This essay is based on the working paper “Enlightenment Ideals and Belief in Progress in the Run-up to the Industrial Revolution: A Textual Analysis” by Ali Almelhem, Austin Kennedy, Murat Iyigun, and Jared Rubin.
Understanding why the modern economy emerged where (Britain) and when (in the eighteenth and nineteenth centuries) it did is one of the great tasks in the field of economics. Several theories have recently been put forth in this regard, focusing on a range of factors, including Britain’s high wages (which encouraged the invention of labor-saving capital—that is, machines that reduce the number of labor hours required to produce goods), abundance of skilled workers, access to coal, and access to the New World and its slave plantations.
One of the more compelling theories, put forth by Joel Mokyr in his 2016 book A Culture of Growth, argues that Enlightenment ideals of progress combined with Britain’s abundance of workers to foster the innovative environment that is the hallmark of the period. Specifically, Mokyr argues that a key cultural change prompted by Enlightenment thinking is that science could be used for the betterment of humankind.
Mokyr supports his theory with much qualitative evidence. After all, culture is notoriously difficult to quantify, so how could one begin to measure the extent to which the type of cultural changes cited by Mokyr are important to Britain’s industrialization? Such quantification is precisely what we do in a new working paper, “Enlightenment Ideals and Belief in Progress in the Run-up to the Industrial Revolution: A Textual Analysis.”
How does one quantify cultural change? While there are numerous proxies for culture that social scientists employ, arguably the most important is language. What were the words people used? In what context? How did this change over time? The answers to these questions provide us with clues regarding what people in the past believed was important and how they viewed the world.
Specifically, with regard to Mokyr’s theory, we want to know how the language of science was changing—if it was in fact changing—in the periods preceding and during Britain’s industrialization. If Mokyr is correct, the language of science should have become more progress oriented in this period. People writing on the subject of science should have seen what they were doing as contributing to the improvement of the lot of their fellow humans.
We test whether this was in fact the case by performing a text analysis on the entire corpus of works published in England between 1500 and 1900 that have been digitized. This selection yielded approximately 173,000 volumes.
The first question we address is “What is a work of science?” This question is not so easy to parse as it sounds. First, while certain works (like those of Newton or Boyle) are clearly science, many others lie in a gray area between science and other fields (including fiction). Second, and more importantly, what we are interested in is the language of science. Thus, even if one was writing about a topic that we would clearly recognize as science today, it is possible (and probable, prior to 1650 or so) that it employed the language of religion, economics, or another field.
Fortunately, text analysis algorithms enable us to overcome these issues. The corpus of books reveals three distinct types of language—that of science, religion, and political economy. We use these distinctions to classify each volume according to the language it uses. No volume uses the language of “pure science” or “pure religion.” Instead, each volume uses a certain combination of each of the three categories.
Then, we utilize standard techniques to determine the sentiment of each volume with respect to its progress-oriented language. This enables us to assign a progress score to each volume. In combination with the type of language the book uses (that of science, religion, and political economy), we test whether (and when) the language of science became more progress oriented.
Our analysis yields three primary findings. First, beginning in the mid-eighteenth century, the languages of science and religion began to become distinct. We simply do not see the same volumes employing the language of science and religion. However, there is clearly a shared language between science and political economy, and there is a different shared language between religion and political economy. Second, and more importantly, the language of science began becoming more progress oriented beginning in the mid-eighteenth century. However, it was not the language of pure science that became progress oriented in this period but those works that shared the language of science and political economy.
This makes sense if changes in language were, in fact, of significance for Britain’s industrialization. It was not the elite intellectuals, such as Newton or Boyle, who invented the machines that characterize Britain’s industrialization; rather, it was the craftsmen and artisans who regularly worked with machines and had specialized skills in manipulating metals, wood, and other materials. It was the works aimed at this audience that became more progress oriented on the eve of the Industrial Revolution.
Third, we find that the change in progress-oriented language was strongest in volumes that also contained a substantial amount of language related to industry. That is, it was precisely those manuals and science-related works that aimed at a broad, industrial audience that were the most progress oriented on the eve of industrialization.
Our results are not intended to suggest that cultural changes are all that mattered for Britain’s Industrial Revolution or all that contemporary countries in the developing world need is cultural change to become innovative. The cultural changes that occurred in Britain were a part of a package of institutional, commercial, political, and demographic changes that propelled the British economy and industry forward.
Nevertheless, these findings have significant implications for how we conceptualize Britain’s Industrial Revolution and, for that matter, the broader role that culture plays in economic processes. They suggest that cultural shifts favoring progress may be important for an economy to become innovative. On the surface, this appears to be an obvious statement. It is difficult to imagine a flourishing, innovative economy in a society actively opposed to innovation. However, understanding what constitutes the elixir of innovation is not an easy task. Our results suggest that one should dismiss the importance of culture at one’s peril.
Read the full paper here.
Jared Rubin is professor of economics at Chapman University.
This essay is part of the Long-Run Prosperity Research Brief Series. Research briefs highlight research that enhances our understanding of the factors that drive long-run economic growth and examine its policy implications.
