The Theology of Consensus | Hacker News Reader

The Theology of Consensus

berkeleyjournal.org

33 pointssmacktoward21 comments

I am a Unitarian Universalist, and like the Quakers consensus is at the core of my faith's practice.

As the president of my congregation, I can attest that consensus can be a difficult and time consuming process. During my time in leadership, I have had to navigate some very difficult and contentious issues. Facilitating an open consensus based process, led to everyone feeling their thoughts and feelings were honored. In the end consensus led to strongly unified decisions.

I know that pure consensus doesn't work in all situations. But it is a powerful tool in the process of decision making. Even when the question is ultimately decided by a vote, including consensus can lead to much greater acceptance of the final decision.

I'm really interested in alternative political or economic organizations... but there's an enormous gulf between those who look at what we have, deeply analyze it, attempt to understand how and why we got where we are, do a 5 Whys-type analysis on the problems, and think about how we might be able to avoid them, even as they are aware that what we have is likely a very strong strange attractor... and those whose approach is basically "Capitalism's, like, evil, man. Let's do whatever that isn't. Consensus, is, like, probably really good. Let's do whatever someone somewhere theorizes might lead to it."

I wish the latter were a strawman, but, alas, at times it's rather accurate. Particularly the part where someone, somewhere comes up with a really nice sounding theory and it makes the leap straight to Accepted Truth, without passing through much or any intervening testing of the theory. You can also see that happening just over and over in the education field, too. Entire massive multi-decade enterprises, if not occasionally multi-century, can be erected in the economics or education fields on things that often turn out to basically have flattered some group of intellectual's fancies, and have little other basis. (Everybody's lists might differ, ahem, but everybody interested in either topic ought to be able to come up with at least a couple they'd describe that way.)

On the direct topic, note that there's really no trick to governing, say, 20 people. Anything works at that scale, especially for a group of self-selected people who came together due to some unifying interest, which is Easy Mode for governance. Your hypothetical governing method isn't even interesting until you're at least bumping up against the Dunbar Number (regardless of whether it is "right" or "true", it's still a decent heuristic), and you're still four or five orders of magnitude away from something that can run a modern society at that point.

I've recently been reading Norman Cantor's really excellent history of the Middle Ages and thinking that we are seriously missing a science that studies forms of human collective organization, which have varied enormously over time.

In business studies there is an extremely minor area called "theory of the firm" that almost no one pays attention to, and it is narrowly focused on modern corporate structure.

In economics there has been some work on cooperatives and partnerships vs corporations, which is why we know that cooperatives suffer extreme diminishing returns as they cross the Dunbar threshold (a fact that is simply flat-out ignored by the alt-community folks you so accurately carricature [1].) There has also been some work done in economics on agency issues in corporations (looting by managers, in particular, especially focused on the savings and loan crisis in the US in the late '80's or early '90's).

Sociologists and criminologists have done a little work on different kinds of collective organization, but no one has attempted any kind of grand cover of all the different ideas and structures humans have employed to solve the problem of getting stuff done in groups of more than a few close friends and family, which includes but is not limited to:

1) military organizations 2) churches and monastaries 3) town corporations 4) craft guilds 5) trade unions 6) civil bureaucracies 7) chartered companies 8) partnerships 9) cooperatives 10) corporations (in the modern, post-1850's sense) 11) criminal gangs 12) collegial organizations (universities)

Reading about the Middle Ages, I am struck by how various social innovations were really about attempting to figure out how to solve the problem of collective organization, and how new institutions that solved certain problems better edged out older ones that failed to adapt (in the later Middle Ages universities and town corporations took over many functions that had been previously handled by monasteries, for example.)

We only have the most scattered and incomplete understanding of any of this, and yet finding effective solutions to the problem of collective organization is the most pressing problem for any human civilization. The disasters of the total state in the 20th century were essentially possible because no one could prove they wouldn't work, and while we are somewhat better off in that regard today there is still an enormous amount of work to be done, to the extent that I think the problem really needs its own field of study, rather than being scattered across economics, business, politics, sociology and history.

[1] It manages to be both accurate and carricature because the real thing is as genuinely grotesque as what you describe. It is a carricature of itself.

> finding effective solutions to the problem of collective organization is the most pressing problem for any human civilization

Threat models need to include active subversion, lest we tar some ideas as structural failures of cooperation rather than structural failures of defense and risk management.

I certainly agree that the model must include that, and I think complete failure to consider this is one of the major reasons the simple answers fail even at small scales. If your model isn't robust even against "Bob is deciding to show off to the women by putting down Phil, and Phil is getting angry", and especially if your model actively shuts down everything that might actually resolve this problem, then you certainly haven't solved governance in any reasonable manner. And that's just one political game we humans play intuitively, without conscious intervention... if that's enough to tear your system apart, you can hardly have been said to have had a system in the first place. Both gaming and active subversion must be accounted for.

Yes, we will need more than spreadsheets to model resonance in the feedback loops (visible and invisible) which influence outcomes within and among complex systems, from single humans all the way to inadvertent collectives. Today, currency and law are used to implement feedback loops, but we have since invented many other forms of expression and composition.

I think we can look a little closer to home. A lot of software 'ideas' are out there, but few people actually test them.

You might like the book Anarchy Unbound: Why Self-Governance Works Better Than You Think. The entire book is dedicated to analyzing historic examples of Super Hard Mode for governance, and how solutions can emerge that reduce violence and other undesirables without what we would recognize as a government or state.

It's a pretty accessible pop-econ book with a compelling premise and a good amount of historical examples and citations.

http://www.amazon.com/Anarchy-Unbound-Self-Governance-Cambri...

I just tried to upvote your post and may have downvoted it instead :( I wish the upvote and downvote buttons were separated horizontally at opposite ends of the post title, instead of vertically by 1mm.

I've still got 1 point on that comment, so either you cancelled out someone else's up vote, or you cancelled out someone else's down vote. :)

I think it's somewhat disingenuous of the article to claim to be a history but maintain an almost entirely adversarial tone. It's clear that consensus has issues at scale, but a more fair analysis might include some settings and constraints where modern consensus-based decision making also does pretty well.

For instance, with large, disparate groups of people intending to accomplish something ambitious, things are bound to get hairy no matter what process you use, and all things considered, the consensus process is probably not helping much or the best choice to use. The article is full of examples of this.

On the other hand, for small-ish groups of people, who aren't trying to pull off any major new projects, but are instead more interested in maintaining some sort of infrastructure and taking on smaller tasks, my experience has been that formal consensus is pretty great. You might argue that almost any process works with a small group of people... if that's your experience, well, maybe you spend time with less ornery and headstrong people than I do.

Living with people is a great example of where consensus can work really well. For small-ish groups, where everyone has to (literally) live with the decision, and especially with the other people who made it, my experience was that the formal consensus process was a pretty great way to ensure that no one felt left out, harbored silent resentment, etc.

When we used a consensus-based process to run a collectively-owned co-op where fourteen or so of us lived, it was pretty great. It didn't necessarily lead to more optimal decisions (though it often did), but it did lead to me realizing that theoretical optimality isn't nearly as important to most people as feeling that their voice is being heard, and feeling included in the process.

Obvious caveats of good faith etc apply, but in practice, it was rarely an issue.

I've found that informal consensus, the tautological type you have when you don't label it, is good. This is just what you have when everyone has hashed something out and found a solution that fits.

But formal consensus is a horrible thing where you're slowly bullied into dropping your objections to something you don't agree with under the guise of everyone agreeing. It's a nightmare because when you're openly bullied you don't have to get up after and tell everyone how wonderful it feels.

I'm okay with being the loser in a vote and to have to do your thing, but to have my thing twisted until I'm told we're all happy. Not.

It's a compliance trick. The person running it or most invested in it can pick almost anything they want and the victims of the process are often very ... cult-like ... in defending the decision. Which has to be the case because consensuses are literally never defendable with technical arguments or the argument would have sufficed.

Interesting history -- but consensus building is really just another name for politics.

Anyone who has participated in an organized protest can tell you that they're chaotic, and when things start to organize, it's usually because one person or faction takes control and starts barking orders. Other factions emerge with other demands, and usually consensus is reached through "back room" dealing.

"Consensus building" as a term has become synonymous with the practice of approaching all members of the group ahead of a decision event, obtaining their approval and input on a plan, and giving them just enough of what they want to get them to agree to it. But it's not about being inclusive -- sometimes the consensus is "Yeah, we all agree on this but John doesn't, so we'll all agree to disagree with him but we're doing it our way."

That is not what consensus is. That's what majority politics is. Two groups that come to agreement through their leadership is not consensus, it's leadership-based agreements. There can be dissension within each of the groups. So you don't have consensus.

Any organization that becomes sufficiently large has leaders to improve efficiency.

> Interesting history -- but consensus building is really just another name for politics.

How so? I would say that politics is the day-to-day actions of governing a country or region. While some political systems are ostensibly based on consensus or at least majority preference, others are not. In fact, I would say that political action becomes less relevant as consensus grows.

This is unconvincing. We might stipulate that consensus is inadequate to governing a nation, but TFA seems explicitly to address the decision-making of activist groups. There's no reason why any particular activist issue should only be addressed by one group (in fact every issue I can think of seems to be addressed by dozens), so if you don't like how Group Z makes its decisions, then just join Group Y, if they'll have you. If it had any examples of a strictly hierarchical-leadership-style group being more effective in some way than a consensus-based group, then TFA might have had some (anecdotal) point. As it is, it merely seems to say "some people like consensus, and others don't, and by the way the former are wrong."

Yeah, consensus isn't the same thing as finding the optimal solution for all involved. For the latter you need a safe way to register opposition and an efficient way to incorporate it until all nodes understand the entire system and respect all the roles appropriately.

> Consensus can easily be derailed by those acting in bad faith. But it’s also a process that is ill-equipped to deal with disagreements that arise from competing interests rather than simple differences of opinion. The rosy idea embedded in the process that unity and agreement can always be found if a group is willing to discuss and modify a proposal sufficiently is magical thinking, divorced from the real-world rough-and-tumble of political negotiation.

Deal with competing interests by letting them compete. Iterate consensus. Don't wait for complete consensus. Try each competing interest and try again.

> “Consensus is a creative thinking process: When we vote, we decide between two alternatives. With consensus, we take an issue, hear the range of enthusiasm, ideas and concerns about it, and synthesize a proposal that best serves everybody’s vision.”

This sounds like the same reasoning that defines climate consensus. Unfortunately, consensus is politics and not science even if you add the word scientific in front of it. I feel confident, though, most people cannot tell the difference.

If one is interested in a more philosophical debate around consensus decision making among non-hierarchical and anti-authoritarian groups, I highly recommend Mark Lance's essay Fetishizing Process: http://www.academia.edu/1110507/fetishizing_process

Reply on news.ycombinator.com