Although dementia praecox or schizophrenia has been considered a unique disease for over a century, its definitions and boundaries have changed over this period and its etiology and pathophysiology remain elusive. Despite changing definitions, DSM-IV schizophrenia is reliably diagnosed, has fair validity and conveys useful clinical information. Therefore, the essence of the broad DSM-IV definition of schizophrenia is retained in DSM-5. The clinical manifestations are extremely diverse, however, with this heterogeneity being poorly explained by the DSM-IV clinical subtypes and course specifiers. Additionally, the boundaries of schizophrenia are imprecisely demarcated from schizoaffective disorder and other diagnostic categories and its special emphasis on Schneiderian “first-rank” symptoms appears misplaced. Changes in the definition of schizophrenia in DSM-5 seek to address these shortcomings and incorporate the new information about the nature of the disorder accumulated over the past two decades. Specific changes in its definition include elimination of the classic subtypes, addition of unique psychopathological dimensions, clarification of cross-sectional and longitudinal course specifiers, elimination of special treatment of Schneiderian ‘first-rank symptoms’, better delineation of schizophrenia from schizoaffective disorder, and clarification of the relationship of schizophrenia to catatonia. These changes should improve diagnosis and characterization of individuals with schizophrenia and facilitate measurement-based treatment and concurrently provide a more useful platform for research that will elucidate its nature and permit a more precise future delineation of the ‘schizophrenias’.