Out just in time for the holidays, a special issue of Smithsonian Magazine lists the “100 Most Significant Americans of All Time.” It’ll cost you $9.99 plus shipping to read the full issue, but you can get a preview from the magazine’s website, which also explains the unique methodology behind the list. Though magazine editors added their own choices, most of the list derives from a Google-like ranking algorithm developed by Google engineer Charles Ward and computer scientist Steven Skiena:
…while Google ranks web pages according to relevance to your search terms, Skiena and Ward rank people according to their historical significance, which they define as “the result of social and cultural forces acting on the mass of an individual’s achievement.” Their rankings account not only for what individuals have done, but also for how well others remember and value them for it.
Their method requires a massive amount of big data on historical reputation. This they found in the English-language Wikipedia, which has more than 840,000 pages devoted to individuals from all times and places, plus data extracted from the 15 million books Google has scanned. They analyzed this data to produce a single score for each person, using a formula that incorporates the number of links to each page, the number of page visits, the length of each entry and the frequency of edits to each page. Their algorithms differentiate between two kinds of historical reputation, what they call “gravitas” and “celebrity.” Finally, their method requires a means of correcting for the “decay” in historical reputation that comes with the passage of time; they developed an algorithm for that, too.
Earlier this week at my personal blog, I asked some U.S. historian friends to weigh in on this methodology. Here’s what our neighbor Jonathan Den Hartog (Univ. of Northwestern St. Paul) had to say:
The methodology strikes me as a reminder of why we need historians. Historians tell stories that make sense of data and experience. This list seems heavy on data but not on interpretation. I would suggest there is not a single category of “significance.” Rather, significance takes shape only when put into context of a specific story. Thus, someone extremely significant in one category might not make such an over-all listing. Significance is not currency that one acquires: it’s only relevant in relation to others. Further, “significance” might not be national or institutional. The single mother who struggles to allow her children to succeed may have attained a level of “significance” in a small circle much greater than many of these individuals. This raises the moral and even Christian question of what counts as significance.
You can see what Jonathan and the other three thought of the actual names on the list here, but we’d love to hear from our students and alumni: As you look at the list, who’s missing? Who should be missing? Whose inclusion on the list was the most pleasant surprise?
– Chris Gehrz