Why Britannica?

Britannica Overview

The Nineteenth-Century Knowledge Project is creating a data set of nineteenth-century knowledge in order to study how knowledge changed over time. This is “official” or dominant knowledge, and I will explain why in a moment. We use historic editions of Britannica as a proxy for what counted as knowledge in the past. Britannica was the most authoritative general reference work of the nineteenth and much of the twentieth century. First published in 1771, it continues in publication today and is the only encyclopedia in any language to survive that 250-year period. It has long been used by researchers to document changes in individual concepts over time, since it provides evidence for when a concept could be called “widely accepted.” But this data has much more to tell us than what happened to individual concepts. Britannica’s continuity gives us a unique opportunity to explore the broader question of how the structure of knowledge changed; it allows us to compare different editions and identify patterns in its transformation. That is the goal of the Knowledge Project.

There have been fifteen editions of Britannica since 1771. Some were reissues with a few new articles, but four relevant to C19 were major editions with new versions of most entries. We based our corpus on these four, which span the time from the French Revolution to WWI.

1. 3rd ed. 18 vols., 1788-1797.

2. 7th ed. 21 vols. 1830-1842.

3. 9th ed. 25 vols. 1875-1889.

4. 11th ed. 29 vols. 1910-1911.

Altogether, these 93 volumes include about 113,000 entries totalling 125 million words.

Britannica was a comprehensive reference source. This immediately distinguished it from the many domain-specific encyclopedias, like the Domestic Encyclopedia (1802), which focused on England and agriculture, or the self-explanatory Brewer's Dictionary of Phrase and Fable (1870). We know Britannica represented all knowledge, too, because it told us so.

Their advertising in 1911 called it "The Sum of Human Knowledge." And they took comprehensiveness seriously. They recruited major figures from every field to explain the newest developments (Kogan, The Great EB [1958], 31 ff).

Knowledge then and now

A remarkable list of Victorian intellectuals, but where are all the women? The first articles by women were not published until the ninth edition in 1889, 118 years after the first edition. The exquisite novelist Mary Augusta Ward (Mrs. Humphry Ward) was asked to write a small critical biography of the Renaissance poet, dramatist, and courtier John Lyly, a male author.

The absence of women authors tells us little about Britannica, actually, but it illustrates how official knowledge works: As social beliefs change, so do the culture’s ideas about what counts as “knowledge.” Nineteenth-century social beliefs determined the gendered selection of authors. They also determined the selection and length of entries in the encyclopedia, where length indicates importance. There were few topics specific to women, something we would expect from a society that discounted women and their contributions. In a similar vein, articles on India and Africa reflected the perspective of the British imperial empire; they represented indigenous people in ways that were overtly racist and imperialist. There were no articles by writers of color in any of these editions. Many of the these articles are deeply offensive to readers today; if you do not believe me, just look at "Slavery" in the ninth edition.

This historical collection is not what we would call “knowledge” today, and that is exactly the point. Knowledge is socially constructed, and it changes over time as social beliefs change. This is by no means a new idea. Karl Mannheim wrote about it in 1936: “The principal thesis of the sociology of knowledge is that there are modes of thought which cannot be adequately understood as long as their social origins are obscured” (Mannheim, Ideology and Utopia 2). Britannica was received as the most authoritative reference source in English because it faithfully represented the idea of knowledge to Victorians, as they understood it. It was an elitist system, racist, and just wrong, but thanks to Britannica’s attempts at comprehensiveness, we now have a tailor-made data set for identifying the assumptions that governed decisions about what counted as knowledge in the nineteenth century, what did not count, and often why.

Temporal change

The offensiveness of many entries in these four editions is heuristic. The Victorians who wrote them were intelligent and well-informed. They were also products of their time, and we can see many of the major preoccupations of nineteenth-century play out in these entries. What was the proper relations between men and women? between upper and lower classes? between urban sophisticates and rustics? between races? between metropolis and colonies? between religion and science? The list is long. We cannot read this material for guidance on any of these topics, any more than we might read the article on "Railroads" for guidance on current engineering standards. But we certainly can read them as authoritative guides to elite nineteenth-century British attitudes and assumptions. The wealth of historical information in this material lies in its comprehensive representation of the of the perspective of Britain's intellectual class—the class that defined what counted as knowledge—not in anything useful it can tell us about women, people of color, religion, or railroads as such. It will tell us a great deal about what non-elites had to put up with, and possibly even why there are so many reports of railroad accidents at the time.

But to view this collection as a static representation of British culture is a mistake. These entries look sometimes bizarre and inexplicable to us because knowledge change over time. And it was changing as well within the 118-year-span of the collection. A dramatic example of this occurs in the entries on "History" in the 7th and 9th editions. A revolution in attitudes toward religion took place between 1842 and 1889, when the two editions were published. In the earlier entry, the topic of history includes geological time, natural history, and human pre-history. Authority for understanding those histories derives from Christian scripture, and any discoveries that might contradict scripture must be reconciled to it. This narrative is replaced in 1889 with a discourse on historiography itself, as if to divorce the entry from its earlier incarnation, replacing it with a more scientific, inductive process. The topic is specifically limited to written history; the formation of the earth is to be found in a major entry, "Geology," which has expanded from 1 to 164 pages in the two editions. Human pre-history has moved to an entirely new topic, "Anthropology," which did not exist in the earlier edition.

Knowledge change today

This example shows that knowledge was in a state of flux within the time period covered by these editions. What can they tell us about the process and parameters of how knowledge changed over time? What relationships existed among discrete changes, like the mutation of "History," "Geology," "Anthropology," and related concepts? The goal of the Knowledge Project is to analyse the changes to what counted as knowledge over the 118-years span of these four editions. We know that knowledge continues to change since then, and that it is changing today in the contemporary world. It is my hope that this data set of content from these dusty reference sources can yield useful information about knowledge change not just in the nineteenth century, but even in some insights that we might see in other times and places, particularly in the current world.