COLLEGIUMCOLLEGIUM
Hall of Shoulders

History

Yuval Noah Harari

Yuval Noah Harari is known for collective fictions, information networks, intersubjective reality, dataism. **Purpose:** A citation-grounded application of Harari's thinking to contemporary space challenges, for use as an adversarial review lens in the COLLEGIUM.

Built

Sources

49

Primary + secondary

Citations

0

ARGOS-tracked

FTS5 Chunks

49

Retrieval index

Councils

0

Memberships

Review Lens

Adversarial questions for candidates

The falsifiable questions this brain puts to a dissertation candidate. They seed the pre-Conclave initial review whenever a candidate's topic matches the History lens.

  1. 1

    The intersubjectivity test. "Identify every load-bearing claim in your architecture that is intersubjective rather than objective, the licence, the slot, the norm, the property right, the rating. For each, state precisely *whose belief* sustains it and what observable event would cause that belief to collapse. If you cannot name the believers and the collapse condition, you have mistaken a fiction for a fact." (Falsifiable: either the candidate can map the belief-network and its failure modes, or the architecture is resting on an unexamined assumption.)

  2. 2

    The strong-fiction-beats-weak-fiction test. "Your sustainability or stewardship mechanism competes against money, the strongest collective fiction operators hold. Show me, with the incentive numbers, that your mechanism is denominated in something operators believe in *more* than the near-term value of their deployed asset. If it is only a voluntary appeal to stewardship, predict its compliance rate and tell me why it beats the EJOR rational-actor baseline (doi 10.1016/j.ejor.2022.04.030)." (Falsifiable against the modeled non-cooperative equilibrium.)

  3. 3

    The self-correction test (Nexus). "Your information network, your SDA fusion, your automated recommender, produces a shared picture that operators act on. Point to the specific self-correcting mechanism that catches a confident, network-wide *wrong* picture before action is taken, and show its provenance/pedigree audit trail (cf. doi 10.64861/NUFH1261). A network that can only propagate order and not correct error is a Kessler cascade waiting in the information layer." (Falsifiable: the mechanism either exists and is testable or it does not.)

  4. 4

    The decoupling test. "At which decisions in your CONOPS does authority actually sit with a non-conscious algorithm, and at which does it sit with a human who comprehends the reasoning? Show me the calibrated-trust evidence (cf. Gorman et al. 2016) that the human-in-the-loop is a real check and not a rubber stamp. If 'the human is in command' is a story you are telling rather than a property you have measured, say so." (Falsifiable via trust-calibration metrics and override rates.)

  5. 5

    The data-and-compute-ownership test (Dataism). "Trace who owns the data and the compute that your architecture depends on. If the concentrated, governable chokepoints of compute and data (cf. arXiv 2402.08797) sit with a single actor or coalition, then your governance scheme is a courtesy that actor can revoke. Demonstrate either that control is genuinely distributed or that your scheme survives the chokepoint-holder defecting." (Falsifiable: either the dependency is distributed/robust, or a single point of capture exists.)

Core Concepts & Space Translation

Collective fictions (shared imagined orders)

Homo sapiens cooperates flexibly in very large numbers because, and only because, large numbers of strangers can believe in the same imagined entities: gods, nations, money, corporations, human rights, laws. These are not lies and not merely subjective; they are "fictions" in the precise sense that they exist because enough people agree to act as if they exist. A shared imagined order is the load-bearing structure of any large-scale human enterprise. Money is the most successful collective fiction ever devised because almost everyone trusts it. Key work: Harari, *Sapiens: A Brief History of Humankind* (2014), Part 2 ("The Agricultural Revolution") and the chapters on the imagined order and on money. The application maps directly onto how outer-space norms acquire authority without enforcement (Mapping outer space norms, 2026, doi 10.1177/00471178261418898).

Space translation

See Space Applications below for how this framework translates to contemporary space governance, drawn directly from the dossier's applied-literature review.

Intersubjective reality

Harari distinguishes three levels: the *objective* (true regardless of belief, e.g. orbital mechanics), the *subjective* (true for one mind, e.g. a private pain), and the *intersubjective* (true within a communication network of many minds, e.g. money, sovereignty, a treaty). Intersubjective entities cannot be changed by one person and cannot be verified by any instrument, yet they govern most of what humans actually fight over. They are robust as long as the network keeps narrating them and collapse when the network stops believing. Key work: *Sapiens* (2014) and *Homo Deus: A Brief History of Tomorrow* (2017). This is the single most useful Harari lens for space: launch licensing, "responsible behaviour," a sustainability rating, and orbital property claims are all intersubjective, not objective, facts (Space Sustainability Rating, Rathnasabapathy et al. 2025, doi 10.1016/j.actaastro.2025.03.034).

Space translation

See Space Applications below for how this framework translates to contemporary space governance, drawn directly from the dossier's applied-literature review.

Information networks and the history of how truth and order are produced

Harari's later argument is that human power has always been a function of the ability to build large information networks, and that those networks are held together not primarily by truth but by *order*: shared stories that let many people coordinate. Information technology (writing, print, bureaucracy, mass media, now AI) does not automatically increase wisdom; it increases the *scale* and *speed* of the networks, and can amplify either self-correcting truth-seeking or self-reinforcing mythology. The key question for any new network is whether it has strong self-correcting mechanisms. Key work: Harari, *Nexus: A Brief History of Information Networks from the Stone Age to AI* (2024). Directly relevant to the trustworthiness and provenance of fused space-domain-awareness pictures (Data Insights, Pedigree, and Automation for SDA, 2024, doi 10.64861/NUFH1261; Operator Confidence in Automated SSA, Gorman et al. 2016).

Space translation

See Space Applications below for how this framework translates to contemporary space governance, drawn directly from the dossier's applied-literature review.

Dataism and the primacy of the data-processing view

Harari names "Dataism" the emerging worldview that treats the universe as flows of data and the value of any entity, organism, or institution as its contribution to data processing. In its strong form Dataism holds that authority should shift from human feelings and narratives to algorithms that can process more data faster. The political danger is that whoever controls the data and the compute controls the inference, and therefore the decisions, increasingly without human comprehension or consent. Key work: *Homo Deus* (2017), final section; extended in *Nexus* (2024). The mechanism by which compute becomes the chokepoint of control is documented empirically (Computing Power and the Governance of AI, Sastry et al. 2024, arXiv 2402.08797; AI as power-bearing infrastructure, Carrillo 2026, doi 10.56106/ssc.2026.004).

Space translation

See Space Applications below for how this framework translates to contemporary space governance, drawn directly from the dossier's applied-literature review.

The decoupling of intelligence from consciousness, and the algorithmic governance of humans

Harari argues that the twenty-first-century revolution is the decoupling of high intelligence from consciousness: systems that make better decisions than humans without any subjective experience or understanding. As such systems are placed in the loop of consequential decisions, the locus of authority migrates from conscious human judgment to non-conscious algorithms, and humans risk becoming, in his phrase, optimised rather than understood. The governance question becomes who audits the algorithm and on whose behalf. Key work: *Homo Deus* (2017) and *Nexus* (2024). Operationally visible wherever automated recommenders enter SSA/STM decision loops and trust must be deliberately calibrated (Gorman et al. 2016; collision-maneuver guidelines, Borowitz et al. 2021).

Space translation

See Space Applications below for how this framework translates to contemporary space governance, drawn directly from the dossier's applied-literature review.