Historians working with an artificial intelligence assistant began to trace the spread of astronomical thought in Europe in the early 1500s.

The analysis contributes to challenging the idea of ​​the “lone genius” of scientific revolutions. Instead, it shows that knowledge of the positions of the stars was widespread and used in a variety of disciplines, researchers report October 23 in Advances in science.

“We can see here the first formation of a proto-international scientific community,” says computational historian Matteo Valleriani of the Max Planck Institute for the History of Science in Berlin.

Valleriani and his colleagues used AI to examine a digitized collection of 359 astronomy textbooks published since 1472, less than 20 years after the first printed The Gutenberg Bibleuntil 1650 (SN: 31.5.05).

These textbooks were used to teach introductory courses on geocentric astronomy—the view of the cosmos that places the Earth at the center and moves outward in sequential spheres. Knowing the positions of the stars was considered important to the study of everything from medicine to Greek and Latin poetry, so introductory astronomy classes were required for all students. Among other things, students learned to use the sun’s position in the zodiac constellations to find the date of an event that occurred in antiquity, before standardized calendars were common.

Studying these texts from the past can give historians an idea of ​​the basic knowledge that most educated people had about the universe and how that understanding has changed over time.

The dataset included 76,000 pages of text, images and numerical tables, many with different fonts, formats and layouts. A historian might be able to analyze a handful of books in a single career. But Valleriani and colleagues wanted to study them all.

“What we wanted to know, in general, is what astronomy students have learned over these 180 years and across Europe,” says Valleriani. “This was humanly impossible.”

The team used machine learning to identify 10,000 separate number tables in textbooks. They then trained an AI model to recognize the individual numbers in the tables. “It was extremely difficult because the tables are not formatted in the same way,” says physicist and machine learning expert Klaus-Robert Müller of the Technical University of Berlin. “Everything is a mess.”

Once the AI ​​extracted all the numbers, it compared the different tables one by one and highlighted the similarities and differences. For example, some textbooks were basically reprints of a previous edition, and their tables were almost identical. Others introduced new ideas or new ways of using astronomical data.

The AI ​​couldn’t tell the researchers what those similarities and differences meant (SN: 8/2/24). But it could give them a place to look for trends or moments of change.

“It goes from using AI as a tool, to help achieve something we were preconceiving, to using AI as a team member, suggesting new solutions that we couldn’t see,” says Valleriani.

A common story about astronomy during this time period is that individual heroes of science, such as Copernicus, Galileo, and Kepler, shook the world by showing that the Earth was not the center of the universe.

But historians of science were moving away from the idea that science is also led by lone geniuses who make great discoveries (SN: 3/5/16). Those discoveries had social, political and cultural contexts and had to be disseminated to some degree in the wider culture.

“When you’re dealing with the scientific revolution, with the triumph of the Copernican worldview, we know the big names,” says researcher Jürgen Renn of the Max Planck Institute for Geoanthropology in Jena, Germany, who was not involved in the new paper. “But in Europe, this was a broad movement. There were many participants.”

One of the team’s main findings is that textbooks printed in Wittenberg, Germany, in the 1530s were widely imitated in other parts of Europe. Similar books that were sold in cities with larger markets such as Paris and Venice created a new homogeneous approach to astronomy.

Valleriani finds this ironic. Wittenberg is best known for being the town where Martin Luther started the Protestant Reformation, which split a new branch of Christianity from the Catholic church.

“It sounds paradoxical,” says Valleriani. “While Wittenberg and the Protestant Reformation were dividing Europe … and creating the backdrop against which the wars arose, at the same time Wittenberg was able to develop a scientific approach to education that was really taken up everywhere.”

There are limitations to this type of research, the team points out. Historical data is always incomplete, and historians must choose a subset of that data to focus on. AI cannot account for this kind of selection bias. Human historians must always be part of the process, the researchers point out.

This work “shows how historians can engage with AI methods in the future and use them intelligently without this utopian or dystopian illusion that they can do the work for you,” says Renn. “They are just a fantastic new tool to help us understand history as a broad flow of human action and human thought, and not just a series of singular events.”