The causes of epilepsy: Changing concepts of etiology of epilepsy over the past 150 years


Address correspondence to Professor Simon D. Shorvon, UCL Institute of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, Box 5, London WC1N 3BG, U.K. E-mail:


This paper provides a survey of the changing concepts of the etiology of epilepsy from 1860 to 2010, focusing on the first two 50-year periods and outlining more briefly major developments in the past 50 years. Among the concepts reviewed in the first 100 years are: the division between predisposing and exciting causes, idiopathic and genuine epilepsy, organic epilepsy, the concept of “cause” being equivalent to “causal mechanism,” Russell Reynolds etiological classification, the neurological taint and theories of degeneration, the self-perpetuating nature of seizures, reflex theories of etiology, autointoxication, heredity and eugenics, epilepsy due to brain disorders, the role of EEG and of hippocampal sclerosis, psychological theories of causation, and the multifactorial view of epilepsy etiology. In the past 50 years, the major advances in studying causation in epilepsy have been: clinical biochemistry, neuroimaging, molecular genetics, studies of mechanisms of epilepsy, better statistical methodologies and classification. A number of general observations can be made: the identification of “cause” is not as simple as it might at first appear; progress in the study of causation has been often erratic and travelled up many cul-de-sacs; theories of causation are heavily influenced by societal influences and fashion, and is also heavily dependent on applied methodologies; the recently explored possibility that the underlying inherited mechanisms of epilepsy are shared with other neuropsychiatric conditions is in effect a reinvention of the concept of the neurological trait, and this has ethical and social implications. Considering and classifying cause in terms of causal mechanism, as was suggested by Hughlings Jackson, is an ultimate goal.

The modern era of epileptology is often dated from the time of Hughlings Jackson (“the father of epilepsy”). As in other topics, concepts of “causation” of epilepsy have changed continuously since that time. In this review, I sketch in brief outline how these concepts have evolved since 1860, in three 50-year periods, focusing on the first two (1860–1910 and 1910–1960) and listing briefly developments in the period 1960–2010. I will outline some major categories of cause as these were then perceived, and end with a number of general observations on the nature of this evolution.

Concepts of Etiology 1860–1910

This was a time when supernatural explanations of the cause of epilepsy were finally put to rest (although their force is evident from the fact that the “demonic controversy” occupies the first eight pages of the chapter on etiology in Sieveking’s book on epilepsy—see below), and a number of interesting hereditarian and biologic concepts of etiology became widely held.

The division of causation into “predisposing” and “exciting” causes

Almost all authors of the period considered epileptic seizures to have two distinct types of causes: an innate predisposition (a predisposing cause) and a precipitating (exciting) cause. This dichotomy has been arguably unjustly neglected in recent times. Although different authors used the terms differently, for instance defining “predisposing” as innate/remote/hidden/underlying and exciting as immediate/provoking/determining/proximate, the essential belief was that there was a predisposition that was inherited, and that in addition there were precipitating causes of varying type. The main source of disagreement was what was to be included in the inherited category (see below).

Sir Edward Henry Sieveking’s Epilepsy and Epileptiform Seizures (1858) provides a snapshot of contemporary thought. He wrote that “no first outbreak of epilepsy ever occurs without the concurrence of the predisposing and the exciting influence.” He considered that the predisposing causes were largely inherited and formed the epileptic diathesis, which he defined rather vividly as follows: “Diathesis may be compared to combustible material of greater or less inflammability, which differs in the facility with which it will take fire, but will infallibly do so if a flame of sufficient intensity is brought into contact with it.” Sieveking considered both constipation and sexual disturbances (particularly masturbation) to be important additional predisposing causes (the latter to be both predisposing and exciting). In his own series of 104 cases, he found exciting causes in 56 cases, 15 of which were psychical (fright, mental work, anxiety) and 41 physical (otorrhea, scarlet fever, dentition, operation, gastric and intestinal derangement, head injury, pregnancy, lead poisoning, puberty, scurvy, uterine derangement, masturbation and venereal excess, intracranial disease (only one case), psoriasis, and spinal rheumatism).

Sir William Gowers in 1881 published his remarkable book, Epilepsy and Other Chronic Convulsive Diseases (Gowers, 1881). The first chapter was devoted to the causes of idiopathic epilepsy. He divided these into two types: predisposing causes and exciting causes, the former being “remote” and the latter “immediate” and Table 1 summarizes his explanation of causes from his own case material. Of his 1665 cases, Gowers considered that an exciting cause could be found in 42% (696 cases), and that these exciting causes were of secondary importance to the predisposing (inherited) causes.

Table 1.   The causes of epilepsy listed by Gowers (1881)
  1. aAmong the acute diseases, Gowers singled out scarlet fever, which he considered especially neurotoxic.

  2. bFright was the most common exciting cause in Gowers’ opinion.

  3. cEmotion was felt to be most important in young adult females.

  4. dGowers was doubtful about the relevance of masturbation.

  5. From Gowers, 1881.

Predisposing causesHeredity (by far the most important: There are few diseases in the production of which inheritance has greater influence)
Other conditions such as syphilis (heterogenous heredity)
Other factors—age, gender
Exciting causesPhysical: Difficulty with labor, birth trauma, febrile convulsions, dentitional (teething) fits usually due to rickets, organic lesions of the brain, acute diseases (measles, scarlet fever,a typhus, typhoid, rheumatic fever, influenza, diarrhea), digestive derangement, intestinal worms, reflex seizures, trauma, asphyxia, lead poisoning, alcohol, renal disease, anesthetics, disturbed menstruation, pregnancy, syphilis.
Emotional: prolonged mental anxiety, fright,b excitement, emotionc,d

In his classic text on epilepsy at the end of this 50-year period, Turner (1907) also considered that the majority of cases were due to an inherited “predisposition” and that an “exciting cause” was present in a minority of cases (Table 2). In Turner’s personal case series, psychical causes occurred in 4.1%, head trauma in 7.2%, acute infective causes in 5.6%, syphilis in 0.4%, and “cerebral birth palsy” in 5.9%.

Table 2.   The causes of epilepsy listed by Turner (1907)
  1. Derived from Turner, 1907.

Predisposing causesHeredity—as part of the neuropathic trait (and associated with other neuropsychiatric disorders and the stigmata of degeneration)
Exciting (determining) causesPhysiologic causes—puberty, catamenia, pregnancy, puerperium, lack of sleep, ingestion of certain foods
Psychical causes—shock, emotional excitement, fear, anxiety, overwork
Pathologic causes—exanthemata and acute infective diseases, organic diseases of the brain (e.g., tumors, syphilis, infantile hemiplegia), trauma to the head, reflex epilepsies due to morbid conditions of various other organs (eyes, ears, nose, stomach, genital organs), auto-infection from the alimentary canal, intoxications, disorders of bodily metabolism, and cerebral palsy

Other authors of the period followed a similar line. Spratling (1904) even attempted to explain the contributions of the exciting and predisposing causes mathematically, writing that: if it took 100 points to induce a seizure in an individual, a predisposition could contribute 60 points and an exciting cause 40 points, whereas if the predisposition contributed only 40 points, it would require an exciting cause to have 60 points in order to reach the “seizure point.”

Idiopathic (genuine) epilepsy and organic epilepsy

Another major division of causation was into “idiopathic” and “organic” epilepsy. Throughout the nineteenth century, most interest was focused on idiopathic epilepsy, sometimes referred to as “genuine epilepsy.” Indeed the seizures arising from organic lesions (roughly approximating to our concept of symptomatic epilepsy) were usually considered not epilepsy at all, but rather a manifestation or symptom of the lesion (see Reynolds, 1861 for instance).

It was universally believed that the main causal agent of idiopathic epilepsy was heredity, but that other conditions, including those considered inherited, also contributed and the term includes the developmental and inherited metabolic disorders, and a variety of other conditions now known to be acquired.

Organic epilepsy was the term used to signify the epilepsies in which there was a clear identifiable cause, either systemic or neurologic. These conditions were of little interest to the physicians of the period. For instance, Gowers defines organic epilepsy as that due to the “many organic diseases of the brain,” which he does not go on to list. Jackson too, although recognizing that epilepsy could be due to tumors or syphilis, for instance, placed far more emphasis on the physiologic predisposition (see below), and his interest in organic lesions was largely from the anatomic point of view. This lack of focus on the organic causes is no doubt in part due to the lack of investigatory tools (only postmortem and surgical neuropathology provided any help here) and also the lack of a systematic classification of neurological disease, and particularly degenerative or developmental conditions. However, by the end of the century, neuropathologic studies had identified a number of organic disorders that were widely accepted to cause “organic epilepsy,” including porencephaly, heterotopy, microcephaly, brain hypertrophy, asphyxia at birth, infantile hemiplegia and cerebral palsy, brain tumors, cerebral trauma causing a cicatrix, cerebral infection such as abscess, and degenerative conditions resulting in cerebral softening.

Russell Reynolds and the first etiologic classification of epilepsy

Sir John Russell Reynolds published his book Epilepsy: Its symptoms, Treatment and Relation to Other Chronic Convulsive Diseases (1861) shortly after Sieveking’s. In this book, cause was carefully considered, and Reynolds also created a classification of epilepsy based on etiology. This classification is, in my mind, a remarkable achievement, and in some ways remains better than later schemes. He classified epilepsy into four categories (Table 3).

Table 3.   The classification of epilepsy devised by Reynolds (1861)
  1. Reynolds considers a disease to be the sum-total of modifications in both function and structure, and the symptoms to be the appreciable changes of these modifications. He considered that the proximate cause of epilepsy was always the same (nutritive abnormalities of the nerve centers). Their essence was abnormal increase in the molecular changes of the nervous centers, and the remote causes induce this increase. He considered convulsive diseases (epilepsy) and convulsions (seizures) to be synonymous.

  2. Derived from Reynolds, 1861.

Idiopathic convulsionsLargely inherited and caused by a disorder of brain nutrition. It was an internal cause—a morbus per se. In this condition, the only deviation from the healthy structure is the nutrition change (the proximate cause of all convulsions)
Eccentric convulsions (sympathetic convulsions)Epilepsy due to some systemic disturbance, which affects the nutritional state of the brain, and which when cured will result in the cessation of seizures. “Eccentric convulsions” can be the result of irritation or can have a “reflex” basis. Causes included teething, worms, and poisons, and when the cause was removed the epilepsy could cease
Diathetic epilepsy (cachectic convulsions)Epilepsy in which the convulsions are due primarily to cachexia or toxemia, and in which the nervous system is “involved in that general nutrition-change which is the essential element of the cachexia itself” and have their basis in a general not specific remote causes. The causes of cachexia he listed included: tuberculosis, scrophula, rickets, syphilis, urinemia, anemia, pyemia, alcoholism, lead-poisoning, typhus, variola, and other exanthemata and diseases that alter nutrition such as pneumonia, carditis and pericarditis, puberty, pregnancy, dentition
Symptomatic epilepsy (form centric diseases)Epilepsy in which convulsions are due to “more or less contiguous structural disease of the brain. Thus, an intracranial tumour, a chronic inflammatory condition of the meninges, softening or disintegration of the brain substance, or any other structural change in the nervous centres… may set up that peculiar interstitial or molecular change which is the immediate cause of convulsion”

“Idiopathic epilepsy”—genuine epilepsy—was he considered, largely inherited and in which there is a change in brain nutrition. He found an inherited tendency in 31% of his cases, a value that included other inherited conditions of the neuropathic taint with epilepsy, including mania, idiocy, paralysis and insanity (see below). “Eccentric epilepsy” was a category similar to the current concept of provoked epilepsy. “Diathetic epilepsy” seems to be a category not previously described in which the causative nutritive changes were not confined to the brain but were part of the general systemic disturbance caused by toxemia or cachexia. “Symptomatic epilepsy” was a term reserved for epilepsies in which there was a contiguous structural brain disease, which is similar but narrower than the usage today. In passing, Reynolds also noted that an “emphasis on masturbation was mistaken,” and that the ideas of “some mysterious entity taking possession of the body” was a theory “long since passed.”

Cause as “causal mechanism”—Reynolds and Jackson

Reynolds also divided cause into proximate and remote categories. The proximate cause is the same in all cases—nutritive abnormalities of the nervous system; in other words, his concept of cause was what today we would consider to be the causal mechanism (and his concept “nutritive” is similar to our use of the terms excitatory or inhibitory). He recognized that the influence of remote causes varied from patient to patient, but that they were all mediated though this defect of nutrition. He also held that the different types of seizure were caused by differing anatomic locations of the abnormal nervous centers, and in all this he anticipated Jackson.

John Hughlings Jackson is generally agreed to have laid the foundations for much of modern epilepsy studies. He considered “a convulsion is but a symptom, and implies only that there is an occasional, an excessive, and a disorderly discharge of nerve tissue on muscles” (1870). In his Lumleian lectures of 1890 (Jackson), he defined nervous discharge as “the physiological fulminate” like the gunpowder in a cannon, and just as gunpowder can store energy that is liberated when firing the gun, so the energy stored in nerve cells could be explosively liberated in an epileptic discharge. His views on causation, however, were heavily influenced by those of Reynolds. He held that the abnormal levels of stored energy were due to deranged “nutrition,” and that this in turn was due usually to congestion of small blood vessels. He equated “cause” with “causal mechanism” and was in general not particularly interested in the question of etiology in the sense usual today; his focus was on physiology. In 1874, in what I consider a seminal article, he wrote that: “The confusion of two things physiology and pathology under one (pathology) leads to confusion in considering “causes.” Therefore, for example, we hear it epigrammatically said that chorea is “only a symptom” and may depend on many causes. This is possibly true of pathologic causation; in other words it may be granted that various abnormal nutritive processes may lead to that functional change in gray matter which, when established, admits occasional excessive discharge. But physiologically, that is to say, from the point of view of function, there is but one cause of chorea—viz. instability of nerve tissue. Similarly in any epilepsy, there is but “one cause’ physiologically speaking—viz. the instability of the gray matter, but an unknown number of causes if we mean pathologic processes leading to that instability” (published in the Medical Press and Circular; see Taylor, 1930).

He did recognize that there were many possible contributing factors that could result in the vascular changes and mentioned tubercle, cicatrix, tumor, syphiloma, or hemorrhagic or ischemic stroke. He also realized that there was often no visible cause.

The neurologic trait and the concepts of degeneration

To understand what was meant by “heredity” in this period, the central importance of two related concepts must be appreciated: that of “degeneration” (dégénérescence) and that of the “neurologic taint.” In the early and mid-nineteenth century, particularly among French writers, the concept of degeneration replaced that of the supernatural as a main focus of interest in the causation of epilepsy (the classic works on this topic were by Morel, 1857, 1860). The concept of degeneration in neuropsychiatric diseases was part of a wider public concerns about social disintegration, the collapsing state of European cultural identify, and fears that rapid population growth and urbanization among the lower classes would sap national intelligence and morality. These fears were also reflected in the artistic movements of the time, criminality, and social science.

Linked to ideas of dégénérescence was the concept that there existed a neuropathic taint (also known by various other terms including neurological taint, neuropathic trait, neuropathic predisposition). According to this theory, a range of conditions including epilepsy were inherited together. Different authorities incorporated different conditions, but most included, with epilepsy, neuropsychiatric disorders of various types such as insanity; mental retardation and general paralysis of the insane; moral degeneration such as was found in alcoholics or the criminal; and sexual degeneration evinced by masturbation, perversion, and sexual excess. This concept was widely accepted among neurologists, although a few, notably Jackson, opposed the idea of mixing up conditions with “no evident pathological connection.”

The two concepts were synthesized into a single theory in relation to neurological disease. Thus, according to Morel, this inherited tendency resulted in a progressive deterioration (degeneration) physically, mentally, and morally, over generations, and this tendency becomes progressively more severe, eventually resulting in the extinction of the line. Moreau (1859) and Falret (1864) introduced the category of the “neuropathic family,” and Féré divided the “neuropathic family” into a psychopathologic arm, which included epilepsy and the major psychiatric disorders, and a neuropathologic arm which included chorea, migraine, and Parkinson’s disease (Féré, 1884). According to these theories, the endowment might for instance cause mild hysteria in one generation, then a more serious epilepsy in the next, and dementia or idiocy in the next.

The notion of degeneration was also linked in this period to the concept of atavism, which had biologic plausibility given the theory of recapitulation popularized by Haeckel in 1866 (“Ontogeny recapitulates phylogeny”). Degeneration was thought to bring out atavistic characteristics (physical, behavioral, and mental), which were, therefore, the signs of the degenerative tendency. Epilepsy was seen as one symptom of degeneration, atavistic in nature, in the progressive downward degenerative spiral.

By the end of the nineteenth century, almost all writings on the inheritance of epilepsy accepted this concept. Echeverria (1873) reported an inheritability rate of epilepsy in 25%, Dejerine (1886) of 66.8%, Binswanger (1899) of 36.3%, and Spratling (1904) of 56.0%.

In parallel to the studies of mental degeneration were investigations of physical stigmata, and particularly physiognomy, and these reached their zenith in the works of Cesare Lombroso. Lombroso’s enduring works include L’uomo Delinquente (Criminal Man; published in 5 editions between 1876 and 1896/7) and Criminal Woman (La donna delinquente e la prostituta et la donna normale (Lombroso & Ferrero, 1893), which are packed with numerous measurements and statistical tables. Lombroso developed a theory of criminality in which he postulated that two thirds of dangerous criminals inherited their criminal traits, and were “born criminals.” These persons also possessed “anomalies” (physical and psychological) resembling the traits of primitive man and animals (and even plants), and were atavistic throwbacks to a primitive stage in human evolution. In the fourth edition of L’uomo Delinquente he expressed the view that epilepsy was an atavistic characteristic and a fundamental component of the criminal type. He supported this by showing that criminals and epileptics shared the same physiognomy, physical and psychological features, and moral deficiency (Table 4). Overall, he wrote that 26.9% of all epileptic men and 25% of all epileptic women have a “full criminal type” from the physiognomic point of view. Lombroso’s theories of criminality had a profound influence on social theory for at least the next half century, and his lasting legacies are the medicalization of aberrant behaviours and the demonstration that social behavior had a biologic basis.

Table 4.   The inherited anomalies shared by epileptics and criminals listed by Lombroso (1876–1896/7)
  1. Derived from editions 4 and 5 of L’uomo Delinquente (Lombroso, 1876–1896/7), with percentages quoted by Lombroso from his own work or that of Cividalli, Tonnini, and Bianchi.

SkullAbnormally large, microcephaly, asymmetric (12–37%), sclerosis, med. occ. fossetta, abnormal indice, large orbital arches, low sloping forehead, wormian bones, simple cranial sutures
FaceOverdeveloped jaw, jutting cheekbones, large jug ears, facial asymmetry, strabismus, virility (in women), anomalous teeth
BrainAnomalous convolutions, low weight, hypertrophied cerebellum, symptoms of meningitis
BodyAsymmetric torso, prehensile feet, hernia
SkinWrinkles, beardlessness, olive skin, tattoos, delayed gray hair/balding, dark and curly hair
Motor anomaliesLeft-handedness (10%), abnormal reflexes, heightened agility (16%)
Sensory anomaliesTactile insensitivity (81%), insensitivity to pain, overly acute eyesight, dullness of hearing, taste, and smell
Psychological anomalies (% in epileptics)Limited intelligence (30–69%), weak memory (14–91%), hallucinations (20–41%), superstitious, blunted emotions, love of animals, absence of remorse, impulsivity (2–50%), cannibalism and ferocity, pederasty (2–39%), masturbation (21–67%), perversity (15–57%), vanity, sloth, passion for gaming, mania/paranoia, delirium, dizziness, delusions of grandeur (1–3%), irascibility (30–100%), lying (7–100%), theft (4–75%), religious delusions (14–100%)
CausesHeredity (of alcoholism, insanity, epilepsy, old parents), alcoholism

This theory was very influential, and Turner (1907) for instance found also signs or“stigmata”of degeneration in 66.5% of his epileptic patients that included: facial deformities (inequalities of the two sides of the face, irregularities of the nose, prognathism or arrested development of the lower jaw, deformities of the hard palate, dental abnormalities, deformities of the ears, deformities of the iris, abnormal arms, mental aberrations, stammering, and astigmatism).

The self-perpetuating nature of epilepsy: seizures beget seizures

Gowers made a unique contribution to “cause” in epilepsy with his theory that “the malady is self-perpetuating; when one attack has occurred, whether as the result of an immediate excitant or not, others follow either without any immediate cause, or after some very trifling disturbance…. The search for the causes of epilepsy must thus be chiefly an investigation into the conditions with precede the occurrence of the first fit.” This concept (“seizures beget seizures”) was also widely taken up (see for instance Turner, 1907), and epilepsy was seen as a process, and the cause of the seizures was the maturation of this process. This topic continues to excite debate.

Reflex theories of causation of epilepsy

The role of reflexes in the causation of seizures was also frequently postulated in this period—following the influential work of Marshall Hall and later Brown Séquard. These were considered by Jackson’s time to be a frequent mechanism of the “exciting causes.” Jackson wrote extensively on this topic. He considered that “irritation” (of various types) could trigger seizures by draining the cerebral centers of their energy. The irritation could arise in the periphery, ears, eyes, teeth, digestive tract, or sexual organs. These conditions were sometimes classified as “sympathetic epilepsies”—being due not to a primary disorder of the brain, but rather to a systemic irritation that triggered a seizure.

In the latter part of the century, a particular and common reflex cause was considered to be eyestrain, particularly in the American literature. Treatment was with eyeglasses and tenotomies (see Friedlander, 2001). Others identified reflexes induced by pain in a limb, by genital stimulation, and by pathologies in the ear or nose. Gowers identified “reflex causes” mediated by irritation of peripheral nerves, visceral or external; pain; digestive derangement; or an “anomalous or indigestible meal.”Turner (1907) summarized the various therapies current at the time, which included surgical excision of traumatic lesions of the peripheral nerves, removal of a tight prepuce in boys, treatment of coexistent diseases of the ears or nasopharynx, and removal of foreign bodies, adenoid growths and polypi to remove the reflex stimuli, and correction of errors of refraction.

After this, the reflex theories of epilepsy fell out of fashion until Pavlov’s demonstration of conditioned reflexes. Pavlov’s theories were favored as the pathogenic mechanism, for instance, in the influential paper on musicogenic epilepsy by MacDonald Critchley, 1937. As time passed, the term reflex epilepsy began to refer only to very specific sensory precipitants, to rare conditions, and acquired a meaning not dissimilar to that of today.

Concepts of Etiology 1910–1960

The great interest in epilepsy, evident in the previous 50 years, diminished in the first decades of the twentieth century. The World Wars diverted attention to other conditions, the focus of advanced neurology moved elsewhere, and there were no great theoreticians of epilepsy in this period. In relation to etiology, two main themes dominated the period: (1) the role of heredity and its disastrous translation into eugenics, and (2) the importance of organic brain disease, reflecting the discovery of radiology and the advances in neurosurgery.

Heredity and eugenics

Eugenics, a term coined by Galton in 1883, began to play a major role in epileptology in the early twentieth century. Many of the early International League Against Epilepsy (ILAE) leaders became active eugenicists—such as Weeks, Munson, Schou, and Lennox—and epilepsy, insanity, and mental deficiency became linked under the microscope of the eugenics movement. Elaborate family trees (pedigrees) were drawn up showing the linked inheritance of epilepsy and the other neuropsychiatric disorders (a landmark paper was that of Davenport & Weeks, 1911). According to eugenic theory, epilepsy was inherited by Mendelian mechanisms, usually in a recessive manner, and by either positive or negative eugenic practices could potentially be removed (or at least minimized) from a population. This resulted in enforced sterilization of individuals with epilepsy, and then in Nazi Germany was used as the theoretical underpinning of the mass murder of the unfit (including people with epilepsy). In the early 1940s, Action T4 a program of killings (“euthanasia”) of the handicapped was inaugurated, based on eugenic concepts, and it has been estimated that between 200,000 and 250,000 mentally and physically handicapped persons were murdered from 1939 to 1945 under this and other “euthanasia” programs. How many persons with epilepsy perished is not known.

Epilepsy as a result of cerebral lesions

This also was the period when neurosurgical pathology was being systematized and when imaging was beginning to visualize the brain in vivo. X-Ray was applied to epilepsy in the first decade of the twentieth century, air encephalography in 1919, and contrast ventriculography in 1925. In parallel, neurosurgery began to expand, based now not only on clinical semiology but also the results of these investigations. This led to a focus on organic theories of causation.

In 1932, Walter Dandy published a landmark paper from the surgical perspective (Dandy, 1932). Dandy took a very surgical viewpoint: “Epilepsy is always regarded as an idiopathic disease. The theories of its causation are indeed so numerous as to reflect seriously upon any exclusive stand concerning its etiology or pathology. However, the writer is confident that there is now assembled from experimental, pathologic, clinical and surgical studies a sufficient number of unquestioned facts to place epilepsy unequivocally upon a pathologic instead of idiopathic basis...the fundamental conception that in every case of epilepsy there is a lesion of the brain can no longer admit of doubt.” He recognized 17 categories of “lesions causing epilepsy” (Table 5), which seem by today’s standards a rather curious mélange, but no doubt reflected advanced neurosurgical opinion of his time.

Table 5.   The 17 categories of brain lesions causing epilepsy listed by Dandy (1932)
  1. Derived from Dandy, 1932.

Congenital malformation and maldevelopment, either general or focal
Syphilis with or without demonstrable gummata or vascular occlusions
Areas of cerebral degeneration and calcification
Depressed fractures
Foreign bodies
Injuries from trauma at birth or subsequently (focal or general)
Connective tissue formation after trauma
Atrophy of the brain after trauma
Thrombosis and embolism
Cerebral arteriosclerosis
Sequelae of obscure inflammatory processes including encephalitis

The emphasis on the organic basis of epilepsy was systematized in the classic neurologic text of the mid-1930s, the three-volume textbook of neurology by SA Kinnear Wilson (published posthumously in 1940; Fig. 8) (Wilson, 1940). Seventy-five pages are devoted to The Epilepsies. Wilson takes the same position as Dandy regarding the organic basis of epilepsy: “Current opinion is…veering round to the view that all epilepsies are symptomatic, inclusive of the variety [idiopathic epilepsy] whose basis still elude search… the cause will eventually be revealed.” Wilson makes the first reference I can find to the term “cryptogenic,” which he feels is preferable to refer to epilepsies of unknown cause.

Wilson thought that inherited epilepsy was uncommon. He cited Myerson (1932), who found a family history of epilepsy among 1,500 inmates of a hospital for epileptics in only 4 families (11 persons), and at the same institution 138 marriages of epileptics resulted in 553 offspring among there was a history of fits in only 10% or 1.8%. As Wilson put it: “The influence of the factor [heredity] is persistently overvalued; in only about one-fifth of my material has [this] seemed to be operative.” Head trauma is extensively discussed, and Wilson referred a 1920s survey of the United Kingdom Ministry of Pension of 18,000 persons with gunshot wounds of the head, finding an incidence of posttraumatic epilepsy of only 4.5%. In civilian cases, he cites figures of between 3% and 21%. He dismisses the importance of “bad teeth, septic tonsils, nasal polypi, refractive errors, phimosis, intestinal worm, and what not,” but mentions seizures during anesthesia (“ether convulsions”), pleural epilepsy (now we would consider this vasovagal), and cysticercosis (an imperial disease as he noted that “although infestation may occur in England, the majority of contracted the disease in Egypt, India, or the Malay states”). He discusses precipitating factors such as cosmic influences (dismissed), sleep, menstrual epilepsy and pregnancy (including eclampsia), and psychical states. Wilson’s chapter on epilepsy is a masterpiece, and in relation to etiology we see a truly modern view emerging for the first time. Others followed this lead. FMR Walshe, editor of Epilepsia 1959–1961, in his standard textbook (1949) wrote: “ is better to speak of ‘the epilepsies’ according to the various known exciting factors than to keep the category of idiopathic epilepsy. Nevertheless, from the practical point of view of diagnosis and treatment the two categories of ‘idiopathic’ and ‘symptomatic’ epilepsy remain useful.... It has always been regarded as a heritable condition, though it obeys no known laws of inheritance, and it is probably that what is inherited—if anything be—is an instability of function in the cells of the cerebral cortex.” As Walshe put it, “nothing is known” of the “’exciting’ causes of idiopathic epilepsy,” and “certainly the heritable causes of epilepsy have been greatly exaggerated in the past, and in consequence severe restrictions upon the liberty of conduct of the epileptic have been imposed in the guise of medical advice” (Walshe, 1949).

This was also a period when the inherited metabolic diseases and chromosomal abnormalities began to be elucidated. Down syndrome was found to be due to chromosomal trisomy in 1959, and the biochemistry of phenylketonuria was identified by Ivar Asbjørn Følling in 1934 (see Christ, 2003).

Psychological theories of causation of epilepsy

Among patients and their physicians it had been long held that epilepsy may be due to psychological causes—notably stress, psychic trauma, and shock. Exactly how “stress” causes epilepsy though has remained obscure, and many theories, some fanciful, have been propounded. In the early twentieth century, psychoanalytical explanations of epilepsy began to be formulated. These theories focused on personality development and predisposition. Clark (1925, 1933) was a leading proponent and published extensively on this topic. He felt that the predisposition to epilepsy was caused by egocentricity, supersensitiveness, emotional poverty, and an inherent defect of adaptability to normal life. The precipitating factors include a loss of spontaneous interest and regression to day-dreaming, lethargy, and somnolence. The epileptic attack occurs when tension becomes very severe and may be viewed as a reaction to escape the intolerable irritation and a regression to a primitive mentality comparable to that of infancy or intrauterine life (see Read, 1920). Another influential psychiatrist, Jelliffe (1918), also wrote on epileptic attacks, holding that these sometimes represented a direct flight into infantile sexuality. Sexual disturbances were also widely held to be the immediate cause of an epileptic seizure, with interesting similarities to what was believed in the mid-nineteenth century and earlier (cf. the ancient proverb attributed to Galen “coitus brevis epilepsia est”). Reich (1937) considered that seizures were repressed libido, and that the seizures were coital equivalents, and similar theories were held widely in psychoanalytical circles (see Glover, 1949 for a comprehensive listing of these). The psychoanalytical theories of causation of epilepsy were widely held right up to the 1940s, although often in the teeth of opposition from the more neurologically inclined.

Autointoxication causing epilepsy

By 1900, another widely held explanatory model of causation was gaining credence: the theory of “autointoxication.” According to this theory, epileptic seizures were caused by toxins produced within the person’s own body (not dissimilar to Galen’s theory of humours and Reynold’s “eccentric epilepsy”). Most believed that these toxins arose in the bowel, either through fermentation or from bacteria. This was backed up by reports for instance of sigmoidoscopy showing “acute angulation of the sigmoid colon” and “impaction of the sigmoid of an inordinate character” (Axtell, 1910), and by radiologic examination with Bismuth showing “coloptosis” (Clark & Busby, 1913; cited by Friedlander, 2001) as well as experiments that included the production of convulsions in rabbits or guinea pigs by injections of blood from epileptic patients. Among the toxins actually responsible, much was written about uric acid.

One illustrative, if extreme, enthusiast was the psychiatrist Dr Henry Cotton, superintendent of Trenton State Hospital, a residential institution for mentally handicapped epileptic or psychotic patients. He decided to eradicate these pockets of infection by surgical means, and after 1916, a program of surgical therapy of extraordinary proportions began, largely for psychotic but also for epileptic patients. In one 12-month period, 6,472 dental extractions were performed, as well as 542 tonsillectomies and 79 colon resections. Between 1918 and 1925, 2,186 major operations were carried out. Dr. Cotton became a national figure, lauded publicly for his remarkable cure rates (85% claimed for psychosis, for instance), and in 1921, the President of the American Medical Association declared that Trenton State was one of the country’s “great institutions… a monument to the most advanced civilisation” (these details are taken from Nevins, 2009).

Electroencephalography (EEG), hippocampal sclerosis, and temporal lobe epilepsy

The discovery of “etiology” in epilepsy is heavily dependent on new technology, and in the early postwar years, the major methodologic advance in epilepsy was of course EEG (Gibbs et al., 1935). EEG boosted the study of inherited epilepsy, for instance by Lennox (see below), but it was quickly recognized that the EEG changes in structural pathologies such as brain tumors were disappointingly nonspecific. EEG did, however, prove invaluable in defining temporal lobe epilepsy and the etiologic importance of hippocampal sclerosis. Gastaut (1953) recognized that hippocampal sclerosis (incisural sclerosis) was often the causal lesion of temporal lobe epilepsy, contrary to the classic opinion of Spielmeyer, still widely held at that time, that the lesion was consequential not causal (see Shorvon, 2006). Gastaut did not, however, identify the association with febrile seizures, and believed that trauma was the most common cause of temporal lobe epilepsy, resulting in contusional damage to the brain as it was compressed against the sphenoid bone or the free edge of the tentorium. He viewed encephalitis as the second most frequent cause (20–25% of cases). The third cause (5% of cases) was obstetrical injury, which in Gastaut’s view resulted in herniation of the temporal lobe over the tentorial edge, causing vascular compression of the anterior choroidal (and other) arteries.

Etiology of epilepsy in 1960: William Lennox and the concept of multifactorial etiology

William Lennox published his classic book Epilepsy and Related Disorders in 1960 (Lennox & Lennox, 1960), and this provides a summary of advanced contemporary thought on the topic of epilepsy etiology. He considered epilepsy to be due to a combination of: (1) genetic (essential) causes; (2) acquired causes; and (3) precipitating causes; and of these the genetic causes are the most important. As Lennox put it “…we personally believe that nature outnumbers nurture. The relative importance of the latter is decreasing because of better control of preventable conditions.”

He followed his nineteenth century predecessors in proposing that in many cases (50% in Lennox’ view) there were “predisposing” (genetic) and “precipitating” (acquired) causes, and indeed draws an analogy with fire in exactly the same way as Sieveking and Gowers. However, it is his “analogies of the reservoir or river” that best encapsulates his thought (these analogies are so vivid that they are both worth illustrating here; Figs. 1 and 2).

Figure 1.

Lennox’s concept of the epileptic threshold: “Causes may be represented as the sources of a reservoir. At the bottom is the already present volume of water, which represents the person’s predisposition, a fundamental cause. But the reservoir is supplied also by streams which represent the contributory conditions, such as lesions of the brain acquired since conception, certain disorders of bodily function and emotional disturbances. Periodic overflow of the bank represents a seizure.” (from Lennox & Lennox, 1960 vol 2, p. 820).

Figure 2.

Lennox’s concept of the etiology of epilepsy: “The genetic watershed is represented…as three generations: parents, grandparents, and great-grandparents [e.g., at A, a paternal grandmother has epilepsy]. A confluence of transmitted traits follows into (and through) the patient…. In addition to these branching streams, there is [an] independent stream which rises in a lake (the uterus). The outlet is the birth canal and below that are contributing streams: infections [e.g., at B, a viral encephalitis], brain trauma from diverse sources, brain tumor, and circulatory disorders. This side stream enters the main stream at the patient level and combines with the genetic influences which had travelled through three generations to make him epileptic. There is then a third stream which enters below the confluence of the two main streams. This represents transient conditions which may precipitate certain seizures in a person already epileptic, or “all set” to be. This evoking circumstance may be physiologic (say at C, hypoglycaemia) or emotional (say at D, a broken wedding-engagement).” (from Lennox & Lennox, 1960. vol 1, p. 529).

According to Lennox, about 20% of epilepsies are purely genetic, 20% are purely acquired, and about 50% are a mixture of both (leaving 10% in which the cause is quite unknown). The genetic epilepsies were predominant, and in his words: “Evidence for this [genetic causation] is relatively simple and convincing: namely, some blood relative who has been subject to seizures which were not the consequence of some acquired brain injury.” Lennox carried out research based on family trees, much as his eugenic predecessors had done, and particularly on twins, and he used EEG as a biologic marker.

The acquired causes, less important in Lennox’s view than the genetic ones, included a variety of conditions, listed in Table 6. In the category of acquired epilepsy (organic epilepsy), Lennox included developmental defects, but realized that these had strong genetic links (“it would be the channel that joins the genetic river system to the uterine lake”). Lennox used the term “hereditary organic epilepsy” to describe inherited genetic conditions that result in structural disorders of the brain. Included here were: Mongolism, tuberose sclerosis (Epiloia), and various malformations of development.

Table 6.   The acquired causes of epilepsy listed by Lennox and Lennox, 1960
  1. In his series of 927 patients with acquired epilepsy, Lennox found the following percentage of causes: paranatal 38%, postnatal 25.2%, infections 19.5%, brain tumors 6.7%, circulatory 5.3% and other causes 5.3%.

  2. Derived from Lennox & Lennox, 1960.

Congenital abnormalities—developmental defects
 Hereditary organic epilepsy
 Mongolism (fetalism)
 Hereditary (Huntington’s) chorea
 Tuberose sclerosis
 Amaurotic familial idiocy
 Various other pathologic states
Intrauterine misfortunes
 Embryonic versus fetal timing of insult
 Placental transmission (rubella, toxoplasmosis, erythroblastosis fetalis)
 Intracranial hemangioma
 The cerebral palsies
Paranatal epilepsy
 Difficulties of parturition
Postnatal epilepsy
 Infection-derived epilepsy (viral, bacterial, bacillary, spirochetal, protozoan, metazoan)
 Posttraumatic epilepsy (including lobotomy, the wounds of war)
 Brain tumors
 Defects of cerebral circulation
 Toxins and intoxications (alcohol, ergot, chemical, lead, radiation)

Lennox also realized that there were other considerations. He cites the use of the term “parahereditary” by continental authors for conditions that were not strictly speaking genetic, but that could alter the sperm or the ovum, and here include alcohol, syphilis, infections, and intoxications.

Etiology 1960–2010

Lennox’s writings mark the end of an era. During the next 50-year period there have been enormous advances in the understanding of the etiology of epilepsy. The underlying etiologies of many cases that would previously be considered “cryptogenic” are now identified. Some advances have been made in understanding the genetic and developmental causes of epilepsy, but these are waters that remain still largely uncharted. The advances made in this period can be only briefly summarized here under six headings:

  • 1 Clinical and molecular biochemistry: The mechanisms of many of the inherited and acquired metabolic disorders are now identified.
  • 2 Neuroimaging: The invention and application of computed tomography (CT) and particularly magnetic resonance imaging (MRI) has of course transformed our ability to visualize structurally or anatomically abnormal tissue (Shorvon et al., 1994). CT allowed the direct visualization of many tumoral and vascular causes of epilepsy, as well as hydrocephalus and some congenital lesions, and MRI of hippocampal sclerosis, the cortical dysgeneses, and subtle small abnormalities. Both technologies have rendered EEG, a technique for studying function, redundant as a tool for establishing structural etiologies.
  • 3 Molecular genetics: The enormous advances in molecular genetics were ushered in by the discovery of the double helix structure of DNA in 1953 and this culminated in the publication of the structure of the human genome (draft publication in 2001). Landmarks developments include the methodology of linkage studies, the development of sequencing methods, karyotype analysis, methods of DNA amplification and analysis, biometrics, the HapMap project, and array technology (Bell, 2010). As a result of these technologies, the genetic basis of almost all of the >200 single gene disorders, which have epilepsy within their phenotype, has been identified. A few of these are rare familial “pure” epilepsies. The immediate challenge now is to unravel the epistatic and epigenetic mechanisms of developmental epilepsy, and the elucidation of postgenomic mechanisms.
  • 4 Mechanisms of epilepsy: A much more advanced, although still partial, understanding of the molecular mechanisms of ictogenesis has been achieved, including membrane function, receptor function, ionic changes, epilepsy at a systems level, neuronal networks, and epileptogenesis. Much still requires to be researched, and the link between underlying cause and mechanism though has still not been explored for many conditions. “Cause” is still generally not equated with “mechanism” in the Jacksonian sense.
  • 5 Methodologies for assigning cause—risk factor analysis: The statistical and methodologic assignment of cause has advanced considerably with an appreciation of the statistical basis of risk factor analyses and the description of etiology in terms of risk factors.
  • 6 Classification: In 1969, the ILAE published its first classification of epileptic seizures, which largely ignored etiology. Since then there have been a series of seizure type and syndromic classifications, and the most recent have begun a division by etiology (Berg et al. 2010).

General Observations

A number of general thoughts arise from this historical survey. First is the importance of societal and nonscientific influences on theories of epilepsy etiology. Examples are numerous. Concerns about degeneration at the end of the nineteenth century for instance were widely discussed in politics, the arts, sociology, and criminality. Eugenic research in epilepsy was primarily driven by economic, political, and social forces. Psychoanalytic thought was found in almost all social discourse, and the similar tendencies are arising now in relation to molecular genetics. It is a delusion of neuroscience that its progress is linear or that contemporary position inevitably is the most scientifically advanced. The awkward reality is that the march of neuroscience has had an erratic course sometimes in a backward direction and veering up many cul-de-sacs. It is partly subjective largely and markedly influenced by fashion and social forces, socioeconomic factors, dominant personalities, and the full gamut of human failings. Science is never neutral or objective, and thus has a social responsibility, a fact often forgotten in the laboratory sometimes with disastrous results as was the case in the 1930s.

Second, it should be realized that clinical neurology, being an essentially applied science, is heavily methodology-driven, and methodology in large part sets the agenda. We know only the etiologies we can measure, and what we cannot measure we cannot know. The introduction of clinical chemistry, EEG, neuroimaging, and neurogenetics has each changed our fundamental perception of etiology, and over the last 150 years, the focus has moved from one category to another, often overstating the importance of that in fashion (the switching interest between inherited and symptomatic causes for instance).

Third, is the importance of genetic factors. Although this is now the focus of intense interest, genetics has disappointingly failed to illuminate causation in the majority of cases. Although it is widely held that epilepsy had a strong hereditarian basis, the “genetic cause” of most cases remains obscure. This relative lack of advance (the missing hereditability of epilepsy; Johnson, 2011) is in marked contrast to the major genetics discoveries in immunologic disease and cancer for instance. The reason may reside in the fact that epilepsy and other neuropsychiatric disorders have their origins in neurodevelopment, and this is where epigenetic and epistatic mechanisms, or even chance, play a much greater role (Johnson M & Shorvon S, 2011, unpublished data). A better understanding of these processes, and not simply continuing searches for genetic polymorphisms, may hold the key to future advance. One interesting aspect of this is the possibility that conditions previously thought of as psychiatric comorbidities may be in fact manifestations of genetic disturbances in the same pathways, and that there may be overlapping or common mechanisms shared by cryptogenic epilepsy, developmental neuropsychiatric diseases, and cerebral degeneration. Further studies in this area might be well worth conducting, but we should be wary of a warning from history. In the early twentieth century, the link of epilepsy to degeneration and to mental disorder resulted in enormous stigma and culminated in eugenic measures to restrict reproduction and ultimately the murder of handicapped persons. Whether similar inquiries would be now be ethical or wise, given the societal consequences of previous genetic inquiries into epilepsy (stigma, sterilization, extermination), is questionable. Such considerations are never far from the surface of genetic enquiry.

A fourth point to note is that currently little focus is placed on the precipitating causes of epilepsy. This was not always so; in the past, there was little distinction between “causes” and “precipitating factors.” It is in fact difficult logically to see any difference, and a reemphasis on the precipitation of seizures, not least on their genetic mechanisms, might be valuable and helpful also in understanding etiology in terms of mechanism.

Another rather remarkable feature of “epilepsy” is that the widely used classification schemes have not incorporated etiology. It is often taught, as a fundamental neurologic canon, that epilepsy is “a symptom” not a “condition” but this is not reflected in our classification schemes. In 1897, Peterson, a leading American epileptologist, wrote: “A classification based strictly on etiology is not possible …in the light of present knowledge, but such a classification would be more scientific and valuable [than other types of classification]” (cited by Friedlander, 2001). Today, such a classification may be viable (Shorvon S, unpublished data), and might make more sense than our current semiologic or electroencephalographic classification schemes (and one can muse what upon what the classification of epilepsy today might look like if neuroimaging had been invented before EEG).

Finally, today we make the distinction between the “mechanisms” of epileptogenesis (increased excitation, defects in ion channels, and son on) and the causes of epilepsy. A more logical classification would be by etiologic mechanism, as pointed out by Jackson, and it is perhaps partly because we classify by “downstream” features that our current classifications are often confused. A focus on “causal mechanisms” might also be profitable in terms of therapy, and certainly it is striking how little the effectiveness of current antiepileptic drug therapy is influenced by current concepts of etiology. The major challenge for etiologic research in epilepsy is now surely to understand the functional and molecular mechanisms underlying epilepsy and to revert to a concept of etiology as mechanism, akin to that of Jackson.


Part of this paper is derived from the introductory chapter of the book Causes of Epilepsy (Shorvon, 2011). This work was undertaken at University College London (UCL) and received a proportion of funding from the Department of Health’s NIHR Biomedical Research Centres funding scheme.


I confirm that I have no conflicts of interest to declare. I confirm that I have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.