University Times

For all the recent talk of globalization, says Patrick Manning, Andrew W. Mellon Professor of World History and director of the World History Center, this planet needs a better understanding of its global past so it can plan better for its global future. That’s why Manning and colleagues created the Collaborative for Historical Information and Analysis (www.chia.pitt.edu) and its project, the world-historical archive, and why he wrote his recent book, “Big Data in History.”

The archive, now under construction but not yet open for public contributions, will benefit not just fellow social science researchers but educational efforts at all levels, Manning says.

He recently examined some top world-history textbooks for grade-school students and found, for example, that the Mediterranean region gets scant notice. “What’s there is about Greece and Rome, the classical world, and after that there is a little bit of a mention of the Ottoman Empire. And that’s that — they just drop it.

“A lot of the history that students learn is highlights rather than the narratives or patterns or something more comprehensive,” he notes. With the archive, whose materials will be readily available to teachers, “students will be able to look at [the Mediterranean] as it changed over time, rather than as if it existed 2,000 years ago and hasn’t existed since.

“I believe looking at evidence over time will show repetition and continuity and show that the issues of today are being fought out not for the first time.”

The archive will have an early concentration on historical subjects where the data can be quantified into numbers most easily: economics, politics, climate and health.

Take the worldwide trade in silver, for instance, which began in the 1500s, with Spanish-run mines in Peru and Mexico, bringing the precious metal first to Europe and then to India and China. “This is a commodity that’s been pretty well studied but nobody’s been able to add it all up yet,” Manning says. “This was an important part of the world money supply, and looking at the world’s total supply and the regional distribution of it will give us a pretty good picture of the world’s economy.” He suspects data collected for the archive may “help to create a picture of India and China as growing economies in the 1600s and 1700s, which is different from the old picture of them as stagnant economies while Europe was growing.” He believes it may even show that the economies of the East were bigger than suspected during the 19th century “Great Divergence” to Europe and North America.

*

As his book outlines, the goal of the archive is “to portray long-term, global change in human society and thereby provide a basis for planning long-term, global policies for the future.

“[A]t the global level we have almost no knowledge of the historical forces and experiences that have unfolded within human society,” he writes. There are “treasure troves” of records, but they are scattered and unconnected — and they may be difficult to quantify for comparison to other databases.

CHIA aims “to seek out relationships among variables that have not previously been suspected” and to find “new empirical results on relationships in global history” as well as to fill in the missing data that can be extrapolated from these new dataset mergers. Ultimately, the project’s intention is to make visual representations of the data in charts, graphs, maps and other illustrations that are easier to construct using software.

CHIA will concentrate on gathering and digitizing data from before 1950, especially data not yet published, or from areas of the world where documentation is scarce. CHIA members will take this crowdsourced data and conduct peer reviews, store it and process it for analysis.

“[G]athering a large number of datasets is not sufficient to produce global data,” Manning writes. “The data need to be merged into a comprehensive, relational data repository.” That means creating universally understood datasets with consistent markers for places, times and topics.

To accomplish dataset mergers, the project will employ software designed here called Col*Fusion, which uses both the data and its metadata. Metadata is a description of the data: At a higher level, the metadata shows where the data came from, who created it, when it was created and in what format and structure (words or numbers, for instance). At the lowest level, the metadata forms the specific definition of the data (miles or kilometers? French or English?)

“We might have more metadata than data,” he says.

*

“It’s of extreme importance to make this something that anyone can go to,” Manning adds. “It has to bridge the gap between availability to advanced researchers and on the other hand teachers and students who are looking for much more basic information.”

When it is open to the public, he envisions a website with a portal for adding data and another for accessing the analyses and illustrations.

“We’re going to be collecting data forever,” he says. “The more effective this archive becomes the more people will be interested in submitting data. Maybe in five years that’ll be a self-sustaining project.”

It also will broaden its focus to less easily quantifiable subjects, such as the world history of cultures, values and ideas: “Any text that’s got a date and a place, you can put it into this archive and search for words that are on it, and out of that, if you do a good job of it, you get the ideas of the author’s — the perspective of the past, in China or Russia, say, rather than just our own” time and place. He believes technology will advance sufficiently in the future to make works of art similarly searchable as well.

The University is being supportive during the archive’s formative stages, Manning reports, funding doctoral or post-doctoral students to work on Col*Fusion, on data visualization capabilities and on digital stewardship (making the public archive a dependable place to store data).

“[T]he project, while complex and expensive, is feasible and highly valuable,” the book concludes, pointing toward current mass efforts to analyze the human genome and climate change that have enjoyed global collaboration. “They snowballed because it was the right time,” Manning says. “So it’s possible, if we do the right work and it’s interesting to people, that this one will snowball.

“People have talked about global views for a long time — religious leaders, military leaders,” he concludes. “But now we’re at a time when lots of people think about the global and make their personal decisions based on what they think is happening globally. So we need some global facts.”

—Marty Levine