Digital Approaches to 19th Century Globalization

By Jim Clifford
First Published on ActiveHistory.ca
The map below drew a lot of attention on Twitter when I posted a few weeks ago in advance of a presentation I gave at an environmental history conference in early July. It was retweeted, not just by friends and fellow environmental historians, but also by Shawn Donnan, a World Trade Editor at the Financial Times. I think it gained traction because it helps visualize something historians and students who take our classes know, but might not be general knowledge: globalization did not begin in the late 20th century with the rise of industrial economies in Asia.

Extensive trade networks predate Columbus and the flow of silver from mines in the Americas through Europe and to China linked and transformed the world economy during the Early Modern period. The scale of global trade and communications has changed significantly over the centuries, but globalization has very deep roots.

Far reaching industrial supply chains date back to the nineteenth century and in a few cases further back. British industrial development relied on importing raw materials from all over the world. Britain was simply too small of an island to supply all of the materials required by the growing factories and it did not have the climate to produce many of the materials required by innovative new industries. By the second half of the nineteenth century many of the products consumed and produced in London originated overseas. These included soap, candles, bread, margarine, marmalade, rubber rain jackets, leather shoes, inks, dyes, paints, fertilizers,  and wooden furniture. These consumer goods were manufactured in factories in the Thames Estuary from raw materials imported, as the map above shows, from Canada, the United States, Jamaica, Peru, Brazil, Spain, West Africa, India, Sri Lanka (Ceylon), and New Zealand, among many other locations. You can zoom into the map and click on locations to see the range of commodities sent to Britain (the locations are rough approximations, as most of the underlying data is at the national level).

Identifying and following industrial supply chains is difficult enough in the present and it is even more complicated for the nineteenth century. I’ve found a lot of information in British archives, but these sources only get me so far. The internet, however, makes it possible to find, organize and read digitized government reports, newspapers, and books from a wide range of sources. The Internet Archive and Google Book contain a staggering number of books from this time period, as they are not under copyright and the big North American research libraries who partnered with these massive digitization projects hold copies in their collections (read Ian Milligan’s post on the scale of the Internet Archive). Jstor and Project Gutenberg provide access to many more 19th century books and journals. Those of us with university affiliations can access numerous other major collections of documents digitized by private companies and not for profit organizations, including millions of pages of British and Canadian government documents found in the House of Commons Parliamentary Papers (HCPP) and Early Canada Online. Altogether the internet has quickly grown into a massive archive with particular strength in nineteenth century English language sources. In all of these cases we can also search and analyse the content of the books and even though computers misrecognize some of the content, this is still a much more powerful tool than a traditional library catalog. Following commodities chains is a little more straightforward with the internet at our disposal.

The map above was created by reading through the “Annual statement of the trade and navigation of the United Kingdom with foreign countries and British possessions in the year 1865” downloaded from the HCPP collection. Using this document I identified the source countries of all of the raw materials coming into Britain that were necessary to supply factories identified on digitized maps of London from the second half of the nineteenth century. This annual series of British trade documents, and others like it for Canada and the United States, are great for raw trade statistics, but they don’t provide much insight into how new commodity chains developed over the course of the nineteenth century. For that we need to analyse the content of many more reports, books and journals. Thankfully historians are not the only researchers confronting the immense amount of information now available digitally on through the internet. Computer scientists and computational linguists are busy researching new ways to identify and extract useful information from the seemingly infinite number of digital words.

Trading Consequences is a collaboration between linguists and computer visualization experts in Scotland and environmental historians here in Canada. We’ve been working for the past two and a half years to adapt their text mining methods to nineteenth century sources. We attempted to teach computers to search for all possible words describing commodities and all location names and then to identify the links between commodity and places. This method is not without its shortcomings and we still have a lot of work to overcome a number of particular challenges when working with scanned historical documents, but by working with millions of pages of text, we managed to extract significantly more information on the changing patterns of nineteenth commodities. Furthermore, we built a number of visualizations that allow you to explore our results and identify documents that could be of further use in your research. Using new tools to explore the huge quantity of historical texts available on the internet makes it easier to explore the complicated history of globalization in the nineteenth century. This collection of software does not write history for us, but it does make it possible to research the history in new and innovative ways alongside our continued work in physical archives.

Further reading:

Andrew Watson and myself have a blog post today over on the NiCHE website discussing our ongoing research into the global commodity chains of the nineteenth century leather industry in Britain, the United States, Canada, and India.

For more a little more detail on Trading Consequences see:  Data mining uncovers 19th century Britain’s fat habit on The Conversation, 3rd April 2014. http://theconversation.com/data-mining-uncovers-19th-century-britains-fat-habit-25064

For a lot more on Trading Consequences see our White Paper: Klein, E., Alex, B., Grover, C., Tobin, R., Coates, C. Clifford, J., Quigley, A., Hinrichs, U., Reid, J., Osborne, N., and Fieldhouse, I. 2014. Digging Into Data White Paper: Trading Consequences, March 2014.

For an accessible and affordable history of early globalization see Robert B. Marks, The Origins of the Modern World: A Global and Ecological Narrative from the Fifteenth to the Twenty-First Century, 2nd Edition, 2nd edition (Lanham: Rowman & Littlefield Publishers, 2006).