The Scientific Review of Mental Health Practice

Objective Investigations of Controversial and Unorthodox Claims in Clinical Psychology, Psychiatry, and Social Work


Attachment Therapies: A Deadly Cure without a Disease?

Review of Attachment Therapy on Trial: The Torture and Death of Candace Newmaker
by Jean Mercer, Larry Sarner, and Linda Rosa

For years, scientifically minded mental healthcare professionals have decried the use of unsupported techniques, especially “fringe therapies.” By and large, these are faddish practices based on theories that run counter to mainstream psychology, and that often closely resemble New Age cults. Most troublingly, such untested therapies are being foisted upon children in our society, who are subject to the vagaries of our laws, mental health systems, and the choices of parents who may be uninformed.

Although the constellation of techniques called attachment therapy (AT) has been implicated in the deaths of several children, the most publicized case was the death in 2000 of Candace Newmaker, a 10-year-old girl who suffocated on her own vomit as a team of attachment therapists and her adoptive mother stood by and ignored her pleas to be released. In Attachment Therapy on Trial: The Torture and Death of Candace Newmaker (2003), Jean Mercer, Larry Sarner, and Linda Rosa tell Candace’s poignant tale, and provide a thorough analysis of the convergence of cultural trends, societal beliefs, and fallacious reasoning that culminated in her death.

Attachment Therapy on Trial is a truly ambitious undertaking. In the introduction, the authors outline the three-part organization of the book: first, a case study of Candace’s life and death; second, a comparative discussion evaluating AT principles in the context of research-based views of human development and psychological scientists; and third, an analysis of the events in the Newmaker case with respect to such causal factors as societal trends and common errors in reasoning. Mercer, Sarner, and Rosa refreshingly refuse to offer even a pretense of a superficially “balanced” perspective on AT. Instead, they adopt a strong anti-AT stance, based on the strong preponderance of evidence that favors such a position.

Candace Newmaker was one of several offspring of a teenage mother (who had herself been born to a teenage mother) and a violent father. Her family had a history of involvement with social services, and her mother was living in a home for mentally ill children when Candace was born. On the surface, it appeared that pediatric nurse Jeane Newmaker would provide Candace with opportunities that her low social class and disorganized family life could not. As occurs in many adoptions, after an initial “honeymoon” period, Jeane began to complain that her new daughter was behaving negatively. Soon, Candace began to see a series of psychotherapists and received a series of diagnoses (including posttraumatic stress disorder and bipolar disorder) and psychiatric medications. During this period, Candace’s school and psychiatrists rarely communicated with each other. Instead, Jeane Newmaker was the sole go-between in seeking services for Candace. Much to Candace’s detriment, Jeane effectively become her caseworker. Though none of Candace’s psychiatrists had suggested that attachment issues were at the root of her emotional difficulties, when Jeane heard through a county social worker about an upcoming workshop at the Guilford Attachment Center in High Point, North Carolina, she chose to attend. Soon, Jeane began to attend group meetings with other worried parents, and took Candace for individual sessions with an AT practitioner.

The treatment she received, holding therapy, was one of a series of sadistic and abusive practices all classified broadly as “attachment therapies,” and was based on the same principles as the techniques that led to her death. During these holding sessions, Candace was grabbed by the face, shouted at, and immobilized. The authors consistently offer concrete examples of the practices involved in attachment therapies. This sort of descriptive frankness is conspicuously absent from materials issued by the Association for the Treatment and Training of Attachment in Children (ATTACh) and other organizations that promote AT. As observers at the trial of therapists Julie Ponder and Connell Watkins, the authors were privy to many hours of testimony and videotape, which provided a rare glimpse into the workings of a fringe practice. In the aftermath of Candace’s case, a quick visit to the ATTACh website revealed that the organization has put its “treatment manual” online, although most specifics revolved around “don’ts” rather than acceptable practices. Holding and restraint are still very much a part of the AT philosophy, although one of the newly posted guidelines reads “The child will never be restrained or have pressure put on them in such a manner that would interfere with their [sic] basic life functions such as breathing, circulation, temperature, etc.” (ATTACh, 2001). Notably absent from the Web site is any statement on Candace’s case, expression of remorse, or condemnation of the actions taken by her therapists. Even the site’s page titled “Recent News Articles” includes only the following headline (from the Orange County Register): “Controversial Attachment—A Couple Says Attachment Therapy, Though Linked to Deaths, Has Brought Their Raging Son Under Control.” Apparently, ATTACh’s media watch person managed to track down one of the few articles issued since the Newmaker case that does not explicitly condemn AT, although certainly not the only news item pertaining to its practices.

As those who have followed this case know, Candace’s real troubles began when Jeane took Candace to Evergreen, Colorado, for 2 weeks of intensive attachment therapy, as recommended by representatives at an ATTACh conference. For 8 days leading up to her death, she was placed in “therapeutic foster care” with Brita St. Clair (though Jeane was in town with Candace). During her stay at Evergreen, Candace also was under the care of AT practitioners Connell Watkins and Julie Ponder. She underwent daily sessions of the same sort of “holding therapy” she had endured in North Carolina, including a session of compression therapy, in which Jeane Newmaker, a robust woman, lay on top of her and licked her face. As brutal as her treatment was during this time, it was purported to offer a respite from the confrontational techniques that ultimately led to her demise. She was to undergo a brief procedure called “rebirthing,” in which she would be tightly wrapped in a flannel sheet symbolizing the birth canal, and push her way out into the world to be “reborn” as Jeane Newmaker’s true daughter.

The authors review the chilling videotaped proceedings that followed. Although the procedure was supposed to last for only 10 minutes, Ponder, Watkins, and Newmaker stood by and taunted Candace as she spent 40 minutes begging for air, vomiting, and choking (all of which were interpreted as active resistance). For another 30 minutes, Candace was silent, while the adults casually conversed, one of the therapists propping herself against Candace’s motionless form. When finally unwrapped, Candace was blue and had no pulse. When emergency medical workers resuscitated her, it was too late: Candace was brain-dead. She remained on life support until her organs could be harvested for transplants.

Rather than simply treating Watkins, Ponder, St. Clair, and Newmaker as the villains, and broadly condemning AT practices, the authors undertake an extremely thorough and well-reasoned comparative analysis of research-based findings in psychology and AT beliefs. First, the authors highlight key differences between AT and “normal” psychotherapy. Although there are several major theoretical schools of psychotherapy, the authors emphasize, there are more common elements than divergences across mainstream approaches. Some key differences between AT and normal therapy practices that are underscored are: (1) an emphasis on transformation rather than communication; (2) a belief that humans can prenatally sense and interpret emotion in the self and others; (3) using touch as a key part of therapy, even with older children; and (4) using physical exertion and removal from schooling as “punishments.”

Perhaps the most disturbing discrepancy between mainstream psychology and AT is the series of faulty assumptions that form the AT understanding of attachments and how they form. Developmental psychologists have found that attachment to familiar people forms between the ages of 6 and 24 months. Research findings consistently indicate that extremely impoverished conditions with little opportunity for social play (like those found in many Romanian orphanages) or frequent caregiver switches during this period are the most likely factors to disrupt attachment. The corresponding DSM–IV diagnosis, Reactive Attachment Disorder, describes 2 characteristic patterns of behavior that arise from such early deprivation: (1) Extremely inhibited behavior that results in a failure to respond normally to social interactions, or (2) Diffuse attachments, as marked by indiscriminate sociability. The authors assert that complete failure of attachment is so rare that it is not well understood, but contrary to some early assumptions by Bowlby (1982) and others, the result would most likely be reflected in relationship difficulties rather than a complete lack of empathy (AT theorists have borrowed a simplified version of this idea from Bowlby). AT theorists believe that poor attachment can result from just about any sort of childhood “trauma,” including run-of-the-mill poor parenting. Once labeled with “Attachment Disorder” by AT practitioners, the child is more or less demonized, and parents are typically told that their child is no more than a manipulative con artist. It is troubling to realize that at this point, the child can effectively do no right. Although most parents end up enmeshed in AT groups because of some degree of perceived negative child behavior, any attempts by the child to ameliorate the situation are seen as trickery. The authors underscore the important differences between the DSM conceptualization of Reactive Attachment Disorder and the AT version. AT practitioners use a simple checklist to diagnose “Attachment Disorder.” This checklist includes many added criteria, most of which more appropriately describe symptoms of other recognized forms of child psychopathology, such as externalizing problems, internalizing problems, and pervasive developmental disorders.

Additionally, AT practitioners underscore the importance of prenatal attachment. According to this scientifically baseless belief, the fetus can receive and internalize the mother’s emotional states, and even early adoption is sufficiently traumatic to disrupt attachment for life (although it is well established that humans cannot remember events that happen in infancy). As the authors point out, this claim completely contradicts the findings of such researchers as Miller, Fan, Christensen, Grotevant, and van Dulmen (2000), who have found no differences in adjustment of early adopted and nonadopted children. Mercer and her coauthors emphasize that another essential faulty assumption underlying AT is “recapitulation,” the idea that a series of events that has gone wrong can be repeated in a guided way that will produce a new and desirable outcome. Alas, rebirthing does not accurately resemble the birthing experience any more than holding therapy resembles the way that human infants form attachments. Regardless of the accuracy of replication, the complexities of human development preclude us from ever stepping in the same metaphorical river twice, as the authors aptly point out.

Nevertheless, thinking that diverges from mainstream theory is not automatically wrong; the burden is simply on AT theorists to produce findings that support their ideas and practices. Part 3 of Attachment Therapy on Trial highlights not only problems with AT and the handling of Candace’s case, but possible solutions. The first problem is the state of scientific support for AT practices. Unsurprisingly, such evidence is largely nonexistent. According to Mercer, Sarner, and Rosa, there are no randomized, controlled studies (published or otherwise) examining the efficacy of AT; additionally, there are no well-conducted case studies that included descriptions of the techniques used. There are a few weak instances of Class II (quasi-experimental) evidence posted by the Evergreen group, all rife with sampling bias and design flaws. Only one study was published in a peer-reviewed journal.

Perhaps the most insightful and entertaining chapter in Attachment Therapy on Trial is the chapter dealing with the series of personal and professional errors that have led to the supply of and demand for AT and similarly appalling treatments. For example, the authors cite the unfiltered flow of sometimes erroneous information disseminated to desperate parents over the Internet, certain insidious aspects of the self-help movement, and parental beliefs about the degree of control one should exert over children as factors that may lead parents to accept AT. Additionally, factors such as a lingering popular belief in the long-debunked concept of catharsis (“If I make a child angry, it releases and ‘destroys’ the anger”), common cognitive errors (e.g., belief in testimonials rather than scientific findings, misinterpretations of causality), and world events (e.g., exaggerated fears of youth violence, worries about the effects of preadoption conditions on child development) have a place in the system of which Candace and her therapists are a part.

Perhaps because it is so provocative, this chapter left me with burning, unanswered questions. The motivations of the practitioners received a bit of attention in a short section on “shadow professionals,” but little explanation was presented other than the fact that such techniques are easy to learn, and come with a handsome certificate that confers legitimacy and authority upon the recipient. These are safe assumptions, but this discussion could have gone further. Why would some individuals presume that they could learn to be psychotherapists after a weekend workshop? Why are some professionals, who should know better after many years of training, drawn to fringe therapies? We may not know the answer to the latter question, intriguing as it may be. Nevertheless, some important points could have been raised about the former. Academic psychologists and those trained in research programs have not managed to communicate effectively the nature of our profession to the general public. Instead, the “psychologists” whom laypersons do typically see in the media often simply reinforce the notion that psychology is just a matter of “good people sense” and that some minimal interaction with “experts” in the field constitutes sufficient training.

It would be an overstatement to call any chapter in Attachment Therapy on Trial weak, but relative to the other chapters, I was most dissatisfied with the chapter that dealt with the approach trained psychotherapists would have taken with Candace. Some important points about the reciprocal nature of all family interactions and adapting parenting to fit the child’s needs were included (the latter point is especially important in light of the AT philosophy that finds the child responsible for all problems). Nevertheless, this would have been an appropriate place to include one important issue that would have made this book even more compelling: how a good psychologist or psychiatrist should go about diagnosing childhood psychopathology. Especially given the authors’ consensus that Candace seemed to be going through nothing more pathological than attention problems at school and mildly oppositional behavior, this subject is important to address. Further, even when Candace was being treated by the mainstream mental health system, she was given a slew of diagnoses and medications that could have aggravated her newly adoptive mother’s confusion. Labeling Candace with such diagnoses as bipolar disorder and posttraumatic stress disorder and prescribing for her antipsychotic and anxiolytic medications not only seems inappropriate, but may have erroneously confirmed Jeane Newmaker’s notion that Candace had severe psychological problems. Mercer, Sarner, and Rosa seem reluctant to indict the legitimate professionals, despite this reflection of how poorly understood and diagnosed childhood mental illness often is, even by those ostensibly qualified.

In sum, Attachment Therapy on Trial: The Torture and Death of Candace Newmaker is an informative, well-researched, and interesting read. This book has the additional advantage of being accessible to educated laypeople, and will hopefully be read by those who are struggling with parenting issues and weighing the options of child treatment. The authors are to be commended for their fine balance of scholarship and passion in their analysis of this heartbreaking episode. Albert Einstein was once quoted as saying “The world is a dangerous place to live, not because of the people who are evil, but because of the people who don’t do anything about it.” Mercer, Sarner and Rosa exemplify those who are doing something about the troubling proliferation of practices that may harm those seeking help.


ATTACh. (2001). ATTACh professional practice manual. Retrieved from

Bowlby, J. (1982). Attachment. New York: Basic Books.

Mercer, J., Sarner, L., & Rosa, L. Attachment therapy on trial: The torture and death of Candace Newmaker. Westport, CT: Praeger.

Miller, B., Fan, X., Christensen, M., Grotevant, H., & van Dulmen, M. (2000). Comparisons of adopted and non-adopted adolescents in a large, nationally represented sample. Child Development, 71, 1458–1473.

Katherine Alexa Fowler, Ph.D. student
Department of Psychology
Emory University

You can read this review in
The Scientific Review of Mental Health Practice, vol. 3, no. 1 (Spring/Summer 2004).
Subscribe now!

  ©2004 Center for Inquiry    | SRMHP Home | About SRMHP | Contact Us |