A Case Study in Historical Critical Psychopathology
Scott A. Baldwin, Department of Psychology, University of Memphis
Daniel C. Williams, Department of Psychology, University of Memphis
Arthur C. Houts, The West Clinic, Memphis, Tennessee.
We thank Rich McNally for providing us with preprints of his work and for discussion of some of the ideas in this paper. Scott Baldwin was supported by a Van Vleet fellowship from the University of Memphis. Correspondence concerning this article should be addressed to Scott Baldwin, Department of Psychology, Psychology Building 202, University of Memphis, Memphis, TN 38152. E-mail: firstname.lastname@example.org.
We describe some basic assumptions of an interdisciplinary approach to psychopathology called historical critical psychopathology. This new approach is contrasted with traditional approaches to psychopathology as found in mainstream mental health literature and textbooks. We illustrate how this new approach can be used in a case study of posttraumatic stress disorder (PTSD). We identify and critically review three periods of the history of PTSD: creation, expansion, and embodiment. PTSD can be seen as a problem diagnosis because the disorder is poorly conceived and subject to ever-expanding scope of coverage. Historical critical psychopathology can be applied to other diagnoses and may be especially useful in anticipating problems and patterns of development in three emerging new diagnoses: traumatic grief, premenstrual dysphoric disorder, and relational disorders.
Posttraumatic stress disorder (PTSD) is a recently identified mental disorder, having been formally named in the third edition of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM–III; American Psychiatric Association, 1980). Since it was first named, PTSD has been surrounded by controversy and debate regarding such issues as: (a) whether PTSD is continuous with previous wartime adjustment conditions (Leys, 2000; Shepard, 2001; Young, 1995); (b) whether PTSD is a unique disorder separable from anxiety and depression symptoms (O’Donohue & Elliot, 1992; Scott, 1990); (c) the extent to which PTSD occurs outside of exposure to the extreme stress and danger of wartime combat (Rosen, 1996); (d) the possibility that monetary incentives could tempt individuals to dissimulate having the disorder (Burkett & Whitley, 1998; Rosen, 1995); and even (e) whether PTSD is a disorder at all (Summerfield, 2001). Despite nearly a quarter century of research on PTSD, controversy about its status as a legitimate diagnosis continues and has spread beyond the United States to include recent lawsuits brought by UK veterans against the British Ministry of Defence for failure to provide assistance for veterans claiming to suffer from PTSD (“Veterans Sue,” 2002). Canadian veterans have filed a similar suit against their government (“Soldiers Joining,” 2003).
The purpose of this article is to add another perspective on the PTSD controversy by applying a new approach to the study of psychopathology we call historical critical psychopathology. Our approach is historical because it emphasizes the historically situated and contingent aspects of mental disorders. We assume that disorders may come and go in the history of mental health diagnosis and treatment, and we focus on the historical discontinuity of disorder appearance and disappearance. Regarding the critical component, we base this approach on a critical reappraisal of key assumptions found in traditional approaches to psychopathology. This skeptical stance stems from a variety of sources in philosophy, sociology, and history of science cited throughout this paper.
We describe PTSD in three distinct but overlapping periods of development: creation, expansion, and embodiment. Throughout the historical presentation we point out the discontinuities between current conceptions of PTSD and previous ways of thinking about war-related mental health problems as well as previous notions of traumatic injury. We first take up the details of post-Vietnam syndrome and show how a few mental health professionals in conjunction with antiwar-activist veterans successfully lobbied the public, the American Psychiatric Association, and eventually the U.S. Veterans Administration to identify the syndrome first called PTSD. This period of creation lasted from approximately 1970 to 1980. Next, we describe the period beginning in 1981 and continuing to the present, when PTSD was expanded to include much more than reactions to wartime combat. This includes an examination of how the criteria for PTSD changed from DSM–III to DSM–III–R to DSM–IV. Additionally, concepts of trauma and what may be counted as stressors have changed, leading to the inclusion of more individuals under the diagnosis. Not surprisingly, this period of expansion was also associated with expansion of various new treatments for PTSD. The third period, embodiment, actually overlaps with the period of expansion, beginning in the mid-1980s and extending to the present. A major feature of the embodiment period is the effort to localize PTSD as a disorder with a physiological signature, to literally move PTSD into the body. This has taken place amid ongoing criticism of the PTSD concept and marks an effort to objectify PTSD as a legitimate disorder with a physiological basis. After all, if PTSD can be shown to be a disorder with characteristic physiological markers, then those who question the validity of the condition as a disorder can be silenced by pointing out that PTSD is, like cancer, a true medical disorder.
Before taking up the specific case of PTSD, it is important to outline some basic assumptions of historical critical psychopathology as an approach that differs from traditional approaches. A fundamental tenet of historical critical psychopathology is that ideas and concepts have embedded social and cultural histories, and the ideas and concepts of historical critical psychopathology are no different in this regard. Our approach arises from a number of different streams of influence and compiles strategies pioneered by a number of investigators, many of whom belong to disciplines far afield from the mental health professions. In the following background introduction to the sources of historical critical psychopathology, we catalog some of the fundamental differences between traditional approaches and our approach.
Basic Conceptual Differences of Historical Critical Psychopathology
For the past hundred years or more, the mental health professions have prided themselves on taking an increasingly scientific approach to mental health problems. One can readily see this reflected in standard textbooks (Barlow & Durand, 2002; Davison & Neale, 2001), where the history of mental health diagnosis and treatment is presented within the Enlightenment credo according to which every day, in every way, we are learning more and more objective information about mental illness, and patients are being treated more and more humanely as a result. The enlightened present of positron emission tomography (PET) scans is contrasted with the bad witch hunts of the not-so-distant past. Silly sounding humoral theories of depression are contrasted with current neurotransmitter and receptor site theories.
Models and Uses of History
Ontology of Mental Disorders
Epistemology and Knowledge Claims
Historical critical psychopathology questions this traditional formulation of the history of mental health fields. The different assumptions of historical critical psychopathology are briefly summarized in Table 1. The basic assumptions or principles are interlocking across three domains: concepts and uses of history, the ontology of mental disorders, and epistemology and knowledge claims. Each is briefly reviewed below.
Models and Uses of History
Mental Disorders Can Appear and Disappear in History
Historical critical psychopathology assumes that there can and will be radical discontinuities in the mental health field so that mental disorders may show up at one time and then disappear at another time. This approach draws upon work in philosophy and history of science, investigations by nonmental-health-professional historians and sociologists, and cross-cultural work in medical anthropology (Berrios & Porter, 1999; Bynum, Porter, Shepherd, & Wellcome Institute for the History of Medicine, 1985; Kleinman, 1987, 2001; Mezzich et al., 1999; Micale, 1993; Micale & Porter, 1994; Porter, 1987, 1988; Scull, 1989). Insights from these domains show how current concepts of mental disorder bear the marks of the current culture of mental health, which is dominated by the idea that mental disorders are a type of medical disorder.
Our Time Is Not Any More Prescient than Previous Times
We are skeptical about the underlying assumption of progress that is written into the canonical histories of psychopathology. The idea of Enlightenment progress provides a kind of guarantee that because we are swept along in the river of scientific advance and humane values, our current condition is by fiat more advanced than that of our predecessors. In other words, by merely being here later in the historical stream of progress we are more knowledgeable and more humane. History is, of course, then reconstructed in this image and according to this story line. Historical critical psychopathology questions this history of mental disorders. Perhaps our knowledge is indeed more advanced, but then again, maybe it is not. For example, a common story in traditional histories of psychopathology is often repeated in undergraduate textbooks: the story of trepanning. The story goes roughly as follows. Thousands of years ago, primitive people cut or broke holes in their members’ heads as a treatment for mental disorders. They believed that the holes released the evil spirits that caused the individual’s problem behaviors. The subtext of this story is something to the effect of “Can you believe how stupid this is?” and ”Look how far we have come.” A rather obvious modern parallel is never mentioned in the textbooks: prefrontal lobotomy. Only just over 50 years ago, Egas Moniz won the Nobel Prize in medicine for inventing the leucotomy, which was the precursor to several other lobotomy procedures (Valenstein, 1986). The theories used to legitimize various lobotomy procedures now appear to have been quite bizarre, perhaps on par with theories appealing to evil spirits inhabiting the skull. The point is that history can be made to appear many different ways depending upon how one arranges and presents the facts. Historical critical psychopathology does not take for granted that our time is any more progressive than previous times and in fact actively looks for evidence that it is not. We are not saying that progress never occurs; only that it is not guaranteed merely with the passage of time.
History Can Be Written to Criticize Rather than Glorify the Present
In canonical histories, historical facts are assembled in such a way to make us look enlightened. In historical critical psychopathology, history is used to question the current state of affairs. Facts are assembled to raise questions about those things we take for granted. For example, several undergraduate textbooks reproduce one of two famous paintings of Philippe Pinel releasing mental patients from their chains (Barlow & Durand, 2002; Davison & Neale, 2001). One painting involving release of men at the Bicetre asylum is entirely fictional (Harris, 2003). The second painting, the one reproduced in textbooks, involves the release of women at Salpetriere, and although the ceremonial scene never actually took place it is not entirely fictional because Pinel had by that time accepted the idea of releasing patients from their shackles, an idea he had at first opposed. The reproduction of painted images of Pinel are often used as a kind of historical record to bolster the claim that Pinel was the father of enlightened and humane treatment of the mentally ill. Recent historical work that challenges the assumptions of standard histories has dubbed this use of Pinel the “Pinel Myth.” Weiner (1994) summarized the key findings as follows:
What is most surprising, given the ubiquitous propaganda for Pinel as the “chainbreaker,” is that, when he initially arrived at Bicetre on 6 August 1793 as “physician of the infirmaries,” Pinel accepted the traditional use of chains to restrain the violent insane as a matter of course. In his “Memoir on Madness” of 11 December 1794, he mentioned without further comment that three of his patients had been shackled for fifteen, twenty-five, and forty-five years respectively. It was his talented assistant Pussin who, evidently on his own initiative, first freed the insane men at Bicetre from their fetters in Prairial, Year V (May–June 1797). At that time Pinel had been gone from Bicetre for two years. The doctor [Pinel] praised Pussin’s exemplary accomplishment in the second edition of the Treatise in 1809, mistakenly giving the date “Prairial, Year VI” for Pussin’s historic initiative, adding that he followed his assistant’s example “three years later” (p. 237).
Pinel was not present when the chains were removed from patients at Bicetre, and he was not the originator of this idea. The idea and the substitute procedure of using a straightjacket came from Pinel’s assistant Pussin, who was not a physician. How, then, was Pinel made the originator of the idea and memorialized for doing so in paintings and textbooks?
At least part of the answer lies in the actions of Scipion Pinel (1795–1859), Philippe’s physician son, and J. E. D. Esquirol (1772–1840), one of the elder Pinel’s most talented students and later a powerful figure in French medicine. For different reasons, both of these individuals advocated a view of the past that memorialized Pinel as the one who broke the chains of the mentally ill and made him out to be a philanthropic reformer who carried out the political agenda of bringing freedom to even the lowest of citizens (Vandermeersch, 1994; Weiner, 1994). The paintings in question were in fact commissioned to convey the theme of Pinel as a hero of the new enlightened order of the French Republic. These images were not painted to provide photographic evidence about historical events.
Ontology of Mental Disorders
Mental disorders are human constructions not yet real in the way that stones and trees are real. We refer to the mental disorder as medical disorder approach as the traditional approach. In the traditional approach, mental disorders are taken as given by nature and as historically and culturally transcendent. In effect, mental disorders are presumed in this traditional approach to be analogous to physical medical disorders. In this regard, we can say that mental disorders belong to the ontology of timeless objects, being the name for natural things rather than names for human-made things. According to this traditional approach, mental disorders are, like physical disorders and like physical objects found in nature, more like things we discover than things we manufacture.
In contrast to this traditional view, historical critical psychopathology recognizes that the “existence” and “reality” of mental disorders is best understood by an ontology that allows for degrees of qualities such as permanence, stability, and historical and cultural transcendence. Foucault was among the first to raise this possibility (Foucault, 1987). More recently, this dimensional ontology is best described by Hacking’s phrase historical ontology and is outlined below (Hacking, 1999, 2002).
Hacking (1999) provides a useful analysis of the dimensional difference between what we have labeled the traditional approach and an alternative historical critical approach. This is summarized in Table 2. The dimensions are bipolar scales where objects of study, mental disorders in this case, are ranked either to the left or to the right in philosophical space. It is important to emphasize that this dimensional parsing of objects of study presents some dimensional extremes and that for any given field of knowledge its objects of study may fall to either side of this dimensional space, with almost none being perfect examples of either extreme. Take the first dimension, for example. In the traditional approach, mental disorders are regarded as inevitable. Once we find that people who display certain unusual behaviors actually have brain wave abnormalities we conclude that this behavior was the manifestation of seizure activity all along, and not, after all, the result of demonic possession. It is the “all along” and the “after all” that conveys this sense of inevitability. We discover epilepsy, and we declare that those who believed the ontology of ghosts and goblins were deceived by their mistaken conventions.
Ontological Dimensions in Historical Critical Psychopathology and Traditional Psychopathology
Source: Hacking (1999).
Another bipolar dimension concerns the status of the language used to describe objects of study. At the nominalistic extreme, words and names are social conventions that bear no special correspondence with the way the world is ordered. In contrast, at the other extreme of this dimension, words and names denote natural divisions of things and correspond to divisions and distinctions found in the joints of nature. We have, for example, more confidence that “tuberculosis” is the name for a state of affairs found in nature than we do for “consumption.” “Consumption” has more the ring of convention and arbitrary naming, whereas “tuberculosis” sounds more like a fact and rigidly designates what is and is not tuberculosis by the presence or absence of some etiologic organism that follows a particular pathophysiology.
The stability of knowledge is important to all scientific fields. This is embedded in concepts of reliability and replication. We regard things as real to the extent that we can repeat them. Histories of mental disorders are often written to demonstrate that a given disorder shows up repeatedly across time. The idea is that if one can demonstrate that a disorder shows up repeatedly across time and across cultures, one has evidence that the disorder is a state of nature rather than a social and cultural artifact due to social mores and conventions. Conversely, when disorders come and go we typically suspect that their instability is indicative of a social rather than natural basis. For example, fugue states and hysterical conversion reactions were once much more prevalent as disorders in Western societies than they currently are, and their relative disappearance is almost certainly due to changing cultural conditions and fashions within the mental health field (Hacking, 1998; Micale, 1993).
The polar opposites of relative dependence on and relative independence from human social conventions are similar to the aforementioned stability dimension. There is a sense in which the objects of study in astronomy do not depend on human social conventions. The planets are there whether we see them or not, and we may mistakenly classify moons as planets, but the very existence of the objects does not depend on our classifications and naming operations. In contrast, mental disorders have less independence from our conventions. What is regarded as a mental disorder may depend very much on social conventions, as the cases of inclusion and exclusion of homosexuality and neurosis with the DSM has shown (Bayer & Spitzer, 1982, 1985).
Hacking’s distinction between interactive versus noninteractive objects is also useful as it designates the extent to which objects of study within a field of inquiry are influenced by the practices and conventions of the investigators. As noted, the planets do not adjust their behavior according to our conventions for naming them. In contrast, people can and do adjust their behavior in response to our conventions for declaring them to have or not to have a mental disorder.
Within historical critical psychopathology, the ontological status of mental disorders is not assumed to be on par with the ontological status of the objects of physics, astronomy, or even most of physical medicine. Rather, we assume that the status of mental disorders is located somewhere within the extremes of Hacking’s multidimensional space, where some disorders are more constructed and some are more natural. In this way, it is possible to look upon the domain of current mental disorders and arrange them within this space. For example, Houts (2001) suggested that the five dimensions of Table 2 be assigned facticity or objectivity ratings from 1 to 5, where the extreme left-hand anchor is 1 (i.e., contingent, nominalistic) and the extreme right-hand anchor is 5 (i.e., inevitable, natural kinds). One can then place diseases and disorders in this space by assigning them a total score based on the sum of the five ranks. Not all of the disorders in the current version of DSM would receive equal scores. For example, social phobia is likely to receive a lower score than Bipolar II disorder, which is, in turn, likely to receive a lower score than dementia due to Huntington’s disease.
Dissociative identity disorder has undergone various permutations but is most likely a condition that falls more to the left side of Table 2 than to the right side (Hacking, 1995). Some critics of this condition have suggested (pun intended) that the very condition of multiple personality is manufactured in the consulting room and is rarely if ever found spontaneously in nature (Borch-Jacobsen, 1997). Where a given disorder falls within the dimensional space of Table 2 may also change over time as, for example, happened in medicine with the transition from consumption to tuberculosis. One of the purposes of taxonomy in medicine is to facilitate this type of transition from disorders defined at the level of signs and symptoms to disorders defined in terms of biological causal mechanisms (Scadding, 1988).
Hacking’s Typology and the Concept of Disorder
The dimensional typology outlined in Table 2 can be applied to the very concept of disorder itself. For purposes of this article we have generally adhered to a concept of disorder that holds as its prototype the medical notion of a broken bodily function that is the cause of the observed signs and symptoms. Using such a standard, Huntington’s disease is a clear-cut disorder and there can be no sense to the notion that Huntington’s disease is constructed by our conventions and naming operations. Huntington’s disease is objective, real, not socially constructed, and found in nature quite apart from our observations and naming activities. We can locate Huntington’s disease in the far right of Table 2. As we noted, not all of what is in the current DSM–IV–TR can be regarded as on the same footing as Huntington’s disease, but instead falls further to the left on Table 2 than Huntington’s disease. Most of the disorders in the mental health field are defined as syndromes and have not been proven to be caused by specific biological breakdowns. Such conditions are nevertheless referred to as disorders, and this has led to a large literature on what the proper meaning of disorder is as well as to speculation on whether or not it is possible to set forth an adequate definition of disorder (Houts, 2001; Lilienfeld & Marino, 1999; Wakefield, 1999a, 1999b). In this paper, we use the term disorder to refer to problems caused by a broken biological function, such as Huntington’s disease. Thus, when we refer to a disorder as “real,” what we have in mind is the extent to which the disorder matches this ideal.
Physiological Data Can Be Socially Manufactured
People frequently are impressed by physiological data in the social and behavioral sciences because readings of the body are often taken as “hard” evidence of a kind not always available in social investigations. There is a certain fascination with physiology, as if when we have measured the body’s response we have somehow reached bedrock. As members of our culture, we are predisposed to view behavior as the result rather than the cause of physiological perturbations. As we show in the case of PTSD, physiological differences between persons with a diagnosis of PTSD compared with those without the diagnosis has been used rhetorically to champion the “reality” of PTSD and to discredit critics. Such a strategy has a long history in the field of psychopathology, and traditional histories feature key examples such as general paresis and the discovery of the syphilitic spirochete. What is often overlooked is that physiological measures can be systematically altered by behavior and surrounding social circumstances. This is especially true for standard psychophysiological parameters of anxiety (e.g. heart rate, skin conductance), but it has also proven true for such measures as brain glucose metabolism and cortisol secretion (Baxter et al., 1992; McKnight, Nelson-Gray, & Barnhill, 1992; Schwartz, Stoessel, Baxter, Martin, & Phelps, 1996). The field of biofeedback is predicated on the observation that, given proper instruction and proper circumstances, ordinary people can deliberately alter aspects of their physiology. Merely taking physiological measures is no guarantee that one has reached bedrock in establishing the “reality” of some presumed disorder because the physiological measures are themselves possibly interactive with participant behavior and surrounding circumstances, not to mention the overarching conceptual problems associated with localizing psychological phenomena in the brain (see Uttal, 2001, for a readable discussion of those problems). From the spinal concussion of railway trauma, one of the many presumed physiological correlates of mental injury from late-19th-century railway accidents (Brown, 1995), to the enlarged ventricles of schizophrenia (Raz & Raz, 1990), the history of nonspecific and noncausal correlations between physiological measures and mental disorders is long and colorful. Historical critical psychopathology is suspicious of the practice of using physiological measures to establish the reality of mental disorders. Perhaps their use is enlightening, but it can also be deceiving.
Epistemology and Knowledge Claims
Traditional approaches to psychopathology are rooted in generally accepted views of knowledge claims found in the natural sciences. To (over)simplify, this approach contends that knowledge is measured by the extent of correspondence between the verbal claims and the conditions of the world. The archetypal image of the process is that of the lone investigator collecting observations and specimens, organizing them, and then having a great insight into the nature of some corner of the universe. However, proponents of this approach often underemphasize the social, historical, and cultural basis of knowledge claims. This social view has become apparent in recent times because of the rise of “big science” and because the field of social studies of science has highlighted the social basis of scientific knowledge (Fuller, 1988, 2000; Mirowski & Sent, 2002). It is important to note that there need be no dichotomy here between a social view of science and traditional views. Several features of the social view are important for historical critical psychopathology, and these are briefly elaborated below because they are important for understanding the history of PTSD.
Knowledge Claims Are Social Accomplishments
This is a rather uncontroversial claim in the sense that knowledge claims are obviously made by people through the public venue of organized science. Successfully staking a knowledge claim requires the backing of a number of supporters to carry out the investigation and to publish the results. Publication is an inherently social process involving passage through the gauntlet of editors and reviewers. That is, scientists must garner and organize resources to produce knowledge claims, and must lobby to get those claims established in reputable sources (Burr, 2002; Latour, 1987; Latour & Woolgar, 1986).
This process is often forgotten in accounts of the history of psychopathology. In historical presentations a kind of anamnesis takes place in which current claims about the state of knowledge regarding some disorder are taken as inevitable and the social basis for current claims is perhaps inadvertently swept under the rug. The point is not that there is a conspiracy to hide such social influences;instead, we are arguing that information about social influences does not make it into historical presentations because the people writing the histories of mental disorders adhere to a point of view as to how history should be written that regards social influences as unimportant. Looking backward, we can find obvious examples of the social basis of knowledge claims that came to be regarded as falsehoods when the social climate changed. For example, drapetomania was a term introduced to describe a mental disorder among slaves in the American South.The chief feature of the disorder was a tendency to flee the conditions of slavery, a medical condition that could be cured by whipping (Cartwright, 1851). The social values and social conditions that produced this knowledge claim are obvious to us at the beginning of the 21st century. What historical critical psychopathology seeks to do differently is to bring this same level of social awareness to the current state of knowledge claims regarding psychopathology. The challenge is: How can we attain that type of perspective outside our social circumstances while standing in the middle of them? To be sure, the distance we can achieve in this regard is never going to equal that afforded by a century of time and associated cultural changes, but some approximation to that distance is not only possible but desirable in order to gain perspective on our current concepts that we take for granted. Cross-cultural perspectives and within-culture perspectives afforded by changes in social climate and values (e.g., homosexuality and neurosis) can serve to question current conventions, although it does not preclude the possibility of reaching bedrock, as has been accomplished for some physical medical disorders.
Epistemological Progress Is Not Necessarily Measured by the Amount of Research Published
Again, this assertion is uncontroversial. The number of publications on a given disorder denotes many different influences, including investigators’ interests, the relative acceptance of the condition as a serious disorder in need of study and treatment, the conceptual resonance of the disorder with expert and other conceptualizations of human behavior and human maladies, and the availability of resources to investigate and treat the condition.
All Mental Disorder Research Serves Purposes of Its Own Time and Place
Related to the aforementioned point, research on mental disorders reflects the historical period and associated cultural values in which the research is conducted. In addition to producing evidence about the disorder and its treatment, research on mental disorders has other rather obvious if mundane functions, such as providing employment for researchers, providing marketing information for treatment purveyors, organizing social groups to advocate for additional attention to a problem, and bringing assistance to those who have the condition.
Mental Disorder Diagnoses Serve Social Functions in Addition to Knowledge Functions
The main scientific purpose of mental disorder diagnosis is to organize knowledge and to facilitate communication among professionals. Nevertheless, diagnoses serve a whole host of other social functions (Houts, 2002). Applying a diagnosis can remove the disturbing person from those being disturbed. Applying a diagnosis can remove ambiguity for a professional and bring social and economic benefits, as well as relief from fear of the unknown, for the individual who is diagnosed. Diagnosis can entitle the diagnosed person to benefits and relief from obligations that would otherwise be imposed.
* * *
In sum, historical critical psychopathology approaches the field of mental disorders from a set of assumptions that differs from those found in traditional approaches. Those different assumptions span issues concerning the use of historical materials, the relative “objectivity” of current concepts of mental disorders, and the social basis of knowledge claims in the mental health field. In what follows, we present a detailed analysis of the concept of PTSD from within the perspective of historical critical psychopathology.
The Creation of PTSD
What we today call PTSD is quite different from both what started out as PTSD and from previous war-related problems (see Leys, 2000; Shepard, 2001; and Young, 1995, for an extended discussion of previous war-related problems). These historical discontinuities are anathema to those who write standard histories of PTSD (Bloom, 2003), but a careful reading of even such standard histories demonstrates that evidence for discontinuity shows through attempts to homogenize that history.
The term post-traumatic stress disorder was introduced by a subcommittee of the DSM–III Task Force in the late 1970s, following extensive lobbying efforts on the part of veterans’ groups, who were assisted by psychiatrists Chaim Shatan, Robert J. Lifton, and colleagues (Scott, 1990, 1993). As antiwar activists, Shatan and Lifton became involved with veterans’ rap groups (self-help groups conducted by nonprofessionals) in the New York City area, where the largest organization of antiwar veterans, Vietnam Veterans Against the War, was headquartered (Burkett & Whitley, 1998). From their participation in these self-help veterans’ groups, Shatan and Lifton began describing what Shatan termed post-Vietnam syndrome (Shatan, 1972). Drawing on psychoanalytic ideas, both [Shatan and Lifton insisted that the problems they described were different from previously described combat adjustment problems. How else could they expect to draw national attention to these problems without making them out to be something new and different that was specifically tied to the nature of the war in Vietnam? This novelty of post-Vietnam syndrome is evident to one reading the original proposals in context, but much of this novelty has been ignored in the history of PTSD thanks to various efforts to homogenize the history to show continuity, stability, and the “always there before we even noticed” nature of PTSD.
The core concept of post-Vietnam syndrome was quite different from what we now call PTSD. Shatan (1972) saw the problem as impacted grief: “The post-Vietnam syndrome confronts us with the unconsummated grief of soldiers ‘impacted grief’ in which an encapsulated, never-ending past deprives the present of meaning” (p. 35). He saw this as akin to some of Freud’s observations about combat adjustment during World War I, but he saw the post-Vietnam syndrome as unique because these soldiers could not mourn. Shatan described the chief features of post-Vietnam syndrome as guilt for those killed or maimed, feelings of being duped and used by society, rage about being manipulated, hatred of others, self-alienation and numbing, and doubt about the ability to love another person. Shatan (1982) went so far as to discount the possibility that war experience interacted with premorbid problems: “In the presence of massive stress, pre-existing psychiatric disorder is irrelevant. The specific stress itself is the crucial predisposition and is generally predictive of the nature and degree of morbidity and symptoms” (p. 1035). Shatan also argued that the unique adjustment problems were different when the stressor was human-made as opposed to a natural disaster. Lifton (1982) echoed similar themes, insisting that this condition was new because death was encountered under morally ambiguous circumstances, as contrasted with the moral clarity of World War II–era veterans. It is worth noting that this list of complaints was taken from a sample of Vietnam veterans who had returned from the war and joined the Vietnam Veterans Against the War organization, yet neither Shatan nor Lifton entertained the possibility that they based their observations on a highly selective sample whose moral and political outlook matched their own.
From Catastrophic Stress Disorder to PTSD
The exact details of how post-Vietnam syndrome was transformed into PTSD remain unclear, but Scott (1990, 1993) provided one of the most detailed examinations to date. Shatan and Lifton formed the Vietnam Veterans Working Group to collect information about post-Vietnam syndrome with the goal of providing evidence for the DSM–III Task Force. A number of individuals and groups cooperated in this effort, and a proposal (Shatan, Smith, & Haley, 1976) was forwarded to a three-person committee on reactive disorders of the DSM–III Task Force. That proposal named the condition “catastrophic stress disorder,” and the committee changed the name to “post-traumatic stress disorder.” Scott (1993) described the proposal as having been adopted “almost exactly as the Working Group had prepared it” (p. 66) for the January 1978 release of DSM–III draft (Task Force, 1978). Interestingly, within the DSM–III draft document, PTSD was placed in a section titled “Reactive Disorders Not Elsewhere Classified,” which also contained various adjustment disorders. At least in the DSM–III draft, the intention was to distinguish PTSD from other disorders not only by placing it in this special category but also in the specific statements regarding differential diagnosis.
The differential diagnosis includes a Depressive Disorder, Generalized Anxiety Disorder, Phobic Disorder, Adjustment Disorder, and Organic Mental Disorder. It differs from a Depressive Disorder, Generalized Anxiety Disorder, and Phobic Disorder in that symptoms of depression, anxiety, or phobia do not dominate the clinical picture; these diagnoses should be made in addition to Post-traumatic Stress Disorder in those patients who meet the full criteria for them. However, it differs from the Adjustment Disorders in that the stressors producing Post-traumatic Stress Disorder are both more severe and outside the range of human experience usually considered to be normal; it also differs from Adjustment Disorders in that the symptoms and complications tend to be more severe, enduring, and handicapping. (Task Force, 1978, p. N:4)
Somewhere between the appearance of the DSM–III draft and the publication of the DSM–III in 1980, PTSD was moved from the section that included adjustment disorders and placed in the section of the DSM–III devoted to anxiety disorders. The only hint given in the DSM–III for this change appeared in the appendix comparing the DSM–III to the previous edition, in which PTSD was compared with traumatic neurosis (p. 378), but this term itself was not contained in the DSM–II.
The term traumatic neurosis does have a history within psychiatry, including its use in the context of World War I shell shock and World War II war neuroses (Leys, 2000). What is striking about the DSM–III Appendix is that none of this literature is cited, but other non-war-related uses of traumatic neurosis are cited. If the intention of the DSM–III authors had been to have PTSD represent continuity with other previous war-related adjustment problems, the obvious comparison should have been with other war-related problems, even though the DSM–II had dropped the DSM–I term gross stress reaction. Such contradictions and inconsistencies are never mentioned in the standard histories of PTSD. To be sure, part of what convinced the DSM–III committee to accept a new diagnosis for these veteran problems was the fact that the working group presented evidence that some mental health professionals at some Veterans Administration (VA) hospitals were recording “traumatic war neurosis” as a working diagnosis along with an official DSM–II diagnosis (Scott, 1993). Nevertheless, one can find no trace of that influence in the published DSM–III.
Once PTSD made it into the official nomenclature, parallel developments within the veteran community and among Washington politicians and VA officials ensured that it would be widely used within the VA system. Scott (1993) summarized the social and political developments as follows:
Legislation for the treatment of alcoholism and drug abuse among Vietnam veterans was introduced in 1971 by Senator Alan Cranston (D-Calif.), a critic of the war. The measure passed in the Senate but died in the House Committee on Veterans’ Affairs. Committee members, staunch supporters of the war, disagreed that Vietnam veterans needed special programs. The American Legion and Veterans of Foreign Wars, organizations dominated by veterans of the Second World War, lobbied aggressively against the measure. The Cranston bill passed the Senate four more times between 1973 and 1978, but each time encountered the same fate in the House committee that it had in 1971. By 1977, Cranston’s bill emphasized counseling to address PTSD and had the support of the VA’s new national director, Max Cleland. In 1979, the House and Senate committees worked out a political deal that allowed the readjustment counseling bill to pass into law. Cleland and his staff designed an Outreach Program to implement the Cranston bill, and by 1981, the VA had established 137 Vet Centers. (p. xix)
These Vet Centers, typically located away from main VA hospital facilities, provided counseling and other services to Vietnam veterans, most of whom were recipients of the newly minted diagnosis PTSD.
Without the veteran connection and the extensive lobbying of veterans’ advocacy groups, it is unlikely that PTSD would have entered the DSM system when it did. PTSD began as a diagnosis to single out Vietnam veterans who were having a variety of adjustment problems so that they might receive the assistance they needed, and the passage of legislation to fund services for those veterans ensured that PTSD would be underwritten as a new disorder.
From those rather specific and somewhat precarious beginnings, the concept of PTSD soon expanded broadly and gradually turned into what we now regard as PTSD. This expansion of the PTSD concept to include more and more people is described next.
Expansion of PTSD
Any segmentation of the history of a concept is necessarily imperfect, and the boundaries between the creation and expansion of PTSD are no exception. In some respects, the seeds for the expansion of PTSD were sown at the beginning of deliberations between the Vietnam Veterans Working Group and the DSM Committee on Reactive Disorders. Whereas the veterans’ groups were concerned with what they believed were difficulties specific to war, their success in getting a new diagnosis required political compromises with affiliated groups seeking to extend the concept to natural disaster victims, holocaust survivors, burn victims, and others who sought assistance for adjustment to various catastrophic life events (Scott, 1993). As early as 1982 Jack Smith, a Vietnam veteran member of the working group and the only lay person appointed to any DSM–III committee, wrote that there was some comparability between reactions of a basketball player who took responsibility for his team’s loss and the reactions of soldiers and emergency medical workers who blamed themselves for loss of life (Smith, 1982). In retrospect, the line between what was evidently traumatic and what was within the range of normal human experience may not have been drawn clearly enough.
Initial formulations of PTSD placed some constraints on what counted as a stressful event, limiting it to such events as war, natural disasters, and debilitating accidents. Over time, however, researchers and clinicians broadened the scope of what could be regarded as a stressor of sufficient magnitude to produce the presumed “psychological trauma.” As we show below, the elusive concept of psychological trauma turned out to be intangible and unconstrained owing to the mistaken application of inference to the best explanation. As a consequence, more people qualified for a PTSD diagnosis. By expanding what could count as a catastrophic event, more people were thereby considered to be at risk for post-traumatic symptoms. This broadening of the scope of PTSD is what we call the expansion of PTSD. In this section, we focus on the expansion that took place following the formal introduction of PTSD in the DSM–III. This expansion has been a complex process. First, we briefly outline the nature of the expansion of PTSD. Second, we take up some of the conceptual issues involved in defining “psychological trauma.” Finally, we review the cultural context in which expansion occurred and the role mental health professionals have played in the expansion of PTSD.
Expanding the Stressor Criterion
The conceptual boundaries of PTSD have been steadily broadening since the term was introduced. McNally (2003) called this expansion “bracket creep” because the conceptual brackets of PTSD have become more and more inclusive, particularly with respect to criteria for deciding presence or absence of trauma. The conceptual structure of PTSD consists of three parts that are organized in a causal chain. First there is a horrible event; second, there is a “psychological wound,” usually described in the form of a traumatic memory; third, there are behavioral and emotional symptoms produced by the psychological wound. Early observers such as Shatan and Lifton placed great emphasis on the first part of the causal chain, the catastrophic event. Indeed, they argued that the horrible events of the Vietnam War were uniquely capable of producing the “psychological trauma” they observed in Vietnam veterans. In other words, there was an inextricable link between such events as the Vietnam War and traumatic symptoms. In the DSM–III the kinds of events were expanded somewhat, but the nature of the qualifying event was constrained as being outside the range of usual human experience and due to “[e]xistence of a recognizable stressor that would evoke significant symptoms of distress in almost everyone” (APA, 1980, p. 238). It was assumed that ordinary human experience was not sufficient to produce the inferred psychic wound.
Over time, however, the link between specific events and psychological trauma became blurred. As we noted, concessions had to be made with regard to the link between the Vietnam War and the traumatic symptoms for PTSD to be included in DSM–III. This was reflected in the changing names from post-Vietnam syndrome to catastrophic stress disorder to the more general posttraumatic stress disorder.
Expansion is also reflected in the changing DSM criteria (APA, 1980, 1987, 1994, 2001). DSM–III criteria required that an individual personally experience the catastrophic event—the link between event and psychic wound still required that a person be present at the horrible event. By the time of the DSM–IV, the wounded person no longer had to be at the site of the horrible event, but could simply “witness” a life-threatening event or “learn about” a loved one whose life had been threatened. There was no longer a distinction between people tortured in a prisoner-of-war camp and people who learned that a close relative was in an automobile accident. Both could qualify for the PTSD diagnosis if they displayed the formerly ambiguous reactions now transformed into symptoms of PTSD.
Such blurring of the link between the horrible event and psychological trauma could be seen as a sign of problems or of progress. If, on the one hand, limiting the stressor criterion to a narrow group of events is inadequate because clear evidence suggests that excluded events cause “psychological trauma,” then broadening the scope of stressful events would reflect a better understanding of what produces psychological trauma—in a word, progress. If, on the other hand, there is little evidence that the excluded events cause “psychological trauma,” then a more general stressor criterion reflects unwarranted expansion. Moreover, if the link between events and “psychological trauma,” whether construed narrowly or broadly, is not clearly established, then the expansion of PTSD reflects conceptual confusion—in two words, big problems. We believe that the link between external events and psychological trauma is not as clear as usually assumed. In fact, the whole idea of psychological trauma is problematic to begin with, and to make matters worse, this troublesome concept has been used repeatedly in erroneous appeals to inference to the best explanation (see below).
Psychological Trauma Is a Metaphor
The fact that psychological trauma is a metaphor is commonly forgotten (Summerfield, 2001). In the original Greek and in medicine, the term trauma refers to a physical lesion or wound (Liddell, Scott, Jones, & McKenzie, 1996). The Oxford English Dictionary (OED) presents this physical wound meaning as primary and cites 17th-century medical references as early usage. A secondary OED meaning referring to psychic injury is attributed to William James, who described psychic traumas as “thorns in the spirit” (Simpson, Weiner, & Press, 1989). James was familiar with the late-19th-century developments in Europe where Freud, Janet, and Charcot had described work with hypnosis to recover traumatic memories and to both remove and create nervous symptoms among hysterical patients. The late 19th century was also the context in which other developments such as railroad accidents led to various speculations about trauma that could not be seen in the form of a physical lesion (Leys, 2000; Young, 1995). By the dawn of the 20th century, the psychologization of trauma was all but complete (Hacking, 1995).
This reification of a hidden, inner wound was not without its naysayers, but as we know, they did not prevail. Hacking noted that one such skeptic was Nietzsche who wrote in On the Genealogy of Morals (1887):
“Psychological pain” does not by itself seem to me to be a definite fact, but on the contrary only an interpretation a causal interpretation of a collection of phenomena that cannot be exactly formulated it is really only a fat word standing in place of a skinny question mark. (Quoted in Hacking, 1995, p. 197)
It is important to note that these first uses of the metaphor of psychological trauma were quite different from our current conceptions of PTSD. For one thing, the emphasis was on dissociative features and the fact that the event was not experienced as traumatic at the time of its occurrence. Further, any claim that a certain type of event could mechanically produce psychological trauma without the active participation and even confabulation of the victim was not consistent with 19th-century theories of trauma (Leys, 2000).
Problems arising from the metaphorical nature of the concept of psychological trauma can be illustrated by comparing two hypothetical situations: a physician diagnosing a physical trauma and a psychologist diagnosing a psychological trauma. As introduced above, there are three parts to the structure of PTSD: the event, the supposed wound, and the symptoms.  When physicians treat physical trauma, they observe the final two parts of the causal chain. They observe the symptoms (e.g., bleeding, shock, screaming) and the wound itself. Moreover, by examining the wound, the physician can draw a strong inference about the cause of the wound—wounds caused by knives look different than wounds caused by blunt instruments. The physician is only a single inferential step away from the cause of the symptoms because the symptoms and the wound are clearly seen. In contrast, the psychologist requires two inferential steps to get to the cause of the symptoms. Indeed, the psychologist observes only the final part of the causal chain, the symptoms. The psychologist does not observe the trauma itself, only the manifestations of the inferred psychological wound. The psychologist must first infer that the psyche is wounded and then must infer that some event produced that psychic wound. That would be similar to asking a physician faced with a bleeding patient but unable to see the trauma to determine whether it is caused by a tumor or an aneurism. The physician can take further action and answer that question by imaging procedures or by autopsy. In contrast, the psychologist has no such recourse.
It is not surprising that the scope of PTSD has expanded given that psychologists have symptoms and behaviors as their only evidence that a “psychological trauma” is present. When people present with PTSD-like symptoms, any event could theoretically produce the symptoms because, after all, they are PTSD-like symptoms. Before PTSD was introduced, PTSD-like symptoms would have been considered ambiguous, nonspecific reactions. Indeed, early critics of PTSD argued that PTSD symptoms were not unique and could be subsumed under other disorders (Scott, 1990). However, once PTSD was set up, peoples’ reactions to bad events were no longer just reactions but were now symptoms of PTSD—evidence of trauma. As Young (1995) noted:
However it is obtained, evidence of a credible etiological experience transforms nonspecific symptoms into tokens of PTSD. Ruminations that would otherwise indicate a mood disorder are now changed into ”reexperiences”; behaviors that resemble common phobias are turned into PTSD ”avoidance behavior”; episodes of irritability are redefined as ”symptoms of autonomic arousal.” The PTSD concept imbues otherwise ambiguous symptoms with a degree of significance that they might not otherwise possess. (p. 120)
In other words, the concept of PTSD provided a framework for inferring psychological trauma from nonspecific symptoms. The inclusion of PTSD in the DSM–III provided a new kind of professional precedent for inferring that horrible events cause psychological trauma, which in turn cause the observed symptoms.
Errors in Inference to the Best Explanation
A troublesome question remains: Are the inferences justified? Our answer is no. Humans do not observe psychological trauma in the same way they observe physical trauma. Appeals to biophysical evidence of psychological trauma are inconclusive at best (see below). Thus, in response to the question of how we know that a psychological trauma has occurred, most researchers and clinicians are forced to state: Because we see the symptoms. Houts (2001) described this inference in this way: “[T]he grounds for knowing that there is a dysfunction [i.e., trauma] is, we see the behaviors [i.e., PTSD-like symptoms] that we see and we cannot imagine any other explanation for the cause of those behaviors except for the explanation that there is a broken process or mechanism inside the organism [i.e., psychological trauma]” (p. 337). This is an appeal to inference to the best explanation, a particular type of argument first identified by Peirce, who called it abduction.
Peirce’s (1901/1956) discussion of abduction helps clarify why this type of reasoning is suspect when applied to PTSD: 
Long before I first classed abduction as an inference it was recognized by logicians that the operation of adopting an explanatory hypothesis—which is just what abduction is—was subject to certain conditions. Namely, the hypothesis cannot be admitted, even as a hypothesis, unless it is supposed that it would account for the facts or some of them. The form of inference, therefore, is this:
The surprising fact, C, is observed;
But if A were true, C would be a matter of course,
Hence, there is a reason to suspect that A is true.
Thus, A cannot be abductively inferred, or if you prefer the expression, cannot be abductively conjectured until its entire content is already present in the premises, “if A were true, C would be a matter of course.” (p. 151–152)
In other words, what is needed is independent experimental evidence that when you perform A, the reliable result is C. In the case of PTSD symptoms, we would need to know that when certain events occur, certain outcomes are reliably produced. In fact in the PTSD case, things are a bit more complicated, because we would have to know that certain events reliably produced psychological trauma and that trauma, in turn, produced the observed symptoms. We do not have evidence of the first type because, for obvious ethical reasons, we have not randomly assigned people to certain horrific events and observed the behaviors that follow. Evidence of the second type is not likely forthcoming because no one knows how to locate independent of the symptoms it supposedly produces.
Additionally, there are few prospective designs in which researchers have information prior to and after a horrible event. Instead, studies are by and large retrospective, at least with regard to whether the horrible event was “psychologically traumatic.” These studies are inappropriate for determining whether psychological trauma was caused by a horrible event. As Young (1995) noted, “research on traumatic memory continues to center on Vietnam War veterans with multiple psychiatric diagnoses and significant psychosocial impairments. Now, decades after the war, this population is entirely unsuitable for investigating the role played by premorbid diatheses, suggestibility, retrospection, and contingency in the production of current symptoms” (p. 137). In short, there is no rigorous way to rule out confounding when the evidence is retrospective and obtained from a population that may benefit quite substantially from claiming the PTSD diagnosis. As others have pointed out, full-time disability due to PTSD is worth approximately $35,000 per year of tax-free income (Burkett & Whitley, 1998, p. 236).
The Examples of Sexual Harassment and Terrorist Attacks
McNally (2003) identified two areas in PTSD research that illustrate how confusion about the metaphorical nature of psychological trauma led to an expansion of PTSD: sexual harassment and the national surveys that followed the September 11, 2001, terrorist attacks. Both examples also illustrate the faulty use of inference to the best explanation. The association between sexual harassment and PTSD attained national prominence in the Paula Jones suit against then president Bill Clinton(Leys, 2000). In a recent article, Avina and O’Donohue (2002) argued that victims of sexual harassment do in fact exhibit the once-ambiguous reactions that have been turned into symptoms of PTSD. Based on logical consistency alone, they pointed out that because these were symptoms of PTSD, then sexual harassment can be construed in a way that meets the DSM–IV stressor criterion. In the case of sexual harassment, researchers and clinicians have observed PTSD symptoms and inferred that sexual harassment produced a trauma that resulted in symptoms. There is no theoretical or empirical reason to believe that sexual harassment is traumatic other than the fact that a few individuals who have been sexually harassed exhibit PTSD-like symptoms. In Peirce’s terms, there is no reason to believe that given sexual harassment, either the elusive psychological trauma or its related PTSD-like symptoms is a matter of course. McNally (2003) underscored this point: “Overhearing obnoxious sexual jokes in the workplace may provide a legal basis for litigation, but it seems unlikely to produce the same psychobiological state of PTSD as violent rape” (p. 232). The problem is obvious. Without any basis for rejecting the presence of trauma, trauma can always be inferred from symptoms of distress. Coupling that indeterminacy with ever-changing social mores as to what is offensive, unpleasant, and legally actionable is a sure recipe for PTSD expansion.
One might argue that sexual harassment could actually produce the same psychobiological state of PTSD as violent rape if the sexually harassed person is predisposed to developing PTSD. This is certainly possible, and some researchers have started to investigate it (see McNally, Bryant, & Ehlers, 2003, for a review). To conclude that the interaction between a predisposition toward developing PTSD and sexual harassment produces PTSD, we would again need to meet Peirce’s criteria. However, the evidence regarding risk factors is preliminary and inconclusive (McNally, Bryant, et al., 2003) and has typically involved people who have been involved in combat, not sexual harassment victims. Consequently, asserting that the interaction between a predisposition toward developing PTSD and sexual harassment causes PTSD would be unjustified and would constitute another misapplication of inference to the best explanation.
A second arena for PTSD expansion has been the national surveys conducted after the September 11, 2001, terrorist attacks (Schlenger et al., 2002; Schuster et al., 2001; Silver, Holman, McIntosh, Poulin, & Gil-Rivas, 2002). In their survey of adults across the United States who were not directly affected by the terrorist attacks, Schuster et al. concluded that the distress symptoms they observed were in fact “trauma-related” (p. 1511). Again it is unlikely that a person in Seattle, Washington, who watched the attacks on television was in a similar state as one who was actually at ground zero in New York City. Nevertheless, the researchers concluded that people across the United States had been traumatized because they observed PTSD-like symptoms—an inference that is logically unjustified.
The Culture of Therapy—December 7, 1941, versus September 11, 2001
Not only is the current culture of therapy unique to our age, it has also facilitated the expansion of PTSD. To gain some perspective on how much our current cultural climate differs from that of half a century ago, we compared popular press coverage of the events of September 11, 2001, with coverage of the events of December 7, 1941.  We examined issues of both Time and Newsweek surrounding these two events. In contrast to the content following the September 11 terrorist attacks, there was little to no mention of concern for Americans’ mental health in the weeks following the attack on Pearl Harbor. No articles warned readers of the negative impact the attacks would have on their emotional health, and no articles on what the military could do to help with soldiers’ emotional health. There was little thought or concern for how people might respond emotionally. An excerpt from the article, “U.S. at War,” published in Time, was typical:
What would the people, the 132,000,000, say in the face of the mightiest event of their time?
What they said—tens of thousands of them—was: “Why, the yellow bastards!”
Hundreds of thousands of them said the same thing in different ways, with varying degrees of expression. In Norfolk, Va., the first man at the recruiting station said, “I want to beat them Japs with my own bare hands.” At the docks at San Diego, as the afternoon wore on, a crowd slowly grew. There were a few people, then a throng, looking intently west across the harbor, beyond Point Loma, out to the Pacific where the enemy was. There was no visible excitement, no hysteria, and no release of words for the emotions behind the grim, determined faces. (p. 17)
No doubt people experienced fear, and it is likely that many soldiers, sailors, and civilians present in Hawaii had nightmares, ruminations, and other sleep disturbances following the attack on Pearl Harbor. What they evidently did not have was a mental health industry and a mental health culture that allowed and even encouraged them to assign particular meanings to those reactions. Again, from the perspective of historical critical psychopathology, what is striking about the September 11 versus Pearl Harbor comparison is that there was no space for PTSD to show up at the time of Pearl Harbor. Instead, what was available was an aggressive, jingoistic reaction to go get the people who did this horrible deed. In the language of historical ontology we can ask: Why do people develop PTSD in 2001 and fail to develop it in 1941?
Along with the confused and confusing concept of psychological trauma and the erroneous use of abductive inference, the expansion of PTSD has been fueled by trends in the mental health field that were set in motion in the late 1960s and continue to the present (Horwitz, 2002; Rieff, 1966). These include the pathologizing of everyday life through research and the ever-expanding the scope of the mental health professions.
The Role of Mental Health Professionals
Throughout the expansion phase of PTSD, the presence of mental health professionals in the world of trauma has increased. Specifically, increased research efforts on trauma have created a new psychotherapy industry focused on treating “traumatized” individuals. This increased attention to trauma has created social and economic forces that have helped fuel the expansion of PTSD.
Research on Trauma
Researchers discuss and investigate trauma more than ever before. As a means for disseminating their ideas, PTSD researchers have founded specialty journals on trauma, such as the Journal of Traumatic Stress, and have held numerous professional conferences dedicated to trauma. To gain a perspective on the size of the increase in PTSD publications, Figure 1 shows a plot of the number of publications by year beginning with 1980 and continuing through 2002. As can be seen, during this expansion phase there has been a sharp and steady increase in the number of publications on PTSD. As evidenced by the outpouring of research following the terrorist attacks on September 11, 2001, researchers are becoming more involved in the lives of those potentially affected by horrible events. As Hacking (1998) has observed about the current culture of trauma research, “One of the incidental hazards of being involved in a mass disaster in America is that you will now be descended upon by traumatologists who will track you down for the rest of your life, to determine the long-term effects of the trauma upon your psyche” (p. 83).
Because of increased competition and payment restrictions imposed by managed-care companies, psychotherapists often need to create their own niche market. In recent years, trauma has proven to be a lucrative market for many clinicians, as evidenced by the increased popularity of such interventions as critical incident stress debriefing (CISD; http://www.icisf.org) and eye movement desensitization and reprocessing (EMDR) treatment for PTSD (http://www.emdr.com), which is one of several treatments advertised as “power therapies” because their proponents claim they produce change more rapidly than traditional therapies. Additionally, there is now an organization called the American Academy of Experts in Traumatic Stress that offers certification in the treatment of various “traumas” (http://www.aaets.org). The lucrative rewards have made it advantageous to “see” trauma in places previously unacknowledged, and has thus influenced the expansion of PTSD.
Two recent issues of Psychotherapy Finances, a newsletter devoted to helping clinicians improve their practice, have touted the monetary rewards of CISD (“There’s a Need,” 1999; “Niche Market Update,” 2000). Psychotherapy Finances interviewed two clinicians who successfully developed practices geared toward trauma-related interventions. One clinician “won’t reveal exactly how much he charges for intervention or training, but says it ranges from $80 per hour to ‘in excess of $150 an hour,’” and added that the pay “‘far exceeds what people get for managed care fees’” (“There’s a Need,” 1999, p. 6). In addition, clinicians can become trained in CISD and become certified CISD trainers for groups such as the International Critical Incident Stress Foundation (ICISF). The ICISF offers various training opportunities that range in cost from $163.00 to $345.00 (ICISF, 2002). Clearly, trauma affords many economic opportunities for clinicians needing or wanting more work. Indeed, some suggest that many opportunities simply go unnoticed: “[W]orkplace trauma isn’t just about bank robberies or shooting sprees covered by the national media. For every high-profile incident, there are thousands you never hear about” (“There’s a Need,” 1999, p. 7).
One clinician attributes her success to the “growing recognition in the business community that such services [debriefing] are essential. There’s a greater understanding that violence is a public health problem. . . . Once smoking became a public health problem, the whole thing shifted. And that’s how you have to look at violence as well” (“There’s a Need,” 1999, p. 7). Indeed, it appears that the success of CISD clinicians depends partially on their ability to market successfully to corporations, emphasizing the need for “traumatized” individuals to undergo debriefing. One clinician “sells his services in three ways: 1) through direct mail; 2) by working the local media—planting stories about his work; 3) exhibiting at trade shows” (p. 6). Another clinician cites evidence that “with proper treatment, employees return to work three times faster after an incident of violence” (p. 7).
Despite these endorsements of the benefits of CISD, the weight of evidence is against such procedures. Most of the evidence suggests that CISD is unhelpful at best and possibly harmful (Bisson, Jenkins, Alexander, & Bannister, 1997; Devilly, 2002; Gist & Devilly, 2002; Kenardy, 2000; Kenardy et al., 1996; McNally, Bryant, & Ehlers, 2003; Small, Lumley, Donohue, Potter, & Waldenstrom, 2000; Wessely, Rose, & Bisson, 2000). Despite good evidence to contrary, clinicians continue to sell CISD. Whether clinicians are intentionally ignoring this evidence or are simply ignorant of it is not important to our point. Instead, our point is that in promoting such procedures, clinicians are selling trauma. They are suggesting that many people, such as those who are victims of bank robbers, will suffer grave psychological consequences if they do not undergo debriefing and that debriefing will help them return to normal life more quickly than if they did not receive the debriefing. The public is told that trauma and its horrible consequences are everywhere. However, most people who have experienced “trauma” do not develop PTSD, but instead exhibit reasonable responses to horrible events (McNally, 2003). The selling of trauma and the continued expansion of PTSD as a malleable disorder is perhaps more pernicious than the errant activities of a few practitioners of CISD. It is interesting to speculate on the counterfactual hypothesis that without the growth and expansion of PTSD, perhaps we would never have seen the various “power therapies” for treating trauma.
As PTSD expanded, the field of trauma studies expanded, and this included the attraction of a number of critics. One of the most interesting periods in the historical trajectory of the PTSD concept has been the more recent turn to “embody” PTSD. The conceptual move, which is an attempt to move the focus of the PTSD controversy away from politics by focusing on the biology of PTSD, took place in a context of continued skepticism about the legitimacy of PTSD as a genuine disorder.
Embodiment of PTSD
Since PTSD’s first appearance in the DSM–III, there has been controversy about its status as a real mental disorder or disease. The need to respond to critics was a major impetus for the development of psychophysiological research into PTSD. Roger Pitman, a major contributor to this research, and his colleagues described the controversial context that spawned their work:
PTSD’s inclusion in the official psychiatric nomenclature in 1980 was greeted with a skepticism that has not entirely disappeared. Many laymen viewed the disorder as a political concession to disgruntled Vietnam veterans. Many psychiatric professionals considered whatever psychopathology PTSD involved to be adequately subsumed under existing categories of anxiety and affective disorders. (Pitman, Orr, Shalev, Metzger, & Mellman, 1999, p. 234)
Supporters of PTSD were forced to defend the reality of PTSD to those who believed that it was a political concession, many of whom were fellow professionals, and to those who believed that it was not distinct from existing disorders.
Those researchers who maintained that PTSD was a real and distinct disorder justified their claims by pointing to the “biological signature of PTSD.” The “biological signature of PTSD” refers to a pattern of psychophysiological responses obtained in a particular experimental paradigm that we explain in detail below. From the perspective of historical critical psychopathology, the simple fact that researchers are attempting to elucidate the biological correlates of PTSD is unremarkable. What is interesting, however, is how the PTSD proponents use and interpret the biological data to substantiate their claims about the validity of PTSD. We have labeled the use of biological data embodiment because researchers have literally attempted to move the debates about PTSD from the social arena of politics into the biological arena of the body. In the terminology of historical critical psychopathology, this is an attempt to move PTSD from the dimensions of a more conventional and socially based phenomenon toward the dimension of a more naturally occurring and inevitable phenomenon. This embodiment period in the history of PTSD began shortly after the publication of the DSM–III, when only a few researchers were investigating the biological correlates of PTSD (Kolb & Mutalipassi, 1982). In the past decade, the embodiment efforts have in many ways dominated research on PTSD (Pitman et al., 1999; Pitman, Shin, & Rauch, 2001; Yehuda & McFarlane, 1995; Yehuda, McFarlane, & Shalev, 1998).
Again, Hacking’s dimensional approach to ontology clarifies the rationale for embodiment. Critics of PTSD argued that it is a product of social and political forces, which would locate PTSD somewhere on the left-hand side of Table 2 and indicating that PTSD is more of a socially based and contingent entity as opposed to a naturally based biological entity. Biological evidence is widely taken to provide the needed evidence that can protect researchers from charges of mere social convention and political expediency. Seen in this way, the embodiment process is an effort to ensure that PTSD is viewed as inevitable, a natural kind (natural entity), having a natural structure for stability as independent from social convention, and as noninteractive. In short, PTSD would be moved from the realm of social convention and invention into the realm of nature and bona fide medical disorder.
Evidence of this characterization of the embodiment movement can be seen in the rhetoric of PTSD researchers who have promoted a psychophysiological methodology. Examples from three of many sources are reproduced below.
The clarity of the findings of biological studies in trauma survivors with PTSD has made it that much harder to ignore and dismiss the persistent and devastating effects of traumatic events in the lives of those who experience them . . . their presence [biological evidence] offers a concrete validation of human suffering and a legitimacy that does not depend on arbitrary social and political forces. Establishing that there is a biological basis for psychological trauma is an essential first step in allowing the permanent validation of human suffering. (Yehuda & McFarlane, 1997, p. xiv–xv)
Psychophysiological research in war veterans and other trauma-exposed populations accomplished over the past 2 decades . . . has arguably been the most important factor in dispelling these prejudices [doubts about the reality of PTSD]. (Pitman et al., 1999, p. 234)
. . . the present study provides empirical support for the presence of objectively measured psychophysiological reactivity to trauma cues as a nomothetic-distinguishing feature of PTSD. (Keane et al., 1998, p. 922)
Such phrases as “objective validation” and “permanent validation” are mirror images of the counterargument that rejects such characterizations of PTSD as an objective entity.
Also noteworthy is the moral tone of such statements. The crusading tone of the embodiment rhetoric not only serves the interest of making PTSD scientifically respectable but also serves a kind of moral interest to protect and defend those who suffer from “psychic trauma.” Indeed, the ideas that PTSD is a political product or that it is best subsumed under existing mental disorders are not simply wrong ideas—they are “prejudices” (Pitman et al., 1999, p. 234). Moreover, Pitman et al. argue that psychophysiological research will put a stop to these prejudices: “Psychophysiological research in war veterans and other trauma-exposed populations accomplished over the past 2 decades . . . has arguably been the most important factor in dispelling these prejudices” (p. 234). The rhetoric of embodiment is consistent with a traditional view of the history of psychopathology according to which objective information about mental illness leads not only to greater enlightenment, but to more humane treatment.
In reviewing the embodiment period of PTSD, we have focused on the classic psychophysiological studies that employed a consistent method, originally developed for studying phobic fear, of presenting recorded audio reenactments of traumatic events while recording such measures as heart rate (HR), systolic and diastolic blood pressure (SBP and DBP), skin conductance response (SCR), and frontalis electromyographical response (EMG). We refer to this methodology as the standard preparation. Inasmuch as our purpose is to illustrate the embodiment period of the history of PTSD, an exhaustive review of several other measurement modalities and preparations, many of which remain speculative, is beyond the scope of this investigation.
Evaluation of the Psychophysiological Studies
The hypothesis in the psychophysical studies is that individuals with PTSD should show greater changes than individuals without PTSD in these psychophysical variables when exposed to trauma-related stimuli. Most of the results have been consistent with this hypothesis and have been replicated across a range of stressful events (Pitman et al., 1999). The results, however, are not as clear as they seem. Two design issues make it difficult, if not impossible, to interpret them: (1) inappropriate comparison groups and (2) expectancy effects.
Inappropriate Comparison Groups
As noted above, the symptoms of PTSD are not particularly unique to PTSD and could be subsumed under other disorders (e.g., anxiety and depression). In fact, what supposedly makes PTSD distinct from other mental disorders is the presence of a traumatic event thought to produce the symptoms of PTSD, including heightened psychophysiological responses of the type studied in that standard preparation. The aim of the standard preparation was to detect a physiological signature of PTSD.
From a logical point of view, to make inferences about the physiological signature of PTSD, one would need to create a comparison group matched as closely as possible to the symptoms of PTSD (i.e., nonspecific anxiety and depressive symptoms), but which also was missing the unique feature of PTSD, namely, exposure to a traumatic event. Such a design would allow strong inferences about what occurs “post-trauma,” controlling for the presence of anxiety and depressive symptoms, rather than what occurs in the presence of anxiety and depressive symptoms controlling for traumatic event. As Pitman et al. (1990) stated: “Non-PTSD psychiatric comparison subjects are important, because it is only by including them that PTSD’s validity as a separate mental disorder can be tested” (p. 49). Additionally, not only should the groups be matched on symptoms, they should be matched on duration and history of symptoms. Chronic anxiety and depression have been linked to numerous biological outcomes (Barlow, 2002). Consequently, if the PTSD group has experienced symptoms for a longer period of time, their heightened physiological response could be attributed to chronic anxiety and depression rather than to PTSD.
The vast majority of standard-preparation studies have used comparison groups matched on exposure to trauma-inducing events rather than symptoms: for example, a comparison group that consists of Vietnam veterans who are not diagnosed with PTSD. However, a few studies included what were referred to as “psychiatric controls” (Malloy, Fairbank, & Keane, 1983; Pallmeyer, Blanchard, & Kolb, 1986; Pitman et al., 1990), the most rigorous of which is Pitman et al.These investigators compared 7 war veterans with PTSD to 7 war veterans without PTSD who had anxiety. Using a one-tailed test of significance not typical for this literature, results indicated that veterans diagnosed with PTSD showed greater SCR and EMG responses when presented with war-related stimuli than veterans not diagnosed with PTSD. The two groups did not differ in their HR responses. The authors concluded that “the findings demonstrate a specific physiologic response to imagery of past events that are presumed to be etiologic in the disorder. Because linkage of symptoms to specific past experience is critical to the definition of PTSD, our results further support the validity of PTSD as a diagnostic entity” (p. 53). Although using a control group in this study was a step in the right direction, it still lacked the important element of matching on the length of time symptoms had been present. Presumably, many of the participants with PTSD had developed the symptoms shortly after returning home from Vietnam. Given that this study was completed in the mid- to late 1980s, these participants could have had these symptoms for 10 to 15 years. Pitman et al. did not provide information on the length of time the participants had had their symptoms, making it impossible to rule out the plausible hypothesis that the heightened physiological reactivity was a consequence of having symptoms of anxiety and depression for a long period of time.
The problem of inadequate comparison groups also afflicted a major study of veterans in which comparison groups differed in other conceptually important ways besides exposure to traumatic events. One of the most extensive and expensive psychophysiological studies ever done was a multisite study conducted through the Veterans Administration (Keane et al., 1998). This study has been described as the most important psychophysiological study ever conducted on PTSD (McNally, 2003). These investigators compared 654 Vietnam veterans with a current PTSD diagnosis to 154 veterans with a lifetime PTSD diagnosis and 340 veterans who had never been diagnosed with PTSD. As in other studies, when presented with trauma-related stimuli, veterans with current PTSD showed greater increases in psychophysiological variables than did veterans without PTSD. The authors compared the groups on several demographic variables and found that the current PTSD group differed significantly on 10 of 13 demographic variables. The current PTSD group was younger, had lower incomes, had held more jobs since military discharge, had more arrests since military discharge, were younger when they arrived in Vietnam, had fewer years of premilitary education, were more predominantly Hispanic, had been married more than once, were nearly twice as likely to be receiving disability benefits, and had a higher percentage of Marine veterans (see Keane et al., 1998, Table 1, p. 918). In addition, the current PTSD group had a higher percentage of participants with a comorbid diagnosis of major depression, panic disorder, alcohol abuse or dependence, antisocial personality disorder, and borderline personality disorder (see Table 2, p. 918).
Given the differences between the groups, the results are difficult to interpret, and it is impossible to attribute the differences to PTSD alone. Keane et al. (1998) did not discuss how these differences did or did not influence their inferences about the relationship between PTSD and the psychophysiological outcomes. Instead, the authors merely attributed the observed psychophysiological differences between groups to PTSD alone. The authors concluded, “Results of this multisite clinical trial provide definitive support for a positive association between psychophysiological responsivity to cues depicting traumatic war-zone experiences and combat-related PTSD” (p. 921). This is a remarkable conclusion given the lack of experimental control in this investigation and the evident multiple imbalances in the groups assembled for study. 
One might argue that despite inadequate comparison groups, the standard-preparation studies are convincing because participants with PTSD tended to respond so strongly to trauma-related stimuli. Additional evidence from the multisite Keane et al. (1998) study demonstrates that this was not always the case. In addition to finding a difference between the PTSD and non-PTSD groups on responsivity to trauma-related stimuli, they did not a find a difference between the groups on a generic mental arithmetic stressor. However, when one compares the physiological responses of the current PTSD group when presented with mental arithmetic to their responses when presented with the trauma-related stimuli, one finds that the physiological responses are two to three times greater in response to the mental arithmetic than to the trauma-related stimuli. This holds across all of the variables: HR, SCR, EMG, SBP, and DBP. Although the research demonstrates a statistically significant difference in physiological responses, we are left to wonder how conceptually significant these findings are if, in fact, mental arithmetic produced a much stronger physiological response than did idiographically tailored and recorded combat scenes designed specifically to replay the trauma that was presumably the cause of the mental disorder.
The Role of Expectancy
As previously noted, PTSD has become part of our culture. When a “traumatic” event occurs, social scientists contact the victims to study their post-traumatic symptoms and encourage them to get psychological help. The popular media and research literature combine to convey an image of “traumatized” people that emphasizes the negative: depression, anxiety, physiological hyperactivity, sleep problems, guilt, and anger. Relatively little is said about survivors’ coping strengths and the positive ways they deal with horrifying situations. This heavy dose of negativity creates the possibility of a self-fulfilling prophecy. For example, in discussion research on stress reactions of Vietnam veterans, LaGuardia, Smith, Francois, and Bachman (1983) stated:
It can be argued that designs that solely emphasize the destructive residuals of the Vietnam experience determine the range and tone of subject responses. If such biases exist, the magnitude of the findings depicting the veteran as an emotionally scarred individual may to a great extent be the result of the experimenter’s predisposition toward pathological interpretations and the use of questionnaires and checklists clearly biased in a direction suggesting maladjustment. . . . It is our concern that the picture of the veteran which emerges from a literature restricting its scope so as to only include questions regarding pathology may not only prove delusive, but may in the long run cause a number of otherwise healthy veterans to fall victim to seeking fulfillment of the socially disabled image. (p. 19–20)
Participants in the standard-preparation studies are not blind to the purposes of the study. Most know their diagnostic status prior to the study, making them fully aware of what PTSD is in the eyes of mental health professionals and what behaviors go with PTSD. It is not clear whether the experimenters are typically blind to condition. Although the stimuli are often presented by computer, we have been unable to locate a study that indicated that experimenters were blind to participants’ diagnostic status when performing intake assessments and preparing participants to experience the trauma-related stimuli. Together, these factors create an experiment that is prone to expectancy effects. This seems particularly problematic given the cultural, economic, and political issues involved in PTSD. We are not saying that the participants deliberately and deceitfully change the results of the studies, but rather that the experimental environment and other social factors can dramatically influence behavior (Snyder & Stukas, 1999).
There is some evidence supporting the role of expectancy in psychophysiological studies. Gerardi, Blanchard, and Kolb (1989) found that when asked to increase their psychophysiological responses, people without a PTSD diagnosis were able to alter their responses to traumatic imagery so as not to be statistically different from participants with a PTSD diagnosis. Gerardi et al.instructed participants with PTSD to try to reduce their physiologic responses to the trauma related stimuli. What they did not explore is whether participants with PTSD could exaggerate their responses to trauma-related stimuli. Although Gerardi et al. explicitly asked participants to alter their physiologic responses and thereby created an unequivocal expectancy, it is not unreasonable to assume that in most physiological studies there is a clear expectation of how participants, particularly those with PTSD, are supposed to respond to trauma-related sounds.
What would happen if one believed one had been exposed to a traumatic event, but had never actually been exposed to the traumatic event? Recent evidence has suggested that the physiological signature of PTSD can be observed in people reporting alien abduction (McNally, Lasko, et al., 2003). Using the standard preparation, McNally, Lasko, et al. found that alien abductees had higher HR and SCR responses than comparison groups when presented with abduction related stimuli. Moreover, a discriminant-function analysis developed previously on Vietnam veteran samples to distinguish physiological responders with PTSD from those without PTSD identified 60% of the abductees as physiological responders.
Although the abductee results are based on only 18 participants and one-tailed tests, they raise important questions about the role of the traumatic event in PTSD. Simply believing an event is traumatic can produce the same responses that putatively provide “permanent validation of human suffering.” One wonders, in light of the abductee evidence, how permanent the effects of a “traumatic” event are, and consequently how permanent the psychophysiological evidence is. Believing that one has been traumatized absent a traumatic event is a far cry from the original descriptions of PTSD in the 1970s and 1980s, when the event took precedence, and, in fact, was what made PTSD unique.
One could argue that years of research have enlightened us about the true causes of PTSD and that the evidence is the reason for the change in emphasis. However, one could just as easily argue that the existence of PTSD has been taken for granted and has not been questioned.
The Future of PTSD
Since it is common for critics of mental disorders to be accused of denying the pain and suffering of those dealing with psychological problems, we believe it necessary to clarify what we are and are not claiming. First, we are not claiming that people diagnosed with PTSD should be ignored or refused help. Clearly, many people now diagnosed with PTSD have problems, some severe enough to warrant formal treatment and many others who need assistance and supportive care. Second, we are not saying that PTSD was and is a hoax. That is, we are not claiming that the originators of PTSD and the current supporters of PTSD are trying to “pull one over” on the scientific community and the general public. Indeed, researchers care about helping people adjust to life after a horrible event. Third, we are not claiming PTSD will never be tied to a specific physiological broken mechanism, the sine qua non of etiologically defined disorder. Rather, we argue that the physiological data currently do not support many of the claims made about PTSD.
We are, however, calling for increased skepticism about PTSD. Our review has led us to believe that PTSD may not be a disorder in the way that tuberculosis and perhaps even schizophrenia are disorders. Indeed, PTSD falls somewhere to the left of center on Table 2, whereas schizophrenia and tuberculosis fall more to the right. Moreover, the expansion of PTSD has not been based on better research, but has been fueled by various social factors and our culture of mental health. This expansion of PTSD has invited or at least assisted the proliferation of “trauma” therapies that have been shown to be ineffective and possibly even harmful.
As mentioned above, people currently diagnosed with PTSD may need help. However, we remain skeptical about the necessity of a PTSD diagnosis as a means for obtaining help, and we wonder whether a PTSD diagnosis is helpful at all or whether it might even be harmful. We recognize that it is difficult, if not impossible, to receive payment from managed-care companies without a DSM diagnosis. Nevertheless, given the conceptual controversies surrounding the PTSD diagnosis, we suggest that it may be time to think of new ways to get people the help they need. Oddly enough, PTSD was originally introduced into the DSM–III draft as a reactive disorder alongside adjustment disorders (Task Force, 1978). In view of what has happened over the past 25 years, this way of thinking about the problems of PTSD makes a good deal of sense even now.
Historical Critical Psychopathology
Our aim was to offer a perspective that highlights problems with theories of mental illness that are often overlooked in traditional historical accounts of psychopathology. If we have been successful, our illustration of historical critical psychopathology demonstrates the usefulness of integrating insights from disciplines outside of psychology and psychiatry into the study of psychopathology. In fact, without insights from sociology, history, and philosophy, we may be unable to identify the theoretical problems that plague our understanding of mental disorder, in which case the problems would remain unnoticed and become further institutionalized. Because of the benefits associated with historical critical psychopathology, we encourage others to adopt this stance at both a micro and macro level. At a micro level, we encourage others to critically examine the historical and social context of other specific mental disorders. At a macro level, we encourage authors of psychological textbooks and editors of psychological journals to consider more seriously the efforts of professionals from other disciplines when deciding what to include in the pages of their books and journals. Although we hold no illusions that doing so will solve all the problems of psychopathology, we believe that it will help researchers and clinicians notice formerly unrecognized problems.
Historical critical psychopathology can also be used to anticipate problems with disorders on the horizon. We briefly highlight three disorders not yet officially recognized as disorders that have been proposed as “new,” distinct mental illnesses: traumatic grief disorder, premenstrual dysphoric disorder (PMDD), and relational disorders. Traumatic grief disorder occurs when the separation by death of a significant other is traumatic, leading to a grieving process distinct from a normal bereavement in terms of duration, degree of impairment, and severity of symptomatology (Prigerson & Jacobs, 2001). PMDD is characterized by a variety of marked disruptions in social or occupational functioning during the last week of the luteal phase of the menstrual cycle, with symptoms ceasing within a few days following the completion of the cycle (APA, 2001). As the name implies, relational disorders are characterized by a pathological relationship between two or more people (First et al., 2002).
The origins of these disorders are strikingly similar to the origins of PTSD. Identifiable groups of people are deemed to be suffering and in need of help. Interest groups, composed primarily of mental health researchers, have proposed diagnostic criteria and are now lobbying for their inclusion in the official diagnostic nomenclature, and these groups are conducting and citing the growing body of literature germane to each problem (for grief, see Prigerson & Jacobs, 2001; for PMDD, see APA, 2001; Endicott et al., 1999; for relational disorders see, First et al., 2002). As with PTSD, increased attention to these new disorders has been followed by development of treatments for these problems (for grief, see Shear et al., 2001; for PMDD, see Lin & Thompson, 2001 and Pearlstein et al., 2000; for relational disorders, see First et al., 2002). This may seem like an unremarkable turn of events, as it is not surprising that clinicians will treat people with problems, regardless of whether the problems are old or new. However, as our review of PTSD showed, the development and dissemination of treatments helps institutionalize ideas about mental illness that are misleading and sometimes even harmful. Because all people experience the death of a loved one, many women menstruate and experience distress, and most people experience some dissatisfaction with their relationships, clinicians would have endless numbers of “disordered” people to treat, particularly if mental health professionals market these new disorders the way they have marketed trauma. Therefore, a heightened level of skepticism is not only warranted, it is desirable.
As these problems are introduced into the diagnostic nomenclature and as research on these problems increases, we predict that there will be an expansion of the scope of each of these problems to include more people under the diagnosis. For example, the current proposal for relational disorders limits them to familial problems. However, it would not be surprising to see relational disorders expanded to include many types of relationships (e.g., coworker relationships), as critics of relational disorders have pointed out (Vendantam, 2002). We are not concerned with whether the expansion of these new disorders is justified. Rather, we want to emphasize that we can bring to bear the insights garnered from critically studying the history of other mental disorders to the expansion of these new disorders. Our review of PTSD suggests that expansion of PTSD was unwarranted, fraught with conceptual confusion, and affected by a host of social and economic factors. Understanding how these problems have influenced the expansion of PTSD can make us aware of how these problems may influence the expansion of new disorders.
All three proposed disorders are controversial, with some already being criticized as misguided. Not surprisingly, the response to these criticisms has been same as the response to critics of PTSD—the rhetoric of embodiment. For example, Steven Beach, a proponent of relational disorders, stated in response to critics that he believed there were “‘genetic underpinnings’ to relationship troubles and that if the new category were created, doctors would find that ‘they are every bit as medical as everything in the DSM’” (Vendantam, 2002; see also First et al., 2002). Likewise, Jean Endicott, a researcher of and advocate for PMDD, stated that PMDD is “a real biological condition for which women seek treatment—and for which effective treatment is available” (Daw, 2002). If it has not already begun, it is likely that such rhetoric will appear in the traumatic grief literature. Although it may be true that relational disorders and PMDD have biological correlates, we remain skeptical of the assumption made by Beach and Endicott that these correlates justify the creation of these disorders. As was the case with PTSD, such claims about the biological basis of disorders are often rhetorical moves aimed at silencing critics. Of course, we recognize that such use of biological data does not bear on the veracity of the claims about the biological basis of traumatic grief, PMDD, and relational disorders. However, the question remains: Are the data sufficient to warrant such claims? As we have shown above, the data were not sufficient with PTSD. Thus, it would be prudent to remain skeptical about such claims with regard to traumatic grief, PMDD, and relational disorders and to carefully examine the evidence to find out how much of these claims are valid and how much is rhetoric without solid support.
The creation of new mental disorders, including PTSD, traumatic grief, PMDD, and relational disorders, is occurring in a cultural context in which many of life’s problems are deemed mental illnesses and psychotherapists are considered the “the legitimate arbiters of suffering” (Horwitz, 2002, p. 80). Historical critical psychopathology can help our discipline maintain a healthy skepticism about such developments in the field of psychopathology, perhaps even fueling better, more useful thinking about mental illness.