The folklore collections amassed by Jean-François Bladé in nineteenth-century southwestern France are problematic for modern readers. Bladé's legacy includes a confusing combination of poorly received historical works and unimportant short stories as well as the large collections of proverbs, songs, and narratives that he collected in his native Gascony. No writer has ever attempted to study any of Bladé's informants in detail, not even his most famous narrator, the illiterate and "defiant" Guillaume Cazaux. Rather than dismissing Bladé as a poor ethnographer whose transcripts do not reflect what his informant Cazaux said, I propose taking Bladé's own confusion about authenticity seriously. This confusion suggests that Bladé was trapped between three competing models that depicted the authenticity of folklore as residing in either the audience or folklorist, or the tradition, or the performer. The texts of Cazaux's legends that Bladé published were not just invented by Bladé, but forged in a dynamic interaction between the folklorist, Cazaux, and the force of tradition. When Cazaux described his beliefs in witchcraft to Bladé, he did not just reveal his own worldview; he also relied on the power of anonymous forces and silence to threaten and coerce the folklorist. The legend texts that Bladé published are not simply monovocal re-writings of some things Cazaux said; they enact a conversation between the two men about place and time. This conversation is a very limited example of an important question that has occupied historians: the "modernization" of the rural population by national forces. Although Bladé and Cazaux had very different backgrounds and education and only knew each other for ten years, their memories are intertwined for posterity.