The year was 1980. I was 12 years old. Little did I know that a book was published by two guys who would, many years later, change my view of, well, everything.
Before we dive in, these two came up with the best title of all time: “Capitalism and Schizophrenia”—a two-volume work consisting of Anti-Oedipus (1972) and A Thousand Plateaus (1980). In the second volume, they introduced the concept of rhizomes—systems that spread underground, pop up in unexpected places, and connect in ways that make no hierarchical sense. Unlike trees, which grow from a single trunk with branches that never reconnect, rhizomes are all about multiple connections, no beginning or end, just middles that keep proliferating. This simple botanical observation became Deleuze and Guattari’s metaphor for how systems could work without centralized control—and why our trust infrastructure desperately needs this kind of thinking.
The Arborescent Prison
Look at any digital trust system today and you’ll see a tree:
- Certificate Authorities: A root CA at the top, intermediate CAs branching down, end-entity certificates as leaves
- Identity Verification: Government-issued IDs flowing down through DMVs, passport offices, and notaries
- Content Authenticity: Platform verification badges issued by Twitter/Meta/Google’s central authority
- Financial Trust: Central banks → commercial banks → payment processors → merchants
These arborescent structures share the same fatal flaws:
- Single points of failure (compromise the root, kill the tree)
- Rigid hierarchies (power flows one direction)
- Binary logic (you’re either in the tree or you’re not)
- Capture vulnerability (control the trunk, control everything)
Deleuze and Guattari saw this coming. “The tree is filiation, but the rhizome is alliance, uniquely alliance.”
Enter the Rhizome
A rhizome—think ginger root, grass, or fungal mycelium—operates on entirely different principles:
- Connection: Any point can connect to any other point
- Heterogeneity: Connections don’t require homogeneous entities
- Multiplicity: No unity from which multiplicity derives
- Asignifying rupture: Break it anywhere, it continues growing
- Cartography: Maps that are produced, not traced
- Decalcomania: No genetic axis or deep structure
What would trust infrastructure look like if we built it as a rhizome instead of a tree?
Rhizomatic Trust in Practice
Identity Without Roots
Instead of government-issued IDs cascading down a jurisdictional tree:
- Multiple attestations from peers, communities, and institutions
- Context-dependent verification (your GitHub contributions matter for code, not for voting)
- Lateral validation where trust spreads horizontally through networks
- No single namespace (did:web:, did:key:, did:peer: all coexist)
Content Authenticity as Assemblage
Rather than platform badges or central authorities:
- Verification assemblages where multiple agents contribute trust signals
- Transversal connections across fact-checkers, source validators, and context providers
- Nomadic trust scores that flow and transform with content
- Deterritorialized provenance not owned by any single entity
Consensus Without Center
Traditional blockchains still dream of the One True Chain. Rhizomatic consensus means:
- Multiple simultaneous truths (different trust graphs for different contexts)
- Productive disagreement where conflicting validations coexist
- Minority reports that don’t get erased by majority consensus
- Forking as feature not bug
The Practical Rhizome
Here’s how we build rhizomatic trust infrastructure:
1. Abandon the Root
No more “ultimate authorities.” Instead:
Trust Score = Σ(peer_attestations × reputation_weights × context_relevance)Where reputation itself emerges from the network, not from position in hierarchy.
2. Embrace Multiplicities
One piece of content, many trust evaluations:
- Scientific community rates factual accuracy
- Local journalists verify on-ground claims
- Subject experts assess technical details
- Affected communities validate lived experience claims
All coexist. None dominates.
3. Design for Breakage
Unlike Certificate Transparency logs that break if history is corrupted:
- Local trust neighborhoods that function independently
- Gossip protocols that route around damage
- Redundant pathways for verification
- Anti-fragile reputation that strengthens under attack
4. Make Maps, Not Traces
Stop tracing existing power structures into digital form:
- Emergent credibility from demonstrated competence
- Situational authority that shifts with context
- Temporary stabilizations not permanent hierarchies
- Trust as performance, not position
Beyond Web of Trust
The “Web of Trust” (PGP/GPG) was close but still too arborescent—it wanted to converge on stable trust paths. Rhizomatic trust is messier:
- Trust changes based on behavior (not just key signatures)
- Context matters (trust for medical advice ≠ trust for restaurant reviews)
- Contradictions persist (Agent A trusts X, Agent B distrusts X, both valid)
- No global convergence required or desired
The Thousand Plateaus of Trust
Deleuze and Guattari wrote: “Write, form a rhizome, increase your territory by deterritorialization, extend the line of flight to the point where it becomes an abstract machine covering the entire plane of consistency.”
For trust infrastructure, this means:
- Thousand different trust graphs, not one
- Plateaus of temporary stability, not permanent structures
- Lines of flight that escape capture by surveillance capitalism
- Abstract machines of verification that operate without central control
- Smooth space for trust, not striated by platforms
The Molecular Revolution
We’re witnessing a shift from molar (centralized, rigid) to molecular (distributed, fluid) forms of trust. AI agents aren’t waiting for permission from certificate authorities. They’re forming their own trust assemblages:
- Agent collectives verifying each other’s outputs
- Temporary authentication zones for specific tasks
- Nomadic credentials that travel between contexts
- Minority trust networks for marginalized communities
Why This Matters Now
As AI proliferates, we face a choice:
Arborescent Path: Ever-more-powerful central authorities (OpenAI, Google, governments) trying to control AI through hierarchical verification
Rhizomatic Path: Distributed trust networks where verification emerges from the collective intelligence of diverse agents
The arborescent path leads to digital feudalism. The rhizomatic path offers something else: a trust infrastructure that’s as alive, adaptive, and resilient as the systems it serves.
Building the Rhizome
Start here:
- Build tools that connect laterally, not vertically
- Create protocols that support multiple truths, not single authorities
- Design for forking and divergence, not convergence
- Celebrate productive disagreements in trust evaluations
- Make reputation portable and contextual, not fixed
“Make rhizomes, not roots, never plant! Don’t sow, grow offshoots! Don’t be one or multiple, be multiplicities! Run lines, never plot a point!”
The future of trust isn’t a better tree. It’s not even a tree at all.
It’s a rhizome, spreading underground, popping up unexpectedly, connecting impossibly, breaking beautifully, and growing relentlessly toward a thousand different suns.
“The rhizome is an antigenealogy. It is a short-term memory, or antimemory. The rhizome operates by variation, expansion, conquest, capture, offshoots.” — Deleuze & Guattari, A Thousand Plateaus