The hypothesis that rapid eye movement (REM) sleep disturbances are the hallmark of posttraumatic stress disorder (PTSD), proposed by Ross and colleagues in 1989, has stimulated a wealth of clinical, preclinical, and animal studies on the role of sleep in the pathophysiology of PTSD. The present review revisits this influential hypothesis in light of clinical and experimental findings that have since accumulated. Polysomnographic studies conducted in adults with PTSD have yielded mixed findings regarding REM sleep disturbances, and they generally suggest modest and nonspecific sleep disruptions. Prospective and treatment studies have provided more robust evidence for the relationship between sleep disturbances and psychiatric outcomes and symptoms. Experimental animal and human studies that have probed the relationship between REM sleep and fear responses, as well as studies focused more broadly on sleep-dependent affective and memory processes, also provide strong support for the hypothesis that sleep plays an important role in PTSD-relevant processes. Overall, the literature suggests that disturbed REM or non-REM sleep can contribute to maladaptive stress and trauma responses and may constitute a modifiable risk factor for poor psychiatric outcomes. Clinicians need to consider that the chronic sleep disruption associated with nightmares may affect the efficacy of first-line PTSD treatments, but targeted sleep treatments may accelerate recovery from PTSD. The field is ripe for prospective and longitudinal studies in high-risk groups to clarify how changes in sleep physiology and neurobiology contribute to increased risk of poor psychiatric outcomes.