the world’s economy has pulled back the curtain, revealing something important: the vast majority of us are victims, and we are all alike. In the wake of this grudge, the Bitcoin protocol promised something new: a cryptographically enforced level playing field, where no one can be treated differently and everyone is the same. If you believe, come join the crowd.
Today, the biggest questions facing Bitcoin are whether the crowd that currently exists today will either continue to grow, hold steadily, or come apart. Not everyone in the crowd is equally zealous about Satoshi's initial mission, but that is hardly a prerequisite: what matters is the density of its members and the absence of distance between them. Latecomers to the crowd can be equally productive members in terms of giving the crowd density and direction. In the case of Bitcoin, the size of the crowd and the list price reciprocally imply one another: one way or another, if you’re betting on the price, you’re betting on the crowd. The appeal to join is straightforward: humans are naturally drawn to a growing crowd. That part comes easy. The more important questions to ask concern the threats to the crowd. What might make the crowd precipitate back into individuals?
There are two common reasons I frequently hear as to why the Bitcoin crowd will inevitably collapse. The first has to do with energy and economics: Bitcoin’s proof of work system requires a net influx of energy and money burned in order to continuously feed the crowd. Critics argue that this makes Bitcoin effectively a Ponzi scheme: the minute it fails to accumulate new deposits, it dies. But that isn’t necessarily so. If a crowd is motivated, it can endure for a very long time, even if that means continually spending outside resources to do so. So long as they remain a crowd, they’ll bear the cost. For the proof of work as Ponzi economics argument to be valid, in my opinion, there needs to be a convincing reason why the collective drain on external resources will be experienced by the crowd as a prompt to precipitate back into individuals. In my view, the opposite is more likely.
The second reason is a more interesting one: the Bitcoin ecosystem is full of scams, hacks, theft, and other bad behaviour that in nearly every way makes it more dangerous than an FDIC-insured savings account. The argument here goes, if you are playing in the Bitcoin ecosystem then you are a potential target. The experience of being scammed, or even just the threat, ought to be enough to turn people against the Bitcoin crowd and back into disillusioned individuals. Here’s the thing, though: the scams and hacks that have happened to date have effectively all been interpreted by the community as attacks from the outside rather than from the inside. If you are a scammer, or a rotten exchange, or whatever, then you are attacking the whole crowd; you’re attacking all of us, therefore you’re not one of us. (This is a great example of the One True Scotsman fallacy in action!) So long as the whole crowd is able to have an allergic reaction and cast you out at more or less the same time, then the crowd is never under real threat.
So what are the real threats then? Well, we ought to be looking for anything that systemically reestablishes distance, differentiation and personal space among the members of the crowd. The whole fight over the identity of who is the “Real” Bitcoin certainly threatened to do so, although it seems like it’s fortunately turned into more of a harmless sideshow rather than a main event. (Although, to be honest, this betrays my relative naïveté in terms of the politics among the Bitcoin mining community, which is obviously crucial for the whole ecosystem functioning. There could be much bigger problems brewing that I don’t fully understand.) But the real kiss of death for Bitcoin or any cryptocurrency like it, in my opinion, is Proof of Stake.
Here’s a hill I will absolutely die on: a cryptocurrency that adopts Proof of Stake cannot survive. Proof of Stake turns the crowd back into individuals. With PoS there essentially is no crowd; only a collection of self-interested voters who are explicitly ranked, incentivized and differentiated by how much stake they have. Proponents of PoS have all sorts of arguments that it ought to “rationally” work, and have invented layers and layers of protections that they feel ought to prevent bad behaviour of various kinds. But they have no way to protect against the most dangerous thing of all, which is the re-differentiation of the individual, because in PoS re-differentiation of the individual is the point!
Why, you might ask, isn’t this also the case with proof of work Bitcoin mining? Because Bitcoin mining happens on the outside. It projects differentiation outwards, into an inert reservoir who does not care: the energy grid. it acknowledges that some sort of objectively scarce resource must be consumed in order to “vote" on which version history of the blockchain is correct, and then wisely places that voting system as far to the outside of the crowd as possible. Proof of Stake, on the other hand, is like the third week of a revolution where the crowd starts to bureaucratize, organize itself, and convinces itself, “we are able to administer ourselves”. Proof of Stake asks, “surely, if we simply create a good incentive structure, the crowd will be able to administer itself?” Your incentive system may be impeccable, but it’s all for nothing, for you no longer have a crowd. You merely have a collection of individuals, all of whom inevitably recover their sense of differentiation and personal space. Even if Proof of Stake works perfectly, it still won’t work.
Bitcoin does not have this problem; not today, anyway. That’s not to say it won’t in the future! But what I can say is that for Bitcoin to succeed, it needs to successfully remain a crowd. I may do a part 2 about this in the future, but in the meantime read Crowds and Power for an extensive historical survey and rumination on the nature of historical crowds: what sustains them, what threatens them, how they work, feel, move, think, want, and act.
Permalink to this post is here: Crowds, Power and Bitcoin | alexdanco.com
This week’s reading links:
So this could be a very big deal. One of Microsoft’s most important products is Active Directory, which lets users have a robust and secure identity that they use to access certain networks or spaces both within and between organizations. We’ve known for a little while that Microsoft has intended to build an open-source, decentralized equivalent of Active Directory; now we know that this project, called Ion, is going to be built directly on the Bitcoin blockchain and will move to the active mainnnet later this year.
Note carefully that the most important word in the headline is Bitcoin. This isn’t some quasi-blockchain vapourware virtue signalling IBM fantasy project, this is Microsoft saying, We believe that the Bitcoin programming language plus the Bitcoin network is the right choice for Microsoft to build its decentralized identity manager. Among other things, it’s a bit of a slap in the face for the Ethereum community. After all, this is precisely the kind of thing that Ethereum ought to be good at, and Microsoft has been an active part of the Ethereum developer and partnership ecosystem for a while; I really have the sense that Microsoft gave the EVM a long, hard look before this and ultimately said no. I mean, I get it; Ethereum doesn’t work. But Microsoft making this kind of commitment to the Bitcoin blockchain is a huge statement of credibility for Bitcoin, not exactly as an all-purpose programming language (which it isn’t) but as a programming language and environment for certain mission-critical things like identity management. For a week full of big Bitcoin-related announcements, this one stands out to me as one of the most interesting. (Honourable mention goes to HTC announcing a new phone that can supposedly run a full Bitcoin node all by itself. Dang!)
This is a really interesting and zeitgeist-y paper that seems like it came out of nowhere: a few days ago, Marco Rubio and his office released a wide-ranging, grand thesis of a document on innovation and investment in corporate America. It’s a very bold paper, and it makes a lot of pretty sweeping claims I’m not totally ready to accept quite at face value the way they’ve explained them. But it’s really interesting! As I tweeted out, never in a hundred years would I have guessed I’d be reading and enjoying a paper by Marco Rubio articulating a left-leaning economic view in defence of Amazon, among other things.
There are problems with it that I take issue with, but I largely sympathize with the core problem that the paper tackles: large American corporations have historically played a role in our economy where they are what Hy Minsky called the “borrower of first resort”: a continual, dependable economic engine for borrowing free financial capital, putting it to work, and then transforming it into production capital that actually does stuff. For various reasons, the paper argues, this is no longer happening to the extent that it used to, because our large corporations are increasingly financialized: they are acting more like banks than they are like businesses, and so our cadence of productive investment is slowing down, and so this means we’re going to get our asses kicked by China, etc etc. Anyway, I am probably going to dedicate a whole issue to talk about this! So stay tuned for that next week.
A phenomenal interactive post / page / playpen (I really don’t know what to call this!) on complex systems, network density, virality, critical thresholds, everything you’ve ever been curious about really, that needs to be experienced firsthand. Click the link.
Lots of people read Scott Alexander at Slate Star Codex. Sometimes we forget that he’s a psychiatrist! So when he writes about brain stuff, or about scientific and medical research in general, it’s worth reading. And this one sure is a doozy.
Short version: there’s a neurotransmitter in your brain called Serotonin, also known as 5-HT, that is involved in mood and affect in various ways. In the 90s, we started studying a gene called 5-HTTLPR, which codes for a serotonin transporter protein that’s involved in maintaining serotonin levels inside and outside of the cell. Scientists found that one particular variant of this gene seemed to be pretty common in depressed people. Interesting! So we started digging around for what else it could be related to. We already had a working theory around depression and the HPA endocrine system, and lo and behold, we found that 5-HTTLPR was related to that. We looked at the amygdala (a part of the brain that’s long been a favourite suspect of depression researchers) and, yep, 5-HTTLPR had a bunch of interactions there too. These studies are hard to do, so they weren’t typically looking at patient populations much larger than in the tens or possibly hundreds of people. But we made up for it in sheer volume of studies that kept coming back with the same confirmation that these associations were real. They had to be real! 400+ scientific studies, in good journals, all confirmed that this could not be a coincidence.
Until, of course, a group of genomics researchers who actually had real statistical sample size power - 600,000+ people - ran a real study, using more modern tools and with no prior assumptions about what genes were believed to associate with what. And, oh boy. As Scott put it: “This isn’t a research paper. This is a massacre.” Everything - everything - associated with 5-HTTLPR may have been just… not true at all; the product of selection bias, multiple statistical comparisons, and other small sins of research that add up to a very big embarrassment.
One of the huge frustrations I have with academic science, especially in biology, is that I have a deep seated belief that I’m not particularly good at explaining that the combination of statistical tools and publication incentives we employ are fundamentally non-suitable for studying complex systems like biology. Like, not “they’re abused a lot” but I mean, they should be entirely discarded and fully outlawed. (I know NNT has railed on this in general, as have many others. Taleb’s a bit of an extremist in that he basically hates all commonly used statistical methods when studying complex systems, which is fair, but I’m happy to settle with something a little less extreme like Bayesian inference.) As Scott puts it: "The problem isn’t that people studied this [and misunderstood it]. The problem is that the studies came out positive when they shouldn’t have. This was a perfectly fine thing to study before we understood genetics well, but the whole point of studying is that, once you have done 450 studies on something, you should end up with more knowledge than you started with. In this case we ended up with less.”
Well, I never thought I’d be linking to pieces from Glamour magazine in this newsletter, but here we are. This is a fascinating story! If you don’t know who Danielle Steele is, she’s a strong contender, page-for-page, for the most successful author in modern history. She writes beach read-y romance novels that aren’t highbrow literature by any means, but her impact and her story as an author are pretty remarkable - from her workaholic nature, her nine kids, or her 1964 typewriter she uses to write her books.
Snippets readers know I love Skiff’s historical accounts of What-Actually-Happened, and this one is no exception. Booking.com is in the pantheon of legendary tech acquisitions, up there with Youtube, Android, and Instagram. If you don’t know why, or if you do, check it out.
And finally, a submission from Snippets reader Orrin Belden on the different types of complexity we see in science:
Have a great week,
*Several of you wrote last week asking whether my employment status had changed again, as I had signed off with “Alex and the team from Social Capital.” So, uh, funny story, that was actually complete muscle memory. After having written that sign-off every week for nearly four years, it was only a matter of time before it snuck its way back into this newsletter, or somewhere else. Anyway, I remain happily on self-created paternity leave, but thanks for asking.