Shunning, tolerance and economic solidarity

I’m about to get very gritty with this post, so I need to preface by saying that I think this is an awesome and amazing project. I’m very excited to see what happens. Okay, please buckle your seatbelt now…

Imagine a future in which circles is used in a very conservative neo-Amish community somewhere. By now cryptocurrency is such old news that even the Amish have embraced it, yet a lot of people don’t really understand how it works because it was invented decades ago.

Well, recently a member of the community called Jeb has been discovered to have purchased some buggy wheels from a previously shunned member. Gossip and outrage follow. This cannot be tolerated. After much bible study, eager consternation and brow furrowed reflection on the part of many righteous community members, Jeb suddenly finds that they and their family are suddenly shunned. All of the circles “trust” relationships they relied upon are suddenly revoked, and as a matter of survival they’re now forced to connect with other shunned people, and more generally with “outlaws” of various circles and communities.

If there’s not a more tolerant yet still high trust community of circles users out there for this person to join, and this is a matter of survival of them and their family, this person will be extremely vulnerable to manipulation and coercion by individuals and groups who don’t necessarily play the money game by the intended rules. For example, it’s not hard to imagine some terrifying and reckless person imposing on everyone a class structure where some people get 1 income, some people get 2 incomes and some people 5 incomes, etc. This could easily be enforced by coercion or physical violence.

Now, I’m tempted to think that using zero knowledge proofs to obfuscate transactions could help solve this problem, but I’m not convinced. And of course adding anonymity would make it harder for communities to catch sybil attacks. I still think anonymity could actually help strengthen economic solidarity overall, especially if there were ways to detect and out sybil attackers, along with processes of accountability that helped to educate and reintegrate them.

Otherwise, I’m concerned that people who don’t “get it” will overrun the intended game theory with their own rules that they think are better and make more sense, or that they feel serious pressure to play.

The deeper issue, to me, seems to be that trust is such a very subjective and contextual thing. For example, there are people who I would trust with my life, but who I wouldn’t necessarily trust not to sleep with my significant other. There are people who I would trust to act in a loving and compassionate way toward everyone around them on a personal level, but who I wouldn’t trust to care much about the stakes or the gravity of tolerating multiple accounts, who might see it as a victimless crime or even as fair depending on who did it. And there are people who I trust to understand the nuances of circles game theory, to understand the vision, but who are stuck in relationships with people or groups where abuse and manipulation are are common and tolerated.

So how do I factor all of that into my decision about who to “trust”? What can I do if someone with community power doesn’t “trust” my moral compass because they saw me purchase some marijuana on the blockchain, and decides they don’t “trust” my economic integrity? What would I do if someone who held power over me threatened me or someone I care about to stay hush about fact that they themselves had three additional incomes?

This seems like a challenge for some deep pedagogy and design thinking, to say the least. Is “trust” the right word here? It seems loaded, in much the same way that facebook’s “friendship” is loaded. But changing the word wouldn’t change the game theory.

When you add the messiness of money (which represents security and survival) on top of all that, I worry that a tool like this could easily point a community head on into facing all of its rifts. Maybe in the end though, I just need to let go and accept that these rifts will be faced some way or another, and that people will meet these challenges when they get there.

As Billy Joel once sang “we didn’t start the fire, it was always burning since the world’s been turning”.

1 Like

That’s an important issue with Circles very nicely explained!

I totally agree, the social aspect of Circles, which makes the individual user depending on his social graph for “money”, together with complete transparency of the Blockchain and the conceptual problem that someone cannot change her identity (ie. account/address) because that’s the base of the social and trust graph makes this aspect vital for Circles.

If zk-snarks work would be a great research task, I cannot see how this works with transitive payments. Maybe some kind of mixer could work. Or incentives which are good enough that each group would have “gateways” which are morally neutral - but this again is something users would have to “trust” in.

1 Like

Regarding this possibility:

For example, it’s not hard to imagine some terrifying and reckless person imposing on everyone a class structure where some people get 1 income, some people get 2 incomes and some people get 5 incomes, etc. This could easily be enforced by coercion or physical violence.

Is it so easy to imagine because that’s the world that we currently live in? Hopefully for most of us the threat of physical violence is abstract and unlikely, but many would say that what the makes fiat currency “real” is the state’s monopoly on the legitimate use of violence, and what makes that monopoly legitimate is the consent of the governed. In your example, that’s the willingness of somebody who gets 2 incomes to trust all 5 circles accounts of somebody in the higher class. Presumably they tolerate this because buying into the hierarchy allows them to collect twice as much as the plebes at the bottom.

But the thing is, despite participating in similarly structured inequality in some abstract way, I don’t really feel like I have consented. And I don’t know what the first steps would be to revoke that consent. But if circles was mediating this class system, I would know what to do: Build a network of trustworthy people, learn to support each other, and revoke trust in untrustworthy people until they’re willing to interact as equals.

Circles takes the “who is willing to support whom” problem and changes it from something you have to guess about into something you can look at. You can look at the trust graph and say: “We’re about to untrust the jerk at the top, are we producing enough food for each other that we can survive without him?” There may have been community support for that kind of thing all along, but with Circles you have more reason to be confident in your assessment of the situation–especially if you’ve been practicing with Circles all along.

I doubt communities that use Circles will see a complete elimination of coercive economic behavior–power structures of that sort have been with us a long time and they’re not going to be easy to rid ourselves of. But overall I think we’ll find that such things are harder to maintain under Circles because there’s a specific, actionable proxy for the consent of the governed. It’s not abstract, it’s a software setting.

Circles-trust may never converge on actual human trust, which as you’ve pointed out is a tricky concept. But we currently all “trust” the central bank because we were born into a system where that’s what everyone else does, and Circles-trust is far closer to real trust than that is. Also, who knows what Circles will look like decades in the future? It seems likely that it will resemble real trust even more by then.

This seems like a challenge for some deep pedagogy and design thinking.

I definitely agree, but the situation with Jeb is more of a “how do we want to conduct ourselves as a community”-issue and less of a “how should the monetary infrastructure work”-issue. So like, more pedagogy and less design. Unless I missed the feature request in your story somewhere :slightly_smiling_face:

I also think that stories are a good way to explore the interesting new world that I hope Circles creates. I have no cred as an author, but I aspire to write fiction one day (scifi, probably). Maybe one part of the pedagogical effort should be a collection of short stories that frames Circles-trust in a way that prepares people to wield it such that Circles-using-neo-Amish communities of the future can deal with Jeb in a more balanced manner.

I don’t know any Amish folk, but I don’t think that shunning is supposed to be a death sentence. So if there’s a future where they’re going to use Circles-trust as part of that practice, then we had better make sure that we’ve set them up to know what they’re doing when they revoke it.

1 Like

Well I haven’t thought of any “infrastructure design” level feature requests, but for example my remark about the usage of the word “trust”, which I think is very loaded and too loosely defined, is a design issue. For example, I could imagine instead building implementations of Circles around the language of “agreements” that you form with peers, agreements that you will each only maintain one income, and that you will hold other peers accountable if you have found them to have multiple incomes. From my perspective it’s much easier to press a button that says “this person broke our very specific agreement” than it is to press a button that says “I don’t trust this person anymore”. The language of “trust” is way overloaded as it’s currently used, I think. The creators of this project understand that “trust” in Circles is implicitly about whether or not you about the sybil attack issue or whatever, but “trust” is a very big word. New users will conflate the Circles concept of “trust” with all kinds of other very deeply charged and personal forms of “trust”.

Imagine that an overall well-intentioned but somewhat ignorant person decides to collect multiple incomes. They don’t mean to scam or hurt anyone, it’s just that they’ve been convinced by a more malevolent individual that gathering multiple incomes is entirely justified for some convoluted reason. Now imagine that this person has been caught and suddenly gets informed via the user interface that 10 people suddenly have “revoked their trust”. Can you imagine the sheer weight of that? I think it could be unbearable for a lot of people. Well and on the flip side, can you imagine using the interface to “revoke trust” in someone you care about, and who you really do believe in could come around, even if they messed up? Not me. I’d be deeply conflicted about pressing a button like that. In fact in a lot of cases I’d probably decide that for now I valued my relationship with this person over this idealized integrity of the circles currency, and would hesitate to “revoke trust” or “confront” them.

If on the other hand the interface provided me with less overloaded language that was suggestive of some well defined accountabilities, I think that would be much easier to deal with. I think there’d still be room for the language of “trust”, as long as it was explicitly connected to those specific accountabilities.