<b><i>Background:</i></b> Cognitive-behavioral therapy (CBT) has played an increasingly important role in psychotherapy for schizophrenia since the 1990s, but it has also encountered many theoretical and practical limitations. For example, methodologically rigorous meta-analyses have recently found only modest overall effect sizes of CBT treatment, and therefore questions have emerged about <i>for</i><i>what</i> and <i>for whom</i> it actually works. <b><i>Method:</i></b> The focus of the present paper is to elucidate the theoretical assumptions underlying CBT for schizophrenia and to examine their consistency with abnormalities of experience and self-awareness frequently reported by schizophrenia patients and systematically studied in phenomenological psychopathology from the beginning of the 20th century. <b><i>Results:</i></b> We argue that a strong theoretical emphasis on cognitive appraisals with only subsidiary attention devoted to affective and behavioral processes - as is characteristic of many forms of CBT - cannot satisfactorily account for the complex subjective experiences of schizophrenia patients. We further argue that certain theoretical strategies widely employed in CBT to explore and explain mental disorders, which involve <i>atomization </i>and, at times, a <i>reification</i> of mental phenomena, can be problematic and may result in a loss of explanatory potential. Finally, we provide a detailed account of how negative symptoms and delusions are conceptualized in CBT and consider the question of how these concepts fit the actual phenomenology of schizophrenia. <b><i>Conclusion:</i></b> We suggest that further advancement of CBT theory and practice can benefit from a dialogue with phenomenological psychiatry in the search for effective psychotherapeutic strategies for schizophrenia patients.