John_Maxwell comments on How much can value learning be disentangled?