But medications do not always behave as we expect them to.
     Two weeks after starting Prozac, Tess appeared at the office to say she was no longer feeling weary. In retrospect, she said, she had been depleted of energy for as long as she could remember, had almost not known what it was to feel rested and hopeful. She had been depressed, it now seemed to her, her whole life. She was astonished at the sensation of being free of depression.
     She looked different, at once more relaxed and energetic--more available--than I had seen her, as if the person hinted at in her eyes had taken over. She laughed more frequently, and the quality of her laughter was different, no longer measured but lively, even teasing.
     With this new demeanor came a new social life, one that did not unfold slowly, as a result of a struggle to integrate disparate parts of the self, but seemed, rather, to appear instantly and full-blown.
     "Three dates a weekend," Tess told me. "I must be wearing a sign on my forehead!"
     Within weeks of starting Prozac, Tess settled into a satisfying dating routine with men. She had missed out on dating in her teens and twenties. Now she reveled in the attention she received. She seemed even to enjoy the trial-and-error process of learning contemporary courtship rituals, gauging norms for sexual involvement, weighing the import of men's professed infatuation with her.
     I had never seen a patient's social life reshaped so rapidly and dramatically. Low self-worth, competitiveness, jealousy, poor interpersonal skills, shyness, fear of intimacy--the usual causes of social awkwardness--are so deeply ingrained and so difficult to influence that ordinarily change comes gradually if at all. But Tess blossomed all at once.
     "People on the sidewalk ask me for directions!" she said. They never had before.
     The circle of Tess's women friends changed. Some friends left, she said, because they had been able to relate to her only through her depression. Besides, she now had less tolerance for them. "Have you ever been to a party where other people are drunk or high and you are stone-sober? Their behavior annoys you, you can't understand it. It seems juvenile and self-centered. That's how I feel around some of my old friends. It is as if they are under the influence of a harmful chemical and I am all right--as if I had been in a drugged state all those years and now I am clearheaded."
     The change went further: "I can no longer understand how they tolerate the men they are with." She could scarcely acknowledge that she had once thrown herself into the same sorts of self-destructive relationships. "I never think about Jim," she said. And in the consulting room his name no longer had the power to elicit tears.
     This last change struck me as most remarkable of all. When a patient displays any sign of masochism, and I think it is fair to call Tess's relationship with Jim masochistic, psychiatrists anticipate a protracted psychotherapy. It is rarely easy to help a socially selfdestructive patient abandon humiliating relationships and take on new ones that accord with a healthy sense of self-worth. But once Tess felt better, once the weariness lifted and optimism became possible, the masochism just withered away, and she seemed to have every social skill she needed.

Tess's work, too, became more satisfying. She responded without defensiveness in the face of adamant union leaders, felt stable enough inside herself to evaluate their complaints critically. She said the medication had lent her surety of judgment; she no longer tortured herself over whether she was being too demanding or too lenient. I found this remark noteworthy, because I had so recently entertained the possibility that unconscious inner conflicts were hampering Tess in her dealings with the labor union. Whether the conflicts were real or illusory, the problem disappeared when the medication took effect. "It makes me confident," Tess said, a claim I since have heard from dozens of patients, none of whom had been given a hint that this medication, or any medication, could do any such thing.
     Tess's management style changed. She was less conciliatory, firmer, unafraid of confrontation. As the troubled company settled down, Tess was given a substantial pay raise, a sign that others noticed her new effectiveness.
      Tess's relations to those she watched over also changed. She was no longer drawn to tragedy, nor did she feel heightened responsibility for the injured. Most tellingly, she moved to another nearby town, the farthest she had ever lived from her mother.
     Whether these last changes are to be applauded depends on one's social values. Tess's guilty vigilance over a mother about whom she had strong ambivalent feelings can be seen as a virtue, one that medication helped to erode. Tess experienced her "loss of seriousness, " as she put it, as a relief. She had been too devoted in the past, at too great a cost to her own enjoyment of life.
     In time, Tess's mother was given an antidepressant, and she showed a modest response--she slept better, lost weight, had more energy, displayed a better sense of humor. Tess threw her a birthday party, a celebration of the mother's survival and the children's successes. In addition to the main present, each child brought a nostalgic gift. Tess's was a little red wagon, in memory of a time when the little ones were still in diapers, and the family lived in a coldwater flat, and Tess had organized the middle children to wheel the dirty linens past abandoned tenements to the laundromat many times a week. Were I Tess's psychotherapist, I might have asked whether the gift did not reveal an element of aggression, but on the surface at least the present was offered and received lovingly. In acknowledging with her mother how difficult the past had been, Tess opened a door that had been closed for years. Tess used her change in mood as a springboard for psychological change, converting pain into perspective and forgiveness.

There is no unhappy ending to this story. It is like one of those Elizabethan dramas--Marlowe's Tamburlaine--so foreign to modern audiences because the Wheel of Fortune takes only half a turn: the patient recovers and pays no price for the recovery. Tess did go off medication, after about nine months, and she continued to do well. She was, she reported, not quite so sharp of thought, so energetic, so free of care as she had been on the medication, but neither was she driven by guilt and obligation. She was altogether cooler, better controlled, less sensible of the weight of the world than she had been.
     After about eight months off medication, Tess told me she was slipping. "I'm not myself," she said. New union negotiations were under way, and she felt she could use the sense of stability, the invulnerability to attack, that Prozac gave her. Here was a dilemma for me. Ought I to provide medication to someone who was not depressed? I could give myself reason enough--construe it that Tess was sliding into relapse, which perhaps she was. In truth, I assumed I would be medicating Tess's chronic condition, call it what you will: heightened awareness of the needs of others, sensitivity to conflict, residual damage to self-esteem--all odd indications for medication. I discussed the dilemma with her, but then I did not hesitate to write the prescription. Who was I to withhold from her the bounties of science? Tess responded again as she had hoped she would, with renewed confidence, self-assurance, and social comfort.

I believe Tess's story contains an unchronicled reason for Prozac's enormous popularity: its ability to alter personality. Here was a patient whose usual method of functioning changed dramatically. She
became socially capable, no longer a wallflower but a social butterfly. Where once she had focused on obligations to others, now she was vivacious and fun-loving. Before, she had pined after men; now she dated them, enjoyed them, weighed their faults and virtues. Newly confident, Tess had no need to romanticize or indulge men's shortcomings.
     Not all patients on Prozac respond this way. Some are unaffected by the medicine; some merely recover from depression, as they might on any antidepressant. But a few, a substantial minority, are transformed. Like Garrison Keillor's marvelous Powdermilk biscuits, Prozac gives these patients the courage to do what needs to be done.
     What I saw in Tess--a quick alteration in ordinarily intractable problems of personality and social functioning--other psychiatrists saw in their patients as well. Moreover, Prozac had few immediate side effects. Patients on Prozac do not feel drugged up or medicated. Here is one place where the favorable side-effect profile of Prozac makes a difference: if a doctor thinks there is even a modest chance of quickly liberating a chronically stymied patient, and if the risk to the patient is slight, then the doctor will take the gamble repeatedly.
     And of course Prozac had phenomenal word of mouth, as "good responders" like Tess told their friends about it. I saw this effect in the second patient I put on Prozac. She was a habitually withdrawn, reticent woman whose cautious behavior had handicapped her at work and in courtship. After a long interval between sessions, I ran into her at a local bookstore. I tend to hang back when I see a patient in a public place, out of uncertainty as to how the patient may want to be greeted, and I believe that, while her chronic depression persisted, this woman would have chosen to avoid me. Now she strode forward and gave me a bold "Hello." I responded, and she said, "I've changed my name, you know."
     I did not know. Had she switched from depression to mania and then married impulsively? I wondered whether I should have met with her more frequently. She had, I saw, the bright and open manner that had brought Tess so much social success.
     "Yes," she continued, "I call myself Ms. Prozac."
     There is no Ms. Asendin, no Ms. Pamelor. Those medicines are quite wonderful-they free patients from the bondage of depression. But they have not inspired the sort of enthusiasm and loyalty patients have shown for Prozac.

No doubt doctors should be unreservedly pleased when their patients get better quickly. But I confess I was unsettled by Ms. Prozac's enthusiasm, and by Tess's as well. I was suspicious of Prozac, as if I had just taken on a cotherapist whose charismatic style left me wondering whether her magic was wholly trustworthy.
     The more rational component to my discomfort had to do with Tess. It makes a psychiatrist uneasy to watch a medicated patient change her circle of friends, her demeanor at work, her relationship to her family. All psychiatrists have seen depressed patients turn manic and make decisions they later regret. But Tess never showed signs of mania. She did not manifest rapid speech or thought, her judgment remained sound, and, though she enjoyed life more than she had before, she was never euphoric or Pollyannaish. In mood and level of energy, she was "normal," but her place on the normal spectrum had changed, and that change, from "serious," as she put it, to vivacious, had profound consequences for her relationships to those around her.
     As the stability of Tess's improvement became clear, my concern diminished, but it did not disappear. Just what did not sit right was hard to say. Might a severe critic find the new Tess a bit blander than the old? Perhaps her tortured intensity implied a complexity of personality that was now harder to locate. I wondered whether the medication had not ironed out too many character-giving wrinkles, like overly aggressive plastic surgery. I even asked myself whether Tess would now give up her work in the projects, as if I had administered her a pill to cure warmheartedness and progressive social beliefs. But in entertaining this thought I wondered whether I was clinging to an arbitrary valuation of temperament, as if the melancholy or saturnine humor were in some way morally superior to the sanguine. In the event, Tess did not forsake the projects, though she did make more time for herself.
     Tess, too, found her transformation, marvelous though it was, somewhat unsettling. What was she to make of herself? Her past devotion to Jim, for instance--had it been a matter of biology, an addiction to which she was prone as her father had been to alcoholism? Was she, who defined herself in contrast to her father's fecklessness, in some uncomfortable way like him? What responsibility had she for those years of thralldom to degrading love? After a prolonged struggle to understand the self, to find the Gordian knot dissolved by medication is a mixed pleasure: we want some internal responsibility for our lives, want to find meaning in our errors. Tess was happy, but she talked of a mild, persistent sense of wonder and dislocation.

My discomfort with Tess's makeover had another component. It is all very well for drugs to do small things: to induce sleep, to allay anxiety, to ameliorate a well-recognized syndrome. But for a drug's effect to be so global--to extend to social popularity, business acumen, self-image, energy, flexibility, sexual appeal--touches too closely on fantasies about medication for the mind. Patients often have extreme fears about drugs, stemming from their apprehension that medication will take over in a way that cannot be reversed, that drugs will obliterate the self. For years, psychiatrists have reassured patients that medication merely combats illness: "If the pills work," I and others have said, "they will restore you to your former self. I expect you to walk in here in a few weeks and say, 'I'm myself again.' " Medication does not transform, it heals.
     When faced with a medication that does transform, even in this friendly way, I became aware of my own irrational discomfort, my sense that for a drug to have such a pronounced effect is inherently unnatural, unsafe, uncanny.
     I might have come to terms with this discomfort--the unexpected soon becomes routine in the world of pharmacology. But Tess's sense of dislocation did not disappear immediately, and her surprise at her altered self helped me to understand the more profound sources of my own concern. The changes in Tess, which I saw replicated in other patients given Prozac, raised unsettling issues.
     Many of these were medical issues. How, for example, would Prozac affect the doctor's role? To ameliorate depression is all very well, but it was less clear how psychiatrists were to use a medication that could lend social ease, command, even brilliance. Nor was it entirely clear how the use of antidepressants for this purpose could be distinguished from, say, the street use of amphetamine as a way of overcoming inhibitions and inspiring zest.
     Other questions seemed to transcend any profession, to bear directly on the way members of our culture see themselves and one another. How were we to reconcile what Prozac did for Tess with our notion of the continuous, autobiographical human self? And always there was the question of how society would be affected by our access to drugs that alter personality in desirable ways.

I wondered what I would have made of Tess had she been referred to me just before Jim broke up with her, before she had experienced acute depression. I might have recognized her as a woman with skills in many areas, one who had managed to make friends and sustain a career, and who had never suffered a mental illness; I might have seen her as a person who had examined her life with some thoroughness and made progress on many fronts but who remained frustrated socially. She and I might suspect the trouble stemmed from "who she is''--temperamentally serious or timid or cautious or pessimistic or emotionally unexpressive. If only she were a little livelier, a bit more carefree, we might conclude, everything else would fall into place.
      Tess's family history--the depressed mother and alcoholic father--constitutes what psychiatrists call "affective loading." (Alcoholism in men seems genetically related to depression in women; or, put more cautiously, a family history of alcoholism is moderately predictive of depression in near relatives.) I might suspect that, in a socially stymied woman with a familial predisposition to depression, Prozac could prove peculiarly liberating. There I would sit, knowing I had in hand a drug that might give Tess just the disposition she needed to break out of her social paralysis.
     Confronted with a patient who had never met criteria for any illness, what would I be free to do? If I did prescribe medication, how would we characterize this act?
     For years, psychoanalysts were criticized for treating the "worried well," or for "enhancing growth" rather than curing illness. Who is not neurotic? Who is not a fit candidate for psychotherapy? This issue has been answered through an uneasy social consensus. We tolerate breadth in the scope of psychoanalysis, and of psychotherapy in general; few people today would remark on a patient's consulting a therapist over persistent problems with personality or social interactions, though some might object to seeing such treatments covered by insurance under the rubric of illness.
     But I wondered whether we were ready for "cosmetic psychopharmacology." It was my musings about whether it would be kosher to medicate a patient like Tess in the absence of depression that led me to coin the phrase. Some people might prefer pharmacologic to psychologic self-actualization. Psychic steroids for mental gymnastics, medicinal attacks on the humors, antiwallflower compound--these might be hard to resist. Since you only live once, why not do it as a blonde? Why not as a peppy blonde? Now that questions of personality and social stance have entered the arena of medication, we as a society will have to decide how comfortable we are with using chemicals to modify personality in useful, attractive ways. We may mask the issue by defining less and less severe mood states as pathology, in effect saying, "If it responds to an antidepressant, it's depression." Already, it seems to me, psychiatric diagnosis had been subject to a sort of "diagnostic bracket creep"--the expansion of categories to match the scope of relevant medications.
     How large a sphere of human problems we choose to define as medical is an important social decision. But words like "choose" and "decision" perhaps misstate the process. It is easy to imagine that our role will be passive, that as a society we will in effect permit the material technology, medications, to define what is health and what is illness.

Tess's progress also seemed to blur the boundary between licit and illicit drug use. How does Prozac, in Tess's life, differ from amphetamine or cocaine or even alcohol? People take street drugs all the time in order to "feel normal." Certainly people use cocaine to enhance their energy and confidence. "I felt large. I mean, I felt huge," is how socially insecure people commonly explain why they abuse cocaine or amphetamine. Uppers make people socially attractive, obviously available. And when a gin drinker takes a risk, we are tempted to ask whether the newfound confidence is not mere "Dutch courage. "
     In fact, it is people from Tess's background--born poor to addicted and dependent parents, and then abused and neglected--who are most at risk to use street drugs. A cynic may wonder whether in Tess's case drug abuse has sneaked in through the back door, whether entering the middle class carries the privilege of access to socially sanctioned drugs that are safer and more specific in their effects than street drugs but are morally indistinguishable in terms of the reasons they are taken and the results they produce. I do not think it is possible to see transformations like Tess's without asking ourselves both whether street-drug abusers are self-medicating unrecognized illness and whether prescribed-drug users are, with their doctors' permission, stimulating and calming themselves in quite similar ways.
     More unsettling to me than questions of definition--licit versus illicit--was an issue raised by Tess's renewed professional success: how might a substance like Prozac enter into the competitive world of American business? Psychiatrists have begun to recognize a normal or near-normal mental condition called "hyperthymia," which corresponds loosely to what the Greeks called the sanguine temperament. Hyperthymia is distinct from mania and hypomania, the disorders in which people are grandiose, frenetic, distractible, and flawed in their judgment. Hyperthymics are merely optimistic, decisive, quick of thought, charismatic, energetic, and confident.
     Hyperthymia can be an asset in business. Many top organizational and political leaders require little sleep, see crises as opportunities, let criticism roll off their backs, make decisions easily, exude confidence, and hurry through the day with energy to spare. These qualities help people succeed in complex social and work situations. They may be considered desirable or advantageous even by those who have quite normal levels of drive and optimism. How shall we respond to the complaint that a particular executive lacks decisiveness and vigor? By prescribing Prozac? In Tess's work, should the negotiators on the union side be offered Prozac, too? The effect of Prozac on Tess's style in her corporate work--and Sam's in his architectural practice--raises questions about how a drug that alters personality might be used in a competitive society.
     Nor is it possible to witness Tess's transformation without fearing that a drug like Prozac might bolster other unfortunate tendencies in contemporary culture. Even Prozac's main effect in Tess's treatment-the relief it provided from social vulnerability--might, in societal terms, prove a mixed blessing. Tess had come for medication treatment only after a prolonged effort at self-understanding through psychotherapy. But I could imagine a less comfortable scenario: A woman much like Tess, abused and neglected in childhood, though not fully aware to what extent and to what effect, seeks treatment in a society that prefers to ignore victimization and that values economy over thoroughness in health care; the woman seems subdued and angry, is discontented for reasons she cannot easily put into words. By what means will her doctor attempt to help her? Would Prozac, alone, be enough?

But my central concern, as I watched Tess's story unfold, involved her personhood. Tess had every right, on the basis of both childhood experience and unhappiness in adult life, to be socially vulnerable in adulthood. But once she had taken Prozac, she--and those who knew her--had to explain her newfound social success on medication. If her self-destructiveness with men and her fragility at work disappeared in response to a biological treatment, they must have been biologically encoded. Her biological constitution seems to have determined her social failures. But how does the belief that a woman who was abused as a child and later remains stuck in abusive relationships largely because of her biologically encoded temperament affect our notions of responsibility, of free will, of unique and socially determinative individual development? Are we willing to allow medications to tell us how we are constituted?
     When one pill at breakfast makes you a new person, or makes your patient, or relative, or neighbor a new person, it is difficult to resist the suggestion, the visceral certainty, that who people are is largely biologically determined. I don't mean that it is impossible to escape simplistic biological materialism, but the drama, the rapidity, the thoroughness of drug-induced transformation make simplicity tempting. Drug responses provide hard-to-ignore evidence for certain beliefs--concerning the influence of biology on personality, intellectual performance, and social success--that heretofore we as a society have resisted. When I saw the impact of medication on patients' self-concept, I came to believe that even if we tried to understand these matters complexly, new medications would redraw our map of those parts of the self that are biologically responsive, so that we would arrive, as a culture, at a new consensus about the human condition.

An indication of the power of medication to reshape a person's identity is contained in the sentence Tess used when, eight months after first stopping Prozac, she telephoned me to ask whether she might resume the medication. She said, "I am not myself."
     I found this statement remarkable. After all, Tess had existed in one mental state for twenty or thirty years; she then briefly felt different on medication. Now that the old mental state was threatening to re-emerge--the one she had experienced almost all her adult life--her response was "I am not myself." But who had she been all those years if not herself? Had medication somehow removed a false self and replaced it with a true one? Might Tess, absent the invention of the modern antidepressant, have lived her whole life--a successful life, perhaps, by external standards--and never been herself?
     When I asked her to expand on what she meant, Tess said she no longer felt like herself when certain aspects of her ailment--lack of confidence, feelings of vulnerability--returned, even to a small degree. Ordinarily, if we ask a person why she holds back socially, she may say, "That's just who I am," meaning shy or hesitant or melancholy or overly cautious. These characteristics often persist throughout life, and they have a strong influence on career, friendships, marriage, self-image.
     Suddenly those intimate and consistent traits are not-me, they are alien, they are defect, they are illness--so that a certain habit of mind and body that links a person to his relatives and ancestors from generation to generation is now "other." Tess had come to understand herself--the person she had been for so many years--to be mildly ill. She understood this newfound illness, as it were, in her marrow. She did not feel herself when the medicine wore off and she was rechallenged by an external stress.
     On imipramine, no longer depressed but still inhibited and subdued, Tess felt "myself again." But while on Prozac, she underwent a redefinition of self. Off Prozac, when she again became inhibited and subdued--perhaps the identical sensations she had experienced while on imipramine--she now felt "not myself." Prozac redefined Tess's understanding of what was essential to her and what was intrusive and pathological.
     This recasting of self left Tess in an unusual relationship to medication. Off medication, she was aware that, if she returned to the old inhibited state, she might need Prozac in order to "feel herself." In this sense, she might have a lifelong relationship to medication,
whether or not she was currently taking it. Patients who undergo the sort of deep change Tess experienced generally say they never want to feel the old way again and would take quite substantial risks--in terms, for instance, of medication side effects--in order not to regress. This is not a question of addiction or hedonism, at least not in the ordinary sense of those words, but of having located a self that feels true, normal, and whole, and of understanding medication to be an occasionally necessary adjunct to the maintenance of that self.
     Beyond the effect on individual patients, Tess's redefinition of self led me to fantasize about a culture in which this biologically driven sort of self-understanding becomes widespread. Certain dispositions now considered awkward or endearing, depending on taste, might be seen as ailments to be pitied and, where possible, corrected. Tastes and judgments regarding personality styles do change. The romantic, decadent stance of Goethe's young Werther and Chateaubriand's Rene we now see as merely immature, overly depressive, perhaps in need of treatment. Might we not, in a culture where overseriousness is a medically correctable flaw, lose our taste for the melancholic or brooding artists--Schubert, or even Mozart in many of his moods?
     These were my concerns on witnessing Tess's recovery. I was torn simultaneously by a sense that the medication was too far-reaching in its effects and a sense that my discomfort was arbitrary and aesthetic rather than doctorly. I wondered how the drug might influence my profession's definition of illness and its understanding of ordinary suffering. I wondered how Prozac's success would interact with certain unfortunate tendencies of the broader culture. And I asked just how far we--doctors, patients, the society at large--were likely to go in the direction of permitting drug responses to shape our understanding of the authentic self.
     My concerns were imprecisely formulated. But it was not only the concerns that were vague: I had as yet only a sketchy impression of the drug whose effects were so troubling. To whom were mypatients and I listening? On that question depended the answers to the list of social and ethical concerns; and the exploration of that question would entail attending to accounts of other patients who responded to Prozac.

My first meeting with Prozac had been heightened for me by the uncommon qualities of the patient who responded to the drug. I found it astonishing that a pill could do in a matter of days what psychiatrists hope, and often fail, to accomplish by other means over a course of years: to restore to a person robbed of it in childhood the capacity to play. Yes, there remained a disquieting element to this restoration. Were I scripting the story, I might have made Tess's metamorphosis more gradual, more humanly comprehensible, more in sync with the ordinary rhythm of growth. I might even have preferred if her play as an adult had been, for continuity's sake, more suffused with the memory of melancholy. But medicines do not work just as we wish. The way neurochemicals tell stories is not the way psychotherapy tells them. If Tess's fairy tale does not have the plot we expect, its ending is nonetheless happy.
     By the time Tess's story had played itself out, I had seen perhaps a dozen people respond with comparable success to Prozac. Hers was not an isolated case, and the issues it raised would not go away. Charisma, courage, character, social competency--Prozac seemed to say that these and other concepts would need to be re-examined, that our sense of what is constant in the self and what is mutable, what is necessary and what contingent, would need, like our sense of the fable of transformation, to be revised.

Chapter 2- Compulsion

As I was becoming acquainted with Prozac, I was consulted by a woman who I thought needed no medication at all but who, from the start, knew better. Julia telephoned because she had read a magazine article I had written about psychopharmacology. A patient described there, a woman who responded to Prozac, reminded Julia of herself. Would I see her and put her on the drug? As Julia elaborated, I was less impressed with any sign of mood disorder than with her frustration at work and home. I suggested that, rather than consider medication, she might speak to a psychotherapist. I referred Julia to a woman social worker who is reliable and charges modest fees.
     In the course of the first two therapy sessions, the social worker came to believe that Julia's problems arose from a perfectionistic style. The evaluation took place just as stories about Anafranil, the first medicine approved for the treatment of obsessive-compulsive disorder, were appearing in the news. The social worker thought it might make sense for me to have a look at Julia after all.
     When I met her, I was impressed, as I had been on the phone, with how well put together Julia seemed. She was pleasant and well spoken and appeared comfortable with herself. Her life had definite form to it: she had completed training as a registered nurse, married, had children, and taken a short-day job at a nursing home, a position that allowed her to be back at the house to greet her children on their return from school.
     But there were problems on every front. She demanded extraordinary control in the household. The beds had to be made just so. The children had to be scrubbed and organized before leaving for school. Julia's husband was uncomfortable with her inflexibility, and she found herself raising her voice to him and the children more than was right. Also, the nursing-home job was beneath the level of her abilities. Julia was not challenged, but she saw no way out. How could she manage her tasks in the house and at the same time tackle a more demanding job?
     These were the sorts of problems I had hoped might respond to a reassessment of her own or the family's needs in therapy. I certainly did not want to prescribe a "mother's little helper," a pill that would allow Julia to feel less frazzled in a domestic setup that, ideally, required better negotiating between spouses or a clearer understanding on Julia's part of her own anxieties over her competency as nurse, wife, or mother. I wanted, in short, to avoid medicating Julia for what looked like marital dissatisfaction.
     I began with the most prosaic questions. Did the housework, child care, and job duties fall on Julia's shoulders in unfair ways? Had Julia considered hiring someone to help with the cleaning? Might she find after-school programs for the children? Would she feel relief if her husband took on additional responsibilities in the house? If he came home earlier, would she be freer to find fulfillment in her career?
     Julia said her problems could not be solved in these mundane, operational ways. Her husband had made these suggestions and others, but she could not let go. She needed to be home in any event, because she disliked disorder. If she were not home, the straightening up would not be done to her satisfaction, and the children would not be neat and scrubbed in the way that pleased her.
     And it was, she said, very much a matter of pleasing her, of her comfort. She was not absolutely compelled to perform any particular task; if there were sufficient reason, she could leave the house with chores unfinished, although she felt better if everything was done, and done according to her standards. I asked about explicit obsessions. Julia did not fear contamination, was not anxious over germs. In fact, she did not have any formed worrisome ideas; she just disliked it when things were left messy. But her style, her preferences, her sense of propriety, her perfectionism were so pronounced that she was continually angry at her children and husband and, given the impossibility of instilling her standards in them, stalemated in her career.

There are many ways of understanding Julia's dilemma, but perhaps it is most instructive to see how the problem was understood in the period before Julia telephoned to request medication.
     Four years into her marriage, and some six years before I saw her, Julia and her husband visited a psychologist for couples counseling. The psychologist, the director of a university program here, found that Julia was unhappy in her marriage. She was not depressed, but she found her husband unresponsive to her needs. Things had to be "just so" to please her, and not only matters of household organization. The husband had to take her out a certain number of nights a month for her to feel loved. If he fell one night shy, she would fault him--but she might not have said openly how many was enough. The psychologist worked with the couple on "communication skills," and he believed they profited. He had never considered a formal diagnosis of depression or of obsessive-compulsive disorder for Julia, although he said she certainly was discontented and had a perfectionistic style.
     Julia's internist also found her anxious and unhappy on occasion, conditions he attributed to her perfectionism and her sense that she was not getting enough attention from her husband. The doctor prescribed antianxiety medication for what he called a "situational reaction associated with depressive overtones," a label for a problem
that does not quite rise to the level of illness but that nonetheless seems to call for treatment. Julia's gynecologist understood the same intercurrent problems as premenstrual syndrome, a condition he tried at various times to alleviate with diuretics ("water pills") and oil of primrose (a plant extract intended to raise prostaglandin-hormone levels), and more antianxiety medication, without apparent effect.
     Julia, in sum, was a patient without a diagnosis, or with bits and pieces of many diagnoses. As a psychiatrist, I was in no better position to categorize Julia's problem than her psychologist, internist, and gynecologist had been. My preference, as I have said, was not to call her ill at all, but to focus on some intimate aspect of the self or the marriage. I wondered about her self-esteem. Where did her need for control, her ineffectiveness in marital negotiations, and her heightened frustration with the children arise, if not from a disorder in self-image? Her perfectionism, if I had to guess, might be a defense against the terrible feelings she anticipated if her imagined inadequacies were laid bare.
     I also considered family pathology. Could the husband be undermining his wife, playing on her anxiety in some way that guaranteed she would tend to home and hearth rather than throw herself into her career? Frequently when an otherwise competent patient becomes dysfunctional in one limited sphere, the greater pathology is in the apparently more flexible spouse.
     But, in order to consider biological treatment, it seemed important to capture Julia's waxing and waning symptoms in a diagnosis. Julia's other doctors had all on occasion noticed a depressive tendency. The current diagnostic system contains a category--"dysthymia"-- for patients who do not quite meet the standards of major depression. But dysthymia applies to people who suffer depressed mood "for most of the day more days than not" for two or more years running, and who when depressed have disturbances of sleep, appetite, energy, concentration, and the like. Julia did not have that sort of disturbance, and her depression was not at all constant. What was constant was her perfectionism.
     Perfectionism makes a psychiatrist think of two diagnoses, obsessive-compulsive disorder (OCD) and compulsive personality disorder. Julia did not meet the criteria for these, either, and she said as much in describing herself. She had read about OCD in the newspaper. "I'm a neat freak," she said, "but I am not at all like that--not that extreme."
     OCD is among the most terrible of psychiatric disturbances. Anyone who has seen a man or woman whose skin is macerated from repeated scrubbings, or who cannot leave a room for fear of germs, or who spends long hours repeating meaningless calculations, or who cannot stop demanding reassurance over an unlikely but paralyzing source of dread, will have a sense of how distinctive and relentless OCD is. Personable, accomplished, interactive with friends, able to do any particular thing she chose --Julia bore little resemblance to the patients a psychiatrist ordinarily labels as having OCD, and she did not fit the standard definition.
     That definition rests on two concepts, the obsession and the compulsion. Obsessions are "recurrent, persistent ideas, thoughts, images, or impulses that are experienced, at least initially, as intrusive and senseless." The example given in the official manual is a parent's impulse to kill a loved child. Julia had no such thoughts. Compulsions are "repetitive, purposeful, and intentional behaviors that are performed in response to an obsession" or in a stereotyped fashion and which are designed to neutralize the dreaded obsession or to prevent discomfort. Here Julia's behavior came closer to the mark, though her actions were more flexible and less strictly compelled than those the definition is meant to indicate.
     Perhaps Julia almost met the criteria: there were weeks when she did two loads of laundry every day; if the floors were dirty, she might stay up late to wash them. But these behaviors appeared as compulsions only when her routine was disrupted. Most days, she was organized enough to do what felt right to her on a schedule she found acceptable. Julia's condition fell, let us say, in the penumbra of OCD.
     Certainly she had a compulsive style. Extremes of style are called, in the insulting language of psychiatry, "personality disorders." Personality disorders have traditionally been thought not to respond to medication. In the case of compulsive personality disorder, the key elements are "restricted ability to express warm and tender emotions," "perfectionism that interferes with the ability to grasp `the big picture,' excessive devotion to work," and indecisiveness--none of which Julia had--as well as "insistence that others submit to her or her way of doing things," which, along with a good many other people, she had to a fair degree.
     I had seen many patients with compulsive personality disorder--the kind who threaten to bore you to death by perseverating on small details of topics whose emotional import is never made clear --and I found Julia entirely unlike them. She was engaging and able to focus well. This having been said, if we had to give a name to what ailed Julia, it would be hard to avoid reference to compulsiveness.

In making a referral to the social worker, I had attempted to define Julia's problem as one of either marital or inner conflict. Julia, however, experienced her disorder as medical. If she were made well, the turmoil in the family would disappear. And that is what happened.
     By the time I met with Julia, reports had emerged that, like Anafranil, Prozac was effective in treating OCD, and it seldom caused the weight gain common with Anafranil. Julia was concerned about her weight, and she identified with the patient I had written about who responded to Prozac. We discussed the risks and benefits of different approaches, but in the end I gave her what she had come for. I wrote a prescription for Prozac and told Julia what I tell every patient, that anti-depressants take about four weeks to work--two weeks to build up a good level in the brain, and then, for unknown reasons, two weeks to affect the illness.

The first week on medicine, Julia reported, was "like night and day." The children behaved more obediently, and when Julia remarked on the change, they told her she was yelling less. Her husband became more cooperative as Julia became more pleasant with him. Then she noticed she had markedly more energy. "I could not have imagined this" was her comment, meaning she did not want me to think she was experiencing a placebo effect.
     I suspected Julia might be experiencing the lift of an amphetamine-like effect, the burst of energy that can arise early in the course of antidepressant treatment. I wrote in her chart, "Good early response," and asked her to return in three weeks.
     By then, the early euphoria had worn off. Julia missed the sense of vitality she had felt in the first days, but she remained moderately improved, on better terms with her family. Obsessive-compulsive disorder often requires higher doses of medication than does depression. Though Julia's was at best a "penumbral" case of OCD, I raised the dose and marked her progress. She reported steady, modest improvement in her mood and in her ability to tolerate messiness. Antidepressants do work this way for some patients--a progressive amelioration of symptoms that does not plateau for months.
     There were ups and downs. Some weeks, Julia reported having been nervous with her children and having yelled excessively. These fluctuations often correlated with particular stressors. For instance, being home on weekends was harder than responding to the structure of work.
     And that structure was changing. First Julia quit her part-time job. She chose instead to do hospital shift work on an on-call basis--a particularly disruptive way to live, but, she felt, the best way for her to re-enter the career path of hospital nursing. She began to specialize in pediatric nursing and found she could enjoy the unpredictability of young children in a way that had been impossible for her in earlier years. She believed that without medicine she could never have taken this step, accompanied as it was by complex caretaking arrangements for her own children and a need often to overlook a degree of chaos in the home.
     Julia felt--much as Tess had--that her life had been transformed. Her relations with her children and her husband were more easygoing, and she was able to tolerate a certain messiness in the structure of her life. Whatever intermittent anxiety and depression she had suffered had disappeared.

Once a patient has done well on an antidepressant for five or six months, I generally try to discontinue it. Julia reached that point early in the spring. She came in then to report she was "doing great--could not be better." She had requested and received a promotion at work and been offered regular hours. In the past, she had applied only for jobs for which she was overqualified: as a registered nurse, she had sought positions advertised for practical nurses. Now she was doing work normally done by nurses with master's degrees.
     She proudly listed the indicators of her improvement. "I left for the hospital even though the beds were unmade! And I was not upset when the children got grease on their new pants. I didn't punish them or make them feel guilty. " Then she told me the biggest news--she was getting a dog.
     Before, the messiness of a dog had been repugnant to her, not to mention the effect on her schedule. Now she was ready. "I can't wait," she said, "even though I'm allergic." She had researched breeds whose fur she could tolerate. The children were ecstatic.

We lowered the dose of medicine, and two weeks later Julia called to say the bottom had fallen out: "I'm a witch again." She felt lousy--pessimistic, angry, demanding. She was up half the night cleaning. And there was no way she could consider getting a dog. "It's not just my imagination," she insisted, and then she used the very words Tess had used: "I don't feel myself."
     I suggested we wait a bit longer to see whether Julia might be experiencing an odd effect related to medication withdrawal, or perhaps--this happened to be the timing--a premenstrual phenom-
enon of some sort. But the next week she saw the social worker, who called me to ask what I had in mind. Julia was back to square one, and none of the external circumstances of her life had changed.
     Julia resumed taking the higher dose of Prozac. Within two weeks, she felt somewhat better; after five weeks, she was "almost there again," with many more good days than bad. She said work had been torture on the lower dose of medicine: "The patients drove me crazy." She had been unable to block out distractions and had been so aware of time pressure that she could never pause and enjoy the children she was tending.
     And, at home, she had been unable to ignore her own children's failings. On the higher dose of medicine, she was once more tolerant. She was again ready to get the dog, and keenly aware that she could not have let a dog into the house when on the lower dose of Prozac. Julia went out of her way to impress on me how much more confident she was, how much more engaged in every facet of her life, when on an adequate dose.
     Her husband had nothing but good things to say about the effect of the drug. Sometimes when she behaved in this more relaxed way, he wondered whether she was buttering him up, trying to get on his good side for an ulterior purpose, so unused had he been over the years to having a wife who could sit with him of an evening without being jumpy and critical.

By this point, Julia had stopped seeing the social worker entirely, and the social worker contacted me to express concern--not about Julia's well-being but about her own adequacy. I asked the social worker how she would have understood the case if medication were not available. As she saw it, Julia's story, and her needs, were not difficult to encapsulate.
     When Julia first came to the office, her distress related to frustration over her stalled career and certain personal issues--unresolved family-of-origin conflicts that had re-emerged in her marriage. Julia's father, a businessman, had been a high-strung perfectionist; the son of a depressed mother, he was the more nurturant of her parents. Julia saw her mother as passive and distant. Julia's older sister seemed to stumble from failure to failure, and as a result, Julia was moved to care for her and to identify with her competent father.
     The conflict in Julia's own marriage, as the social worker formulated the case, involved gender-role conflict. Identifying with her father, Julia secretly, or even unconsciously, felt herself to be more competent than her spouse. At the same time, not least for her own sense of security, she wanted to maintain the illusion that her husband was like her father, strong and decisive. The social worker saw Julia's obsessionality--and her paralysis in career and home life--as an expression of inner conflict over control in the family; she was torn by a wish to let her husband take the lead, opposed by repeated urges to barge in and do things right. These same conflicts emerged in her handling of the children, whom she pushed hard while telling herself she was giving them their head.
     The social worker saw Prozac as having had an interesting effect on these conflicts. It had tipped the balance in favor of assertiveness, allowing Julia to make it clear to her husband what she needed and why; at the same time, it made her less urgent, which allowed her husband to do things at his own pace, a condition under which he appeared quite competent. Before she began her Prozac treatment, Julia had obsessed over which bedspread to buy her daughter. All were imperfect, because the real problem was that Julia disliked the paint color in her daughter's room, a subject she had been reluctant to raise with her husband. On medication, Julia simply asked her husband to repaint the room and then waited patiently for him to complete the task. Once the painting was under way, Julia had no trouble selecting a bedspread. In this interaction, both husband and wife were able to exert control in their different ways.
     I asked the social worker why she felt guilty. The medication had done what she would have wished to accomplish with her psychotherapy: it had facilitated an improvement in the family dynamics. The problem, for the social worker, was that this change came about without any increased self-knowledge on Julia's part. I said that evidently insight had not been necessary. This comment did not allay the social worker's concern. She believed that medication-induced change, unaccompanied by growth in self-understanding, was inferior to what psychotherapy has to offer.

To Julia, the story was entirely different. As so often happens, the pill reified the illness. If there was a chance in the world that Julia might see her difficulty in adjusting to married life as anything but a result of a "biological disorder characterized by compulsiveness and depression," her relapse and rescue by the increased dose of medicine ended it. Once the drug kicked in, she had visited the social worker only infrequently, and then with skepticism. "If I had been on Prozac," she said, "I would not have needed to see the marriage counselor either."
     I found I had little desire to cling to my earlier hypothesis of family pathology--the competent wife secretly undermined by the threatened husband. Even my sense of her perfectionism as a defense against low self-esteem was shaken, although I was beginning to wonder whether medication could perhaps provide self-esteem. Certainly on medication Julia was able to make major adjustments in her life with no sign of inner conflict. Her husband was enthusiastic when she moved on with her career. If anything, the response of Julia and her family to the medicine made the various scenarios conjured up by psychotherapy seem hypercritical and ungenerous.

But what did Julia "have"? Since her condition responded to a medication that can treat OCD, do we want to say she had OCD?
     This decision is consequential. Large numbers of patients who visit doctors with psychological complaints are not "diagnosable." To make them diagnosable would mean expanding the current schema, and thus calling many more people mentally ill. Whether we want to change our view of mental illness and whether we want to make medication response a deciding criterion are interesting questions, with humane arguments available both for and against expansion.
     If we do say that Julia has, say, an incompletely expressed case of OCD, we will be recapitulating in biological psychiatry the history of psychoanalysis. Once reserved for the most obviously ill patients, "obsessional" and its contrasting counterpart, "hysterical," came as the period of psychoanalytic dominance progressed to be applied to people's social styles. The advent of biological psychiatry originally resulted in a severe restriction of the use of such terms; but, with the discovery of new biological treatments, the operational definition of OCD is expanding once again, in part because what responds like OCD comes to be called OCD.
     "Obsessionality" and "compulsiveness" are now used by those who treat illness with medication to encompass what in earlier days would have seemed mere personal idiosyncrasy. Increasingly, what was once the penumbra of OCD is fully in its shadow. But the expanded disease has its own penumbra. Now there will arise questions about how patients slightly less compulsive than Julia should be categorized and treated.

To see how far this new penumbra extends, I want to return to Julia's decision to contact me for treatment. Julia called because she identified with a patient I had described in a magazine article. That patient was Tess. I had described Tess as "a hard-working executive so attentive to detail in her professional life that she found little time to socialize Those few words struck a chord with Julia, convinced her that she and Tess--very different women--might have something biologically in common.
     And it's true that I had treated Tess for something rather like perfectionism. When Tess responded to imipramine but remained stuck in love and work, I began to wonder whether what held her back might bear some relationship to OCD.
     Tess had no compulsions. She was, to be sure, obsessed with a hurtful lover; otherwise, she was not obsessional even in the colloquial sense. But she was driven to an unusual degree. What distinguished Tess was her success under impossible conditions, her determination, and her insistent and effective nurturance of others. She was, one might say, almost too giving. I don't know at what point I began looking at that goodness from the odd perspective of the biological diagnostician; I can only say that there are in pharmacology, as in psychotherapy, important moments when the clinician suddenly sees the patient afresh. In one such moment, I began to re-examine, to recategorize, those traits that made Tess special.
     I have characterized clinical psychopharmacology as an impressionistic art. The doctor listens to a patient and, on the basis of the patient's story and the empathic response it evokes--bizarre biological probes, so qualitatively different from the usual blood, urine, and spinal-fluid tests--the doctor attempts to make an assessment of the state of the patient's neurons. The pharmacologist assumes that the complex constellation of behaviors and feelings a patient reveals reflects a simple physiological state. On the basis of the extraordinary, unique shape of a patient's life, the pharmacologist asks such questions as "Is this a disorder that is likely to respond to a drug that treats OCD?"
     What an odd thought this is--"dedication to others less fortunate" as a form of aberrance that can lead a doctor to choose one medication over another. But, in the inexact process of extrapolating back from symptom and behavior to chemistry, the psychiatrist takes every bit of help he can get. We are not beyond grasping at straws, and I grasped at this one: I wondered whether Tess had a touch of the obsessional about her.

I began to ask Tess about her strengths as one might ask other patients about their weaknesses. How had she managed to raise her siblings so effectively? She answered, "You don't understand. I had no choice. The only other possibility was to disappoint everyone who counted on me, and I could not bear that."
     We are accustomed to thinking of compulsiveness as a disorder or an annoying style of relating to the world, and it can be. But some of the characteristics of compulsiveness--the deep sense of responsibility, the vigilance, and the attention to detail--are also virtues. Sociobiologists have speculated that compulsiveness survived in the human species because it was a competitive advantage for our ancestors' tribes to contain one or two members who were prudent and driven in the extreme. Certainly Tess's family survived because of her inability to tolerate failure.
     I am morally certain that I would not have had these ruminations if Prozac did not exist. Because an antidepressant likely to be useful for compulsive patients was available, it made sense to ask whether Tess's strength of character could be a manifestation of the same biological constellation that in other people shows itself as compulsiveness. Equally, it was because her dedication might be, in physical terms, something like OCD that--with so many tried and true medications available--I turned to a new and relatively untested drug for Tess.
     In clinical pharmacology, contemporary technology plays a dominant role in shaping ideology. What we look for in patients depends to a great degree on the available medications. That Tess's depression was accompanied by what could be construed as compulsiveness was of interest only because this trait might be an indicator of something we could now treat.
     Who Julia is--whether she is a fully functional woman with marital troubles or a slightly handicapped woman adjusting uncomfortably to reasonable constraints--is largely a function of drug development. We may decide on similar grounds whether Tess's dedication is a moral or a psychopathological trait. How we, as observers of our fellow men and women, look and listen, how we categorize, how we understand the tensions between people and their predicaments, is in part a product of the available means of influence. The interaction between a drug and cultural norms does not require the use of medication by the people we are assessing but can result from the mere availability of a substance that colors our beliefs about deviance and how it is produced.

Tess's self-understanding revolved around a quality she identified as seriousness. Once on medication, she explained her newfound success with men in simple terms: "I am less serious."
     Seriousness covers a lot of ground. For most of her life, Tess did not allow herself to seek out pleasure. She focused on duty, and heeded the warning of a strong "superego," or conscience, that always put work before play. On medicine, her ability to attract men owed something to her increased flexibility, to a more generous sense of permission to enjoy. The problem earlier may have been as much one of unwillingness to attract men as inability. These traits, which might loosely be termed compulsive, disappeared when Tess took Prozac. And of course we may attribute her new ability to forget her married ex-boyfriend Jim to a true anticompulsive effect of drug treatment.
     We have earlier considered Tess's transformation in terms of the alleviation of chronic depression. Here is a somewhat different way of conceptualizing her responses to medication: the imipramine had handled her acute depression. The Prozac cured her of a masked form, or variant manifestation, of compulsiveness.

Tess's recovery parallels Julia's. In both instances, the very traits targeted as compulsive disappeared on the appropriate medication. The question then becomes just how far we are willing to "listen to drugs." Will we want to expand the definition of OCD or compulsive personality disorder to include someone like Tess, who has no compulsions whatsoever and who meets none of the explicit criteria of illness except, perhaps, exaggerated devotion to work? Though we may resist it, the temptation is there. Surely, in a colloquial sense, part of what happened to Tess when she took Prozac was that she became less compulsive.
     Facing someone like Tess, I think we are drawn in two directions. One is to stretch the scope of illness to encompass her character traits. Another is to say we have found a medication that can affect personality, perhaps even in the absence of illness--only now, instead of restricting our powers to the depressive-to-manic continuum, we are considering whether we may not also be able to influence what we might call the obsessional-to-hysteric continuum.
     Either way, we are edging toward what might be called the "medicalization of personality." Or perhaps, once we say that traits on both the depressive-to-manic and the obsessional-to-hysteric continua respond to medication, we are over the edge. Those two spectra cover a good deal of what makes different people distinctive. It is not only Tess's "seriousness" whose biological underpinnings are likely linked to compulsion or depression. If seriousness is subject to chemical influence, we can imagine a large collection of pairs of opposed traits that will be as well: contemplative/action-oriented, rigid/flexible, cautious/impulsive, risk-averse/risk-prone, masochistic/ assertive, by-the-book/by-the-seat-of-the-pants, deferential/demanding, and many others. The first element in any of these pairs might equally be associated with depressive or obsessional leanings and might equally be a candidate for drug treatment.

The extension of our reach beyond depressed/manic to obsessive/ hysteric is significant. The obsessional-to-hysteric continuum was once a mainstay of psychiatry. Toward the middle of this century, most relatively healthy patients, those with what was then called "neurosis," were discussed for treatment purposes in terms of whether they were more obsessional or more hysterical. Every person can be understood as sitting somewhere on this spectrum.
     We recognize the flavor of compulsiveness even in the absence of a single symptom. Think about Bert and Ernie on "Sesame Street." They represent extremes in the diverse styles of healthy children. Bert has a fixed, serious, even worried look, and he is decidedly more reliable and less spontaneously playful than Ernie. Children identify with Bert because they love order; they identify with Ernie because they love mischief. Every child has mixed affinities for discipline and innovation, noise and quiet, group activity and solitude. I am hardly suggesting treatment for Bert or Ernie. But I suspect we would be near the truth if, putting aside such formal labels as "obsessional" and "hysterical," we were to say that what Prozac did for Tess was to shift her from a personality like Bert's to one more like Ernie's. Only Ernie would make or enjoy three dates a weekend; only Ernie would venture the gift of the red wagon.
     This broad view of obsessionality, in which any affinity for the Apollonian virtues as opposed to the Dionysian suffices to make the diagnosis, gives some sense of what it might mean to introduce a medicine that can affect minor degrees of a trait that exists along a broad continuum, extending from illness to health.

We may not be convinced that Tess was compulsive. But even extending the definition of OCD to include Julia (and I think, especially taking into account the way her symptoms returned when her medication dose was lowered, many psychiatrists today would consider her to suffer from something like OCD) raises interesting questions for me. I recall, in my own childhood, having been specially scrubbed and warned against messy play before visiting certain demanding older relatives. Those relatives had grown up in Germany, where extremes of neatness and order were the cultural norm. Their homes were, every day, more tidy than any homes I have seen since even on special occasions. Julia reminded me of the wives in those families; perhaps my comfort with them explains my reluctance initially to accept Julia's behavior as symptomatic of medical illness.
     Those wives, if my childhood perceptions are accurate, were not conflicted about their perfectionism, nor did their husbands seriously challenge it, though it was a matter for teasing and banter. I doubt the wives would have gone to bed with the floor or clothes dirty, but neither would their schedule often have required them to do so. There was in those families, I suspect, a certain male comfort in being the better-acculturated, more flexible spouse--the cock of the
walk--while the wife assiduously tended the homefires. This arrangement inevitably led to a certain amount of marital unhappiness, but I would say that for the most part the couples managed to make their way through life contentedly.
     Once Julia responded to medication, I found myself wondering whether those more contented perfectionists of years past would have responded to Prozac similarly. There are reasons for thinking not. To be neat in a culture that prizes neatness may bespeak a very different, less aberrant biological state than maintenance of the same behavior in a culture that has adopted different values. And if those perfectionistic housewives could have been relieved with medication of some of their need for order, their lives might not have been improved so much as made complicated in interesting ways. To say something less speculative: whether a particular behavioral style like perfectionism is deviant is very much a matter of cultural expectations, and that culture can be as broad as a nation and as narrow as a twosome.
     This particular contrast--the contented and discontented perfectionist--gives rise to further thoughts about the notion of the "mother's little helper." Mother's little helpers were pills--Miltown, amphetamine, barbiturates, Librium, and Valium were the most popular and widely available in the fifties and early sixties--that were used to keep women in their place, to make them comfortable in a setting that should have been uncomfortable, to encourage them to focus on tasks that did not matter. I cannot think of the phrase even today without hearing it in Mick Jagger's sneering tones.
     In Julia's story, the mother's-helper role is most clearly played by the various antianxiety pills given her over the years. Those medicines allowed her to perform her housekeeping tasks with a diminished, but still substantial, level of anxiety. The failing of those medicines was not that they did not work well enough but that they worked the wrong way altogether. The point, in retrospect, was not to make Julia less anxious but more bold.
     Prozac's status in Julia's treatment is more complex. At the most obvious level, it was the opposite of a mother's little helper: it got Julia out of the house and into the workplace, where she was able to grow in competence and confidence. I see this result often. There is a sense in which antidepressants are feminist drugs, liberating and empowering. In this scenario, it is the failure to prescribe medication that keeps the wife trapped, apparently by her own proclivities. We may even want to say that nonbiological therapies, like the couples counseling, though apparently aimed at change through understanding, are in fact palliative and likely to lead only to a slightly more tolerable form of inertia.
     It is hard not to see Prozac in these stories as the opposite of a mother's little helper. But the memory of my fastidious relatives makes me want to include a small caveat, a reminder that we might want to maintain awareness of how culture-bound this reading of events is. After all, should a person with a personality style that might succeed in a different social setting have to change her personality (by means of drugs!) in order to find fulfillment?
     Even Tess's success falls under a similar caveat. I have in mind a recent remark by John Updike: "Masochism is as unfashionable now as aggressiveness was twenty years ago. . . ." If we see Tess's transformation as a victory, it's because of a change in mores, because we value the assertive woman and shake our heads over the longsuffering self-sacrificer. Perhaps medication now risks playing a role that psychotherapy was accused of playing in the past: it allows a person to achieve happiness through conformity to contemporary norms. This accusation is the "mother's-little-helper" label in modern colors.
     We may have difficulty entertaining such a point of view, because cultural expectations have shifted so decisively. We can hardly imagine wanting to do anything other than relieve Tess's suffering by freeing her from her addiction to sadistic men. We can hardly imagine wishing for Julia that she find more fulfillment in her well-kept home. But to say this much is to excuse us as a society for failing to find a satisfying, growth-enhancing niche for women with obvious strengths and two rather common forms of personality organization.
     Certainly our valuation of compulsiveness in men has undergone a change. One has only to consider Phileas Fogg, "the most punctual man alive," who nonetheless had the spunk and resourcefulness to travel around the world in eighty days. For decades, the eccentric and fastidious Englishman was at once a figure of fun and of admiration--he ruled the Empire. It took the work of such writers as Edmund Gosse and the Bloomsbury group to begin to make tenderness a male virtue, overattentiveness to work a failing, and eccentricity an aspect of fatherly tyranny rather than masculine charm. (Indeed, from the I930s through the 1960s, an influential critique of capitalist society held that it created and rewarded the "anal character"--compulsive, hoarding, and industrious--while repressing sensuality and spontaneity.)
     In the everyday practice of medicine, and in the everyday valuation of human success and suffering, it is fruitless to try to maintain the viewpoint of cultural relativism. Here is the physician's compulsion, and perhaps society's as well: once we have seen Julia recover or Tess become "better than well," we inevitably assess their personality styles as handicapping forms of minor mood disorder. The operational definition of wellness must be in relation to the demands and goals of our society, here and now. Once we have seen the joy on patients' faces, we can only be grateful for the availability of more powerful and specific medication. But the awareness that what we are altering is a personal style that might have succeeded in a different, and not especially distant, culture may make us wonder whether we are using medication in the service of conformity to societal values. Indeed, experience with medication may make us aware of how exigent our culture is in its behavioral demands.

The reader may still be puzzling over a different question: whether Tess and Julia and Sam had something "really" like compulsiveness or "really" like depression. Not only do depression and OCD have penumbras and penumbras-of-penumbras, but these larger areas of shadow often overlap. People who are pessimistic tend to be cautious, and vice versa. Moreover, the effectiveness of Prozac for both conditions may lead us to wonder whether the conditions are related.
     Here is another important aspect of listening to drugs: responsiveness to medication can influence our thoughts about which illnesses are distinct and which overlap. How doctors divide up mental illness may seem an issue merely internal to psychiatry, but for decades the debate over the continuity or separateness of mental illnesses has colored our understanding of the way human beings are related to one another.

The most basic diagnostic distinction in psychiatry is that between manic-depressive illness and schizophrenia, the disorders defined and declared to be separate by the father of modern diagnostic (or descriptive) psychiatry, Freud's contemporary Emil Kraepelin. At the turn of the century, Kraepelin showed that manic-depressives have a different course of illness from that of patients suffering from schizophrenia; he assumed that both diseases had a biological basis. By mid-century, many American psychiatrists were prepared to ignore Kraepelin's distinction and, indeed, to discard almost all diagnosis. As late as 1963, Karl Menninger wrote that "we tend today to think of all mental illness as being essentially the same in quality, although differing quantitatively and in external appearance."
     This declaration was part of an egalitarian manifesto, the assertion that the well and the ill differ primarily in the degree of trauma they have suffered, and secondarily in the strength of their natural constitutions. This spectrum theory of mental illness arose from psychoanalysis. As Donald Klein, a formidable critic of the spectrum theory, put it, "The predominant American psychiatric theory was that all psychopathology was secondary to anxiety, which in turn was caused by intrapsychic conflict. Psychosis was considered the result of such an excess of anxiety that the ego crumbled and regressed, and neurosis, the result of a partially successful defense against anxiety that led to symptom formation." The well and the mentally ill differed only in the degree of anxiety they bore; and therefore the same treatment, the diminution of inner conflict via psychotherapy, was applicable to all ailments and all people. The spectrum theory was part of a broader psychology that emphasized the qualities people have in common.
     Disregard for diagnosis was an American phenomenon. Though admirable in its demand that people be seen and treated similarly, it led to peculiar contrasts with observations in Europe. Considering diagnosis a mere administrative requirement, American psychiatrists had begun calling all seriously ill patients schizophrenic, a practice Menninger encouraged. The result was international data showing that New York had more schizophrenics and fewer manic-depressives than did London. This discrepancy was then treated as reflective of real phenomena, and theories were generated to explain it: perhaps urban violence in New York caused schizophrenia, whereas the calm and dull life of London was conducive to mood disorders. Racial theories were also advanced.
     At last an epidemiological study was conducted, using uniform criteria (the British criteria, based on Kraepelin's distinction) to diagnose patients in the two cities. The landmark "U.S.-U.K. study," published in 1972, concluded that, "In spite of the gross differences in the diagnostic statistics produced by the hospitals of the two cities, in spite of the profound social and cultural differences between the cities themselves . . . when uniform diagnostic criteria are employed the diagnostic distributions of patients entering hospital in New York and London are to all intents and purposes identical." The apparent differences in proportions of illness were due entirely to differences in doctors' diagnostic practices. But which diagnostic system was superior? That question would be answered by a drug, lithium.

The story of lithium has the quality of legend. Lithium is an element of the periodic table, where it sits just below sodium. Like sodium, lithium readily forms salts. Early in the century, lithium bromide had been used as a sedating tranquilizer (hence our term "bromide" for a commonplace saying), but lithium fell out of favor in the 1940s, when it was used in an uncontrolled way as a sodium substitute for cardiac patients, some of whom died. At just this inauspicious time, in 1949, the Australian John F. J. Cade, "an unknown psychiatrist, working alone in a small chronic hospital with no research training, primitive techniques and negligible equipment," discovered that lithium salts were a remarkable specific treatment for manic depression.
     Cade's discovery is often characterized as serendipitous. Cade had found that the urine of manic patients was especially toxic to guinea pigs, and he was looking for the responsible substances. He thought one might be uric acid, and he began experimenting with lithium urate, not because of any psychiatric properties of lithium, but because lithium urate was the most soluble salt of uric acid. To Cade's surprise, far from being toxic, the salt protected guinea pigs against the urine of manics, and it also sedated the animals, effects Cade found were due to the lithium. He immediately tried other lithium salts on himself and, when they proved safe, on ten hospitalized manic patients, all of whom recovered, some almost miraculously. The discovery of lithium as an antimanic agent resulted from one man's curiosity and powers of observation and deduction.
     Because of the cardiac deaths, as well as Cade's lack of renown in the profession, the use of lithium for mania spread slowly. But by the late 1960s, doctors once more considered lithium to be a reasonably safe drug. It was also understood that lithium can treat and prevent recurrences of manic-depressive illness but is only rarely effective for schizophrenia. Once lithium's safety and specific efficacy for manic depression were accepted, diagnostic distinctions mattered in a way they had not before. At the same time, pharmacologic outcome could guide diagnosis.
     This reasoning was precisely circular: since diagnosis was needed to predict medication response, medication response should determine diagnosis. It seemed, for the most part, that lithium treated all manicdepressive illness and nothing else; and no other medication treated manic depression. That is, lithium conformed to a one-drug/onedisease model of pharmacology, a model so aesthetically pleasing as to be irresistible. Lithium responsiveness confirmed the Kraepelinian model of manic depression and caused American psychiatrists to expand their use of the diagnosis. Lithium had performed an extraordinary "pharmacological dissection," defining for all the world the boundaries of a particular disorder.

The success of lithium set off an explosion of precise psychiatric diagnosis. In a few decades, American psychiatrists went from using only two diagnoses, neurosis and schizophrenia, to using hundreds.
     Lithium and the one-drug/one-disease model had an enormous influence on the minds of physicians. Lithium made it look as if medications would be splitters-definers of illness. But, sadly, there has never been another lithium. Most subsequent medications have been lumpers, and none more so than Prozac. Within a couple of years of its introduction, Prozac was shown to be useful in depression, OCD, panic anxiety, eating disorders, premenstrual syndrome, substance abuse, attention-deficit disorder, and a number of other conditions.
     The firm link between one drug and one diagnosis has become an ideal model which even lithium no longer fits. With an effective medication available, American psychiatrists became such enthusiastic diagnosers of manic depression that today only half of the patients who receive that diagnosis respond to lithium, and two or three other drugs are in common use for the illness. And lithium is now being used to treat other forms of disturbance.
     Medications, it is increasingly understood, alter neurochemical systems. They do not treat specific illnesses. And the proliferation of illnesses has become so disturbing that the cutting edge of research involves attempts to elucidate links between them.
     OCD and dysthymia, for example, are classified in contemporary psychiatry as discrete entities, one related to anxiety and the other to depression; but a countervailing movement, based in part on observations of drug effects, characterizes them as related disorders. Our confusion over just what the medication is working on in Sam, Tess, and Julia suggests that diagnostic specificity may have its limitations. Especially in mildly disturbed or near-normal patients, syndromes that should be distinct overlap. As a drug prescribed for these fairly healthy patients, Prozac casts a spotlight on the indeterminateness of diagnosis. This boundary-blurring constitutes an unanticipated--humanistic--effect of listening to drugs: like psychoanalysis, drug response can emphasize commonality, and the futility of attempts at mechanistic categorization. Tess and Julia and Sam share something very much like "neurosis," psychoanalysis's umbrella term for the mildly disturbed, the near-normal, and those with very little wrong at all.
     What is especially noteworthy about the blurring of boundaries is its source. For decades, the thrust of biological psychiatry--not only because of lithium, but in response to evidence from brain scanners, genetic studies, and research on neurotransmitters--has been to bolster the discrete-disease model of mental deviance and to undermine the spectrum concept. Thoughtful people may have anticipated that the pendulum would some day swing the other way, but not, perhaps, that the new challenge to distinctions among illnesses, and between health and illness, would come from one of the fruits of biological psychiatry, the psychotherapeutic drug.

Aganist Depression
Peter D. Kramer
Chapter 2- Return

Often, depression abates by imperceptible degrees. But occasionally change is sudden--darkness into light. Medical practice contains its share of drama. A coma patient wakens. A stroke victim, mute since the event, speaks. Oncologists may witness the spontaneous remission of a seemingly terminal cancer. But for the doctor, little rivals this particular return to life, depression's end--especially if the episode has been a long one.
     The psychiatrist's relationship to the patient is intimate. If psychotherapy plays a role in the treatment, the two will have sat across from each other, week after week, perhaps for years. By virtue of the disorder, the patient may have been guilt-ridden and scrupulous, disclosing all. The inquiry reveals emotion and its origins, immediate and distant. The doctor may have traced the patient's inner life more minutely than his own-the coloration will be subtler, the dark corners more closely examined. And then the day arrives.
     I was walking from your office to my car, and just like that, the depression ended. In an instant, the sense of living in the world returned. I sat in the driver's seat and let sensations fill me. A passerby rapped on the window and asked if I was all right. I said.- "At long last."
     Or, I got out of bed, and I realized it was not with me. At breakfast, I told myself, this can't be. Breakfast! When last have I been able to stomach food in the morning? I poured myself a bowl of the kids' cereal.
      Or, I buzzed Bill's office, to tell him. He was annoyed at the interruption. In the past, I have tried to humor him, to define myself as healthy when I knew I was not. How could I convince him that this time was different? "Bill, " I said, `I'm coming by. It can't wait for lunch. I want you to see me now.
     Memory of these moments sustain us, in the bleak hours. If we can keep this patient alive, if she can cling to the structures that support her, the happy day may come.
     Lifting is a verb patients use. They speak of depression rising, like a fog. My mind was clouded, they say. Then they comment on an insidious quality depression has: it dulls awareness of its own force. Depression damages the ability to assess the self. I did not appreciate how distorted my thinking was.
     Equally, the report may be in the passive voice, of a burden having been lifted. A weight off the back, a load off the mind. Again, a past distortion of awareness is apparent. Until it was removed, I had no notion how heavy it was.
     For me as well, the gravity of depression is evident in the lightness, the unburdening that I feel when someone I am treating recovers.
     Imagination is a weak instrument. Restored, the patient will be more vital and less predictable than the person I had been awaiting during the months of depletion. It's like the difference between thinking of a friend and spending time with him. If he is quick and clever, on his return, those traits will be more marked than you had remembered them to be. And he has other dimensions. You forgot that calm look he gives you, the one that throws you back on your own resources.
     The contrast between a patient depressed and a patient recovered is the contrast between absence and presence. The depressed lack roundedness. Their interests are narrow, their repertoire of behaviors is limited, their account of their own past life has become repetitive and stale. Rarely do the depressed surprise. Willpower is missing, and spontaneity. Depression is the opposite of freedom.
     When depression lifts or is lifted, the person who emerges will appear shockingly strong, full, human. There are shadings and dimensions. It is not a case of "better than well" but of "back to myself." She has returned.
     If I am a reductionist, these dramatic and welcome recoveries from depression are a reason. They are utterly convincing. To see the patient healthy is to understand how gravely impaired she was these many months.
     Lately, in the era of aggressive psychopharmacology, the sharp transformation has become more common. Last week, the medication kicked in. The particular recovery I have in mind took place fifteen years ago, before my immersion, before Prozac, not long into my tenure as a private practitioner here in town. These abrupt transitions were rarer then. Each was burned into memory-but this one for an additional reason.

Margaret was a patient whose depression was visible in the body. In the waiting room, she would sit slouched on the furthermost chair, staring at the floor, ignoring my entrance, moving only when I spoke. As her appointment approached, I would be braced for an hour of fatigue, pain, and frustration.
     And then one Tuesday, there she was, nearby, upright, eyes open, smiling warmly. There is no mistaking the glorious truth, in these instances. The affliction is gone. Entering the consulting room, Margaret pinched herself, the universal sign of being here against all expectation.
     With Margaret, the role of medication was unclear. She was taking antidepressants only because she had seemed yet worse without them; it had been months since I had adjusted her dose. Months, too, that the psychotherapy had meandered. Though it took place in the midst of treatment, the change was something like spontaneous.
     How extraordinary to have Margaret before me. Her voice was brassy where it had been wavering. Her vivacity, the pace of her speech,
the alertness of her mind--every aspect of her demeanor and behavior contained a lesson about the power of depression to distort identity. For all that I had opposed the depression, actively, I had underestimated its effects. This error is common. Even family members come to forget what the depressive was once like, or to disbelieve memory.

Margaret's restoration took place over the course of an evening. She walked me through it:

     I was sitting at dinner with Gregory. He was talking office politics. My mind drifted to my own work. I thought of the moves that one of my colleagues has been making against me. I've mentioned her, Callie, the head of planning. It turns out that Callie's been scheming to absorb my staff. I was at the kitchen table, nodding at the right spots while Greg went on. I found that I had conceived a plan to derail Callie, and what is more, I knew I would carry it off. Which made me realize that I cared. I took a bite of the vegetable. It was awful. When had I stopped paying attention to cooking?
     Does that place the moment? Between my preparing the eggplant dish--poorly--and my thinking about Callie an hour later? There I was across from Gregory, half listening, and I saw that results mattered--the job, the meal, what would happen with the two of us.
     I said to Gregory, "I'm better."
     How could he know what I meant?
     I kept repeating, "No, look. I'm back."

Margaret filled me in on her new life, for that is what she called it. She spoke of her appreciation for this man who had stuck with her. She was amused at his flaws, the ones that only weeks ago drove her to despair.
     She said, "I've been working on the consultation project--marking up the interim reports. I can't believe I let my staff get so far off track. And that Callie--how did I give her the chance to walk all over me? I'm at work on my plot, to ease her out. I'm such a sly fox."
     Margaret had described herself as hard-nosed. But I had never seen her exercise her powers. As Margaret had insisted all along, she had not been herself--in the particular sense that we ascribe to ill health. She had been laid low.
     That was how the session started, with my seeing Margaret, well, for the first time. Later in the same hour, she pulled the rug out from under me. To understand that brief confrontation requires some context.

Margaret had first approached me two years before. Intent on making a good impression, she had arrived prepared to present her life history in compact fashion. She gave an opening summary. She had notes. She would have done well at a business meeting, except that she could not sustain her composure. Within minutes, Margaret was in tears. She had come so far. How was it that the depression had returned?
     Margaret was then in her late forties, five years into a successful second marriage. She was close again to her twenty-three-year-old daughter, Kate, from whom she had been estranged. Margaret's work life was back on track. She should have taken satisfaction in it. But nothing seemed to matter.
     This episode of depression was Margaret's second or third, depending on how you assess her early adult years. Is the tendency inborn? Margaret asked. Margaret's father had been alcoholic, and Kate went through a patch of alcohol abuse.
     Margaret described a negative fairy-tale childhood--Cinderella or Snow White. She had one brother, a cheerful bruiser whom her mother, the strong figure in the family, had favored at every turn. As a girl, Margaret had been steady and meticulous; the flamboyant mother faulted Margaret for her self-control, her pedestrian mind. Margaret's father had been loving, if spineless. He died while she was in college. About then, Margaret began to falter.
     Margaret became not so much sad as passive. She stopped protecting herself from possessive men. The abuse she allowed was verbal, the treatment she had gotten from her mother.
     This pattern is one that our culture produces with regularity. Controlling men are drawn to vulnerable women. Many sorts of fragility suffice: low self-worth, drug abuse, social anxiety. But depression is a prime offender. As the history emerged, I came to suspect that Margaret had been depressed in her early twenties. Certainly, she had had trouble defending herself.
     She moved forward, to graduation and then to public health school, all in a sort of daze. A young professor, ten years her senior, reached down to rescue and wed her.
     The marriage seemed just right for Margaret. Career, daughter, time with friends. The unpleasantness of the college years had been erased. The husband knew how to navigate life--to protect Margaret from her mother, to engineer Margaret's leap from academia to private sector work at a health insurance company. With the help of the marriage, Margaret enjoyed the pleasures of adulthood. The very qualities her mother had mocked--straightforwardness, attention to detail--served Margaret well.
     And then Margaret's life went off the rails again. The husband's affairs came to light. For years, he had been carrying on with students. Now he had gotten a young woman pregnant. He wanted to move in with her. He listed Margaret's shortcomings. Her tastes were prosaic. She was uninvolved in the life of the mind. She was not a playful partner. Margaret recognized these complaints as boilerplate--what a man says to justify running off with a young student. But because they echoed the mother's taunts in Margaret's childhood, the accusations shook her confidence. Was she lacking? And what of her having chosen this man? Could she trust her judgment ever again?
     When the abyss opened, Margaret turned to a male friend for comfort. The husband put Margaret out the door, using her brief affair as justification. And then Margaret's mind stopped functioning. She could not make choices. She did not eat or sleep. She began puncturing her wrists, using knitting needles. That impulse, toward self-destruction, led to overwhelming guilt. She was a mother with an adolescent child.
     This first (or second) episode, in Margaret's thirties, lasted twenty months, full force, and then lingered for another two years. Margaret went through the motions. She rented an apartment and set it up for visits from her daughter. At work, Margaret thought she was holding up
her end, but co-workers found her distracted. Always a slight woman, she became alarmingly thin. The daughter stayed away, more disgusted by her mother's single brief affair than her father's many protracted ones. Before Margaret could mobilize herself, the husband had moved off with Kate, to the university where his pregnant graduate student had found a job. At her own work, Margaret was effectively demoted. It became clear that much of her social life had turned on an intact marriage and her husband's university status.
     Nothing touched Margaret's mood disorder. Antidepressants gave only side effects. The medications had been administered in spotty fashion, by psychiatrists who favored psychotherapy but then threw up their hands in the face of this unengaged depressive. Where therapy helped, for Margaret, was in creating a clear awareness that the flaws she ascribed to family members were not imagined but real, constant, and damaging to her development. The mother had been cruel, the father weak, the brother heedless. The husband was every bit as selfcentered and treacherous as Margaret remembered him to be. The therapy threw into question Margaret's automatic assumption that the disappointments and rejections she encountered arose from her own essential worthlessness. Yet, like the medication, these insights seemed to have had little effect on the duration of the illness.
     In its own good time, the depression waned. Recovered, Margaret relied all the more on caution. It is said that depressives are especially perceptive. Some are. But Margaret was a poor judge of character, and she knew as much. She misinterpreted or ignored social cues. Her own emotional responses were broad, rather than nuanced. She did not fit any standard feminine ideal--she liked fiddling with data, and with organizational charts. Boyfriends found it a challenge to "make a woman of her" by eliciting her vulnerabilities. Because she was prone to prolonged decompensation, Margaret could not afford to make further mistakes that put her at risk of disappointment or betrayal. She became cautious with men.
     Gradually, Margaret reassembled the elements of her life: the career, the relationship with the daughter, surviving friendships. The workplace had served Margaret especially well. Certain colleagues had tried to protect Margaret during her years of compromised functioning. She came to rely on these women more generally, shifting the center of her social life away from the university community.
     A friend introduced Margaret to Gregory, a civil engineer with his own business and his own story of betrayal in marriage. For a while, the relationship was tentative, centered on group activities. In time, Margaret decided that Gregory was trustworthy. The wedding was an important symbolic event, attesting to the possibility of new starts.
     By her own account, since her recovery Margaret had led a contented life, built around work, motherhood, marriage, and friendships. The new episode of depression came on for no apparent reason.

In her sessions with me, Margaret had been slow to reveal the extent of her symptoms. She was puncturing her wrists again, without knowing why. She was not eating or sleeping. Life was uniformly flat, except for the fluctuation in her feelings of pain.
     Margaret and I talked weekly. We worked with medications, the few that were available back then. Research shows that psychotherapy and medication help to ameliorate depression and that the two in combination work better than either alone. Margaret's response to treatment was modest and sometimes paradoxical. Psychotherapy gave Margaret a feeling of security; in this new state, she felt freer to obey her impulses and attack her forearms with knitting needles. I raised doses, added adjunctive medicines, to no effect--until one day, a particular combination worked, in this sense: Margaret lost the compulsion to injure herself.
     Substantial elements of the depression remained. For Margaret, her apathy was a form of torture. Her enthusiasm had defined who she was.
     Margaret had little self-protective impulse. She would as soon have died as lived on--although she did fear being "found out" in her depression. She struggled to produce a semblance of normal affect for Gregory and for her staff at work. She wanted her daughter to feel loved; but for Margaret, in the midst of the prolonged depressive episode, many of the components of love--the upwelling of feeling, the hopes for the future--were hard to access.

Treated with medication and psychotherapy, even second and third episodes of depression should moderate within weeks. When depression lasts, psychiatrists worry. We have seen too much; we know how corrosive mood disorders can become, how capable of obliterating the self.
     My first clinical encounters with depression involved frightening afflictions. I did my training in the 1970s. Psychiatric instruction in both medical school and residency began on the inpatient ward. If patients were hospitalized more liberally back then, still, ward populations were dominated by the gravely debilitated. Medications were used sparingly. Illnesses ran their course.
     In its late stages especially, depression had a typical look. Blank stare, downcast eyes, knitted brow. Head supported by a hand, or face masked by one. This expression is one that artists have depicted for centuries, what doctors call the facies of the illness. Depression also has a habitus--the body slumped and inert, muscles flaccid. I saw the face and posture of depression repeatedly in my training--not only on the psychiatric ward but also in the general hospital and in the outpatient clinic. End-stage depression was common.
     The face and the posture are products of the career of depression. Reviewing the chart of the depleted inpatients, you would see notations that stretched back over decades. Accounts of recurrent hospitalizations and incomplete recoveries. And then, hidden in the back of the folder, perhaps, notes from the initial interview at an outpatient clinic. They would contain a description of someone resembling a "good prognosis" patient of the sort who might be admitted to bring hope to an embittered ward--or assigned to a beginning student for treatment in psychotherapy.
     Tortured immobility was a destination for the depressed, a terminus ad quem. A few would commit suicide. Many would get better. Some would trade depression for the nonspecific "neurosis" that in those years was the most common psychiatric diagnosis. But a number were headed toward this bleak fate. Working with the sickest patients first gave every student a picture of the arc of depression. Even for the young, responsive depressives, it can end here.

Today, we encounter the face of depression less often. If I believe that treatment works--and here I do not mean any one treatment in particular, but the psychiatric enterprise, everything we throw at depression--it is in part because the end stage of depression is rarer. Now what we see, down the road, is a subtler picture: aimlessness, diminished functioning, spotty memory and poor concentration, dense apathy, tears and a pervasive sense of suffering. The symptoms tend to wax and wane. The depressive may find a good hour or two in the day, and enough energy to fake it for a while. That is the progress we have made-from the horrific to the terrible.
     Still, this patient could be the one we lose for good and all. And so when depressive symptoms settle in to stay, doctors exercise a tendency Freud warned against, therapeutic zeal. With Margaret, the less the medication helped, the more intent I was on making a difference with psychotherapy. But it was hard to find the path.
     Margaret complained of listlessness. Did the symptom bear interpretation? Many depressed patients are listless, and they come with diverse histories. Listlessness is part of the syndrome. Yes, Margaret was listless today, in the face of a particular challenge, but she was listless many days, without apparent reason. Rarely was I optimistic that interpreting this lethargy now--its origin, its quality--would move us ahead.
     Often I tried to provide what any injured person needs: hope, steadfastness, support. I might choose simply to turn down the volume, to tiptoe past apparent meaning. Margaret would complain of her inability to feel what a mother should feel toward her daughter. Sometimes, I would refer to the hurt Margaret had felt when Kate chose to go with the ex-husband. But often I would favor a response from within the medical model: Yes, I would say, that's a symptom of depression. In the midst of an episode, it is hard to muster warm feelings or the energy to convey them. That's the illness, I would say. I have seen patients recover from worse.
     I stood alongside Margaret, in solidarity with her and in opposition to her ailment. Rather than interpret self-accusations, I would make the case for setting them aside: It does seem that way, when you're depressed. My goal was to put distance between Margaret and her negative certainties.
     Depression skews the proportions of good and bad recalled in self and others. The I has always been needy and useless; they have always been justified in their rejection of the damaged, sinful, demanding self. It would have been folly to take every one of Margaret's outcries at face value--except as expressions of her current state. Besides, Margaret had done some of the heavy lifting of a traditional psychotherapy during her earlier bout of depression. Now, her self-explorations were repetitive and mechanical.
     Our sessions functioned as a repository for pain. Because Margaret brought it here, she could face the rest of the week. My role was to help Margaret to hold on to her job and her marriage. To help her sustain the relationship with Kate. To keep Margaret bound to life.

My training had been in depth therapies, the ones that attend to unconscious conflict. And here, depression sets traps for the psychotherapist. The depressed appear to be ideal candidates for treatment. The depressed are impaired in their ability to sustain feeling. Many have experienced early trauma and later violations of trust. In the midst of an episode, the depressed have ready access to a library of negative memory. They are caught between choices, as if torn emotionally, mired in uncertainty.
     I knew that in her good years Margaret had been less introspective and more action-oriented than the troubled figure before me now. But how could I sit with Margaret and not explore the obvious questions--whether she was paralyzed by ambivalence about the people and roles that defined her life?
     Had Margaret's attachment to Gregory been weakened by mistrust of men in general? When she gave her daughter encouragement, did the effort arouse difficult memories from Margaret's own early years with the critical mother? Surely Margaret's enthusiasm for her career was alloyed by recollections of its origins in the awful years on campus and the failed first marriage.
     For much of the last century, depression was understood to arise from a particular type of inner struggle. When love is intertwined with hatred so shameful it cannot be acknowledged, the result is a sapping of energy and an inhibition of action. We dislike or fear or disdain our husband, daughter, mother, and career, though we tell ourselves that we should love them merely. This unresolved ambivalence gives rise to guilt, self-hatred, and finally depression. Therapists had moved some distance from that set of beliefs. All the same, it was hard not to view depression as Margaret's secret, forceful critique of her circumstances--marriage, motherhood, career, the whole investment in a "fulfilling" modern life, every element of which had already revealed its shortcomings to her in prior years.

Empathy is a tool of psychotherapy. Mirroring the patient's mood state with great exactness can be stabilizing, soothing, reassuring. But which emotions shall we mirror?
     In therapy, I found myself filling in for Margaret. She might report that Kate had received a promotion at work. You must be proud, I would say.
     In truth, Margaret was all but untouched. Proud, yes, if one counted a thin version of pride, achieved with difficulty. This aspect of depression is one of its most painful: alienation from feelings that accord with one's values, the inability to treasure the things one should hold dear. It was when I mistook this difficulty that Margaret was most likely to rebuke me. That I of all people would attribute pride to her! When what I was treating was an absence of feeling.
     And then I might say--how hard, to be unable to care!
     That sort of empathy could throw Margaret into a tailspin. Was she an uncaring mother? In her depressed state, Margaret was so prone to self-accusation that she was open to accepting any hypothesis, as long as it put her in the wrong. Psychotherapy depends on the patient's being able to correct the therapist: No, it's not that way. Margaret would correct me only if I complimented her.

But she corrected me now. We were at the end of the session in which Margaret had announced her recovery. Was I expecting thanks, for having pulled her through? Margaret's priority was to set the record straight.
     She looked me in the eye and said, "I'm annoyed with you, for pushing me so hard. You wanted too much from me--at work, or with Gregory. You deal with depression every day. Didn't you know how much each effort cost me?"
     These are the two images that stick in memory: Seeing Margaret recovered. And then, as I sit self-satisfied before her--having her face me head-on and bring me to attention. I had done it wrong, the project of psychotherapy.
     Margaret's complaint was not entirely fair. Because I worried about her career and marriage, I had made behavioral demands. I had encouraged her to persevere at work, even though the job had lost all meaning for her. I like to see patients sustain the semblance of normal life. Sometimes it pays to ask for miracles.
     But then she added a second charge: In here, too.
     I understood the complaint and agreed with it. I had pushed Margaret too hard emotionally. Not often. In a handful of sessions over the course of two years, I had stretched to make connections--asked Margaret if that was the way it was. She had said, yes, exactly, it's just that way--even upped the ante. But the dark mixed feelings, about Gregory and Kate especially--those were illness. Ordinarily--in health--Margaret was an enthusiast. If Margaret had appeared ambivalent about her husband or daughter, that was because she had been suffering apathy and pessimism. When you simply cannot experience warmth or attachment, then past disappointments and humiliations will play a disproportionate role in your mental landscape.
     As our session wound down, I recalled another patient, a man whose guilty ruminations disappeared when his depression ended. I'm burning my diary, he told me. I was like a torture victim confessing to fantastic crimes he never committed. The intrusion of that memory helped me understand the source of Margaret's resentment. She wanted to know why, in our discussions, I had granted an impostor--the depression--such standing. I had been negotiating with an occupying government, of Margaret's mind, while the legitimate ruler was in exile.

Often, the therapist's task is straightforward. Imagine a troubled man who is hostile toward women, in ways that interfere with his own happiness. Secretly, let us suppose, he fears women, because he considers himself needy and inadequate, and therefore vulnerable--and so he has constructed this defense, contempt. His hunger for attachment gives rise, paradoxically and against his own interest, to aggression. Here, the therapist must select the emotion that requires acknowledgment, so that in time the whole picture will come into focus. The therapist might say, "You despise women," or "You fear them," or "You adore them," and in each instance, the therapist would be right. Those emotions attach to the man; however contradictory, they are his, all of them.
     Margaret's claim was that her case was mostly not of this sort. The emotions--those we had discussed during the months she was depressed--were externally imposed, by the onset and then the persistence of an illness. My empathy had been misplaced, my interpretations ill-founded. The guilt--that was not Margaret, nor the indifference either.
     Past disappointments might have injured Margaret and made her vulnerable to recurrences of depression, including this seemingly spontaneous one. Arguably, before her prior course of psychotherapy, memories of past harm exercised whatever troubling effect undigested experiences impose on the mind. But those same memories did not, in any meaningful sense, cause or sustain the disabilities Margaret bore during this latest episode, the one that arrived out of the blue.
     Margaret never had substantial misgivings about motherhood, career, or her current marriage. A disease had robbed her of feelings that were properly hers and imposed alternative ones. Margaret was the one who experienced those depressive feelings and reported them in psychotherapy; but they did not arise from, or had only weak roots in, the psychology that was hers in health.
     Later, reading a Philip Roth novel, I came across a passage in which a political progressive recalls his mistaken thoughtfulness about the behavior of a mindlessly vicious right-wing politician: "In the name of reason, you search for some higher motive, you look for some deeper meaning--it was still my wont in those days to be reasonable about the unreasonable and to look for complexity in simple things." In time, I felt that way about my approach to Margaret's depression. I had attributed too much meaning to an arbitrary opponent.

In the following weeks, I came to know Margaret, healthy. She was a steady, reliable, highly competent woman. She was even-tempered and frank--not an ounce of neuroticism in her. There was an ease in Margaret's loves, for her daughter and for Gregory. Her work excited her; what others would have called dull, she found absorbing.
     I have referred to a roundedness that can seem to be absent in the depressed. Elements of a self are missing. As onlookers, we may be tempted to fill them in, using our imagination to supply intentions, wishes, or beliefs, in order to provide a coherence that is lacking. In the process, we may create false drama; the depressive runs a similar risk, of ascribing to herself motivations (especially guilt-inducing ones) that must be present, to explain feelings or their absence.
     Recovered, Margaret was an altogether less complex person than she had appeared in the midst of her depression--which was all to the good, in her view and my own. She was straightforward and vibrant, straightforward and whole. The rounded, the immediately present person was a largely untroubled one.
     Margaret felt no guilt--that change alone would have been remarkable. Where her emotions were contradictory, she forgave herself, leaning on the truisms of the day. Do I sometimes hate my daughter? Look, it's a generation of ingrates. Who doesn't hate these kids sometimes? The negative feelings were not one pole of a paralyzing ambivalence. They were passing breezes. Margaret adored her daughter, simply, directly. Margaret's emotional structure was less convoluted than any psychological theory of depression would have predicted.
     She was sunny, frank, jocular. She called 'em as she saw 'em. She could stand to be tough. She was untroubled, not a ruminator.
     If Margaret was inclined to move on, I was not. For me, the shaky psychotherapy contained a moral, that depression was yet more pathological than I had made it out to be. One collects these lessons--long before my scrap and shard project, I was gathering patient stories, as every doctor does. Those uplifting, most welcome recoveries, when I was lucky enough to witness them, seemed often to reinforce the point: depression is more distorting than we imagine, more self-estranging, more other.

Superficially, depression sometimes resembles passion, strong emotion that stands in opposition to the corrupt world. This impression can arise from the solidity of the symptoms; depression looks like a sitdown strike. Or it can arise from depressives' tendency to act impulsively. Who would puncture her arms but a woman of passion? In truth, the puncturing is an attempt to feel anything at all. Depression is passion's absence.
     Simply to name emotions--you feel such guilt--is to lend them legitimacy. With Margaret, in employing the most basic elements of therapy-empathy, tentative interpretation, the search for meaning--I had in effect sided with the illness and against the person Margaret was in health. The feelings I had underscored for Margaret were foreign to her. She experienced them, she reported them, but there is a sense in which they were not hers.
     I don't mean to exaggerate my doubts over Margaret's treatment. The episode of depression we worked on together ended more quickly, and with less destruction in its wake, than the one that preceded it. I was open to the possibility that I had done Margaret some good. I am pointing to one of many experiences that convinced me, cumulatively, on an intimate basis, that depression is best understood as disease, for good and all.
     One wants to take care to avoid negotiating with impostors. Seeing Margaret in the weeks after her recovery reminded me how much I do,
in fact, appreciate self-confidence and contentment. It was wonderful to hear of her slashing her way through the thickets of the business world.

Many depressed patients, between episodes, remain moody and introspective. Margaret had no such tendency. Her personality style left the period of depression sharply demarcated. If a bout of depression arises without evident causation, if the mind-set it produces is distinct from the patient's usual attitudes, if the episode ends cleanly--that sequence contains some of what we mean when we say that depression is pathology. In a discussion about eradicating depression, anyone inclined to favor that project would include cases like Margaret's.
     But think about Margaret's prior encounter with mood disorder, following her discovery of her first husband's treachery. That episode was prolonged and disruptive, but its causes were entirely obvious. Does the episode's being understandable influence our attitude toward it? And what about the college years? Margaret's difficulties then fit into any number of available categories: mourning, immaturity, adolescent adjustment. That episode, if it was one, would cause trouble, in the eradication discussion; but in retrospect, we may feel compelled to count it as illness. And with good reason on additional grounds: even early, mild depression can be distorting of memory and identity. Even first episodes can spawn impostors--that is, traits that are foreign to the self.
     Disease is a concept that we understand through usage. Epilepsy is a disease-not each seizure, but the underlying condition, the tendency to experience convulsions, and the sweep of recurrences. Similarly for asthma. If the fifth, serious bout is asthma, then so was the first, mild one. This model holds especially strongly for depression, which is progressive. Each episode confers a greater liability to the next; each is a risk factor for a lifetime of chronic and recurrent mood disorder. By this logic, late and clear-cut episodes cast a shadow backward, inviting us to treat every episode vigorously, to halt illness in its tracks.

As new research evidence emerged about the harm depression does, I became ever more wedded to the notion that depression is disease altogether, across a wide range of presentations. But alternative views of depression, the ones I have called romantic, are so ordinary a part of our assumptions that they can persist even in the face of decades of work with patients. In my years of immersion--traveling, addressing different audiences--I discovered that outside the office, my own understanding of depression remained muddled.