In brief: We aim to demonstrate how distributed worlds — with institutional facts — are encoded and stored in distributed neural structures via ritual. As a slogan “ritual builds worlds”.
Intuition: An institutional fact is a computational structure represented in two or more intelligent agents that modifies their individual perception and action model. Its effects can be measured in group capacities of distributed long term (multigenerational) memory, coordination, communication and prediction in the wild (E.g. Lera Boroditsky’s spatial intelligence anecdote).
Approach: Blend the complex systems analysis of Religion and Ritual (Rappaport), Computational neuroscience (Friston), and contemporary analytical philosophy (Searle)
Flowchart: Ritual -> Distributed Generative Models for Active Inference -> Ontogeny of Institutional Facts (Measurable via patterns of choice behavior in groups of ritualized agents)
Definition of world: We might define the worlds being built as collections of possible action pathways.
This is literally the definition of action policies used in AI — reinforcement learning algorithms as well as in Friston’s Active Inference framework.
It echoes a popular quote from Wittgenstein: "The world is the totality of facts, not things"
What are you trying to do? Articulate your objectives using absolutely no jargon.
We aim to understand and naturalize —using adaptive, cognitive, ethnographic and cybernetic methods— the ancient, essential, and constructive role ritual plays in developing humanity
By naturalizing the role of ritual for humanity’s distributed “world building”, we will be following through on Sir John’s expectation that “scientific revelations may be a gold mine for revitalizing religion in the 21st century”
How is it done today, and what are the limits of current practice?
Research methods and tools developed in Anthropology and Social Science today lack sufficient Computer Science and Data Science talent and fit-for-use advanced modeling frameworks
This has left many areas of study (e.g. ritual and language) deprived of the remarkable progress over the last 10-15 years in machine assisted data analysis techniques (e.g. text modeling, video extraction, semantic processing) and tools for realistic and large scale inference and simulation (e.g. Active Inference, Free Energy Principle, Open AI Gym)
What is new in your approach and why do you think it will be successful?
An exceptionally motivated team with a radically Neuroscience and AI oriented approach, teaming with the wisest researchers in the Anthropology of Religion and Culture, to unlock the multimillenial trove of ethnographic data on ritual in culture, that is being aggregated and is waiting for at-scale activation.
We expect to find important scientific revelations by applying newly developed, but proven, machine learning and simulation tools from AI (e.g. Natural Language Processing – Watson Jeopardy, realistic multi-agent, and symbolically enabled ABMs – Free Energy Principle) to the vast amounts of messy, but highly pertinent, ethnographic data on socioecological ritual practices (e.g.
Who cares? If you are successful, what difference will it make?
Communities of practice will enable bridge building between religious and scientific thought leaders, ultimately fostering the mainstream adoption of Ritual as "globally vital technology"
A two way benefit by integrating the study of ritual into Cognitive Sciences and Engineering disciplines
With the computational treatment of Ritual becoming a recognized area in CS, AI and Cognitive Science, while also helping retool an important area of study in Anthropology and the Social Sciences
What are the risks?
The work is ignored and the sciences continue to drift apart from traditional sources of knowledge, including timely and practical insights on socioecology, passed down to us from thousands of societies
How long will it take?
POC (proof of concept): 3 months
Pilot: 9-12 months
Production: 2 years (where a growing community of practitioners will develop deliverables including scholarship, tools, and material for public outreach)
Integration: 3 years (a rollout period where toolkits of practice will be integrated into various centers of research to develop regional communities of practice and implementation)
How much will it cost?
POC: $60,000 (Viability oriented simulation mock ups, Wise counsel - experts meeting and video archive, Searchable Bibliography and Artifacts)
Pilot: $150,000 (ABM demonstrations, Research paper, Community use cases and interviews)
Production: $1,200,000 (4 Work-streams: FEP, NLP, Pulotu++, Community Engagement)
Integration: $2,400,000 (Establish communities of practice at 2-4 leading universities, including Brain-measurement, Anthropology 2.0, AI 3.0, CBT & Wellness)
What are the mid-term and final “exams” to check for success?
Entrance test: End of Pilot
Midterm: Proof of value and buy-in from prestigious scientific journals
Final: Communities of practice established at leading universities