What Print Changed: The Reformation, Science and the Fixity Debate

The first lesson in this course covered how the press worked and spread. The harder question is what it changed. The most influential answer came from the historian Elizabeth Eisenstein in The Printing Press as an Agent of Change (Cambridge University Press, 1979).

Eisenstein argued that print altered the conditions of intellectual life in ways that hand copying could not:

• Dissemination: the same text reached many more readers in many more places.
• Standardisation: every copy of an edition was identical, down to its page numbers, diagrams and tables.
• Preservation: with hundreds of copies in circulation, a text was far less likely to be lost.
• Fixity: a stable text could be cited, compared, corrected and built on.

Key idea: Her claim was not that printing caused any single event. It was that print changed the infrastructure of knowledge, and that movements such as the Reformation and the Scientific Revolution took the shape they did because they grew up on that infrastructure.

The Reformation is the strongest case for Eisenstein's view. Mark U. Edwards Jr., in Printing, Propaganda, and Martin Luther (University of California Press, 1994), assembled the numbers:

No earlier religious reformer could have produced anything like this. Jan Hus, whose ideas overlapped with Luther's, was burned at the Council of Constance in 1415, a generation before the press, and his movement stayed largely regional.

Gotcha: Edwards's often-quoted total of 3.1 million copies of Luther's works between 1516 and 1546 rests on an explicit assumption of one thousand copies per printing. It is an order of magnitude, not a count.

Share of German-language pamphlet editions, 1500 to 1530, from Edwards (1994), citing Köhler's estimate of about 10,000 editions.

Luther's output was not mostly large theological books. It was short works that a printer could set in days and sell cheaply, and much of it was in German rather than Latin. Compare the Latin-dominated trade of the incunabula period, when 72% of books were in Latin (Dittmar, Quarterly Journal of Economics, 2011).

That format had consequences for the printers as well as the reformers:

1. Low risk. A short pamphlet tied up little paper and type, so a printer could reprint a title that was selling within days and drop it when demand faded.
2. Reprinting without permission. There was no effective copyright; a successful pamphlet was reprinted by printers in other cities, which multiplied its reach at no cost to the author.
3. A reading public beyond the clergy. Vernacular pamphlets reached townspeople who could read German but not Latin, and could be read aloud to those who could not read at all.

In practice: The printers were not neutral. Pamphlets were what sold, and Edwards's figures show the market rewarding whoever could supply them fastest.

Eisenstein's second case was science. Before print, a diagram, a star table or an anatomical drawing degraded each time it was copied by hand, because copyists were rarely experts in what they copied. A printed edition gave every reader the same figure on the same page.

Two books from 1543 are often used to illustrate the point: Copernicus's De revolutionibus, with its tables and planetary diagrams, and Vesalius's De humani corporis fabrica, with its detailed woodcut illustrations of human anatomy. Readers in different countries could compare the same numbers and the same images, and argue about them precisely.

Print also changed how errors behaved:

Key idea: In Eisenstein's account, the gain was not that printed books were free of mistakes. It was that shared, identical mistakes could be found and fixed cumulatively.

The strongest challenge came from the historian of science Adrian Johns in The Nature of the Book (University of Chicago Press, 1998). Johns did not deny that print mattered. He denied that fixity was a property of the technology itself.

His evidence came largely from the early modern book trade, where he found:

• Piracy was routine. Unauthorised editions, often altered or carelessly set, circulated alongside authorised ones, so readers could not assume two copies of a title said the same thing.
• Trust was social. Readers judged a book by who printed and sold it, and by the reputation of the booksellers' networks behind it, not by the mere fact that it was printed.
• Reliability was built over time. Stable, trustworthy texts depended on institutions, from licensing and trade regulation to learned societies and their printers, that took generations to develop.

Definition: Technological determinism is the view that a technology's properties directly produce social outcomes. Johns's critique is a case against reading print this way: the same press could produce a reliable scientific text or a pirated, corrupted one, depending on who ran it.

The disagreement is about mechanism and timing more than about whether print mattered.

Many historians now read the two as complementary. Print created the possibility of stable, widely shared texts; whether a given text realised that possibility depended on the trade and the institutions around it.

A useful way to hold this debate is to sort the claims by how well supported they are.

Try it: When you meet a claim that a new medium caused a social change, ask which row it belongs in. Is it a count, a mechanism with evidence, or a story that fits the timing? The same sorting works for arguments about radio, television and the internet.