SOURCE: " 'Documents in Madness': Reading Madness and Gender in Shakespeare's Tragedies and Early Modern Culture," in Shakespeare Quarterly, Vol. 42, No. 3, Fall, 1991, pp. 315-38.

[In this essay, Neely argues that modern analyses of the cultural construction of madness have been prone to "misreadings of the past" due to a failure to historicize their own position. She then examines how representations of madness in Shakespeare's tragedies function within wider cultural contexts.]

Olivia: How now? Art thou mad?
Clown: No, madam, I do but read madness.
                (Twelfth Night, 5.1.293-94)¹

If others had not been mad, then we should
  be.
   (Shoshana Felman quoting Georges Bataille
                          quoting William Blake)²

This essay begins to investigate the continuities and discontinuities between the above epigrams. In the Twelfth Night exchange Olivia accuses Feste, her licensed fool, of madness; he defends himself against the charge by declaring that he is rather an interpreter of madness, referring literally to the letter he is reading from the supposedly mad Malvolio, figuratively to his fool's role as a satirist of human folly, and at a deeper level to his apt inscription of madness in Malvolio, the ambitious Puritan social climber and foolish would-be lover of Olivia. In the second quotation Shoshana Felman, in the epigram to her book Writing and Madness, identifies herself with the madness that is her subject in a quotation which enacts the intertextuality espoused by contemporary theorists. Feste inscribes madness to thwart Malvolio's desires and reads madness to dissociate himself from it; Felman reads madness to associate herself with it and to license desire.

As the epigrams imply, madness is a conundrum to those who would study it. It is a material condition that, to be understood, must be read, made sense of, inscribed into discourse.³ As Michael MacDonald has aptly noted, it is "the most solitary of afflictions to the people who experience it; but it is the most social of maladies to those who observe its effects." ⁴ Today, as in the early modern period, it is detected by laypersons before it is referred to doctors. Because it is "theoretically indeterminate,"⁵ it must be defined and read from within some framework; its definitions and therapies are always constructed from a particular historical moment and within a particular social order, influenced by and influencing that order. The final difficulty of reading madness—implicit in the two epigrammatic exchanges—is that in the act of doing so, one dissociates oneself from it or associates oneself with it, and in either case becomes disqualified as an interpreter. To read madness sanely is to miss the point; to read madness madly is to have one's point be missed. In this essay I want to begin to examine why, how, and with what consequences madness was read and represented in England in the early modern period by focusing on how representations of madness in Shakespeare's tragedies function within wider cultural contexts.

It has long been recognized that England in the period from 1580 to 1640 was fascinated with madness, although some aspects of this obsession have been over-estimated or misreported. The signs of its fascination are to be found in the treatises on the topic by Battie, Bright, Jorden, Wright, and Burton; in the theatrical representations of madness in the plays of Kyd, Shakespeare, Dekker, Middleton, Fletcher, and Webster; in the large numbers of patients who consulted such well-known doctors as Richard Napier and John Hall (Shakespeare's son-in-law) with symptoms of mental distress; and in the widespread references to and representations of Bethlem, or Bedlam, the popular name for Bethlehem Hospital, the main institution in England in this period which confined the insane. Bedlam, according to a 1598 visitation report made a couple of years before Hamlet and Twelfth Night were written, contained only twenty inmates: nine men and eleven women (or perhaps ten of each). The thirty-one inmates listed in a 1624 report caused overcrowding in the institution, which was tiny, "loathsomely and filthely kept," and badly mismanaged. The term "Bedlam" was in widespread use in early modern England not so much because of the impact of the institution itself (which had been in existence as a hospital since about 1330 and may have started accepting disturbed patients sometime before 1403, when a visitation record reports the presence of six men "mente capti") but because it had become a code word in Elizabethan and Jacobean culture for the confused, charged, and contested topic of madness.⁶

Madness, a concept in transition in the period, begins to be read/constructed/experienced differently in the sixteenth and seventeenth centuries than it had been in the Middle Ages (where it marked the intersection of human and transcendent) or than it will be in subsequent eras. In the eighteenth century, it will become, as Michel Foucault claims, the mark of unreason, the symbol of the animal side of human nature that needs confinement and restraint; in the nineteenth century, insanity (now the preferred term ⁷) becomes identified with hereditary degradation and immorality and is to be rectified by "moral treatment" or domestication. In the latter half of the twentieth century, philosophers, theorists, the anti-psychiatry movement, and investigations into the chemical basis for mental disorders have collapsed the boundaries between mad and sane, mental and physical, real and illusory, that were being constructed in the Renaissance.

This twentieth-century breakdown of partitions is apparent in both medical practice and philosophical theory. In the 1960s the clinical and theoretical work of Thomas Szasz, R. D. Laing, and the anti-psychiatry movement argued that mental illness was a myth used to bring disruptive behavior under control, a "sane" reaction to oppression in the family and in the culture.⁸ Current public policy mandating the deinstitutionalization and "mainstreaming" of the mentally distressed similarly (though with different motives) loosens boundaries between the sane and the insane. Current research and recent therapies stress the biochemical basis of and pharmacological treatments for mental distress, resplicing mind and body. Likewise, for literary theorists and philosophers, reading madness functions subversively to blur boundaries, to put the verb "to know" in quotation marks, as Shoshana Felman notes.⁹ Post-structuralist philosophers of radical skepticism like Derrida and Lacan, denying the possibility of a unified subject with continuous identity, of a coherent language that can ever say what it means, of "true" knowledge of the world, erase the boundaries between madness and sanity that were constructed in the Renaissance and strengthened and policed in the Enlightenment. Most influentially, Michel Foucault's Madness and Civilization critiques the Age of Reason for exploiting the discourse of madness and the confinement of the mad to erase reason's antithesis, unreason.

Because current theories and therapies of madness work to deconstruct what the early modern period worked to construct, misreadings of the past are likely. Too often, analyses of the cultural construction of madness, like those of Foucault and Elaine Showalter, fail to historicize their own position and to distinguish it from that of earlier periods. Both these influential accounts are weakened by inadequate knowledge of periods before that on which they focus (Foucault focuses on the eighteenth century, Showalter on the nineteenth); by a conventional, hence inaccurate, view of historical periodization; by a refusal to make sufficient distinctions between aesthetic representation and other sorts of historical data; and by a failure to fully gender the subject of madness. For Foucault there are only madmen; for Showalter there are only madwomen.¹⁰

In the early modern period the discourse of madness gained prominence because it was implicated in the medical, legal, theological, political, and social aspects of the reconceptualization of the human. Gradually madness, and hence sanity, began to be secularized, medicalized, psychologized, and (at least in representation) gendered. In the Middle Ages, madness was seen as the point of intersection between the human, the divine, and the demonic. It was viewed alternatively or simultaneously as possession, sin, punishment, and disease, and it confirmed the inseparability of the human and transcendent.¹¹ By theorizing and representing madness, the Renaissance gradually and with difficulty began to try to separate human madness from the supernatural (from demonic and divine possession, as does Edward Jorden's treatise on hysteria, The Suffocation of the Mother); from the spiritual (from doubt, sin, guilt, and rational suicide, as does Timothy Bright's Treatise of Melancholy); from witchcraft and bewitchment (as does Reginald Scot's Discouerie of Witchcraft); from frauds who imitated these conditions (as does Samuel Harsnett's Declaration of Egregious Popish Impostures); and from the sheerly physical (as do Jorden and doctors like Richard Napier and John Hall in their diagnoses of epilepsy and menstrual disorders), and began to try to map the normal, "natural," and self-contained secular human subject. Splitting the supernatural from the natural, and attempting to define what remained, the period began to separate mind from body, man from woman, insanity from both sanity and from other types of aberrance such as poverty, heresy, and crime. We can watch the linked aspects of this multifaceted process unfold in treatises on melancholy, hysteria, and witchcraft, in medical and legal practice, and in the drama.

In his A Treatise of Melancholy Timothy Bright (at the time a doctor and subsequently an Anglican priest) provides elaborate classifications of madness and recommendations for treatment that serve, by complex distinctions between the spiritual and the psycho-physiological, to subordinate the former. The treatise is written in the form of a letter to a male friend, addressed as "M", who is suffering from what we would call depression. Designed to cure M, the letter advises him on how to distinguish between spiritual doubt and the disease of natural melancholy. Spiritual doubt, caused by the sense of sin and the incomprehensible and inexpressible loss of God's favor, is to be cured by penitence, prayer, and faith. Spiritual consolation is the subject of the longest of the treatise's forty-one chapters.¹² The rest of the treatise outlines an etiology of melancholy that explicates the elaborate interactions between the soul, mind, passions, and body, on the one hand, and, on the other, the animal spirits that unify them. Natural depression is caused by the unnatural excess or combustion of natural melancholy, the cold dry humor or black bile that, when burned, causes such symptoms as passivity, unsociability, fury, stupidity, paranoia, lust, anger, mania, but especially sorrow and fear. Bright's recommended treatment (remarkably familiar) is healthy diet, exercise, sleep, and good friends.

In Bright's treatise, however, the careful distinctions between spiritual and physiological melancholy repeatedly collapse. Both states are characterized by the same symptoms: hallucinatory terror and unreasonable sadness. Natural melancholy predisposes one to spiritual doubt while spiritual doubt exacerbates the pathology of the black bile. Both the medical therapy, based on diet and rest, and the spiritual cure, dependent on faith and grace, are designed to relieve the loss of self-worth that characterizes equally both forms of the disease. The effect is to merge the two kinds of melancholy and to subordinate the spiritual causes and cure to the psychological and physiological ones. The gender of M, the respectful scholarly tone of Bright's letter/treatise, and the identification of the disease with spiritual doubt all point to the associations of melancholy with the fashionable, the upper class, the literate, the masculine—associations that become yet more prominent in Robert Burton's The Anatomy of Melancholy (1621).

While Bright's treatise strives unsuccessfully to distinguish spiritual guilt from natural melancholy, Edward Jorden's landmark treatise, A Briefe Discourse of a Disease Called the Suffocation of the Mother, sets out to distinguish bewitchment from insanity (and, indirectly, to legitimize licensed physicians). It is directed at Jorden's fellow members of the College of Physicians, who, as trained and experienced doctors are, he claims "best able to discerne what is naturall, what not naturall, what preternaturall, and what supernaturall,"¹³ and who might therefore be called upon, as Jorden had been, to testify on the status of the victim's symptoms in witch trials. If these symptoms are diagnosed as natural in orgin, the result of hysteria, the accused witch is acquitted (as over half were¹⁴); if they are found supernatural, she (or, infrequently, he) is convicted. The diagnosis is a difficult one to make because the symptoms of bewitchment and hysteria are identical. Hysteria was caused, traditional medicine believed, by the pathology of the diseased and wandering womb, and hence it was primarily although not exclusively a disease of women: "The passiue condition of womankind is subiect vnto more diseases and of other sortes and natures then men are: and especially in regarde of that part from whence this disease which we speake of doth arise," Jorden declares.¹⁵ One internal cause of the disease, Jorden claims with some reticence, is retention of menstrual blood or sperma (which women were believed to have) due to sexual frustration or the suppression of the "flowers," the menstrual periods. The origin of the fantastic and disconnected symptoms of the disease—swooning, paralysis, choking, convulsions, numbness, delirium, epilepsy, headaches—is the wild peregrinations of the uncontrollable uterus and its capacity to corrupt all the parts of the body. One recommended cure is marriage, which institutes regular sexual relations and thus aids in evacuation of fluids and brings the wild uterus under a husband's control. In spite of the tendency of such an analysis to identify hysteria as a disease of women, Jorden does not explicitly draw this conclusion and refers without comment (as do other writers) to men who suffer from "the mother."¹⁶

This association of hysteria with women, especially women of the upper classes, incipient in the early modern period, is present as well in Robert Burton's compendious Anatomy of Melancholy. As the all-male frontispiece of the book suggests, Burton associates melancholy especially with male scholars, philosophers, and geniuses like Democritus and himself, although its causes and symptoms are multitudinous and its sufferers are everywhere. But when he defines the "Symptomes of Maides, Nunnes, and Widowes melancholy," he associates this type with "fits of the mother," which he represents as linked with marital, sexual, and class status, associated with sexual frustration, and cured by sexual satisfaction: "For seldome shall you see an hired seruant, a poore handmaid, though ancient that is kept hard to her worke, and bodily labour, a course country wench troubled in this kinde." Those who are "prone to the disease" are "noble virgins, nice gentlewomen, such as are solitary and idle, live at ease, lead a life out of action and imployment, that fare well in great houses and Ioviall companies, ill-disposed peraduenture of themselues, and not willing to make any resistance, discontented otherwise, of weake judgement, able bodies, and subject to passions." Like Jorden, Burton recommends marriage as a "remedy."¹⁷

Jorden's Discourse not only aims to forestall mistaken diagnoses of bewitchment but also to expose "impostures" who only pretend to have the symptoms. Reginald Scot's ironically titled and cogently argued The Discouerie of Witchcraft (1584) is written by this Justice of the Peace to deny the supernatural powers of witches themselves, attributing their behavior, including their voluntary confessions, to the effects of melancholy or hysteria. This diagnosis, of course, has the effect of continuing the secularization of witchcraft by medicalizing witches' behavior. (Witchcraft had begun to be secularized when its disposition was consigned to civil courts by a 1542 statute.) Samuel Harsnett (an ambitious chaplin to Bishop Bancroft) joined the established church's coordinated campaign against Catholic and Puritan exorcists in his A Declaration of Egregious Popish Impostures (1603), which attacks illegal Catholic exorcism rituals, exposing both possession and exorcism as instigated insanity—fraud.¹⁸ These contexts may help to explain why the drama of the period often focuses on distinctions between feigned and actual madness and represents tests, like Claudius's test of Hamlet, to uncover fraud.¹⁹ While witchcraft prosecutions continue to take place in England until 1680, these treatises and others function to medicalize the behavior of witches and the bewitched and to call the trials into question. In these areas—bewitchment, possession, witchcraft—madness is becoming a psychological alternative to conditions formerly defined as supernatural in origin and treatment.

On the new stages of the public theaters, Shakespeare, following Kyd in revising classical and Senecan tragedy, in Hamlet, Macbeth, and King Lear shapes a new language for madness and provides one important site for its redefinition.²⁰ The plays, by representing both madness and the process of reading madness, theatricalize and disseminate the complicated distinctions that the treatises theorize. In the drama, as in the culture outside it, madness is diagnosed by those who observe it—both specialists and laypersons. Their readings enable the drama's audience to participate with them in distinguishing madness from sanity and from madness's look-alikes—loss of grace, bewitchment, possession, or fraud. Since madness, like its imitations, is extreme, dislocated, irrational, alienated—separated both from the self who performs and the spectators who watch—the diagnosis is difficult. In Shakespeare's plays that make this diagnosis, the speech of the mad characters constructs madness as secular, socially enacted, gender- and class-marked, and medically treatable.

Although the importance of madness in the period's drama, especially in that of Shakespeare, has long been acknowledged, and although literary historians have outlined its anatomy and traced its occurrences,²¹ there have been few recent attempts to understand its rhetorical structure and dramatic function in Shakespeare's tragedies, or its wider cultural significance. Take, for example, responses to Ophelia and to Lear. A. C. Bradley sums up, at the beginning of the twentieth century, two centuries of views of and visual representations of Ophelia in madness as beautiful, sweet, lovable, pathetic, and dismissible.²² More recently, feminist critics, challenging this interpretation, have read Ophelia's madness as either her liberation from silence, obedience, and constraint or her absolute victimization by patriarchal oppression.²³ In responses to King Lear, traditional critics often see Lear's madness as a means to illumination and self-knowledge.²⁴ Significant contemporary analyses, in opposing the humanist optimism of these earlier interpretations, oddly pass over Lear's madness without notice. Stanley Cavell's influential monograph, "The Avoidance of Love: A Reading of King Lear," bypasses the long period when Lear is, as he puts it, "stranded in madness." Stephen Greenblatt's important new historicist essay, "King Lear and the Exorcists," reinterprets Edgar's feigned madness but ignores Lear's actual madness. Jonathan Dollimore, rather than seeing radical theatrical or social implications to Lear's madness, dismisses it as "demented mumbling."²⁵ None of these critics, representing various current theoretical approaches, reads madness closely in the plays. None asks, as I do here, how its linguistic construction, its gender-coding, and its dramatic functions participate in cultural needs, practices, and attitudes.

Shakespeare, prefiguring Foucault's analysis, dramatizes madness primarily through a peculiar language more often than through physiological symptoms, stereotyped behaviors, or iconographic conventions.²⁶ This characteristic speech is both something and nothing, both coherent and incoherent. Spectators, onstage and off, read this language, trying to make "sense" of it, translating it into the discourse of sanity. Shakespeare's language of madness is characterized by fragmentation, obsession, and repetition, and most importantly by what I will call "quotation," which might instead be called "bracketing" or "italicization."²⁷ The mad are "beside themselves"; their discourse is not their own. But the voices that speak through them are not (even in the case of Edgar's parody of possession) supernatural voices but human ones—cultural ones perhaps. The prose that is used for this mad speech (although it includes embedded songs and rhymes) implies disorderly shape,²⁸ associates madness with popular tradition, and contributes to its colloquial, "quoted" character. These quoted voices, however, have connections with (or can be interpreted to connect with) the mad characters' pre-mad gendered identity and history, their social context and psychological stresses—as well as with larger themes of the plays and of the culture. The alienated speech allows psychological plausibility, thematic resonance, cultural constructions, and social critique. Using it, Shakespeare represents distinctions between female hysteria and feigned male melancholy in Hamlet, between supernatural witchcraft and natural alienation in Macbeth, and between feigned possession and natural madness in King Lear.

Onstage characters mediate this pregnant, mad discourse, showing us how to translate it in ways made explicit by the anonymous Gentleman in Hamlet who introduces Ophelia, Shakespeare's first extended "document in madness":

She … speaks things in doubt
That carry but half sense. Her speech is
 nothing,
Yet the unshapèd use of it doth move
The hearers to collection; they yawn at it,
And botch the words up fit to their own
 thoughts,
Which, as her winks and nods and gestures
 yield them,
Indeed would make one think there might be
 thought,
Though nothing sure, yet much unhappily.
                                 (4.5.6-13)²⁹

The speech here described is painful, unshaped unsense that can be "botched" up into shape by an audience's perceptions. Ophelia's alienated discourse invites a psychological, thematic, and gendered interpretation. It resituates sacred material in a secular, psychological context, and she and Hamlet act out distinctions between feigned and actual madness and between rational and mad suicide, distinctions that the culture was gradually establishing.

Ophelia's madness is represented almost entirely through fragmentary, communal, and thematically coherent quoted discourse. Through it, rituals elsewhere involving the supernatural are appropriated and secularized. Ophelia recites formulas, tales, and songs that ritualize passages of transformation and loss—lost love, lost chastity, and death. These transitions are alluded to in social formulas of greeting and leave-taking: "Well, God dild you," "Good night, ladies, good night" (11. 42, 73); in religious formulas of grace and benediction: "God be at your table!" "God 'a' mercy on his soul! / And of all Christian souls, I pray you" (11. 44, 198-99); in allusions to folk legends or tales of daughters' metamorphic changes in status: tales of the "owl [who] was a baker's daughter" (11. 42-43) and of the master's daughter stolen by the steward.

Her songs likewise enact truncated rites of passage. Love and its loss are embodied in the song of the "truelove," imagined with a cockle hat, staff, and sandals, all icons of his pilgrimage. She sings of Valentine's Day loss of virginity when a maid crosses a threshold both literal and psychological: "Then up he rose and donned his clothes / And dupped the chamber door, / Let in the maid, that out a maid / Never departed more.… / Young men will do't if they come to't, / By Cock, they are to blame" (11. 52-55, 61-62). This imagined deflowering preempts and precludes a marriage ritual. The other songs mourn a death and represent the concrete markers of a spare funeral ritual—a flaxen poll, a bier, a stone, no flowers. They enable Ophelia to mourn her father's death, enact his funeral, encounter his dead body, and find consolation for her loss: "He is gone, he is gone, / And we cast away moan" (11. 196-97). Into this central loss and its rituals, Ophelia's other losses or imagined losses—of lover, of virginity, of "fair judgement"—are absorbed. Her distribution of flowers to the court is an extension of her quoted discourse, an enacted ritual of dispersal, symbolizing lost love, deflowering, and death. A secularized cultural ritual of maturation and mourning is enacted through Ophelia's alienated speech.³⁰

Ophelia's madness, as the play presents it, begins to be gender-specific in ways that later stage representations of Ophelia and of female hysterics will exaggerate.³¹ Her restlessness, agitation, shifts of direction, her "winks and nods and gestures" (1. 11) suggest the spasms of "the mother" and show that madness is exhibited by the body as well as in speech; gesture and speech, equally convulsive, blend together: Ophelia "beats her heart, / Spurns enviously at straws" (11. 5-6). The context of her disease, like that of hysteria later, is sexual frustration, social helplessness, and enforced control over women's bodies. The content of her speech reflects this context. Laertes's anguished response to Ophelia as a "document in madness"—"Thought and affliction, passion, hell itself, / She turns to favor and to prettiness" (11. 187-88)—shows how the reading of madness's self-representation can aestheticize the condition, mitigating both its social critique and its alien aspects. In a similar fashion Gertrude narrates Ophelia's death as beautiful, natural, and eroticized, foreshadowing later representations of it and representations of female hysterics as sexually frustrated and theatrically alluring. The representation of Ophelia implicitly introduces conventions for reading madness as gender-inflected.

Gender distinctions likewise begin to take shape in the contrasts between Hamlet and Ophelia. Although Ophelia in her mad scenes can be seen to serve as a double for Hamlet during his absence from Denmark and from the play,³² Hamlet's madness is in every way contrasted with hers, in part, no doubt, to emphasize the difference between feigned and actual madness. His discourse, although witty, savage, and characterized by non sequiturs and bizarre references, almost never has the "quoted," fragmentary, ritualized quality of Ophelia's—as we are instructed: "Nor what he spake, though it lacked form a little, / Was not like madness" (3.1.164-65). Significantly, the one time it is "like madness"—that is, like Ophelia's speech—is after the encounter with his father's ghost, when Hamlet must abruptly reenter the human, secular world of his friends. The "wild and whirling words" (1.5.133) that he utters to effect this transition are quoted truisms and social formulas for parting which are incoherently deployed:

And so, without more circumstance at all,
I hold it fit that we shake hands and part:
You, as your business and desire shall point
 you,
For every man hath business and desire
Such as it is, and for my own poor part,
Look you, I'll go pray.
                                  (11. 127-32)

After this moment of dislocation he announces a plan to feign madness, to "put an antic disposition on" (1. 172); and he is able to "go in together" (1. 186) with his friends, reuniting himself with the world of human fellowship and sanity, although he is himself marked by the remembrance of the Ghost's "commandment" (1. 102).

The stylistic distinction between Hamlet's feigned madness and Ophelia's actual madness is emphasized by other distinctions. Henceforth in the play, Hamlet is presented as fashionably introspective and melancholy while Ophelia becomes alienated, acting out the madness Hamlet only plays at. Whereas her madness is somatized and its content eroticized, Hamlet's melancholy is politicized in form and content. Caused purportedly by Claudius's usurpation of the throne and by his father's commandment, it manifests itself in social criticism, and it is viewed as politically dangerous. Ophelia must be watched, contained within the family, within the castle; Hamlet must be first contained and later expelled to England to be murdered. By acting out the madness Hamlet feigns and the suicide that he theorizes, the representation of Ophelia absorbs pathological excesses open to Hamlet and enables his reappearance as a sane, autonomous individual and a tragic hero in the last act. There he appears detached from family and from sexuality, seemingly freed from passivity and loss of control, capable of philosophical contemplation and revenge, worthy a spiritual epitaph and a soldier's funeral; his restored identity is validated—symbolically as well as literally—over Ophelia's grave: "This is I, / Hamlet the Dane" (5.1.257-58).

The contrast between Ophelia's mad suicide and Hamlet's contemplated one represents in drama the distinction the period was required to make between calculated suicide (felo-de-se), a religious sin and a civil crime, and insane self-destruction (non compos mentis). When the act was judged self-murder, the deceased's property was seized by the state and Christian burial was not encouraged.³³ Madness, however, rendered suicide innocent and permitted conventional inheritance and burial. The secularization of suicide and that of madness reinforced each other. The play enacts these distinctions without choosing sides. Whereas Hamlet's calm contemplation of suicide would render the act on his part a sin (of despair) and a crime (as he recognizes with his reference to the "canon 'gainst self-slaughter" [1.2.132]), Ophelia's suicide is described by Gertrude as accidental ("an envious sliver broke" [4.7.173]), passive, involuntary, mad. In England in the period, drowning was the most common means of suicide for women and the cause of death that made distinctions between accident and volition most difficult.³⁴ The play keeps various possibilities in suspension. Gertrude's representation of Ophelia's death neither condemns it on religious grounds nor explicitly condones it on medical/legal grounds. Instead she narrates it without interpretation as a beautiful, "natural," ritual of passage and purification, the mad body's inevitable return to nature:

          Her clothes spread wide,
And mermaidlike awhile they bore her up,
Which time she chanted snatches of old lauds,
As one incapable of her own distress,
Or like a creature native and indued
Unto that element.
                                  (11. 175-80)³⁵

Later the issue of Ophelia's death is reopened when the lower-class gravedigger and the priest skeptically challenge the "crowner's" warrant and argue that it is aristocratic prerogative that permits Ophelia's Christian burial.

In Macbeth, Lady Macbeth's suicide has none of the purifying and involuntary aspects of Ophelia's, and its meaning is not interrogated. But it occurs following a state of gendered alienation represented through quoted discourse with similarities to Ophelia's. The alienation of Lady Macbeth in sleepwalking is, like Ophelia's, psychologized, represented by means of quoted speech, read by representatives of the community, associated with symbolic purification, and it culminates in suicide. Her breakdown embodied in sleepwalking is contrasted with Macbeth's enraged, bloody, "valiant fury" ("Some say he's mad" [5.2.13]). But the division between her powerful will in the early acts of the play and her alienated loss of it in the sleepwalking scenes, her connections with and dissociation from the witches, and their bifurcated representation all construct—and blur—other distinctions associated with madness: those between supernatural and natural agency, diabolic possession and human malevolence.

Lady Macbeth's sleepwalking, like Ophelia's madness, occurs after an absence from the stage, is presented as a sharp break with earlier appearances, and is introduced by an onstage spectator. When sleepwalking, Lady Macbeth quotes, in the form of proverbial commonplaces ("Hell is murky" [5.1.38]) and chilling pseudo-nursery rhymes ("The Thane of Fife had a wife. Where is she now?" [11. 44-45]), her own earlier words (or perhaps thoughts) and Macbeth's. She refers to Duncan's murder, Banquo's ghost, and the death of Lady Macduff all in the mode of advice and comfort to Macbeth ("No more o' that, my lord, no more o' that" [1. 46]). She narrates Macbeth's bloody acts, talks directly to him although he is not present, and acts out her own complicity by "washing" her hands to remove the smell and sight of the blood that taints them. This quotation has the effect of distancing the discourse from its speaker and inviting a reading. But it is less communal and thematic, more personal and psychologized than Ophelia's. The doctor explicitly reads Lady Macbeth's state as religious despair, not as demonic possession or physical breakdown—in Bright's terms, as spiritual rather than natural melancholy: "More needs she the divine than the physician" (1. 77).

The witches and Lady Macbeth, as Peter Stallybrass has argued,³⁶ are indirectly identified with each other by their gender, by the structure and symbolism of the play, and by their parallel roles as catalysts to Macbeth's actions. They function as cultural scapegoats for the unnaturalness, disorder, and violence let loose. But the play also implies contrasts between Lady Macbeth and the witches, and these produce disjunctions between the natural and the supernatural. The witches' supernatural ambiguity is contrasted with the "natural" ambiguity of Lady Macbeth's sleepwalking scene. In their early appearances they are described as ambiguously male or female, as on the earth but not of it; they speak equivocally (but not madly). Lady Macbeth, when sleepwalking, is in a state that combines "the benefit of sleep" with "the effects of watching" (5.1.11-12); "Her eyes are open," "but their sense are shut" (11. 26-27). The witches are dramatized in connection with some of the conventional accoutrements of witchcraft belief: familiars, submission to Hecate, spells, potions, fortune-telling, and successful conjuring. In contrast Lady Macbeth's attempted (and unsuccessful) invocation is to spirits that seem more natural than supernatural: they "tend on mortal thoughts" and "wait on nature's mischief (1.5.41, 50). She does not ask directly for help to harm others as witches typically do, but only for a perversion of her own emotions and bodily functions: "fill me … top-full / Of direst cruelty. Make thick my blood" (11. 42-43). In contrast the witches plot to cause the magical kinds of harm to others conventionally associated with witches' maleficium: interference with livestock, weather, and male sexuality.

The witches are, then, ambiguously associated with and dissociated from Lady Macbeth.³⁷ Their own representation is likewise bifurcated. They are ambiguously "natural" and supernatural. They are represented partly as the disgruntled outcasts of Scot's Discouerie, partly as the agents of harmful activities like those charged in English witch trials, and partly as devil-possessed like the witches described by Continental witch-mongers in the Malleus Maleficarum (c. 1486). In the opening scenes they seem to invite Scot's psychological interpretation (statistically supported by Alan Macfarlane's social, structural analysis³⁸); they appear to be frustrated, melancholic women who, on the margins of society, get back at those who have disregarded them by muttering curses and plotting revenges—"I'll do, I'll do, and I'll do" (1.3.10)—and hence attracting blame and punishment. However, they do have familiars and seem capable of preternatural travels, so are not represented merely as social misfits. In their later appearances (3.5 and 4.1), although their theatrical power is diminished, the witches are endowed with all the paraphernalia of demonic possession from Continental witchlore. They serve Hecate (in what may be a later, non-Shakespearean addition), use illusion to influence Macbeth, mix a "charm" made from the noxious parts of animals (and humans).³⁹ Macbeth "conjures" them by their "profess[ed]" supernatural powers (4.1.50-61). The effect of these representations of an alienated Lady Macbeth and divided witches, ambiguously connected with each other, is to create a continuum of alienation and malevolence in the play, which blurs the boundaries between natural and supernatural agency, among witchcraft of English or Continental sorts, antisocial behavior, and madness. This continuum has made it tempting to ask of the play just as the period (through witchcraft prosecutions and through reading madness) was asking: who is to blame? Who or what is the source of harm and evil? The questions produce conflicting and incompatible answers, as they did in the period. The continuum of malevolence blurs the question of agency in the play as it blurs the question of the ontological status of "witches." It reproduces the period's "hovering" between contradictory belief systems and conflicting attributions of causality and agency: God and the devil, madwomen and witches, castrating wives and ambitious tyrants.

To understand the complicated responses and flexible practices that such uncertainty created, and to place Shakespeare's tragedies against contemporary attempts to categorize madness, it is helpful to look briefly at the medical practice of Richard Napier and at the 1598 and 1624 Bedlam censuses. Napier was a doctor, a minister, and an astrologer who from 1597 to 1634 treated about sixty thousand patients in Great Linford in northern Buckinghamshire, taking notes on each consultation. Two thousand and thirty-nine of these patients from all social classes consulted him for mental disorders, and these cases are analyzed in the epidemiology of mental disorder constructed by Michael MacDonald in Mystical Bedlam: Madness, Anxiety, and Healing in Seventeenth-Century England. Thanks to MacDonald's superb, detailed, and gendered analysis, Napier's practice becomes a site where definitions, distinctions, and gender-coding in mental ailments can be explored. Like theorists and playgoers, Napier strove to distinguish between the similar symptoms caused by possession, bewitchment, and mental or physical disorders; he worked hard to do so but was often at a loss.⁴⁰ His cures, designed to fit the disorder, were eclectically magical, medical, astrological, and spiritual; to some patients he gave advice, to most purges, to a few amulets or prayers or exorcisms.

Women consulted Napier for all causes more often than did men (ratio: 78.8 men to 100 women); they consulted him more for mental disorders than did men (ratio: 58.2 men to 100 women, similar to that reported in England today) and reported suffering almost twice as much stress as men (ratio: 52.3 men to 100 women). Most of Napier's female and male patients suffered mental distress and depression from the same causes: courtships (23.6 percent), marital problems (17.6 percent), bereavements (17.5 percent), and debt (12.9 percent).⁴¹ The reasons why women are over-represented in Napier's practice, especially in consultations for mental distress, are as complex and difficult to analyze as why women visit doctors more than men do today and report more depression. Then as now it may be connected with their vulnerability to diseases of the reproductive system, their need therefore to see doctors more, and the stress that family life under patriarchy puts on them.⁴²

However, although more women came to Napier with symptoms of mental distress, there is not much difference in the percentages or even the numbers of men and women identified as suffering extreme forms of mental disturbance—i.e., madness. (Similarly, recent findings by medical historians and sociologists show that while, today, women see doctors more for depression, insomnia, and other imprecisely identified types of mental distress, they do not suffer from extreme pathological states like schizophrenia more often than do men and, contrary to earlier claims, are not more likely than men to be institutionalized for mental disorders.⁴³ MacDonald's raw statistics show a similar pattern. Patients who report extreme symptoms—symptoms associated with mania as opposed to melancholy and designated by terms like "mad," "lunatic," "mania," "frenzy," "raging," "furious," "frantic"—are rare. There are more cases for women in almost every category (because there are more women in the sample), but the percentages are virtually identical and the absolute numbers not that different. For example, of the 2,039 patients, 34 of the men (or 5 percent) and 54 of the women (or 4 percent) are designated "mad"; 25 of the men (or 3 percent) and 21 of the women (or 2 percent) as "lunatic." There is 1 man with mania and 7 men and 3 women with frenzy. Men are more likely to be designated melancholy or "mopish," a milder form of melancholy (in accord with the early modern period's male coding of this disease—which is regendered female in the nineteenth and twentieth centuries), whereas women more often "take grief," "grieve," and are more often "troubled in mind"; both men and women are tempted to and attempt suicide in about equal rates, but women are more often tempted to kill their children or, uniquely, tempted to kill either their children or themselves.⁴⁴ Napier never identifies the fits of the mother as mental disturbance but connects it with strictly physical symptoms like menstrual cramps. And "sexual urges" are a symptom of only one (male) patient.⁴⁵

In Napier's report of his practice, while women suffer more mental disturbance than men, the gendering of types of madness is only hinted at, foretold, much less apparent than in such dramas as Hamlet. What stands out is Napier's attempts to categorize madness, to distinguish it from supernatural visitations and from physical maladies. Another set of documents of the period also shows tentative movement toward division by gender, but here, too, the reading must be cautious. These are the 1598 and 1624 censuses of Bedlam, included in visitation committee reports to Bridewell Hospital, which administered the facility.⁴⁶ The reports give the names of the inhabitants and some of the following data: source of admission (from Bridewell, the lord mayor of London, or private parties); length of stay (from Neme Baker, twenty-five years in the 1598 census, to Thomas Denham, fourteen days in the 1624 census); source of maintenance (guilds, individuals, parishes, colleges, other hospitals); indications of social class and context (in the 1598 census, when such information was more frequently noted, inhabitants included "Welch Elizabeth", "Rosse an Almes-woman", "Edmond Browne one of the Queenes Chappell", and "Anthoney Greene fellow of Penbrooke Hall in Cambridge"). Both censuses usually list patients with the longest tenure first, but the 1598 census is divided between admissions from Bridewell and from elsewhere, and the 1624 census is divided up into men (18) and women (13), and comments are made on the seriousness of the condition (probably because even at these small numbers the place was overcrowded, and the committee wished to reduce the number of those confined). The designations for the men speak to their administrative status; they are termed "fitt to bee kepte," "not fitt to bee kept," or to be sent to "some other hospitall," "home to his wife," "to Hull from whence hee came." Only two of the men, who are "Idiots," have their illness specified, and none are called "mad." In contrast the women are explicitly characterized as "very ill," "madd," "very madd," "a mad woman," "something idle headed," "fell madd." (Eight of the 18 men are designated fit to be kept and 9 to be sent elsewhere; 7 of the 13 women are to be kept and 4 are to be removed to other care; the dispositions of 1 man and 1 woman are not specified.⁴⁷) These no-doubt-unconsciously chosen designations suggest a tendency to identify the women with their illness and the men with their institutional disposition.

While the stage does not associate madness more with one class or gender than another, in King Lear, as in the records of Richard Napier and of Bethlehem Hospital, madness and distress are conceived of as treatable illnesses with mental and physical components. By underlining the distinction between Lear's natural madness and Edgar's feigned supernatural possession and by including two cures, one physical (administered by a doctor) and one mental (administered by Edgar, a layperson), the play contributes to the secularization, psychologizing, and medicalization of madness and extends conventions for representing it.

Edgar, victimized by his bastard brother, Edmund, assumes the speech of demonic possession as a role—as a disguise.⁴⁸ Quotation in his speech is, in effect, quadrupled. Disinherited Edgar speaks in the voice of Poor Tom, the Bedlam beggar, who speaks in the voice of the devil, who quotes Samuel Harsnett's melodramatic exposure of the drama of bewitchment and exorcism.⁴⁹ Tom's mad speech, like Ophelia's, is made up of quoted, that is culturally and psychologically resonant, fragments, but his discourse incorporates differently inflected cultural voices. His speech embeds song fragments—"Through the sharp hawthorn blows the cold wind"—bits of romance—"But mice and rats, and such small deer, / Have been Tom's food for seven long year"—formulaic commandments and proverbial sayings—"obey thy parents; keep thy word's justice," "Keep thy foot out of brothels, thy hand out of plackets" (3.4.45, 136-37, 79-80, 95-96). These quotations transmit a theological discourse of sin and punishment in which Poor Tom is an emblematic fallen Christian, a "servingman, proud in heart and mind," "hog in sloth, fox in stealth, wolf in greediness, dog in madness" (11. 84, 92-93). Embodying the seven deadly sins, especially those of pride and lust, he represents, like traditional madmen, guilt over and punishments for these sins; he is led by the "foul fiend" "through fire and through flame, through ford and whirlpool, o'er bog and quagmire," and "eats the swimming frog, the toad, the todpole, the wall-newt and the water" (11. 51-52, 127-28).

This mad discourse functions variously. It provides Edgar-as-Tom with a coherent characterization by permitting him to express and conceal his victimization and (it has been argued) his suppressed desire for self-punishment and revenge.⁵⁰ It functions dramatically to trigger, mark, and counterpoint the specific moment of Lear's own break with sanity, which occurs decisively at his emotionally apt but logically groundless identification with Poor Tom at line 62: "What, has his daughters brought him to this pass?"⁵¹ The disguise allows the disinherited Edgar, by identifying with the middling or lower sorts and by adopting their speech and beliefs, to participate with the Fool and naked Lear in the reversals of class and status that pervade the play. But always Edgar's quoted religious discourse is rendered theatrical, both because the discourse is feigned and because it is constructed through quotation of Samuel Harsnett, who himself narrates possession as theatrical role-playing instigated by the suggestion and rehearsal of the exorcists. By appropriating for Poor Tom a "documented fraud," the spuriousness of Edgar's madness is emphasized, possession and divine retribution are mocked through mimicry, Lear's contrasting madness is marked as "natural," and the Church's attempt to outlaw exorcism is furthered. At the same time, the surviving belief in possession, perhaps especially prevalent among middle and lower ranks, is represented onstage. While Greenblatt sees these rituals as "emptied out,"⁵² I would say rather that in this mad discourse their sacred meaning is resituated: morality, guilt, suffering, and punishment are understood within human, psychological parameters.

In stark contrast to Edgar's feigned delirium of sin, guilt, and divine punishment, Lear's madness is staged as "natural," as psychologically engendered, and as obsessed with secular revenge and justice. It is rooted in obvious physical and psychological causes: his exposure to the cold and storm in old age, his mistaken banishment of Cordelia, his other daughters' betrayals, his encounter with Poor Tom. His alienation is rendered on a continuum with his sanity from which it gradually emerges. He is metaphorically described by Kent as "mad" in the first scene, notes the onset of delirium himself, specifies his malady with medical precision as "hysterica passio"—the fits of the mother, defined, ingeniously, as his rising heart rather than his wandering womb (2.4.55-56). As he loses control of his children and his kingdom, he feels weak, vulnerable, a victim of feminine and feminizing hysteria.⁵³ But once he is beside himself, his madness grows more aggressively satiric. He is restored to sanity by conventional remedies, conventionally applied by a doctor—herbal medicine, sleep, clean garments, music, and the presence of Cordelia.

The construction of Lear's mad discourse, like that of Ophelia's, involves fragmentation, formula, depersonalization, the intersection of communal voices, and secularized ritual. Like Ophelia, he uses tags of social formulas incongruously: "We'll go to supper i' th' morning," "Give the word," "Pull off my boots: harder, harder: so" (3.6.83; 4.6.92, 173). But more often, rather than being transected by quoted voices, Lear envisages hallucinatory cultural dramas in which he is both narrator and participant. Whereas Poor Tom acts out guilt by presenting himself as poor and persecuted, Lear defends himself against guilt by acting as prosecutor: "cry / These dreadful summoners grace" (3.2.58-59). His hallucinations of the rituals of secular trial and judgment expose their fraudulence. His scenarios expose civil punishment as fraudulent just as Edgar's Poor Tom role implicitly exposes demonic punishment as fraud. In the enacted mock trial (found only in Quarto Lear), Lear plays the judge who will "arraign" (3.6.20) his absent daughters, Goneril and Regan, for their crimes against him while Edgar, Kent, and the Fool serve as jury. But the ritual, like those in Ophelia's songs, is aborted, and the judge humiliated, barked at by dogs (11. 61-62).

During Lear's encounter with Gloucester in 4.6, his identification with the prosecutor can no longer protect him; he is given fantasy scenarios of justice undone by the corruption of female sexuality and the complicity of the judge. In his first fantasy Lear as judge will "pardon that man's life" because all are guilty of copulation centered in the "sulphurous pit" of female sexuality, the domain to which the fiend is metaphorically confined in Lear's discourse (4.6.126-29). Whereas Edgar's feigned supernatural madness locates lust in himself—"[I] served the lust of my mistress' heart, and did the act of darkness with her" (3.4.85-87)—Lear's natural madness displaces it onto women and their judges. In Lear's second fantasy, following a series of reversals, the punisher and the punished become indistinguishable: the constable who whips the whore "hotly lusts to use her in that kind" for which he whips her (4.6.162). These fantasies simultaneously expose Lear's own habit of persecuting others to conceal his own guilt and provide a critique of the operations of a class-determined system of justice. Social status and the costumes that the period prescribed to mark it control guilt, judgment, and punishment: "Through tattered clothes small vices do appear; / Robes and furred gowns hide all. Plate sin with gold, / And the strong lance of justice hurtless breaks; / Arm it in rags, a pygmy's straw does pierce it" (11. 164-67). With justice presented, like the theater, as a matter of costumes, its fraudulent nature is revealed.

The impertinent madness of Lear, like that of Edgar and the Fool, serves, as Robert Weimann suggests, to provide satiric "disenchantment" of conservative values and hierarchies supported by those in power: "The Prince of Darkness is a gentleman" (3.4.141). Ophelia's madness, although Weimann does not discuss it, functions similarly to disenchant domestic values: she "marks" the falsehood of love, the emptiness of religious formulas, the betrayal of men. She narrates the arbitrariness, instability, and corruption of love and the family as Lear narrates those of justice and the state.⁵⁴ But the theatrical, fragmented, and psychologized discourse of madness, while it allows these critiques, also italicizes and distances them.

Edgar in disguise not only provides critique and counterpoint but is the vehicle of another inversion as he becomes a "philosopher" to King Lear and caretaker for his father, Gloucester. With each, Edgar employs a traditionally recommended remedy for delusion and despair, a strategy that Burton and others record and which Foucault calls "continu[ing] the delirious discourse."⁵⁵ In this strategy the delusions of the mad are complied with and extended through theatrical representation in order to undo them. This strategy further naturalizes madness and brings it under human control while testifying to the real power of theatrical illusion and the longstanding awareness of the theatricality of madness. Friends fraudulently extend the delusions of the mad to manipulate them toward a cure. The most frequently cited example of this is a story of a melancholic man who, believing himself dead, refused to eat. Friends costumed themselves as dead men and consumed a banquet in front of him to demonstrate that the dead eat; he then ate too and recovered. A more bizarre example is that of a man who refused to urinate, believing that if he did, he would drown the world; friends set fire to the house next door and prevailed on him to put it out lest the town burn. "So he pissed and was by that means preserved."⁵⁶ Less ingenious strategies involve physicians or friends curing patients who complain of toads or snakes in their bellies by administering emetics and slipping the animals into the vomit basin. Similarly, when Lear imagines himself barked at by dogs, Edgar exorcises them for him through a song in which he impersonates a dog (3.6.64-72). Later he more elaborately "trifle[s]" with his father's "despair" to "cure" it, engineering Gloucester's mock suicide and the mock exorcism of his (and Edgar's own) demons to save his father from actual suicide. In this performance of possession and exorcism, the rituals of the supernatural are appropriated and secularized, and used by humans to reverse human self-alienation just as they are in Renaissance treatises on melancholy, medicine, exorcism, and witchcraft.

Edgar's uses of the illogic of madness in the service of logic and sanity, like Feste's claims that he but reads madness to exonerate himself from the charge of being mad, demonstrate how the purpose of reading madness, propounding definitions, and prescribing cures is usually to dissociate oneself from the condition and to regulate its disruptiveness. In these Shakespeare tragedies, as in the treatises and the medical practices, the representation of madness permits a restoration of normality, a restoration in which madmen and mad-women participate differently. The disguise of Poor Tom is abandoned, Gloucester eschews suicide, and Lear is returned to sanity. The mad women characters in tragedy, however, are not cured but eliminated. Ophelia is reabsorbed into cultural norms by her narrated drowning and her Christian burial. The report of Lady Macbeth's suicide, abruptly announced in the play's final lines, reduces the supernatural to a simile to vilify and dismiss her as a "fiendlike queen, / Who, as 'tis thought, by self and violent hands / Took off her life" (5.8.69-71).

Likewise, in the culture, constructions of madness tended to support established institutions in preserving the status quo. Preferred treatments were those undertaken by Anglican ministers, not Catholic exorcists or Puritan enthusiasts, by licensed practitioners, not quacks. These practitioners tended to favor outcomes that sustained conventional social hierarchies, and these too had different impacts on men and women; Napier, for example, viewed wives who wanted to leave brutal husbands, children who resisted their parents, servants who did not obey their masters, as mentally unstable and was severe with them. But the mad could be recuperated because they were not seen as inhuman; hence they were not usually isolated, confined, or ostracized. They might be subjected to purges and bleeding (like all ill people), drugged sleep or music therapy, or might be coaxed back, through their own delusions, into the rituals of everyday life. Such treatments, however, did not yet segregate them from human community as did the eighteenth century's institutionalization, the nineteenth century's "moral treatment," or the twentieth century's romanticization or pharmacological normalization.

If the discourse of madness, in the short run, promoted normalization and supported the status quo, in the long run it had the capacity to contribute to changing constructions of the human and hence to cultural change. The distinctions established in this discourse helped redefine the human as a secular subject, cut off from the supernatural and incomprehensibly unstable and permeable, containing in itself a volatile mix of mind and body, of warring and turbulent elements: "For seeing we are not maisters of our owne affections, wee are like battered Citties without walles, or shippes tossed in the Sea, exposed to all maner of assaults and daungers, even to the overthrow of our owne bodies."⁵⁷ Such images opened up a new range of questions about and possibilities for human beings.

The theater, by representing and disseminating madness, contributed to its changing constructions and its destabilizing potential. Shakespearean tragedy, drawn to madness perhaps because of its inherent theatricality, represented madness by a conventionalized speech that was successful (and imitated) by virtue of its excessiveness, its rich imagery and associations, its verbal inventiveness, its multiple functions: psychological, thematic, satiric, theatrical. By providing a language for madness, the theater contributed to the process whereby it was becoming a secular, medical, and gendered condition. The Elizabethan theater is, at its origin, as C. L. Barber has suggested, a place apart, a space where the sacred is reconstituted in the human,⁵⁸ and madness is, as we have seen, one place where this reconstitution is especially apparent. The secular human characters this stage represents are inevitably gender- and class-specific in ways that the hierarchical "dramatis personae" or "names of the actors," introduced in seventeenth-century editions, inscribe. Gender distinctions may be especially rigid because of the absolute division between adult actors who play men and boy actors who must self-consciously perform femininity, drawing on gender stereotypes to do so—as the instructions to the Page in the Induction of The Taming of the Shrew suggest. This may be one reason why madness shows signs of gender-markings in the theater earlier than in medical treatises or in the visual arts. Even while representing stereotyped or conservative formations, the theater may participate in change. As Steven Mullaney has shown, it is a place apart from the established state as well as from the established church, situated in the Liberties alongside unruly neighbors: taverns, bearbaitings, brothels, and the empty leper houses that Foucault (wrongly) imagines will soon fill up again with madmen.⁵⁹ By constructing a language through which madness can be represented, the popular theater facilitated the circulation of the discourse; by italicizing the language of madness, it encouraged its interrogation and transformation.

Although these Shakespeare plays represent madness as a condition to treat, italicize, or eliminate, and although the gender distinctions they initiate can still prove oppressive to women, their representations of madness can be vehicles for social critique achieved through unsettling productions or indecorous interventions by performers. Hamlet's feigned madness and Lear's natural madness can be performed and read as social critique (as in Grigori Kozintsev's 1970 film of King Lear or in the Studio Theater of Moscow's 1989 production of Hamlet). Ophelia's madness can be politicized by an actress who might represent the hysterical female body now as an eroticized and aestheticized object of desire and repulsion and now as an agent of uncontrollable voice, desire, pain, and rage (as in Ange Magnetic's "Ophelie Song" [1989], an "opera minimal" derived from Ophelia's songs).⁶⁰

The complexities of reading the discourse of madness in Shakespeare and his culture reveal the difficulty and necessity of historicizing: that is, of trying to understand one's own position and that of one's subject(s) in today's culture in relation to the construction of the subject(s) that emerged in early modern culture, of trying to tease out disjunctions and connections. In particular this project reveals that the shape of gender difference cannot be assumed but must always be reformulated in specific cultural and historical contexts. Reading the discourse of madness provides powerful lessons in the gradual and erratic progress of cultural change and in the complex and not fully retrievable interactions between dramatic texts and other cultural documents. The theater does not just reflect, contain, or subvert the cultural realities in which it is embedded. But finding the right metaphor for the relationship is hard. Perhaps, in the context of this essay, it is appropriate to note that the playwright, like the mad, expresses inner conflicts, quotes cultural voices, speaks through disguises, enacts emotions visually and verbally, performs for diverse audiences, and is protected from harm because playtexts are illusions. These playtexts, moreover, like other "documents in madness," both do and do not belong to the authors who generate them, and they are read, performed, and used by others in the service of their own sanity.

Notes

¹ All citations of Shakespeare plays will appear in text and refer to The Complete Signet Classic Shakespeare, Sylvan Barnet, gen. ed. (New York: Harcourt, Brace, Jovanovich, 1972).

²Writing and Madness (Literature/Philosophy/Psychoanalysis), trans. Martha Noel Evans and Shoshana Felman (Ithaca, N.Y.: Cornell Univ. Press, 1985), p. 11.

³ W. F. Bynum, Roy Porter, and Michael Shepherd, eds., The Anatomy of Madness: Essays in the History of Psychiatry, 3 vols. (London: Tavistock, 1985), Vol. 1, p. 7.

⁴ Michael MacDonald, Mystical Bedlam: Madness, Anxiety, and Healing in Seventeenth-Century England (Cambridge: Cambridge Univ. Press, 1981), p. 1.

⁵ Andrew Scull, Social Order/Mental Disorder: Anglo-American Psychiatry in Historical Perspective (Berkeley: Univ. of California Press, 1989), p. 8.

⁶ See Patricia Allderidge, "Management and mismanagement at Bedlam, 1547-1633" in Health, medicine, and mortality in the sixteenth century, ed. Charles Webster (Cambridge: Cambridge Univ. Press, 1979), pp. 141-64, esp. pp. 153, 143. This essay and a subsequent one by Allderidge, "Bedlam: fact or fantasy?" in Anatomy of Madness, Vol. 2, pp. 17-33, correct the inaccuracies and fantasies of Bedlam scholarship, especially those of the standard history, E. G. O'Donoghue, The Story of Bethlehem Hospital: From its Foundation in 1247 (New York: Dutton, 1914). Although many Bedlam inmates were released, some were incarcerated for periods of twenty years or more, and numbers and turnover were small; there could have been few actual Tom o' Bedlams wandering the countryside.

⁷ The OED records this shift in a cautionary paragraph following the first definition of mad ("Suffering from mental disease; beside oneself; out of one's mind; insane, lunatic"), which prescribes: "The word has always had some tinge of contempt or disgust and would now be quite inappropriate in medical use or in referring sympathetically to an insane person as the subject of an affliction." Insane, from the Latin root insanus, means not sound, not healthy, not curable, and does not come into widespread use until the eighteenth century, when it appears first in medical and legal contexts. Madness, the earlier term, is not the opposite of not-mad but on a continuum with it.

⁸ Szasz, The Myth of Mental Illness: Foundations of a Theory of Personal Conduct (New York: Harper and Row, 1974); Laing, The Divided Self: A study of sanity and madness (London: Tavistock, 1960).

⁹ p. 12.

¹⁰ Michel Foucault, Madness and Civilization: A History of Insanity in the Age of Reason, trans. Richard Howard (New York: Tavistock, 1967), and Elaine Showalter, The Female Malady: Women, Madness, and English Culture, 1830-1980 (New York: Pantheon, 1985). Showalter looks only at women's experience of madness and only after 1830, and the category of gender is missing from Foucault's large intuitive canvas. The discussion of the period from the Middle Ages to the end of the seventeenth century is the most sketchy and least supported part of his book (at least in the English translation), for his concept of the modern centralized state does not make sense of early modern institutions. Mental institutions like Bedlam often developed early out of medieval hospitals; unlike leper houses, they attempted cures and declared patients recovered. Confinement of the mad is also more varied, more historically continuous, and more complicated in its representations, aims, and consequences than Foucault or Showalter allows. But Foucault's intuitions about the transformation of the madman from supernatural voyager to secular case study are useful, as are Showalter's analyses of the associations among women, madness, and sexuality which developed in representations of madwomen. For criticism of Foucault by an historian, see H. C. Erik Midelfort, "Madness and Civilization in Early Modern Europe: A Reappraisal of Michel Foucault" in After the Reformation: essays in honor of J. H. Hexter, Barbara C. Malament, ed. (Philadelphia: Univ. of Pennsylvania Press, 1980), pp. 247-66. For criticism of Showalter by an historian of medicine, see Nancy Tomes, "Historical Perspectives on Women and Mental Illness" in Women, Health, and Medicine in America: A Historical Handbook, Rima D. Apple, ed. (New York: Garland, 1990), pp. 143-71. I am grateful to Nancy Tomes for allowing me to read her review essay in manuscript form before its publication.

¹¹ See especially pages 45-55 in Judith S. Neaman, Suggestion of the Devil: The Origins of Madness (New York: Anchor Books, 1975), a study of the medical, theological, legal, and social contexts of madness in the Middle Ages. See also and

¹² In the original edition of Bright (London: Thomas Vautrollier, 1586), as in the 1969 Theatrum Orbis Terrarum facsimile, this chapter, number 36, pages 207-42, is misnumbered chapter 30.

¹³ Jorden's treatise (London: Iohn Windet, 1603) is available in a 1971 facsimile reprint from Theatrum Orbis Terrarum; quotation at fol. C1^r. See Michael MacDonald's reprint of Jorden's pamphlet in Witchcraft and Hysteria in Elizabethan London: Edward Jorden and the Mary Glover Case (London: Routledge, 1991).

¹⁴ Keith Thomas, Religion and the Decline of Magic (New York: Scribners, 1971), pp. 451-52.

¹⁵ fol. B1^r.

¹⁶ fols. F4^r, G3^r, F4^v-Gl^r, H1^r. Jorden is the first to find the source of hysterical symptoms in the brain as well as in the uterus. See Ilza Veith, Hysteria: The History of a Disease (Chicago: Univ. of Chicago Press, 1965), pp. 122-23.

¹⁷The Anatomy of Melancholy, What it is. With all the Kindes, Causes, Symptomes, Prognostickes, and Severall Cures of it, 4th ed. (Oxford: J. Lichfield, 1632), pp. 202, 204. In fact, according to MacDonald's statistics, although far larger numbers and percentages of women came to Napier to report distress in connection with courtship, love, sex, and marriage negotiations, most of these sufferers were untitled (Mystical Bedlam, Table 3.6, p. 95; see also ). Perhaps aristocratic women suffered less stress in matters of courtship and marriage because they had little or no choice in the matter.

¹⁸ For discussion of the political climate that produced Jorden's and Harsnett's pamphlets in 1603, see Thomas, pp. 482-86; Stephen Greenblatt, "Shakespeare and the Exorcists" in Shakespearean Negotiations: The Circulation of Social Energy in Renaissance England (Berkeley: Univ. of California Press, 1988), pp. 94-128; MacDonald, Witchcraft, pp. vii-lxiv.

¹⁹ Two other plays of the period that contain scenes in which characters undergo a test for madness are Dekker's The Honest Whore, Part I (1604) and Middleton and Rowley's The Changeling (c. 1623).

²⁰ Ascriptions of madness occur elsewhere in Shakespeare, beginning with Titus Andronicus, The Comedy of Errors, and Twelfth Night and concluding with the extended portrait of the Jailer's Daughter in The Two Noble Kinsmen. Her characterization has connections with Ophelia's and with that of the madwomen and groups of madpersons in other Jacobean plays, for example, Dekker's Honest Whore, Part I, Webster's Duchess of Malfi, Fletcher's The Pilgrim, Middleton and Rowley's The Changeling. Such representations will be the subject of another essay.

²¹ Robert Rentoul Reed, Jr., Bedlam on the Jacobean Stage (Cambridge, Mass.: Harvard Univ. Press, 1952); Lawrence Babb, The Elizabethan Malady: A Study of Melancholia in English Literature from 1580-1642 (East Lansing: Michigan State Univ. Press, 1951); and Bridget Gellert Lyons, Voices of Melancholy: Studies in literary treatments of melancholy in Renaissance England (London: Routledge, 1971). The discussion closest to mine is Lillian Feder's analysis of Lear's madness in Madness in Literature (Princeton, N.J.: Princeton Univ. Press, 1980), p. 6 and pp. 119-46.

²²Shakespearean Tragedy: Lectures on Hamlet, Othello, King Lear, Macbeth (New York: St. Martin's Press, 1985), pp. 132-33. As a result of this attitude, Bradley did not give Ophelia's mad scenes the detailed analysis that he is elsewhere known for.

²³ These interpretations of Ophelia replicate feminist theorists' polarized interpretations of the association between women and madness. For positive readings of the textual representations of the connection, see Sandra Gilbert and Susan Gubar, The Madwoman in the Attic (New Haven: Yale Univ. Press, 1979); of Ophelia, see Carol Thomas Neely, Broken Nuptials in Shakespeare's Plays (New Haven: Yale Univ. Press, 1985), pp. 103-4. For the negative aspects of the connection, see Showalter, The Female Malady, and for an extended discussion of representations of Ophelia, see Showalter, "Representing Ophelia: women, madness, and the responsibilities of feminist criticism" in Shakespeare and the Question of Theory, Patricia Parker and Geoffrey Hartman, eds. (London: Methuen, 1985), pp. 77-94. Showalter discusses how different periods represent Ophelia according to their stereotypes of female insanity.

²⁴ See, among many examples, Robert Bechtold Heilman, This Great Stage: Image and Structure in King Lear (Baton Rouge: Louisiana State Univ. Press, 1948), pp. 173-223; Paul A. Jorgensen, Lear's Self Discovery (Berkeley: Univ. of California Press, 1967), pp. 78-82.

²⁵ Cavell, Disowning Knowledge In Six Plays of Shakespeare (Cambridge: Cambridge Univ. Press, 1987), pp. 39-124, esp. pp. 50, 74, and 77); Greenblatt (cited in n. 18, above), pp. 94-128; Dollimore, Radical Tragedy: Religion, Ideology and Power in the Drama of Shakespeare and his Contemporaries (Chicago: Univ. of Chicago Press, 1984), p. 193.

²⁶ For Foucault, language constitutes madness; "Language is the first and last structure of madness" (p. 100 [cited in n. 10, above]). Since madness is unreason, the "delirious discourse" (p. 99) that constitutes it is the inverse of reason but, in effect, identical with it. It involves "sedimentation in the body of an infinitely repeated discourse" (p. 97), "the language of reason enveloped in the prestige of the image" (p. 95). "It is in this delirium, which is of both body and soul, of both language and image, of both grammar and physiology, that all the cycles of madness conclude and begin" (pp. 100-101).

²⁷ I take the notion of italicized writing from Nancy K. Miller, "Emphasis Added: Plots and Plausibilities in Women's Fiction" in The New Feminist Criticism: Essays on Women, Literature, and Theory. Elaine Showalter, ed. (New York: Pantheon, 1985), pp. 339-60. She extends Luce Irigaray's analysis of women's special relation to the mimetic (in This Sex Which Is Not One) and defines italics as a modality of intensity, intonation, and emphasis that characterizes women's writing (p. 343).

²⁸ A. C. Bradley notes, in Shakespearean Tragedy, that Shakespeare invariably uses prose to represent abnormal states of mind like madness or Lady Macbeth's somnambulism (pp. 335-37). I am indebted to Lars Engle for bringing this discussion to my attention.

²⁹ The phrase "document in madness" occurs at 4.5.178. Other mad characters are given equally precise and explicit introductions: see the conversation between Lady Macbeth's waiting woman and the doctor (5.1.1-20) and Edgar's commentary as he disguises himself as Poor Tom in Lear (2.3.1-21).

³⁰ Joan Klein, "'Angels and Ministers of Grace': Hamlet, TV, v-vii," Allegorica, 1, 2 (1976), 156-76, reads Ophelia's madness closely and attends to the cultural lore that she draws on. But whereas she sees Ophelia's role as providential, as a minister to Hamlet, I see religious references as split off from their theological context in her mad speech. Much of the attention devoted to Ophelia's speeches has been directed toward identifying the referents of her songs, especially the "truelove," and determining to which characters the songs are addressed. My analysis suggests that it is not possible to pinpoint a single referent or audience since the discourse's referents are multiple and are both personal and cultural. See Peter J. Seng, The Vocal Songs in the Plays of Shakespeare: A Critical History (Cambridge, Mass.: Harvard Univ. Press, 1967), pp. 131-56, for a summary of commentary.

³¹ Cf. Showalter, Female Malady (cited in n. 10, above).

³² Joan Klein sees Ophelia as Hamlet's surrogate and minister, and Lyons sees her as mirroring aspects of Hamlet's melancholy (pp. 11-12), but I see her as a "dark double" who, in Gilbert and Gubar's sense, acts out what is repressed in Hamlet.

³³ Some form of Christian burial might be possible, even in cases of suicide; cf. Michael MacDonald, "Ophelia's Maimèd Rites," SQ, 37 (1986), 309-17, esp. pp. 314-15. For other discussions of suicide, see Michael MacDonald, Mystical Bedlam (cited in n. 4, above), pp. 132-38; "The Inner Side of Wisdom: Suicide in Early Modern England," Psychological Medicine, 7 (1977), 565-82, esp. pp. 566-67; "The Secularization of Suicide in England 1660-1800," Past and Present, 111 (May 1986), 52-70; see also

³⁴ MacDonald, "Ophelia's Maimèd Rites," p. 311, and "Inner Side of Wisdom," p. 567.

³⁵ Immersion is both conventional to the iconography of madness and a traditional cure for it. Cf. Foucault (cited in n. 10, above), pp. 162, 166; Basil Clarke, Mental Disorder in Earlier Britain: Exploratory Studies (Cardiff: Univ. of Wales Press, 1975), pp. 229-30.

³⁶"Macbeth and witchcraft" in Focus on Macbeth, John Russell Brown, ed. (London: Routledge, 1982), pp. 189-209.

³⁷ I see the relationship between the witches and Lady Macbeth as more ambiguous and unstable than does Janet Adelman ("'Born of Woman': Fantasies of Maternal Power in Macbeth" in Cannibals, Witches, and Divorce: Estranging the Renaissance, Marjorie Garber, ed. [Baltimore: Johns Hopkins Univ. Press, 1987], pp. 90-121). I do not see their relationship as an "alliance" (pp. 97, 98) either literal or symbolic, nor do I find the witches or Lady Macbeth as unstintingly malevolent or powerful as Adelman does. In fact the witches wish Macbeth to fail while Lady Macbeth wishes him to succeed, and their relation to the supernatural is quite different from hers. Both the witches and Lady Macbeth lose what power they have by the end of the play, though Adelman never discusses the implications of Lady Macbeth's somnambulism and suicide. Whatever power each has exists only contingently; neither the witches nor Lady Macbeth have agency or control except through Macbeth.

³⁸Witchcraft in Tudor and Stuart England: A regional and comparative study (London: Routledge, 1970).

³⁹ Thomas, in chapter 14 of his book (cited in n. 14, above), discusses how Continental views of witchcraft, conceived as a heresy marked especially by a pact with the devil, were only gradually and incompletely filtered into England, where witchcraft was defined more usually as harmful activities. The fact that the witches in Macbeth are also called "weird women" (3.1.2) and compared with "elves and fairies" (4.1.42) emphasizes their shifting representations. If Hecate and the songs from Middleton's The Witch were later interpolations, somewhat at odds with the earlier portrayal of the witches, this strengthens my claim that the witches are ambiguously portrayed, reflecting the conflicting ideas about witches in the period. For arguments that 3.5.39-43 and 4.1.125-32 are interpolations, see Macbeth, The Arden Shakespeare, ed. Kenneth Muir (London: Methuen, 1951), pp. xxxv-xxxviii. That the witches are dramatically more powerful early in the play when presented more naturalistically may also be connected to the weakening of beliefs in possession and witchcraft in England.

⁴⁰ MacDonald, Mystical Bedlam, pp. 189-217. John Hall, a successful doctor who practiced at the same time as Napier (1600-1635) in nearby Warwickshire and who appears to have been more Puritan in his religious beliefs, and more of an apothecary and less of an astrologer than Napier, treated a similar range of disorders. Analysis of his casebooks shows that his patients presented similar symptoms of mental disorder in similar ratios. In his published cases (included in Harriet Joseph, Shakespeare's Son-in-law: John Hall, Man and Physician [Hamden, Conn.: Archon Books, 1964]), Hall treated 70 men and 109 women; 13 of the men (or 7 percent) and 39 of the women (or 22 percent) showed signs of emotional disorder as analyzed in John G. Howells and N. Livia Osborn, "The Incidence of Emotional Disorder in a Seventeenth-Century Medical Practice," Medical History, 14 (1970), 192-98. These figures are based on only a small sample of Hall's cases, which were published to disseminate his recipes for purges, not to explicate his patients' symptoms. "Emotional disorder" is somewhat more broadly defined by Howells and Osborn than by MacDonald.

⁴¹ MacDonald, Mystical Bedlam, pp. 35-40, 72-75.

⁴² MacDonald, Mystical Bedlam, pp. 35-40; Tomes (cited in n. 10, above), pp. 145-46. For a discussion of the gender distribution of psychiatric illnesses in twentieth-century London, see Michael Shepherd, Brian Cooper, Alexander Brown, and Graham Kalton, Psychiatric Illnesses in General Practice (London: Oxford Univ. Press, 1966), pp. 164-66; for American statistics, see essays cited in note 44, below. The self-reporting and diagnosis of women's mental distress depend on a difficult-to-unravel conjunction of factors including vulnerability to gynecological ailments, women's self-images, gender-role socialization, medicine's construction of diseases, the nature of diagnoses, and wider cultural trends.

⁴³ Tomes, pp. 146-47, and her numerous sources, especially Noreen Goldman and Renee Ravid, "Community Surveys: Sex Differences in Mental Illness," and Deborah Belle and Noreen Goldman, "Patterns of Diagnoses Received by Men and Women," both in The Mental Health of Women, Marcia Guttentag, Susan Salasin, and Deborah Belle, eds. (New York: Academic Press, 1980), pp. 31-55, 21-30.

⁴⁴ Cf. MacDonald, Mystical Bedlam, pp. 243-45.…

⁴⁵ p. 244.

⁴⁶ The two Bedlam censuses I cite are reproduced in Allderidge, "Management" (cited in n. 6, above), pp. 152-53, 158-60.

⁴⁷ The removal of more men may merely indicate that the distribution of space in the facility makes the confinement of similar numbers of men and women patients a convenience; hence more men are designated removable. I cannot tell whether Bedlam was sex-segregated as some later asylums were.

⁴⁸ This use of madness as disguise derives perhaps from Kyd's Spanish Tragedy and is common in other Jacobean plays, for example The Changeling and The Pilgrim. William C. Carroll, "'The Base Shall Top Th'Legitimate': The Bedlam Beggar and the Role of Edgar in King Lear" SQ, 38 (1987), 426-41, analyzes the period's identifications of Tom o'Bedlams as feigning, lower-class con men. While this may not be the only Poor Tom stereotype, it does add associations with feigning at another level to Edgar's role-playing.

⁴⁹ This is the point developed by Greenblatt in "Shakespeare and the Exorcists" (cited in n. 18, above). Kenneth Muir, "Samuel Harsnett and King Lear," Review of English Studies, 2 (1951), 11-21, finds over fifty separate fragments from Harsnett embedded in the play, many of them connected with the role of Poor Tom.

⁵⁰ Janet Adelman, in her introduction to Twentieth Century Interpretations of King Lear (Englewood Cliffs, N.J.: Prentice Hall, 1978), pp. 1-21, has a fine discussion of the role and language of Poor Tom and the ways in which this disguise allows Edgar to protect and preserve himself. In contrast, William Carroll sees the Poor Tom disguise as a source of pain and suffering for Edgar as well as a release from them (p. 436).

⁵¹ Although my students have long been unable to identify this moment and have refused to accept it as marking a decisive break with sanity, Lillian Feder (p. 132 [cited in n. 21, above]) and Paul Jorgensen (p. 80 [cited in n. 24, above]) concur. The definitiveness of Lear's delusion is emphasized by his four-times-repeated claim that Tom's daughters are to blame for his state: "Didst thou give all to thy daughters?" "What, has his daughters brought him to this pass?" "Now … plagues … light on thy daughters!" "Nothing could have subdued nature / To such a lowness but his unkind daughters" (3.4.48, 62, 66-67, 69-70). This theatrical moment manifests one of the places where the boundary between sanity and madness was defined and crossed.

⁵² Greenblatt, pp. 117, 119.

⁵³ Coppélia Kahn, "The Absent Mother in King Lear" in Rewriting the Renaissance: The Discourses of Sexual Difference in Early Modern Europe, Margaret W. Ferguson, Maureen Quilligan, and Nancy J. Vickers, eds. (Chicago: Univ. of Chicago Press, 1986), pp. 33-49, argues that Lear's madness results from his rage at maternal deprivation and that it enables him eventually to accept his own vulnerability. While this argument seems partly valid, I see both the causes and uses of Lear's madness as more complicated.

⁵⁴ Weimann uses the range and scope of Hamlet's and Lear's mad speech to exemplify the flexible alternation possible in Renaissance popular theater between the illusionistic locus position, staging dialogue of the psychologically naturalistic character, and the nonillusionistic platea position, staging monologue which draws on popular tradition, induces audience identification, and permits social critique (Shakespeare and the Popular Tradition in the Theater: Studies in the Social Dimension of Dramatic Form and Function, ed. Robert Schwartz [Baltimore: Johns Hopkins Univ. Press, 1978], pp. 120-35 and 215-20). This flexibility also reveals "the twofold function of mimesis ('enchantment' and 'disenchantment'), which we have seen to be so fundamental a part of traditional popular drama" (p. 132). More recently, in "Bifold Authority in Shakespeare's Theatre," SQ, 39 (1988), 401-17, Weimann again uses the "impertinent" language of Hamlet and Lear to define the bifold authority generated by the language and play space of the Elizabethan theater (pp. 410, 416). This highly particularized form of discourse perhaps cannot stand as the theatrical norm, but Weimann's analysis does get at the combination of individual psychology and cultural discourse that I argue characterizes this speech. Although Weimann (curiously) does not discuss Ophelia's madness, it functions in many of the same ways. She too speaks impertinently, proverbially, bawdily, disturbingly; she too is both actress and character, partly an object of the audience's gaze, partly a spokesperson for their contempt for Claudius and his court. Ophelia, as much as (or perhaps even more than) Lear, "disrupts the authority of order, degree, and decorum" ("Bifold Authority," p. 417).

⁵⁵ p. 188 (cited in n. 10, above).

⁵⁶ Clarke (cited in n. 35, above), p. 226, quoting Du Laurens. He describes such ingenious cures as part of "the folk-lore tradition of the profession" (p. 222). He discusses (pp. 222-23, 226) cases cited by Levinus Lemnius in The Touchstone of Complexions (trans. Thomas Newton [London: Thomas Marsh, 1576], pp. 150^v-52^r) and by M. Andreas Du Laurens in A Discourse of the Preservation of the Sight: of Melancholike Diseases … (trans. Richard Surphlet [London: Felix Kyngston, 1599], pp. 100-40). See also and

⁵⁷ Jorden, fol. G2^V.

⁵⁸ With Richard P. Wheeler, The Whole Journey: Shakespeare's Power of Development (Berkeley: Univ. of California Press, 1986), pp. 20 ff.

⁵⁹The Place of the Stage: License, Play, and Power in Renaissance England (Chicago: Univ. of Chicago Press, 1988), chap. 2; Foucault, pp. 3-7.

⁶⁰ The Studio Theater performed this Hamlet at the University of Illinois at Urbana-Champaign, 12 February 1989. "Ophelie Song" was a co-production by Ange Magnetic and Mon Oncle d'Amerique, collaborated on by French director Antoine Campo and American choreographer Clara Gibson Maxwell and produced in 1989 in Paris, in New York, and at the Edinburgh Fringe Festival.