How do we know what historical figures really looked like?

Posted by Dave on May 20, 2010 | 5 Comments

Compare these two pictures of George Washington. Which one is more accurate?

Most Americans would probably guess the picture on the left. (Left-image: Gilbert Stuart “Vaughan” portrait of George Washington. Source: Wikimedia Commons. Right-image: Charles Wilson Peale portrait of George Washington. Source: Metropolitan Museum.) But how would you really know? Gilbert Stuart is a more famous portraitist than Peale, having painted the famous incomplete Athenaeum portrait of Washington. Are you just biased because you’re more familiar with Stuart’s work?

It might seem that there is no way to know—after all, there are no photographs of Washington—but Eric Altschuler and Ahmed Meleis believe they have come up with a method to assess a portrait artist’s accuracy in representing his or her subjects. While there are no photographs of the first five U.S. presidents, John Quincy Adams was photographed. He also had his portrait painted by both Stuart and Peale. Unfortunately, the photograph was taken late in Adams’ life, much later than the portraits were made. So Altshuler and Meleis used computer software to artificially “age” the portraits:

The Stuart portrait turned out to be much closer to the photograph than the Peale portrait. Other Stuart portraits similarly match with photos, while Peale’s work consistently falls short. Altschuler and Meleis say that Stuart is a sort of “Rosetta Stone” for images, and we can use Stuart’s portraits to learn what historical figures really looked like. Stuart was a prolific painter, who indeed painted many presidents and other historical figures, so this is a valuable insight.

Altschuler and Meleis go even further, suggesting that older painters who use a style similar to Stuart’s might also be relied on to give accurate representations of their subjects. You could potentially go back quite far in history, comparing portraits of individuals that both Stuart and another artist painted. If the portraits offered similar depictions, then you could say that other artist was also an accurate portraitist. The process could be repeated indefinitely. Of course, it could be problematic as you went further back in history, since the connection between the portraits and the photographic record would be progressively more remote.

Altschuler, E.L, & Meleis, Ahmed (2010). What Did the Early United States Presidents Really Look Like?: Gilbert Stuart Portraits as a “Rosetta Stone” to the Pre- Photography Era. Poster presented at the Vision Sciences Society Meeting, Naples, FL.

Comments