In the early 1970s, he acquired a small landscape painting and attributed it to Vincent van Gogh. The attribution was disputed. The Van Gogh Museum in Amsterdam, which maintains the official catalogue of Van Gogh’s work, did not accept it. The painting existed in a state of contested attribution for decades — accepted by Hannema, rejected by the Van Gogh Museum, discussed in the literature as a case of either correct identification or mistaken attribution by an elderly scholar with a history of authentication errors.
In 2010, the Van Gogh Museum reversed its position and accepted the painting — *Sunrise at Montmajour* — as a genuine Van Gogh. The painting, acquired for a modest price as a disputed attribution, was now worth millions.
The parallel to the van Meegeren case was not lost on observers. Hannema had been wrong about the Emmaus — catastrophically wrong, in a way that cost the institution he ran an enormous sum and embarrassed the Dutch art establishment. He had been dismissed, in part, for the consequences of that error. And then, decades later, he had been right about a Van Gogh that the Van Gogh Museum had rejected — right when every institutional authority said he was wrong.
The two cases together do not resolve the question of whether expert judgment can be trusted. They complicate it in productive ways. Expert judgment failed spectacularly in the van Meegeren case because the forgery was specifically designed to defeat it. It succeeded, eventually, in the Hannema-Van Gogh case, though only after a long delay and against institutional resistance. The lesson is not that experts are reliable or that experts are unreliable. It is that the authority of expert judgment is both real and limited, and that the institutional protocols surrounding authentication need to account for both the genuine value and the specific limitations of connoisseurial expertise.
