Diagnostic systems, phenotype models, and theories of etiology incorporate propositions on the underlying nature of psychosis and schizophrenia phenotypes. These propositions, whether implicit or explicit, are that the distributions of the phenotypes, or the phenotype experiences themselves, are dimensional or categorical. On one hand, evidence on the epidemiology of schizophrenia phenotypes suggests symptom phenotypes may not be bound by conventional diagnostic thresholds but instead may blend imperceptibly with subclinical, statistically frequent experience, supporting continuum viewpoints. On the other hand, evidence on the population structure suggests a latent categorical structure; the population may be composed of two types of people. However, both sets of evidence are beset by methodological limitations that point unequivocally to the need to move beyond current diagnostic conceptualizations, observation, and anamnesis of psychosis, and toward responsive and scientifically refutable formulations of schizophrenia.