Sandro Botticelli's Ideal Portrait of a Lady, painted in around 1475 in Florence, is one of the gems of a huge show,The Renaissance Portrait From Donatello to Bellini, which just opened at New York's Metropolitan Museum. The exhibition looks at the very beginnings of people-pictures in Western art, gathering some of the greatest portraits ever made, with one stunning image after another. But here's its big surprise: The Renaissance Portrait proves that we still know almost nothing about how portraits work, how they were made, why they were made, or even what a portrait really is. Can almost any picture of a person's face count as a portrait, or does a portrait need to have been made through direct contact with its sitter? How idealized—or caricatural—can a picture be, and still count as a portrait? Can we ever say for sure what's a "realistic" portrait versus a deceptively flattering one? Were portraits commissioned mostly to show other people what the patron looked like and stood for, or to establish that patron's own sense of self, in an age where mirrors were small and flawed? Or were portraits commissioned simply so viewers could take pleasure in the fine art of portrayal? We don't even know how people sat for their portraits, or for how long—if and when they did at all. Even a portrait as famous and heavily studied as Botticelli's, which is on loan from Frankfurt's Städel Museum, provokes almost every one of these questions.
In this slideshow, see the range of Renaissance objects that can seem to count as portraits, and the range of problems they arouse.