Gutenberg's Second Half: What the Printing Press Actually Decentralized

Printing genuinely moved the power to copy from the few to the many. The licensing regimes, religious wars and nation-states that followed carry the honest lesson: a decentralized medium is not a decentralized society.

The comparison everybody reaches for is Gutenberg. Whenever a new technology promises to take a function away from a gatekeeper, the printing press is called as the character witness: proof that this has happened before, and that it ended well. Half of that is true. Around 1440 a Mainz goldsmith assembled a working system for casting metal type, and within sixty years presses were running in 282 towns across Europe, leaving some thirty thousand distinct editions on the record. That is a real decentralization, and the sceptics do not get to erase it. The capacity to make identical copies at scale left the scriptorium and the chancery and became something a tradesman with capital could own.

What the character witness leaves out is everything that happened next. On this series' reading, the two centuries after Mainz are not a story of knowledge flowing outward until the old powers gave up. They are a story of a contest the center kept losing tactically and winning structurally: licensing regimes built faster than the presses, wars fought over what the presses printed, and a state system that ended up reaching into its subjects' lives more thoroughly than any pope had managed. Decentralizing a medium redistributes the fight. It does not decide it.

Begin with the evidence that print genuinely shifted something. Testing the question econometrically in 2014, the economist Jared Rubin found that European cities with at least one printing press by 1500 were at minimum 29 percentage points more likely to be Protestant by 1600, using distance from Mainz as an instrument against the obvious objection that presses appeared where reform was already brewing. Whatever else is arguable, the machine moved a continent's religion.

Now the part the triumphal version omits. In early 1525, German peasants put their grievances into the Twelve Articles, a document printed roughly twenty-five thousand times in two months and carried into villages whose inhabitants mostly could not read it themselves. What followed was the largest popular revolt in Europe before 1789 and as many as a hundred thousand peasants dead. That May, Martin Luther, whose own pamphlets had made the revolt thinkable, published Against the Murderous, Thieving Hordes of Peasants and told the princes to strike. The settlement came thirty years later at Augsburg in 1555: each prince would fix his territory's confession, Lutheran or Catholic and nothing else, and dissenting subjects could sell up and emigrate, a right that did not reach the serfs. That is the political outcome of the first century of cheap printing. Not a public sphere. A rule assigning the religion of everyone in the Empire to whoever held the land under his feet.

Meanwhile the censorship machinery went up fast. Rome printed a papal Index in 1557 and a revised one in 1559 banning the entire works of some 550 authors; the Tridentine Index followed in 1564, a standing Congregation of the Index was created in 1571, and the list was not formally abolished until 1966. England worked through the trade as much as the church: a royal charter to the Stationers' Company in 1557 that made a guild the enforcer of its own monopoly, a Star Chamber decree in 1586 confining printing to London plus one press each at Oxford and Cambridge while leaving the licensing pen with the Archbishop of Canterbury and the Bishop of London, a ban on newsbooks in 1632 after foreign diplomats complained about English coverage of the Thirty Years' War, and the Licensing Act of 1662, under which the licenser Roger L'Estrange reduced the newspaper press to a single official gazette. None of this is a footnote to the printing revolution. It is the printing revolution's second half.

The instructive episode is what the winners did. Parliament abolished the Star Chamber in 1641, in a constitutional fight with a king it would be at war with a year later. On 14 June 1643, the same Parliament passed its own ordinance for the regulating of printing, restoring pre-publication licensing and handing enforcement back to the Stationers in exchange for their monopoly. Milton's Areopagitica, published in November 1644, was aimed at those men, not at the king. Even Milton drew his line short of the principle: "I mean not tolerated popery, and open superstition." English licensing finally lapsed in 1695, two and a half centuries after Mainz, and not because anyone had been persuaded. The Commons simply declined to renew a law that, by their own reasons, had "in no-wise answered the End for which it was made."

Nor did print only arm the edge. Benedict Anderson argued in Imagined Communities, published in 1983, that print-capitalism standardized vernacular languages and allowed strangers who would never meet to imagine themselves as one people, each reading the same newspaper on the same morning in a ceremony performed, as he put it, "in the lair of the skull." The institution that harvested that imagining was the nation-state, which went on to conscript, school, register and tax at a scale the manuscript era could not have contemplated. The press dispersed the means of copying and helped assemble the most effective centralizing machine yet built. Both are true, and an account that reports only the first is selling something.

A further correction cuts at the technology story itself. Elizabeth Eisenstein's The Printing Press as an Agent of Change, published in 1979, credited print with dissemination, standardization and preservation, treating fixity as a property the machine supplied. Adrian Johns replied in The Nature of the Book in 1998 that fixity was made, not given: reprinting brought piracy and corruption, degrading even Galileo's printed diagrams across successive editions, and the trust readers came to place in print was a social achievement built over generations. Part V of these papers made the same argument in Ostrom's vocabulary. The virtues we credit to a technology are usually achievements of the institutions built around it, and institutions are where power settles.

Keep the scale honest. As late as 1820, on Our World in Data's estimates, roughly one person in ten worldwide could read and write. For most of the period in which the press was supposedly handing knowledge to the many, the many still needed someone to read it to them. So use Gutenberg as case law, not as a mascot. The record says a real decentralization can produce mass violence nobody intended, hand its largest dividend to a new center, and take two and a half centuries to secure the freedom it is remembered for. The test Part VI set is not whether a technology distributes copies. It is which way power flows, and whether the people under it can verify, amend and leave. Mainz passed the first test within a decade. Europe spent two hundred and fifty years on the second, and the argument over who may license a press has never closed. It has only changed venue.