Hans Rosling, who died earlier this month, made data come to life in ways that were simultaneously astonishing, intuitive, elegant and, ultimately, simple. What is remarkable about Rosling’s landmark TED Talk of 2006, titled The Best Stats You’ve Ever Seen, is that there is very little data in that presentation that had not already been available for many, many years. Rosling’s genius was threefold. He first wondered why people held biased views that were no longer backed by data. Secondly, he decided that the problem was not paucity of data but the way in which this data was represented and visualized. Thirdly, he decided that the trick was to see all this data not as static markers but as moving, animated visualization.
There is perhaps one demand-side reason why Rosling’s work resonates with so many people, especially on the Internet. In many ways, we live in an age of the tyranny of measurables. Everything from the performance of companies to the performance of nations is ranked and measured. Good companies are measured by their stock prices. Good countries are measured by their “Global Competitiveness” or “Ease Of Doing” business rankings. With plummeting trust in news, widespread commodification of opinion and comment, and a global distrust of any kind of filter, it is little wonder that we are in such thrall of measurables and data and statistics. These are, as it were, the hardest things to lie about. Data is king.
Many historians too, and this might come as a surprise, extensively use data and quantitative techniques in their research. How can this be, you ask? Aren’t the humanities all about sitting in dark rooms, making up Marxist stuff in tweed jackets and then picking up a government salary at the end of the month before emailing your students to go on strike?
Not quite. Historians have been formally grappling with quantitative methods and statistical techniques, as we recognize them today, since at least the 1960s. The first journal devoted to this type of “Quantitative History” was Historical Methods Newsletter, established in 1967. There are now many others. Indeed, these days it is a rare historian who does not have to grapple with some form of quantitative technique in his work.
Of course, this does not automatically legitimize this branch of work more than any other branch of historical scholarship. It is important to not fetishize numbers in a spreadsheet because numerals somehow smell truthier than alphabets. Having said that, these methods can often yield results of great value to the historian. Let me briefly cite one such example.
In the early 2000s, Roman Studer, then a doctoral student at Oxford, began to collate the market price of wheat and rice in 46 cities in India between 1700 and 1914. This was a formidable task. It is not only difficult to find such data, but more so to find information across unbroken periods of time. Eventually, Studer settled on 70 “price series” for wheat and rice. He then began to analyse this data to understand something fairly straightforward: the levels of integration of all these markets. He investigated how prices fluctuated across these markets, whether these fluctuations were correlated, and how this correlation changed with the distance between markets.
A paper he published in 2008, titled India And The Great Divergence: Assessing The Efficiency Of Grain Markets In Eighteenth- And Nineteenth-Century India (goo.gl/zxLKOL), in which he summarized his findings, makes for fascinating reading: “Before 1850, these year-to-year price movements seem completely unrelated in regions that were distant from one another; even the massive price spikes did not coincide.” Thus, while prices shot up astronomically in Bengal during the 1770 famine, they remained stable in northern and western India. It was only from 1850 onwards that markets seemed to show signs of integration and efficiency: “Not only did prices inflate similarly in all three regions during this later period, but the short-term variations also looked very similar. This feature contrasts starkly with the totally asynchronous price movements observed for the early period.”
But what is the broader point of this result? Studer’s research feeds into the Great Divergence debate that continues to enthrall many historians. This is the one in which historians (and anybody else in the mood really) argue over why western Europe developed before the rest of the world in the period leading up to the Industrial Revolution and colonialism. To gravely generalize, the debate is waged between one side that says this Divergence was due to the unique characteristics of the West, and another that says that these characteristics existed elsewhere until the onset of colonialism ruined it for the Indians, the Chinese and others.
Studer’s results aligned with the first camp. His point was that grain markets in India remained highly unintegrated until the onset of colonial rule. Thus, the lack of sophisticated markets in India up to that point, in comparison to Europe but also some parts of China, suggests that India’s economy was already underdeveloped even before the British arrived on the scene.
The debate continues. And regardless of how his results make you feel, Studer’s paper is well worth reading.
Thus, data is not the purview of the hard sciences alone. If there is anything we should take away from the brilliant work and infectious enthusiasm of Rosling, it is that data can be useful to people from all walks of life. Including historians.
You can tell a lot just from the price of rice.
Déjà View is a fortnightly conversation on history. Read Sidin’s Mint columns at www.livemint.com/dejaview