In this paper I reply to Spence's criticism of Schafer's action language theory as applied to work with patients. Spence assumes that such an application will replace the analytic situation with adversarial confrontation. This would supposedly come about largely through badgering the patient to produce reasons for his behavior and blaming him both for the behavior and for not producing the right reasons for it. He also foresees an overly specific and desiccated language replacing the metaphors and allusions that accurately portray experience. Spence's concerns rest on his assuming that Schafer's ideas would be applied clinically in ways that no one need do even if employing the ideas theoretically.