Causal Entropic Forces and the Drive to Maximize Future Freedom

In the 2013 paper “Causal Entropic Forces” (Physical Review Letters), Alexander D. Wissner-Gross and Cameron E. Freer propose a physical principle: systems tend to act in ways that maximize their future causal influence: that is, they behave so as to keep as many possible future histories open as possible.

This is a generalization of entropic forces. Instead of maximizing thermodynamic entropy in the present, intelligent/adaptive systems act to maximize the number of accessible future causal paths. The paper shows that this single principle can spontaneously produce tool use and social cooperation in simple physical systems: behaviors we normally associate with intelligence and life.

From this perspective, many behaviors we associate with life and intelligence (exploration, tool use, cooperation, learning) emerge naturally as strategies to keep future options open. The universe contains a thermodynamic pressure toward systems that can influence more of their own future.

This principle has informed work in AI and robotics as a way to think about adaptive behavior without explicit task-specific rewards.

The paper drew both popular interest and substantial criticism. Gary Marcus and Ernest Davis (The New Yorker, 2013) argued that the physics is speculative, that the simulations are handpicked toy problems, and that there is no evidence the approach is competitive with domain-specific AI methods. An arXiv comment by E. Canessa (2013) claimed priority for related force-entropy relations in a simpler statistical-thermodynamic model. Behavioral-science commentary raised the charge of “interdisciplinitis”: restating “keep options open” without engaging existing psychological literature.

Citation counts remain respectable; the directional intuition continues to appear in intrinsic-motivation discussions. The isomorphism to a universal physical law of intelligence is not established. Identity Engineering keeps that status explicit.

Causal entropic forces are one line in a broader portfolio that points in a related direction:

• →Empowerment (Klyubin, Polani, Nehaniv; Salge et al.): information-theoretic controllability of future states.
• →Active Inference / Free Energy Principle (Friston) and Predictive Processing (Clark): agents that minimize free energy effectively preserve and expand the capacity to act under uncertainty.
• →Possible Selves and narrative forwardness (Markus & Nurius; Ricœur / McAdams): cognitive and narrative representations of possible futures.

IE takes the shared directional claim as a normative design goal (maximize future optionality under trajectory coherence), not as a derived physical law. Full positioning: Causal Entropy as Design Telos.

• Causal Entropic Forces as one telos-inspiration: The framework borrows the directional intuition and applies it at the level of personal identity. A central design aim is to increase the number and quality of accessible future paths while preserving recognizability.
• Identity Stem as optionality maximizer under constraint: A strong, coherent Stem with high Mass and well-curved geometry is a system that can exert more influence over its own future without dissolving trajectory coherence.
• Questions as Probes for optionality measurement: Precise questions function as instruments to assess how open or closed one’s current identity geometry is with respect to future possibilities, and to deliberately open new paths.
• Explicit uncertainty: Whether any of these lines can ground IE beyond analogy remains open. IE keeps that uncertainty public (see Critical Gaps, especially issue #33).

Position in the series: Wissner-Gross remains the most formally mathematical physics-style anchor in the Foundations sequence. He is no longer presented as the single scientific foundation of the telos. Identity Engineering draws on him as one structuring analogy inside a multi-source portfolio and remains transparent that the isomorphism is not established.