Stag Hunt and Shared Vision

As requested by @rufuspollock I’m capturing some of the WhatsApp discussion around the Stag Hunt theme here. Includes some familiar names, as well as a newcomer (welcome @Nymrod !)

The discussion began with a reframing of the Prisoner’s Dilemma as a Stag Hunt: the central problem is not simply selfish defection, but the risk of ambitious cooperation when trust, common knowledge, shared commitment, or confidence in the goal are not yet present. “Rabbit hunting” was framed as the safer, risk-dominant option under uncertainty, while “stag hunting” requires more developed coordination infrastructure, trust, and credible shared vision.

From there, the conversation moved into the deeper problem of competing visions. Several participants argued that people do not come to collective projects merely as neutral contributors; they arrive with their own hopes, strategies, identities, and “fully worked out plans.” This makes coordination difficult because collaboration often asks people to loosen attachment to their own preferred future.

A major theme was vision dropping: the maturity to let go of one’s cherished plan so that it can fertilize a richer shared vision. This was not presented as self-erasure, but as a way for individual insight to become part of something larger. Related ideas included “ontological commoning,” nested goals, non-attachment, and the need to hold one’s own contribution without clinging to it as the one true path.

The discussion also explored the metaphor of building a cathedral. A collective may agree that it wants to build something grand, but still faces the meta-problem of deciding on the blueprint. Not everyone is an architect, yet many kinds of contribution are necessary. This raised questions about hierarchy, legitimacy, postmodern pluralism, shared ethos, and how a group can converge on a goal without domination.

A tension emerged between subtle alignment work and tangible action. Some argued that deep collaboration requires philosophical, relational, and ontological groundwork; others pressed for concrete experiments rather than endless talk about how to contribute. This led to the proposal of small, low-risk collaborative projects as a practical test of coordination.

Overall, the thread circled around one core question: How can people with strong, diverse visions become capable of ambitious cooperation without collapsing into hierarchy, fragmentation, endless deliberation, or premature action?

Highlights / extracts

“It’s simpler & more reliable for people to work autonomously towards their own projects (‘hunting rabbits’). Coordinating for bigger projects (‘hunting stags’) is higher risk, takes greater trust, & the stags actually have to be in the forest.”
— Melody Half-elven

“The point… is not ‘always choose rabbit,’ but ‘do not moralize rabbit as defection before you have done the work required to make stag a sane choice.’”
— David Joseph

“Our visions don’t have to be identical, in fact they mustn’t be, but we can and must cultivate the ability to flow in dynamic harmony with one another’s visions and work.”
— Ganga Braun

“People don’t get together ONLY to contribute — they want their visions to come to life.”
— Martin

“We who come to spaces such as this often come with our hands full of gifts, though that interferes with us bringing our hands together.”
— Nymrod G

“This situation demands of us all the maturity and faith to ‘drop our vision’.”
— Nymrod G

“This isn’t a dilemma about defection. It’s a dilemma about maturity.”
— Nymrod G

“Trust that rather than your ideas, you are the gift that you are bringing into the space.”
— Nymrod G

“The most beautiful vision you can imagine is ever partial, with room to grow, and for other people’s visions.”
— Melody Half-elven

“How does a collective decide which vision to follow? I’d liken this problem to deciding on the blueprint of a cathedral.”
— David Joseph

“Not every player on the team can be the goals and shots leader. Some can focus on assists. Others can just defend or do other sorts of set up.”
— Robert Bunge

“We need to start contributing towards certain tangible goals rather than endlessly talk about how to contribute towards certain tangible goals.”
— Martin

“In the dialogical juxtaposition of all those fully worked out plans, a less fully worked out plan — but most probably a richer one — is very likely to emerge.”
— Robert Bunge

“Moving toward a goal can be counterproductive if the goal itself is out of alignment with… the Good, the True and the Beautiful.”
— Nymrod G

“Not detachment in the sense of devaluing one’s own contribution, but rather not clinging on, not seeing it as the one true path.”
— Simon Grant

“A Collaboration mindset doesn’t need to invoke ‘subordination’ or ‘submission’… perhaps think instead of nested goals.”
— Margeret Heath

“Mostly, work with love, and an openness towards how the unifying Vision, and shared Ethos might express itself.”
— Margeret Heath

Possible forum framing

A useful framing for the post might be:

From Stag Hunts to Vision Dropping: Notes on Trust, Cooperation, and Collective Action

The discussion begins with game theory, but quickly opens into a richer inquiry: under what conditions does ambitious cooperation become rational, mature, and alive? The recurring answer is that trust alone is not enough. A group also needs shared ethos, practical experiments, better coordination infrastructure, and the maturity to loosen attachment to individual visions so that a genuinely collective vision can emerge.

Original chat log: ChatGPT - Vision Dropping and Cooperation

3 Likes

Looking over all the different contributions just now, it strikes me that any of us might hunt rabbits, stags, elephants, or any number of creatures simultaneously, and we probably already do that all the time! Why? Because any of us likely participate in a large spectrum of social groups, each of which are at very different scales. In some groups (like a nuclear family), our roles may be quite large and essential. In other groups, we likely participate to a much lesser degree. Why not do all of it all at once?

I used to be central player in my faculty department, but now I’m retired and just a bit player, answering the occasional email. My Second Renaissance role is expanding lately, mostly because I have time and interest for it. Some of my other volunteer gigs are quite niche - like the bird sanctuary I support a few times a year by working a couple hours to pull invasive species out. If I skip a turn at that, no one misses me at all, but they thank me when I do yank out a wad of thistles or blackberry vines. It all matters, in its own way.

So I guess if someone has a stag to hunt, I could be there for it. And not necessarily as the big guy with the stag gun. Maybe the guy who hauls the cooler to camp and keeps the drinks on ice. Or whatever. Meanwhile, yeah, my own rabbits will all be well and truly hunted. There is nothing either/or about any of this. Second Renaissance is big enough to support quite a wide array of projects. If some passion or another is rising the fore for anyone reading this, please do consider just advancing the project, opening it up to support of all types, and see what comes of it!

1 Like

Nice idea, Robert.

One thing I’ve been working on is the creation of a community, offline, centered on self -sufficient as well as incorporating a new way of being. A while ago I recall Life Itself may have had such a community - near the French Pyrenees, if I remember correctly?

What’s become a little more clearer to me in this journey is I may not necessarily wish to create a community, but rather to join an existing one. Perhaps the idea of creating something afresh even comes more from ego than that of making impact. I’ve also struggled with the practicalities of keeping animals - sheep, ducks, goats (the goats not so much, as they are basically like dogs). Worst of all is dealing with death (one baby duckling died and two others were killed by my partner’s dog, with which I still feel a lot of resentment about).

So going back to the topic: what is the dynamics and vision of the Life Itself community? And, in terms of if I were to create a community from scratch, how do I know if that is really the best route? And if it is, what true foundational principles should it be based on? Up to now my focus has been on the practical ones (location, quality of land, potential for relative comfort and security, ability to conduct some element of research, etc).

Another thing: I do see an increasing trend of communities arising, or in the process of getting started. Why isn’t there more of a shared platform for them all and ability to learn from mistakes before they happen?

Hey Justin,
Regarding the community questions, please check out this thread where we will discuss one idea for handling the “interstitial space” between communities, as part of a larger vision.

1 Like

Doing this kind of recapitulation here is a new, and most welcome idea to me. One of the big problems with forums, and even more so with single-threaded messaging, is that it is hard to know what has been going on, either for newcomers or for those of us who take a break. Using LLMs to summarise can help fill a vital role in cohering the conversations. But, how do we judge the quality of such summaries? Does one LLM service help more than others? How much human supervision is needed, or, what is a good balance between LLM and human here? (that question is mainly for @dvdjsph)

2 Likes

For pragmatic reasons (guiding students who are being thrust willy-nilly into an AI-infused workforce), I’ve been refining my own relationship with AI. Also, I had to write a couple textbook chapters on the topic, so no hiding from research on that! Anyway, for me, Gemini is like a very handy reference librarian. I get some idea and then send Gemini to the “stacks” so to speak to come back with dozens of references and to organize the review of the literature about whatever it is. For example, yesterday I decided there is very likely a correlation between Donella Meadows’s leverage theory and the UTOK layered, emergent ontology. Moreover, I got this intuition by listening to a debate between @rufus and Brendan Graham Dempsey. So that’s four large bodies of work - Meadows, UTOK, 2R white papers, the considerable output of BGD - all of which I have read to some extent in the original. But to create my own summary of all that, including selected quotes and footnotes, would be a full summer project. Gemini handled it quite well in a minute or two! But let’s be very clear, I am very much the senior author of the piece. Sadly, I will never converse with Donella Meadows, but I have conversed directly with UTOK’s Gregg Henriques, Rufus, and BGD. So my moral and intellectual compass about the confluence of all these ideas is vary far removed from AI slop. It’s very human, very intuitive, very relational. If Gemini has served me rancid droppings on a shiny platter, I would have sent it all straight back! As it happened, it very much advanced the cause of what @dvdjsph calls “throughput”. To sum up, LLMs can serve us well. But it’s on us to know what we are doing in the first place.

All you say here is very reasonable, @RobertBunge , but what I was asking is (I emphasise the precision here) how do we judge? Not “is it good?” or “is it bad?”, but how do we assess its quality? Method. Methodology. And yes, for sure, sounds like excellent practice to treat yourself as the senior author, with the AI as the librarian or research assistant. I’m inviting us to move up a level in this conversation.

If anyone wants to play the LLM game at home, here is a lengthy list of my current intuitions about the Meadows-UTOK idea, expanded in many, many directions. Free free to run your own correlation studies about any of this! I’d like to see what different LLMs, guided by different human operators, have to say about the entire business!

Some Conjectures:

  • As being emerges through the UTOK ontological planes (Matter-Energy to Culture-Person) it both complexifies at the systems level and incorporates richer Wilberian interiors.
  • Meadows low-leverage action works on grosser levels of the UTOK hierarchy (Matter-Energy) and Meadows high-leverage action works at more subtle levels (Culture-Person).
  • The 2R white papers are thus supported in their thesis by both Meadows and UTOK.
  • Maslow’s Hierarchy stacks in a comparable way.
  • Personal growth proceeds from the gross to the subtle.
  • UTOK’s internalization process continues to emerge and deepen as it explores the components of human mind, personality, and consciousness.
  • As UTOK arrives at Mind 4 - the meta-cognitive observer - it identifies the interface of the human with the spiritual.
  • This is comparable to the non-dual teachings of Ramana Maharshi, among others.
  • Thus aligning with all manner of traditional theology, we might say ultimate “leverage” lies with God.
  • Paradoxically, maximum leverage requires maximum surrender.
  • Following Sri Aurobindo, Thich Nhat Hahn, Bahá’u’lláh, and many other spiritual teachers, the subtle energies of the encounter with the spiritual in the depths of the human interior must then be directed into both personal and social embodiment.
  • This provides the basis for a rigorous 2R action theory, informed by all the original sources in my list of conjectures above.

First of all, all the hard-core, professional/technical, thoroughly late-capitalist, LLM fans I follow (to learn the technicalities of it all) are consistently flagging “taste” and “judgment” as qualities humans must bring to add value to the AI-infused workplace. These sources can scarcely be accused of excessive spiritual idealism! So your questions about “how to we judge”?" are very much on point.

My most up to date extended answer is implied by the list of conjectures posted by me elsewhere in this thread. High level human judgment is the fruit of a developmental, spiritualizing, divinizing process - all the Wilberian “ups”. To know the Good, one must embody the Good.

I think there are two sides to this: defining norms, and giving feedback. Of course, anyone can come up with their own way of assessing quality, but I’d prefer a way to allows us to easily express what we find good or bad about the output.

But first we’d need to be clear about what we’re actually assessing. This thread received a few likes. These could be intended for the LLM output, the idea to create a thread like this, or possibly other bearers of feedback.

If we wanted to make threads like this better in the future, we could 1) define a prompt that is used for the summary 2) define a set of normative properties we’d like outputted threads to have and 3) define an easy way to give feedback on the output based on these normative properties.


For example, normative properties for a thread summary might look like:

Property Promise Example assessment
Faithfulness “No participant’s view is misrepresented” “Nymrod’s ‘vision dropping’ was summarised as self-erasure, which he explicitly rejected”
Coverage “No major thread of argument is omitted” “The cathedral blueprint problem got one line; it deserved more”
Salience “Quotes selected reflect the thread’s centre of gravity, not its outliers” “The Martin quote felt peripheral; the Melody quote was a better anchor”
Framing neutrality “The summary does not advance one participant’s position over others” “The opening framing favoured the game-theory reading over the relational one”

Feedback could then be as lightweight as tagging a specific property: “+faithfulness, -salience” on a post, or leaving a one-line note referencing the property by name. This borrows from promise theory’s logic: the promise is declared in advance, and an assessment is a simple claim that a specific promise was or wasn’t kept — separating the what from the how well.

The advantage over a simple like/dislike is that it gives the prompt-writer something actionable to iterate on.

1 Like

God save the peripheral!