Power built these institutions to serve itself — and mostly, it worked
Athens built one. So, on the current evidence, did China — a Warring States kingdom on the far side of the known world, with no plausible route of contact to Athens at the time. India, the Jewish communities of Babylon, and the Islamic world each built one too, though not, it turns out, independently of each other or of Greece: each sits somewhere on a documented or highly plausible line of transmission running back to the same Hellenistic root. What looks at first like five civilisations arriving separately at the same institutional answer turns out, on closer inspection, to be one genuinely separate invention and one long chain of inheritance — a different puzzle, and arguably a more interesting one.
A word on the evidence before either strand is followed further. Nothing that follows claims the certainty of a laboratory result. Ancient history is reconstructed from fragments — inscriptions, later chronicles, the occasional surviving text — and even a figure as central to Section 2 as Socrates has no fully independent, contemporaneous documentary record confirming the standard account of his life and death; what survives comes through Plato, Xenophon, and Aristophanes, none of them a neutral witness. This essay does not attempt to settle questions historians themselves have not settled. It attempts something more modest and, for its purposes, more useful: a reasonable chronicle, built from the best current evidence, of how institutions dedicated to gathering and transmitting knowledge developed into the thing the modern university has become — and of what, across that development, kept recurring often enough to look like a pattern.
Plato's Academy in Athens dates to 387 BCE. Jixia Academy, in the Chinese state of Qi, was founded or restored around 318 BCE according to Sima Qian's Records of the Grand Historian — though some historians trace its origins earlier still, to Duke Huan of Qi in the mid-fourth century BCE. The two institutions have no plausible connection: organised contact between China and the Mediterranean world would not begin for another two centuries, with Zhang Qian's mission west around 130 BCE, and even that was indirect and overland rather than the sustained exchange either institution would have needed to influence the other. Whatever pushed Qi's rulers and Athens's philosophers toward the same institutional answer, they arrived at it without knowledge of each other.
The remaining three traditions in this essay's original frame do not share that isolation, once each is examined rather than assumed. Nalanda, in India, grew from the fifth century CE into a residential centre for logic, medicine, and Buddhist philosophy — but by then, the Indo-Greek and Greco-Bactrian kingdoms established in Alexander's wake had already spent three or four centuries fusing Hellenistic and Buddhist thought across Gandhara and Bactria, producing the hybrid tradition known as Greco-Buddhism. The Jewish academies of Sura and Pumbedita, founded in 225 CE and around 259 CE in Sassanid-era Babylon, sit further from that influence — direct Seleucid Greek rule in Mesopotamia had ended some three and a half centuries earlier — but the region had absorbed Hellenistic administrative and intellectual habits for generations before that, even as a newly assertive Persian dynasty pushed back against them. Al-Qarawiyyin in Fez (859 CE) and Al-Azhar in Cairo (970 CE) show the clearest line of all: both were founded after the Abbasid translation movement had already spent fifty to a hundred years rendering Aristotle, Plato, and the Neoplatonists into Arabic, to the point that a tenth-century reader in Baghdad had, in one historian's phrase, roughly the access to Aristotle that an English reader has today.
None of this makes India's, Babylon's, or the Islamic world's contribution derivative or lesser — what each tradition did with an inherited or absorbed idea is its own considerable achievement, and Section 2 takes each on those terms. But it does mean the puzzle this essay actually has evidence for is narrower, and sharper, than "why did five civilisations independently invent the university." It is this: why did a shared institutional idea, wherever it travelled and wherever it arose from nothing, keep landing in the same relationship to power almost every time — and what does the one clear case of true independent invention add to, or complicate about, that pattern?
There is a complementary answer to this puzzle, sitting underneath the political one rather than competing with it. Before any civilisation built an institution to serve power, gathering to exchange information for its own sake may already have been a pleasure with roots considerably older than any human civilisation at all.
House sparrows offer a documented, rather than merely anecdotal, example. Ornithological records show that after a feeding bout, house sparrows gather in deep cover and join together in quiet, conversational chirping — a behaviour distinguished in the literature itself from territorial song, alarm calls, or mating display, and given its own name: social song (Nivison, 1978, cited in Birds of the World, Cornell Lab of Ornithology). A related but more agitated version of the same behaviour occurs at the communal night roost. The sparrows do not merely tolerate proximity while nattering. The gathering itself appears to be sought out, a discrete activity separate from feeding, breeding, or safety.
The pattern is not confined to birds, and it is worth seeing it repeat in animals with no close evolutionary relationship to sparrows at all, because a pattern that shows up once is a coincidence and a pattern that shows up independently across unrelated species starts to look like something more fundamental. Guinea pigs have a vocal repertoire that splits cleanly by function: "chutting" is produced as guinea pigs explore calmly together, without alarm, understood in the research literature as a quiet contentment and location signal exchanged between animals rather than a call directed at any threat or mate. This sits alongside, and is clearly distinguished from, the sharp "tutt-tutt" warning call used to signal danger and the "rumba," a mating-specific vocalisation performed with an accompanying hip-swaying display (Bays, in Veterian Key, 2016; PetMD, 2022). Three separate vocal channels doing three separate jobs, one of which exists for nothing more than calm social contact.
Sperm whales make the same case with considerably more scientific weight behind it. Shane Gero's research on sperm whale "codas" — patterned sequences of clicks — shows the whales producing them specifically during social interaction at the surface, often in duet-like exchanges between individuals, and never during deep-water foraging dives, where the same animals instead produce an acoustically related but functionally distinct signal: echolocation clicks, used for hunting rather than socialising (Gero, Whitehead and Rendell, 2016). Codas carry identifying information at the level of the individual, the family unit, and the wider vocal clan, and are learned and culturally transmitted within a social group rather than produced instinctively. The same animal, using two related forms of the same basic tool, keeps one version for finding food and an entirely separate version for maintaining social bonds with no other apparent purpose.
The human case carries real theoretical weight behind it. The evolutionary psychologist Robin Dunbar has argued that primate grooming exists primarily for social bonding rather than hygiene, and that human language itself evolved as a more efficient substitute for grooming once group sizes grew too large for one-to-one physical bonding to sustain every relationship (Dunbar, 1996). On this account, idle chatter and gossip are not a degraded or trivial use of language. They are close to language's original evolutionary function — the mechanism that made large-scale human cooperation possible at all.
This pleasure in gathering to exchange information was never confined to idle chatter, and long before any formal institution existed to house it, humans had already built sophisticated, non-written ways of storing and transmitting knowledge that genuinely mattered. Walter Ong's foundational study of oral cultures shows that societies without writing stored knowledge in the only medium available to them — the human mind — using rhythm, repetition, formula, and narrative specifically because these patterns are what make large bodies of information memorable and accurately transmissible across generations (Ong, 1982). Ong draws on the classicist Eric Havelock's work on Homer to make the point concretely: the Iliad and the Odyssey functioned as vast mnemonic repositories of practical, legal, and cultural knowledge, not merely as entertainment. The clearest and most rigorously documented case available is Aboriginal Australian songlines, in which geography, law, genealogy, astronomy, and ecological knowledge are encoded jointly through song, story, dance, art, and ceremony rather than through any single medium, and verified as functionally accurate — usable to navigate real country and locate real water — after tens of thousands of years of purely oral transmission (Kelly and Neale, 2020). This was not primitive mythology standing in for knowledge. It was knowledge, held and exchanged through performance rather than text, doing genuine intellectual work long before anyone thought to write it down.
Seen this way, the ancient university was not a wholly new invention so much as an intellectual sophistication of something humans were already doing, and doing well: holding information collectively, exchanging it through gathering, and using that exchange to develop new ideas and concepts rather than simply preserve old ones. What changed with Plato's Academy, Nalanda, and the rest was not the underlying impulse — that impulse, on the evidence above, may be older than language itself — but the medium and the method: formal argument in place of formula, text in place of song, a designated institution in place of a route walked and sung across open country. The university's real inheritance, on this reading, is not from any single ancient tradition of scholarship. It is from the much older and more universal human habit of gathering to hold what is known, and from that holding, building something new.
The university, on this reading, may be the elaborated, institutionalised, human version of the sparrows' bush: a place built, ostensibly, for the formal transmission of knowledge, but sustained in practice, generation after generation, common room after common room, by something older and simpler than any curriculum. This does not compete with what the rest of this essay has to say about power capturing these institutions. It may explain why humans were so willing to keep building and defending them regardless of who controlled them — a biological throughline running underneath the political one, present before the argument turns to what power did with the gathering place once it existed.
Not a brisk survey. The interesting question is what each institution thought it was for, in its own terms, and whose interests that purpose actually served.
Plato's Academy was founded in the shadow of an act of state violence. Socrates, Plato's teacher, was tried and executed by the Athenian democracy in 399 BCE, officially for impiety and for corrupting the young — a direct exercise of political power against a man whose method was relentless public questioning. Plato left Athens for roughly a decade afterward, then returned and founded the Academy in 387 BCE. On the surface, this looks like the cleanest possible case of knowledge founded in defiance of power. The scholarship complicates that reading considerably. S. Sara Monoson's revisionist study of Plato's relationship to Athenian political life argues it was unorthodox rather than straightforwardly hostile (Monoson, 2000). More tellingly, Plato's own political conclusion, laid out in the Republic, was not that knowledge should check power from outside. It was the philosopher-king: the argument that knowledge should become power directly. The Academy's founding moment stood against the regime that killed Socrates. Its founder's actual political philosophy pointed the opposite way.
The Jewish academies of Sura and Pumbedita present a case of partnership-with-friction rather than opposition. Robert Brody's study of the Geonic period — the first English-language survey of the era in nearly a century — traces the relationship between the academies and the exilarch, the hereditary political head of the Babylonian Jewish community who claimed Davidic descent and represented the community to the Abbasid caliphate (Brody, 1998). The exilarch originally funded both academies and appointed their heads as subordinate spiritual figures. This was not an independent check on power. It began as a creature of it. Real friction did eventually emerge — rival exilarch candidates were backed by competing academies in the 830s, and exilarchs are recorded as sometimes appointing weaker, more compliant heads specifically to keep the academies subservient. Over roughly four centuries, the scholarly authority of the academies' heads gradually superseded the exilarch's inherited political authority as the community's real source of leadership. Knowledge did eventually out-compete hereditary power here, but it took four centuries, and it never fully escaped the entanglement it started inside.
Nalanda is the plainest case of capture in the entire set. Founded around 427 CE by the Gupta emperor Kumaragupta I, its survival for nearly a thousand years depended on an unbroken chain of royal endowment: later Gupta emperors extending support, King Harsha in the seventh century transferring the revenue of roughly a hundred villages to sustain it, the Buddhist Pala dynasty gifting entire villages for its upkeep centuries later. The historian Gregory Schopen's broader body of work on Indian Buddhist monasticism — built deliberately on inscriptions, monastic legal codes, and archaeology rather than religious doctrine — shows these institutions operating as sophisticated economic actors deeply entangled with worldly power, not as detached contemplative communities standing apart from it (Schopen, 1997). An institution this dependent on royal patronage for a millennium has a structural incentive to avoid seriously destabilising the power sustaining it, whatever its formal teachings claimed to value.
Al-Azhar is cleaner still. It was established in 970 CE by the Fatimid Caliph al-Mu'izz li-Din Allah explicitly as an instrument of da'wa — missionary propagation of Ismaili Shi'i doctrine — countering the rival Sunni Abbasid caliphate and asserting the Fatimid imam's claim to infallible religious authority (Daftary, 2007). Even the institution's name is dynastic propaganda, honouring Fatima al-Zahra, the ancestor from whom the Fatimids claimed descent. Unlike Nalanda, which was patronised from outside, Al-Azhar was founded directly by the regime it served, for that regime's own doctrinal survival. The clearest demonstration of how completely knowledge-function and power-function were fused here came a full two centuries later: when Saladin ended Fatimid rule in Egypt in 1171, he did not destroy Al-Azhar. He simply expelled its Ismaili scholars and repopulated it with Sunni ones, converting the same physical institution to the opposite doctrine almost overnight. The building and its authority survived intact. The ideology it served changed completely, the moment the regime controlling it changed.
Jixia Academy is the one institution in this set built without any contact, direct or inherited, with the Hellenistic world discussed in Section 1 — and it is also, in its way, the starkest case of capture in the whole essay. Traditionally dated to 318 BCE under King Xuan of Qi, though likely restored rather than founded outright — earlier Tian rulers, including Duke Huan, appear to have supported scholars at Jixia a generation or two before — the academy gathered as many as a thousand scholars from every major school of the period, Confucian, Taoist, Legalist, Mohist, and Naturalist among them, and gave its most senior figures the title of "grand master," along with mansions, stipends, and the standing of court officials (Sima Qian, Records of the Grand Historian). Some historians read the academy's founding as straightforward statecraft: a way for the Tian clan, which had seized the Qi throne in a coup that included the killing of a brother, a nephew, and a mother, to buy the prestige that patronising a golden age of scholarship could confer. Whatever the founders' motives, the arrangement was frankly transactional — scholars debated freely and were not required to reach doctrinal agreement, but they were retained, housed, and paid by the throne they served, and their standing rose and fell with the fortunes of the dynasty funding them.
The clearest evidence of how completely the academy's fate was tied to its patron came at the very end. Jixia did not decline through internal intellectual exhaustion, doctrinal capture, or a rival institution outcompeting it. It ended because the state of Qi did: when Qin conquered Qi in 221 BCE, the last of the Warring States to fall, Jixia's scholars scattered and the academy simply ceased to exist. No successor regime inherited it, absorbed it, or repurposed it the way Saladin repurposed Al-Azhar two governments later. The institution and its patron state died in the same year — the cleanest demonstration available anywhere in this essay of just how completely a knowledge institution's survival can depend on the power that built it.
Five case studies now, not four, and the addition changes the shape of the argument rather than just its length. Only one of the five — the geonim's slow four-century divergence from the exilarch — shows knowledge institutions moving even partially away from the power that built them. Three more show capture that held: in one case for a thousand years, in another surviving a complete change of regime and doctrine without missing a step. Jixia shows the sharpest version of capture of all — not resistance withheld, but total dependency, ending the moment its patron state did. Oxford and Cambridge's monastic colleges, established from 1277 onward to train monks in theology on behalf of the medieval Church, sit alongside these cases rather than apart from them: knowledge deployed in service of an existing religious authority, not against it.
This has to be the essay's starting assumption going into what follows, not a surprising twist introduced partway through. Institutions built to house and transmit knowledge are built and sustained by power far more often than they resist it — a pattern that held regardless of whether the institution's founding idea travelled from elsewhere or, as in Jixia's case, arose with no outside contact at all. The interesting question is not whether this happens. It almost always does. The interesting question is what makes the rare exception possible at all.
This is where the essay's argument earns whatever weight it is going to carry, because it is the single case in the whole set where knowledge genuinely destabilised the power that had built the institution housing it.
When Henry VIII broke from Rome and Thomas Cromwell's dissolution of the monasteries removed the religious houses embedded within Oxford and Cambridge in the late 1530s, the universities' near-exclusive claim to theological authority — the thing that had justified their existence for four centuries — could no longer be taken for granted. The historian Diarmaid MacCulloch's account of this period makes clear the outcome was neither smooth nor guaranteed: applications to Cambridge fell by roughly half in the years immediately following the dissolution, and a 1545 Act of Parliament came within a hair of authorising the Crown to liquidate the universities' own accumulated wealth outright, driven by Henry's late-reign financial desperation rather than by any religious logic at all (MacCulloch, 2018). The universities survived that moment by circumstance as much as by strategy.
What they survived it into is the essay's real point. Cromwell's own injunctions of the period pushed the universities toward a broader humanist curriculum — readers in Greek and Hebrew, support for scholars beyond canon law, a genuine widening of what counted as worth teaching. That widening had already been under way for longer than the dissolution itself: scholastic disputation, and the humanist scholarship layered onto it across the preceding century, had been training students inside Oxford and Cambridge in exactly the intellectual habits — close textual argument, the comparison of authorities, the willingness to press a claim to its logical conclusion — that would eventually be turned against the theological orthodoxy those same habits had been built to serve. By the time Luther's ideas reached England, the universities' own training had already sharpened the tools used to question the institution that had provided it.
This took centuries of quiet accumulation to produce, and it happened exactly once in the five traditions examined here. That is not a flaw in the argument. It is the argument. Institutions built to transmit an orthodoxy do not reliably produce people who question it — Nalanda didn't, Al-Azhar didn't, the Academy's own founder argued against the very possibility in principle. Sometimes, after a long enough accumulation of the intellectual habits knowledge transmission requires, they do anyway. Seen one way, this looks like an institution specialised for one narrow purpose — theological monopoly — facing a genuine shift in its environment and surviving not by holding its ground but by adapting into something adjacent: less a revolution than a lineage that diversified fast enough to avoid the extinction its over-specialisation might otherwise have guaranteed. Section 6 returns to this mechanism directly. For now, the point stands on its own: that this happened even once, against how consistently it failed to happen elsewhere, is what makes the Reformation case worth an entire section rather than a footnote.
The temptation, arriving here, is to conclude that knowledge reliably challenges unsustainable power, and that the Academy, Nalanda, Al-Azhar, and the Babylonian academies were simply slower or less successful examples of a pattern the Reformation completed. The evidence does not support that conclusion, and the argument is stronger for admitting as much plainly rather than forcing four contrary case studies to fit a tidier story.
What the case studies actually show is that power has been remarkably successful at building knowledge institutions that serve it — successful nearly every time it tried, across every civilisation examined here. Education's disruptive potential is not the reliable mechanical outcome a simpler version of this argument might assume. It is a low-probability outcome that occasionally survives centuries of institutional effort built specifically to prevent it. When it does survive, the consequences are large precisely because producing that outcome at all is so difficult. The Reformation matters not because it proves a rule. It matters because it is the exception the rule was built to prevent, and prevention failed exactly once.
Section 5 described the Reformation as a rule of capture failing, once, after centuries of near-total success elsewhere. That framing holds at the scale of the whole story, but it starts to strain the moment the mechanism is examined closely, because "failing" implies something was gripping tightly enough to fail. What follows complicates that picture rather than replacing it: control was real at some scales and closer to absent at others, and the Reformation's significance survives that complication rather than depending on a single, uninterrupted grip finally slipping.
If knowledge institutions are usually safe for the power that builds them, and only rarely escape that control, a real question follows: why does power keep building them at all, rather than dispensing with formal education altogether and relying on more directly controllable forms of training? Evolution offers a useful frame for this, but evolution is not the settled, single-voiced theory it is often assumed to be, and the essay is better for using two competing accounts of it rather than one.
Richard Dawkins's gene-centric view, popularised in The Selfish Gene, holds that evolution has no plan and no agent: organisms are, in his own framing, survival machines built by genes that persist purely because they happened to work, with no foresight involved anywhere in the process (Dawkins, 1976). Denis Noble, an Oxford physiologist — and, notably, Dawkins's own doctoral examiner in the 1960s — has spent much of his later career arguing this gets cause and effect backwards. Genes, in Noble's account, are servants of the cell and the organism, not masters of it; organisms actively harness variation and randomness to adapt, in a way that looks considerably more like purposeful, functional behaviour operating at the level of the whole organism than blind chemistry operating at the level of the gene (Noble, 2006). The two men have been debating this in public for years, including a recent head-to-head at the How the Light Gets In festival, and biology has not settled the argument between them.
Tested against the case studies rather than asserted in the abstract, both lenses turn out to fit — just not the same cases. Al-Azhar reads as Noble, not Dawkins. Its founding was not blind trait-selection accumulating without a plan. A caliph and a state deliberately built an institution for a specific, known doctrinal end, with clear purpose and foresight behind the act — agency, in Noble's sense, operating at the level of the whole institution rather than emerging from nowhere. Nalanda reads as Dawkins. No single dynasty planned a thousand years of dependency; it accumulated, endowment by endowment, across successive unrelated regimes, closer to blind path-dependency than to any one actor's design. Section 1a's sparrows are the cleanest Dawkins case in the whole essay — social song persisted because it worked, with nothing resembling planning behind it. The Reformation is the case that resists a single answer, and may be the essay's most interesting result because of it. No one planned for scholastic training to eventually produce scholars capable of dismantling the orthodoxy it was built to serve — that outcome accumulated blindly, the way Dawkins would expect. But once humanist scholarship was under way inside Oxford and Cambridge, individual scholars were acting with clear purpose and agency in pushing the curriculum where they did, exactly the picture Noble would recognise. Section 4 described this as a lineage diversifying fast enough to avoid the extinction its over-specialisation might otherwise have guaranteed — and that description now reads as a Noble-shaped outcome (functional, adaptive, purposeful at the level of the whole institution) produced by a Dawkins-shaped process (blind accumulation, no single actor tracking the outcome across generations). The same historical episode looks Dawkinsian at one scale and Noble-ian at another, depending only on how closely the observer chooses to look.
Jixia is the essay's cleanest case of both lenses operating in sequence rather than in competition. Its founding — a calculated bid for legitimacy by a dynasty needing prestige, deliberately recruiting a thousand scholars with mansions and stipends — is Noble through and through: purposeful, agential, built by rulers who knew exactly what standing they wanted to buy. Its ending is pure Dawkins: no one chose to let Jixia die, no rival institution out-competed it, no doctrine displaced it; it simply stopped when the state carrying it was eliminated by a fitter competitor. The same institution is Noble at its birth and Dawkins at its death — which may be the more general rule these case studies point toward: institutions of this kind are usually built with purpose, and lost by accident.
This suggests the university's persistence is not well explained by either lens alone. At the scale of a single founding act — a caliph, a king, a Cromwell issuing injunctions — power visibly knows what it is doing, in the purposeful, agential sense Noble describes. At the scale of centuries — an institution drifting into total patronage-dependency, a curriculum's slow accumulation of habits nobody was tracking — the outcome looks far more like blind, undirected accumulation, in the sense Dawkins describes. The risk of producing an occasional Reformation is not a cost any single founder knowingly accepted, the way Dawkins's frame would predict for anything this unplanned. But nor is it pure accident with no agency involved anywhere in its production, since real people, with real purposes, kept shaping the institution's direction at every scale small enough for a single mind to act on. The most honest answer may be that both men are right, at different resolutions of the same very long story.
Every institution examined here — whether it arose with no outside contact at all, as Jixia did, or grew from a documented or richly plausible line of inheritance running back to the Hellenistic world, as Nalanda, Sura and Pumbedita, and Al-Azhar all did in their different ways — ended up in nearly the same relationship to power: built or sustained by it, and rarely able to escape its hold fully. That grip looked different depending on where it is examined: purposeful and deliberate at the scale of a single founding act, a caliph or a king who knew exactly what he wanted the institution to do; closer to blind accumulation at the scale of centuries, where no one was tracking, let alone controlling, what an institution's slow habits of teaching might eventually produce. Only once, at that second, slower scale, did the knowledge outgrow the power that housed it, adapting into something adjacent rather than being replaced.
That is not evidence that knowledge reliably challenges unsustainable power. It is evidence that it can, rarely, unpredictably, after a very long wait, in exactly the gap between what a founder plans and what centuries of unwatched accumulation eventually produce. Every civilisation examined here built one of these institutions anyway — some scale of the decision was always deliberate, even where the outcome, centuries later, was not. Underneath the politics and the accumulation both, in the quiet chatter of a common room or a communal roost, something older and simpler may explain why the niche was worth occupying in the first place: gathering to exchange information about the world, and about each other, appears to be one of the oldest pleasures there is. The university is one particular, elaborate, human expression of it. It has just never quite belonged only to the people gathering there.
See also: AI and Higher Education (YoungFamilyLife, 2026), which shares source material with this essay's account of medieval Oxford and Cambridge and the Reformation, and Academia and the Marketplace (YoungFamilyLife, 2026) for the modern continuation of the argument that knowledge institutions are shaped by the financial and political power sustaining them.
Topics: #WhyUniversity #IndependentEnquiry #HistoryOfUniversities #HigherEducation #AnimalBehaviour #PowerAndKnowledge
© 2026 Steve Young and YoungFamilyLife Ltd. All rights reserved.
All original content on YoungFamilyLife — including essays, articles, frameworks, and other written material — is the intellectual property of Steve Young and YoungFamilyLife Ltd, unless credited to another source. Content is developed collaboratively using AI assistance to research sources and refine structure, while all intellectual authorship, original insight, and professional expertise remain those of the author.
No part of this website may be reproduced, distributed, or transmitted in any form without prior written permission, except brief quotations for noncommercial use.
For permission requests, contact: info@youngfamilylife.com