Anton Dziatkovskii · ORCID 0000-0001-7408-3054
Preprint, 7 September 2026. Version of record: Zenodo, DOI 10.5281/zenodo.22639712. License CC BY 4.0.
Full text (PDF, 4 pages, 166 KB)DOICode: claw-consensus
Decentralized autonomous organizations (DAOs) increasingly require principled governance design rather than ad-hoc parameter choices. The leading cybernetic lens applied to DAOs is Stafford Beer's Viable System Model (VSM), recently mapped to decentralized organizations by BlockScience (Zargham, Nabben). We propose a distinct cybernetic grounding drawn from the Soviet/Russian school of homeostatics (Gorsky) and Anokhin's theory of functional systems. We decompose governance into three coupled control circuits — a main steady-state circuit, an adaptive circuit, and a protective self-compensation circuit — and show that this tripartite model is a simpler, implementable subset of VSM's five subsystems with different coupling assumptions. We map the three circuits onto concrete on-chain mechanisms (Ethereum's slashing and inactivity-leak as negative feedback; dual-governance and circuit breakers as the protective circuit) and use the model to predict governance failure modes, which we compare against observed DAO crises. We additionally sketch a computable governance-stability score derived from a typed actor-relation matrix (developed in companion work), and exhibit the three circuits as executable guards in a dependency-free multi-agent consensus protocol (claw-consensus) offered as a reference implementation beyond the on-chain setting.
DAO governance · cybernetics · homeostatics · viable system model · multi-agent systems · blockchain
Dziatkovskii, A. (2026). Homeostatic Governance: A Gorsky-Anokhin Alternative to the Viable System Model for Decentralized Organizations. Preprint. Zenodo. https://doi.org/10.5281/zenodo.22639712
@misc{dziatkovskii2026homeostaticg,
author = {Dziatkovskii, Anton},
title = {Homeostatic Governance: A Gorsky-Anokhin Alternative to the Viable System Model for Decentralized Organizations},
year = {2026},
publisher = {Zenodo},
doi = {10.5281/zenodo.22639712},
url = {https://doi.org/10.5281/zenodo.22639712},
note = {Preprint, CC BY 4.0}
}
One text, one DOI: the PDF on this page is the Zenodo file byte for byte. Please cite the DOI.