I respond to two key issues raised by Bernstein and Frankel. One concerns the complex and potentially useful impact of misdemeanors on the treatment process. Without, however, minimizing this dimension of misdemeanors, I focus instead on how we deal with instances when we fail our patients by deliberately placing our own needs ahead of theirs. Bernstein raises the possibility that we are most likely to commit misdemeanors when we embrace an idealized Winnicottian model. I disagree, suggesting that all theoretical positions exclude some aspect of the analyst's personhood. Ultimately, we cannot escape the conflict between the analytic ideal and the reality of our nonideal humanity.