Writings about epileptic seizures date back 3,000 years to Mesopotamia and India.¹⁸ In 400 BC, Hippocrates attributed epileptic seizures to disorders of the brain, rather than to supernatural forces and, 500 years later, Galen distinguished idiopathic seizures, which originated within the brain itself, from sympathetic seizures, which had external causes.²² Nevertheless, throughout this time until the 19th century, very little was understood about the causes of epileptic seizures or, indeed, the variety of their manifestations. Until this time, for all intents and purposes, only generalized tonic–clonic convulsions were considered to be epileptic seizures. Esquirol, in 1815, however, discriminated between grand mal and petit mal seizures¹² and, in 1827, Bravais described focal motor seizures.⁵ Hughlings Jackson is credited with recognizing, in the 1860s, that a variety of focal symptomatology could be epileptic at a time when most workers considered only generalized convulsions—believed to originate in the medulla oblongata—to be manifestations of epilepsy.^11,21 Jackson also recognized that it was easier to understand the mechanisms of epilepsy by studying focal ictal events, and his clinical pathologic correlations correctly established the cerebral cortex as the site of origin of focal epileptic seizures. This work helped define the location and function within the human cortex and led directly to the development of resective surgical therapy for epilepsy.¹¹

With the advent of the electroencephalograph (EEG), characteristic electrographic ictal discharge patterns were identified to clearly distinguish among grand mal, petit mal, and psychomotor seizures.¹⁴ Because linked-ear references were used, however, all these patterns were believed to be generalized until so-called bipolar montages revealed that the location, and not the pattern, of the electrographic discharge differentiated various focal ictal events.¹⁶ Despite the description of hippocampal sclerosis in patients with epilepsy as early as 1826,⁴ and Hughlings Jackson’s observation that lesions in the hippocampal region occurred in patients with tasting movements and “dream state,”¹⁵ it wasn’t until the mid-1900s that epileptologists began to accept the fact that most “psychomotor” seizures originated in mesial temporal structures.¹⁷ This fact was further substantiated by the observation that psychomotor seizures could often be successfully treated by surgical resections of the temporal lobe.^2,20

During this period, worldwide, multiple terminologies developed that made sensible discussions and communication increasingly difficult. In 1970, the International League Against Epilepsy (ILAE) created the first classifications of epileptic seizures and epilepsies.¹³ They clearly distinguished ictal events from the disorders with which they were associated, and ictal and interictal EEG patterns, anatomic substrate, etiology, and age were included in this definition of seizures. Seizures were broadly divided into partial, beginning in a part of one hemisphere, and generalized, which were bilaterally symmetrical without local onset (Table 1). Partial seizures were further divided into those with elementary symptomatology, referring to signs and symptoms mediated by eloquent areas of neocortex, and complex symptomatology, referring to signs and symptoms mediated predominantly by mesial temporal limbic structures. The term “complex partial seizure,” therefore became synonymous with the older terms, “psychomotor seizure” and “temporal lobe seizure.” This seizure classification was criticized because, at the time, insufficient anatomic and pathophysiologic information was available to provide a basis for classification, and there was controversy about some of the generalized seizure types. These perceived problems were rectified in 1981, with the second ILAE Classification of Epileptic Seizures (Table 2).⁶

Despite persistent controversy, the 1981 Classification of Epileptic Seizures has gained general acceptance and is widely used. An ILAE Task Force has been reviewing this classification since 1997, but no new classification has yet been proposed.^7,9,10 Although flawed, it has been agreed that this useful construct should not be abandoned until a clearly better version can be devised.

The 1981 Classification of Epileptic Seizures was based on expert consensus, analyzing video-recorded seizures and considering their semiology as a sequence of ictal events developing in time. In addition to the clinical phenomenology, the ictal and interictal EEG was considered, but it was decided that there was still insufficient information about pathophysiologic mechanisms and anatomic substrates to warrant a more diagnostic approach.⁶ The partial/generalized dichotomy was preserved, but the distinction between simple and complex partial seizures was based entirely on impairment of consciousness. Thus, the original intent of the concept of complex symptomatology indicating behaviors mediated predominantly by limbic structures was lost. Neocortical seizures with impaired consciousness became complex partial seizures, whereas limbic seizures without impaired consciousness became simple partial seizures. This was, and continues to be, a source of controversy and confusion. On the other hand, the 1981 classification acknowledged the evolution of partial signs and symptoms and recognized that simple partial seizures could evolve into complex partial seizures, and that both could evolve into secondarily generalized seizures. This classification also differed from the 1970 classification in that absences were divided into typical and atypical, infantile spasms were deleted because they were considered to be a syndrome and not a seizure type, and akinetic seizures were deleted because their existence was questioned.