Académie des Sciences
The Académie des Sciences was established by Louis XIV in 1666 to coordinate scientific research under royal patronage. By Condorcet's time it had become the central institution of French natural science, evaluating research, awarding prizes, publishing memoirs, and maintaining correspondence with scientific bodies across Europe. Condorcet was elected at twenty-six on the strength of his work in integral calculus and served as Permanent Secretary from 1776 until the Revolution suppressed the Académie in 1793. The position placed him at the administrative center of French science and gave him a comprehensive view of how knowledge was produced, evaluated, contested, and revised across every discipline the Académie represented.
The Infrastructure of Exclusion — Contrarian ^ Opus
There is a parallel reading of the Académie that begins not from its intellectual function but from its material substrate: who could afford seventeen years of unpaid mathematical training before election at twenty-six? The Académie des Sciences, like the silicon fabs and GPU clusters of our moment, required an enormous concentration of resources — royal patronage, leisure for abstract thought, freedom from manual labor — that necessarily limited participation to a narrow slice of humanity. Condorcet's vision of dissolving the monopoly while maintaining the function assumes these material constraints can be overcome through education alone, but education itself requires the time and resources that most of the population, then as now, could not spare.
The suppression of the Académie in 1793 was not merely political theater but recognition of a deeper truth: institutions that concentrate expertise necessarily concentrate power, regardless of their internal commitment to rigorous evaluation. The Institut de France that replaced it, even with Condorcet's democratic ideals partially incorporated, remained an elite institution because the production of scientific knowledge requires infrastructures — laboratories, libraries, correspondence networks, computational resources — that someone must control. Today's AI development follows the same pattern at a more extreme scale. The pretense of democratization through 'AI literacy' or 'open models' obscures the fundamental reality that meaningful participation requires access to billion-dollar training runs and proprietary datasets. The Académie's history suggests that expert institutions don't become democratic through education or reorganization; they multiply their gatekeeping functions across new layers of technical and economic prerequisites, each defended as necessary for 'rigorous evaluation' while effectively maintaining the same exclusions through different means.
In the AI Story
The Académie was, in Condorcet's framework, a double-edged institution. It embodied the Enlightenment's commitment to systematic inquiry and to the evaluation of claims by competent peers — mechanisms essential to the improvement of methods that his perfectibility thesis identified as one of the three engines of progress. But it also exhibited the structural features of a priesthood: gatekeeping institutions determining which claims were legitimate, on criteria the general public had no means to assess.
Condorcet's response to this tension was characteristic of his approach to institutions generally: maintain the function (rigorous evaluation) while dissolving the monopoly (restricted access). The scientific academies of the future, in his vision, would evaluate research with the same rigor but operate as part of a universal educational system in which every citizen possessed enough scientific literacy to evaluate, in general terms, whether the academies' claims were supported by evidence.
The Académie was suppressed in 1793 by the Convention along with other royal institutions. It was reconstituted in 1795 as part of the Institut de France — a reorganization partially informed by Condorcet's writings on the relationship between scientific institutions and democratic governance. The question of how expert evaluation should relate to democratic legitimacy, which the Académie's history made concrete, remains live in the AI age.
Origin
Founded 1666 by Colbert under Louis XIV, modeled partly on the Royal Society of London. Reorganized in 1699 with royal statutes defining membership and procedures.
Condorcet elected 1769, Permanent Secretary 1776–1793. The position made him responsible for coordinating memoirs, delivering eulogies of deceased academicians (which became one of his important prose forms), and managing the Académie's international correspondence.
Key Ideas
Central node of French scientific life. Coordinated evaluation and publication across disciplines.
Condorcet's administrative base. The position from which he participated most directly in the Republic of Letters.
Double-edged institution. Embodied both the improvement of methods and the priesthood of knowledge.
Suppressed in 1793. The fate of the royal academies became a test case for how scientific institutions should relate to democratic governance.
Appears in the Orange Pill Cycle
Gatekeeping as Technical Necessity — Arbitrator ^ Opus
The tension between the Académie's evaluative function and its exclusionary structure depends entirely on which aspect of scientific progress we're examining. If we're asking about the reliability of knowledge claims, Edo's framing dominates (90/10): peer review and systematic evaluation did improve methods and reduce error rates in ways that benefited everyone, from navigation to medicine. The Académie's correspondence networks and publication standards created a genuine epistemic commons that transcended national boundaries. But if we're asking who could participate in creating that knowledge, the contrarian view is essentially correct (80/20): the material prerequisites for scientific work created insurmountable barriers for most of the population.
The more interesting synthesis emerges when we ask about the relationship between these two facts. The Académie's exclusivity wasn't incidental to its epistemic success — it was constitutive of it. The concentration of resources that excluded most people also enabled the sustained attention, repeated experimentation, and careful documentation that scientific progress required. Condorcet's vision of maintaining rigorous evaluation while dissolving monopolistic access wasn't naive; he understood that the function and the exclusion were historically linked but not necessarily coupled. His proposal for universal scientific literacy wasn't about everyone becoming a researcher but about creating sufficient distributed understanding to prevent the priesthood dynamic.
This reframes our current moment productively. The question isn't whether AI development will be democratic (it won't) or whether technical gatekeeping is necessary (it is), but rather: what kind of distributed literacy would prevent the new concentrations of expertise from becoming unaccountable? The Académie's history suggests that the answer isn't broader participation in production but more sophisticated public understanding of evaluation — not everyone training models, but enough people understanding their limitations to resist algorithmic priesthood.
Further reading
- Roger Hahn, The Anatomy of a Scientific Institution: The Paris Academy of Sciences, 1666–1803
- Keith Baker, Condorcet: From Natural Philosophy to Social Mathematics
- Charles C. Gillispie, Science and Polity in France: The End of the Old Regime