Despite the abundance of information on cerebral malaria (CM), the pathogenesis of this disease is not completely understood. At present, two nonexclusive dominant hypotheses exist to explain how the neurological syndrome manifests: the sequestration (or mechanical) hypothesis and the inflammatory hypothesis. The sequestration hypothesis states that sequestration of Plasmodium falciparum-parasitized red blood cells (pRBCs) to brain capillary endothelia causes obstruction of capillary blood flow followed by brain tissue anoxia and coma. The inflammatory hypothesis postulates that P. falciparum infection releases toxic molecules in the circulation, inducing an imbalanced systemic inflammatory response that leads to coagulopathy, brain endothelial cell dysfunction, accumulation of leukocytes in the brain microcirculation, blood brain barrier (BBB) leakage, cerebral vasoconstriction, edema, and coma. However, both hypotheses, even when considered together, are not sufficient to fully explain the pathogenesis of CM. Here, we propose that the development of acute liver failure (ALF) together with BBB breakdown are the necessary and sufficient conditions for the genesis of CM. ALF is characterized by coagulopathy and hepatic encephalopathy (HE) in a patient without pre-existing liver disease. Signs of hepatic dysfunction have been shown to occur in 2.5–40% of CM patients. In addition, recent studies with murine models demonstrated that mice presenting experimental cerebral malaria (ECM) had hepatic damage and brain metabolic changes characteristic of HE. However, the occurrence of CM in patients with mild or without apparent hepatocellular liver damage and the presence of liver damage in non-CM murine models indicate that the development of ALF during malaria infection is not the single factor responsible for neuropathology. To solve this problem, we also propose that BBB breakdown contributes to the pathogenesis of CM and synergizes with hepatic failure to cause neurological signs and symptoms. BBB dysfunction would thus occur in CM by a mechanism similar to the one occurring in sepsis and is in agreement with the inflammatory hypothesis. Nevertheless, differently from in the inflammatory hypothesis, BBB leakage would facilitate the penetration of ammonia and other toxins into the brain parenchyma, but would not be sufficient to cause CM when occurring alone. We believe our hypothesis better explains the pathogenesis of CM, does not have problems to deal with the exception data not explained by the previous hypotheses, and reveals new targets for adjunctive therapy.