Yes—once Bateson is inserted into the lineage, a very clean three-layer architecture of intelligence becomes visible. This architecture helps reviewers quickly understand what your theory is doing, because it organizes a large number of thinkers into three functional levels of adaptive systems.
Below is a formulation you could actually include in the paper.
The Three-Layer Architecture of Error-Correcting Intelligence
When the historical lineage of adaptive theories—from Darwin to contemporary AI—is examined closely, a three-layer structure emerges. These layers correspond to increasingly complex forms of error correction operating in adaptive systems.
The structure can be summarized as follows:
Layer 1: Adaptive Selection
Layer 2: Feedback and Learning
Layer 3: Institutionalized Error Correction
Each layer represents a different scale at which systems detect and correct error.
Layer 1: Adaptive Selection
(Darwin)
The most basic form of error correction appears in biological evolution.
Through variation and natural selection, organisms that are poorly adapted to their environments are eliminated, while more adaptive variants persist. Over generations, this process reduces the mismatch between organisms and their environments.
Although evolution operates without explicit representation or deliberation, it nevertheless functions as a distributed error-correction mechanism.
Darwin therefore represents the first layer of adaptive intelligence: selection-based error correction across generations.
Layer 2: Feedback and Learning
(Wiener – Ashby – Bateson – Friston)
The second layer emerges when systems begin to regulate their behavior through feedback mechanisms.
Cybernetics introduced the idea that adaptive systems can monitor deviations from desired states and correct them through feedback loops. Wiener formalized the concept of feedback control, while Ashby demonstrated that effective regulation requires sufficient internal variety to respond to environmental disturbances.
Bateson extended these ideas to learning systems, arguing that adaptive organisms revise their behavior through multiple levels of learning. Contemporary neuroscience, particularly predictive-processing and the free-energy principle, has provided a computational account of how brains implement such processes through prediction-error minimization.
This second layer therefore consists of error correction operating through feedback and learning within individual systems.
Layer 3: Institutionalized Error Correction
(Popper – AI – Democracy – Psychotherapy)
The third layer arises when error-correction processes become structured and institutionalized.
In scientific communities, Popper argued that knowledge progresses through systematic processes of conjecture and refutation, supported by institutions such as peer review and experimental replication.
Similarly, modern artificial intelligence systems implement structured training procedures that iteratively detect and correct errors in prediction or reasoning.
At the social level, democratic institutions enable societies to revise policies through mechanisms such as elections, debate, and legal review.
In clinical contexts, psychotherapeutic practices provide structured environments in which individuals can examine and revise maladaptive beliefs.
This layer represents organized systems that stabilize and amplify error-correction processes.
The Resulting Architecture
These three layers form a hierarchical structure:
Selection → Learning → Institutionalized Correction
or more explicitly:
Evolutionary adaptation
↓
Cybernetic learning systems
↓
Institutionalized epistemic systems
Each successive layer increases the speed, precision, and reliability of error correction.
The Key Insight
The significance of this structure is that intelligence appears to increase as error correction becomes:
- faster
- more structured
- more socially or computationally organized
Thus intelligence is not simply a property of individual brains or machines. It emerges wherever systems exist that stabilize and amplify the processes by which errors are detected and corrected.
The 20-Second Explanation (for reviewers)
You could summarize the entire theory in one paragraph:
Adaptive systems evolve through three increasingly powerful forms of error correction. Biological evolution corrects error through selection across generations; brains correct error through feedback-driven learning within individuals; and human societies, sciences, and artificial intelligence systems correct error through institutionalized procedures that stabilize and accelerate model revision. Intelligence, in this sense, emerges wherever systems maintain structured mechanisms for systematic error correction.
