For more than a century, epilepsy was characterized as a chronic disease, with little chance of remission or cure. It was also considered a progressive disease in which seizures led to more seizures. Experimental work in animals provided additional support for the notion that seizures could beget seizures. However, the earliest clinical observations in humans were based on highly selected, largely refractory patients. Furthermore, the experimental work in animals bore little relation to naturally occurring seizures and epilepsy in humans. Evidence from multiple sources regarding the nature and natural history of seizures and epilepsy in humans has repeatedly demonstrated that in most cases of occurrence of seizures itself does not influence the long-term outcome of epilepsy. Consequently, interventions to prevent seizures early in the course of a seizures disorder do not alter the natural history of seizure disorders with respect to whether remission will occur in the long term. That outcome is largely predetermined by other factors, many of which are not currently amenable to intervention. In some rare syndromes, deterioration is progressive. In these instances, it is the underlying syndrome, not the seizures, that is primarily responsible for the deterioration. In addition, extremely prolonged seizures (which are rare) may also directly cause damage. These are exceptions rather than, as previously believed, the rule. The available human data strongly suggest that seizures do not beget seizures and that epilepsy in humans is usually not a progressive disorder.