Self-Blame and Suicidal Ideation Among Combat Veterans

Introduction

Suicidal ideation, and attempted and completed suicide are a serious problem in the United States military. The rate of completed suicide among personnel in different branches of the military was in the range of 11 to 14 per 100,000 per year from 1980 to 1992 (Helmkamp, 1995). Among Army personnel, it was stable from 1990 to 2003, and then nearly doubled from 2003 to 2010, at which time it was about 21 per 100,000 per year (Ritchie, 2012). The problem of attempted and completed suicide in the military has been confirmed in a series of studies and reviews (Allen, Cross, and Swanner, 2005; Anestis and Bryan, 2012; Bush, Reger, Luxton, Skopp, Kinn, Smolenski, and Gahm, 2013; Gradus, Shipherd, Suvak, Glasson, and Miller, 2013; Kang and Bullman, 2009; Mancinelli, Lazanio, Comparelli, Ceciarelli, Di Marzo, Pompii, Girardi, and Tatarelli, 2003).

Multiple factors have been identified that increase the risk for suicide among military personnel. These include: posttraumatic stress disorder ([PTSD] Guerra and Calhoun, 2011; Jakupcak, Cook, Imel, Fontana, Rosenheck, and McFall, 2009; Sher, Braquehais, and Casas, 2012); childhood trauma (Perales, Gallaway, Forys-Donahue, Spiess, and Millikan, 2012); lifetime trauma (Belik, Stein, Asmunson, and Sareen, 2009); life stressors and emotional distress in the 24 hours preceding attempts (Bryan and Rudd, 2012); intimate partner problems and job stress (Logan, Skopp, Karch, Reger, and Gahm, 2012); depression and alcohol problems (Pietrzak, Goldstein, Malley, Rivers, Johnson, and Southwick, 2010); low vitamin D levels (Umhau, George, Heaney, Lewis, Ursano, Heilig, Hibbein, and Schwandt, 2013); type of military occupation (Trofimovich, Reger, Luxton, and Oetjen-Gerdes, 2013); deployment-related physical pain (Pietrzak, Russo, Ling, and Southwick, 2001); male gender, prior self-harm and young age (Mahon, Tobin, Cusack, Kelleher, and Malone, 2005); guilt, shame and negative internal states (Bryan, Morrow, Etienne, and Ray-Sannerud, 2013; Bryan, Ray-Sannerud, Morrow, and Etienne, 2012; Bryan, Rudd, and Wertenberger, 2013); and combat exposure (Bryan, Hernandez, Allison, and Clemans, 2013).

The purpose of this paper is to describe my clinical understanding of suicidal ideation in combat veterans, and my therapeutic approach to it, in the hope that the approach might prove useful in our efforts to help wounded warriors. My aim is to provide some tools and strategies that therapists can use in their work with this population. The setting in which I work is a civilian hospital-based Trauma Program and the veterans are inpatients who have been admitted primarily for suicidal ideation. This is a biased sample, because the veterans are suicidal inpatients being treated at a civilian facility, but I believe that general lessons about suicide in the military can be learned from it. The veterans have not been selected for study in any way, and are part of my clinical caseload. The total number of veterans worked with over two decades is about 70, of which about 50 saw combat in the Iraq and Afghanistan wars and the remainder in Vietnam. The ages of the veterans ranged from early twenties to mid-sixties. The case examples in this paper are all composite cases and are not based on any single individual. My role in the Trauma Program was to conduct single cognitive therapy sessions with patients to identify core cognitive errors and negative self-schemas, the psychodynamics of their core conflicts, the core issues on which they needed to focus, and the main therapeutic tasks on which they needed to work during their admission.

In my clinical experience, veterans who are suicidal almost always describe intense self-blame related to combat. The self-blame drives both the PTSD and the suicidal ideation, although numerous factors are involved in precipitating and maintaining both. Additionally, in my clinical sample, most combat veterans who are suicidal report serious chronic childhood trauma in the form of sexual abuse, physical abuse, family violence, emotional neglect, loss of primary caretakers, and community violence. Each veteran reported some but not all of these forms of trauma. The pattern of self-blame has usually been established in childhood prior to entry into the military.

The Locus-of-Control Shift

The basic principle underlying my therapeutic approach to suicidal ideation in combat veterans is the locus-of-control shift (see Ross, 2007 and Ross & Halpern, 2009). According to this model, the chronically abused child processes information through a developmentally normal cognitive apparatus, namely, the mind of the magical child. The magical child is at the center of the universe, causes everything that happens in his or her world, and has a magical power to make things happen. This is the developmental stage of primary narcissism and concrete operations.

In a healthy family, the child learns basic trust and develops self-esteem because he or she receives consistent love and nurturance. The chronically abused and neglected child, however, learns a very different lesson: I am bad, I am causing the abuse, and I deserve to be abused. The child shifts the locus of control from inside the parents, who are really responsible for and causing the abuse, to inside the self. This is not an option; rather it is the inevitable outcome of the cognitive processes of the magical child, according to the model.

The badness of the self is repeatedly reinforced over the course of years by what the perpetrators say and do, and by what they do not do, which is provide consistent unconditional love and acceptance. This ingrained belief becomes a core negative self-schema by adulthood, and continues to be reinforced by negative self-talk, adult abusive relationships, and self-destructive behavior. This is not simply a theory; survivors of severe, chronic childhood trauma frequently state as a fact that they are bad, unworthy, and caused and deserved their abuse.

The locus-of-control shift happens automatically because of the child’s cognitive processes. Once it is reinforced, entrenched and deeply ingrained, the locus-of-control shift serves a defensive function: I am no longer trapped, helpless and overwhelmed, unable to stop or escape the trauma, in fact I am in charge, I am making it happen, and I can make it stop if I decide to do so. This is a developmentally protective illusion of power and control that buffers the child from the full emotional impact of the trauma. The fact that the abuse and neglect continue year after year proves that the child is bad for not deciding to be good and not deciding to make the abuse stop, which further proves the badness of the self.

The locus-of-control shift can be heard in the speech of the battered spouse who explains that her husband has beaten her because he was under stress and because she did not perform up to his standards as a wife, mother, and homemaker. She decides to leave the battered spouse shelter and return to the marriage, expecting that if she improves her performance, the abuse will stop. The locus-of-control shift can be heard in the victim of a date rape who explains that the rape is her fault because she went to the bar, accepted the drive home, accepted a drink, allowed him to kiss her, wore the wrong clothes, or did some other normal innocuous behavior.

The locus-of-control shift frequently contributes to suicidal ideation in combat veterans.

Precombat Trauma

In the inpatient setting, which is a subsample of all suicidal combat veterans, the suicidal veteran frequently describes severe, chronic childhood trauma consisting of varying combinations of sexual abuse, physical abuse, verbal and emotional abuse, neglect, absent caretakers, and family and community violence. Sometimes, in addition to these sources of self-blame, veterans carry great shame about their racial and ethnic backgrounds if they grew up in gang neighborhoods, if their parents were illegal immigrants, if single-parent families were the norm in their subculture, if their parents were uneducated or illiterate, or were convicted felons, or if they were poor as children. In such veterans, the locus-of-control shift takes the form of the belief that the self is bad and unworthy because of the person’s race, culture, socioeconomic background, or ethnicity.

In such a case, I might start a session, after taking a brief psychosocial history, by repeating back to the person a core negative self-belief the veteran had stated as a fact: I’m useless and bad because my mother was an illegal immigrant. I might ask why there are very few people illegally migrating from the United States to Mexico, the answer being that opportunities are often better in the U.S. It is perfectly understandable that the veteran’s mother wanted to avail herself of opportunities in the U.S. I then discuss the fact that in the political, marital and financial reality in which the veteran’s mother found herself she had only two options: stay in an abusive marriage in Mexico, or flee to a migrant job in the U.S. She could not afford a lawyer, could not get a job in the U.S. in advance, and had no relatives in the U.S. to sponsor her. She did what she could to better her life and the lives of her children.

I then challenge the veteran to review this situation: given this reality, was it more accurate to see his mother as a disgraceful illegal immigrant, or as a courageous mother who overcame great odds to make a better life for her children? Is it a disgrace to be the son of such a woman, or should it be a matter of pride? Once established in the U.S., the mother made a choice to move out of a gang-ridden neighborhood before her son reached puberty, despite the many barriers to moving up the socioeconomic ladder. This resulted in the veteran being able to finish high school and enter the military.

I might then ask the veteran if it is shameful to be in the military, knowing that the veteran feels pride and patriotism about his military service. I ask him (or her) how easy it is for a middle-class Anglo-American to make the journey from birth to military service, compared to the journey he and his mother made. Isn’t his journey the more remarkable one? Depending on the individual case, I might then use a paradoxical intervention, saying that I guess it is true that Hispanic people are intrinsically inferior to White people. Of course, the veteran will disagree with this, and I will then point out that this is what he had just been telling me; however, in response to the paradoxical intervention, he vehemently disagreed with his own core negative self-belief.

A general tactic, irrespective of content, is to get the veteran to state his attitude towards other veterans who have committed similar “crimes,” such as being Hispanic, not saving a buddy, or not saving a civilian. The attitude expressed by the veteran is invariably kind, compassionate, understanding and forgiving. I then ask the veteran to explain the difference between himself and these other soldiers. Why does the suicidal veteran hold himself to a different standard? Is this fair to him? Would it be fair to hold a buddy to that standard? The justification for the disparity in standards always returns to the locus-of-control shift.

Suicide as Murder of the Self

I not infrequently explain to patients that every suicide is a murder mystery: who killed whom? The three options are: the self killed the self, which is always true; the self killed the other, and the other killed the self. One, two, or all three of these murders could be going on at the same time. The same logic applies to suicidal ideation and attempted suicide. The self killed the other summarizes the dynamic in which the suicidal person plans to get revenge on the perpetrator of abuse or neglect by killing him or herself: I’ll be dead and then you’ll be sorry, or, I’ll be dead and then you’ll see what you did to me.

In response to this motive, which is stated by the veteran, I might ask: Who wins when a victim of childhood sexual abuse commits suicide? Do we score one for the victims? No, all that has happened is that trauma has claimed another life. The victims win when the survivor steps out of the shadow of the perpetrator, no longer has his or her feelings and self-perception determined by what the perpetrator said and did, and is free of the dance with the perpetrator. In any case, if the veteran commits suicide, the perpetrator will just use that as proof that the person was always crazy and the allegations of abuse false. So really, suicide does the perpetrator a favor.

The idea here is to turn the rage at the perpetrator outward, instead of directing it towards the self, and to satisfy the need for retaliation by denying the perpetrator the satisfaction of seeing his victim commit suicide. That is when the victim wins.

The third murder is when the self kills the other. This summarizes the dynamic of identification with the aggressor, and acting out the aggressor’s hostility towards the self on the self. Again, the perpetrator wins. In this motive, the victim is acting as the agent of the perpetrator, and carrying out his wishes in the perpetrator’s absence. The self is agreeing with the perpetrator’s evaluation of the self. This is an expression of loyalty to the perpetrator, and of unresolved ambivalent attachment to the perpetrator.

I often ask the veteran to imagine that he and I have formed a commission to decide whether it is acceptable for adults to sexually abuse children. In order to make a determination, we go to jails and ask imprisoned sex offenders whether they should be forgiven and released. They all say, “Yes.”

Would we then recommend releasing all these sex offenders? No. We know they are just lying. We don’t value their opinions at all. I might repeat the story using imprisoned bank robbers. The strategy is to get the veteran to state forcefully that offenders are just liars, then point out that the veteran has been agreeing with them by threatening to kill the self. It is the victims who are to blame in the minds of the perpetrators.

Placing Oneself on Death Row

A variation on this therapeutic strategy is to ask the person what it takes to get put on death row: The answer, murder in the first degree. I usually review the facts that murder in self-defense and killing an enemy in combat are not crimes. Then I ask whether a person can be placed on death row for stealing candy from a store? How about stealing a car? Robbing a bank? Rape? No, only capital murder. The veteran agrees that sentencing a person to death is far out of proportion as punishment for most crimes (if necessary, I point out that this discussion is not about the death penalty, or whether the veteran agrees with it; the goal of the conversation is to help the veteran recover).

I then explain, unless I have done so earlier, that feeling guilty means that one feels guilty about something, which in turn means one has committed a “crime”—whether it be a legal, moral, spiritual or ethical crime. Then I ask the veteran what crime he or she has committed that warrants being put on death row? The veteran is effectively judge, jury, prison warden, parole board and executioner all rolled into one.

Once the veteran agrees that he or she has not committed first degree murder, and therefore, does not deserve to be executed (suicide = execution of the self), I ask what sentence he or she deserves. I then role play a parole board hearing (with the veteran as both prisoner and parole board) which is a kind of gestalt technique for having the person examine his beliefs from the outside perspective. Usually, the veteran answers that the sentence has not yet been served. The veteran feels as if he deserves more years of punishment in the form of depression, active PTSD, lack of intimate relationships, and lack of a place and purpose in life. The prison is built of symptoms. The goal of the conversation is to get the veteran to agree that he or she has suffered enough, or in fact too much, has served a full sentence, and deserves immediate parole. This means taking the self off death row and making a serious commitment to saying “no” to suicide.

In the civilian trauma survivor the crimes usually include causing or not stopping childhood abuse, not being a perfect parent as an adult, and other offenses that are not even crimes. In the combat veteran the crimes may include failing to save a buddy, surviving when others did not, killing civilians, disappointing the commanders, or not fulfilling other combat and deployment-related actions. The dynamic is the same as the civilian’s, but the content is different.

The Euthanasia Model of Suicide

Suicidal trauma survivors frequently endorse what I call a euthanasia model of suicide. I tell them that suicide is a coin with two sides. One side is the euthanasia side. When a person has an aged dog and takes the dog to the vet to be put to sleep, this is done out of love and kindness. The dog is blind, has bad arthritis, has cancer, and has lost control of his bladder. The purpose of the euthanasia is to end the pain and suffering and help the dog pass on to a more peaceful place, whether that be extinction or dog heaven.

Similarly, the traumatized veteran describes euthanasia as a motive for suicide: the veteran is too tired, too discouraged, in too much emotional and/or physical pain, too depressed, has no vision of a life and no future, and just wants it all to end.

I say that I understand that side of the coin, but I want to look at the other side. On the other side of the coin, the suicide is a violent, angry attack on a human being, who just happens to be the self. It is murder not euthanasia. I might ask the veteran to imagine that we both are police officers and are called to a domestic violence scene. A huge, enraged man with a knife is threatening to kill his wife. Do we think that the man is motivated by being happy with his wife? No, clearly the emotion driving the behavior is anger.

The same thing is true when the self is threatening to kill the self. Even if the person is out of touch with the anger, it is an angry attack on a human being. Without the anger, the murder would never take place. I might then try to get the veteran to look at him or herself through someone else’s eyes. I ask if the veteran loves her children. What if her husband had a mistress who was walking up the driveway with a loaded gun, intent on killing her children? Of course, the veteran would do anything and everything to protect her children, including taking a bullet herself or killing the mistress. I ask the veteran why she would bother protecting her children? Because she loves them. What if the mistress was coming up the driveway, intent on killing the veteran and then taking over as the children’s mother? Again, the veteran would do whatever it takes.

So, if a person was walking up the driveway, threatening to kill your children’s mother, you would do whatever it takes to prevent that? The veteran usually agrees, “Right.” What if the person walking up the driveway threatening to kill your children’s mother was you?

This often shocks the veteran out of her cognitive errors. If not, I might have to review the locus-of-control shift-based rationalizations for why it is OK to kill the self, why the children would be better off without the self, how they’d get over it, and so on. I will ask the veteran if she thinks a mother’s suicide will increase or decrease the children’s risk for suicide? Depression? Substance abuse? Getting into bad relationships? Believing they are unlovable and not worth sticking around for? Is the veteran sure she wants to give her children these gifts?

The goal is to reframe the suicide as the murder of her children’s mother, rather than a kind euthanasia. I might ask the veteran how her children would vote if I called them up and asked them if they wanted their mother to kill herself? Of course they would vote “no.” So, what’s wrong with them? Are they delusional? No, of course not, the mother will respond. Well, if they’re not delusional, then who is right? Whose evaluation of the veteran is more accurate, the veteran’s or her children’s? Is the veteran sure she wants to keep agreeing with her perpetrator’s evaluation of her, and disagreeing with her children?

Survivor Guilt

Survivor guilt is a common condition among suicidal combat veterans. The self should have died and the buddies should have lived. In cases of survivor guilt I may go over some details of combat scenes to establish that who lived and who died was purely a matter of luck. If there is difficulty getting agreement on this point, I will review in more detail. The basic intervention is to get the veteran to look at himself through his buddies’ eyes. They were robbed of the chance to have a life. What would they want for him? How do they feel about him? I have never received anything but a positive evaluation of the self from this inquiry.

I then ask the veteran whether he loves his buddies? Yes. Does he honor them? Yes. Honors them enough to honor them by not committing suicide, by living for them since they can’t live for themselves? Reframing suicide as dishonoring fallen veterans is a powerful disincentive. It will override any motive to kill the self in order to get revenge on commanders or the bureaucracy of the military (the self killing the other).

Self-Blame for Death of a Fellow Soldier

Self-blame for the death of a fellow soldier is common among suicidal combat veterans. I have seen it in artillery crew, ground combat troops, logistics personnel responsible for truck convoys, and survivors of mortar and missile fire in base camps. The most common scenario is having a buddy dying in a fire fight. How is this the veteran’s fault? I get the veteran to walk me through the details leading up to the fatal bullet’s hitting the buddy.

In a typical, composite example, a small group of soldiers out on patrol start to take enemy fire. One colleague was killed and the veteran survived. How is this the veteran’s fault? I’ve found that it is always some version of, “I should have known.” The veteran claims he either should not have changed places with the buddy, or should have, depending on what happened. Or, the veteran should have told his buddy to go left, right, forwards, or backwards. I will then explain two things: hindsight is perfect, and Terr’s (1992) concept of omens, which she formulated after interviewing victims of the Chowchilla bus kidnapping.

Five years after the bus kidnapping, in Terr’s follow-up report, a child might explain, “You know, Dr. Terr, I should have known there was something wrong that day, and I shouldn’t have gotten on the bus. Every day, the newspaper boy leaves the paper on the same side of the sidewalk. That day it was on the other side. I should have known something was wrong, and I shouldn’t have gotten on the bus.”

The omen is the magical sign of the impending trauma that the person missed, thereby missing the opportunity to avoid or prevent the trauma. This is a variation on the locus-of-control shift: the omnipotent child self failed to spot the omen and take evasive action. The problem with omens is that they have no objective predictive value. It is all magical thinking.

I will then agree that it is in fact true: If the child had not gotten on the bus, she wouldn’t have experienced the trauma. If the veteran had made sure the buddy was elsewhere, the buddy would have survived. Those are facts. But that is hindsight. With the knowledge and information that the veteran had at the time, what error in judgment was there? How could the veteran know the path of the bullet beforehand? The veteran may have erred by changing places with his buddy, but the bullet could just as easily have hit the other location. In that case, the buddy would have died because the veteran did not change places. I go into as much detail as necessary to make the point.

I might also ask: Do soldiers die in war? Maybe death is just something that happens in war, and it is not the veteran’s fault. But the core point is the defensive function of the self-blame: If the buddy’s death was the veteran’s fault, and if the veteran could have prevented it, then the veteran was not powerless and helpless to prevent his friend’s death. The problem, in reality, is not that the veteran is guilty; it is that he is innocent. It’s a tradeoff between feeling guilt and powerlessness. However, realizing this, deeply, means feeling the powerlessness, helplessness and horror of combat; it means feeling the sadness and grief. Self-blame and the illusion of power and control are good places to hide from grief.

Murder by the self is the veteran’s punishment for failing to protect his fellow soldier.

Self-Blame for Deaths of Civilians and for Atrocities

Self-blame for deaths of civilians and for atrocities follows the same dynamics, but the intensity is greater. This circumstance can involve both errors of omission and errors of commission. Sometimes the veteran has actually committed a war crime. I have never treated such a person. In some cases, the self-blame might be accurate and warranted. In that case, I would inquire as to whether a true hero dodges responsibility or faces up to it. What would facing up to responsibility mean, and how could the veteran make amends? I would suggest that he perhaps become involved in charitable volunteer work, donate to an overseas relief fund, or provide support to other veterans with similar experiences. I would also inquire about mitigating circumstances, and whether the trauma, stress, and anger of war might lead to acting out during combat. I would compare the veteran’s attitude towards himself to his attitude towards any buddies who acted out in a similar way, with the expectation that his attitude towards them would be far more forgiving. I might then inquire about what a dead buddy would want for him now, even knowing about the veteran’s misconduct, war crime, or atrocity.

I might also ask whether the motto, “Support the Troops,” applies to all veterans, including him or herself. If not, why is he or she the only exception? Any veteran who is not proud of his or her service should consider whether disagreement with U.S. foreign policy, the validity of any given war, the decisions of generals, or even regrettable acting out by soldiers erases the principle of Support the Troops. Supporting the individual veteran does not mean condoning any and all behavior in the field, but on the other hand, what is the best course of action now? If the veteran would not execute a buddy after deployment, no matter what the buddy did, why is murder of the self warranted?

Although these approaches are not based on clinical experience with this subgroup of veterans, they are included for completeness, and in case they might be of use to therapists working with such individuals.

Self-Blame for Being Raped by a Fellow Soldier

I have worked with a series of female veterans who reported rape by fellow soldiers. In my experience, the dynamics and self-blame are the same as in civilian rape cases. The discussion here is not about whether the veteran should take action. What to do is a separate question—report the rape, or do something else. The goal of the clinical conversation is releasing the veteran from the self-blame. Why does the veteran want to let the perpetrator “off the hook?” If the self is to blame for the rape then there is nothing to do, except blame and punish the self, and the perpetrator is absolved of all responsibility.

After asking the veteran how she was to blame—was she too attractive, too seductive, too something, or she did do this or not do that—I might then ask whether she thinks it is acceptable for prostitutes to be raped. They are out in public selling their bodies for sex. The answer the veteran invariably gives is that no woman deserves to be raped, rape is always a crime, and there is never a justification for it.

So why is the veteran an exception to this rule? The answer goes back to the locus-of-control shift.

Self-Blame for Failure to Protect Loved Ones Post-Deployment

Sometimes the veteran returns home to his family hypervigilant in his desire to protect his family from any and all harm. Protecting the family is a very reasonable goal, and to the veteran is evidence of being a good husband and father. The problem is the extreme hyperarousal and post-traumatic stress disorder (PTSD). From a fear-circuitry perspective, this could be due to brain dysregulation and a fear response that has gotten stuck in combat mode. The combat caused the fear, which caused the hypervigilance, which is part of the PTSD.

In my experience, there is usually an active theme of self-blame in such cases. The veteran failed in his duty to protect his buddy, who died, and now he has made a vow never to fail in his duty to protect his family. The problem is not a current, objective threat to the family, it is self-blame for past events. When the veteran is able let go of the self-blame, then he can relax his hypervigilance, and his PTSD can begin to go into remission. In this model, the intrusion and hypervigilance symptoms of PTSD are driven by core negative self-beliefs. The self failed in combat and deserves to be punished. The PTSD is part of the punishment.

The flashbacks are a replaying of the trauma to spot the omens one missed in the original event, to be able to spot them in the future, and to prevent future trauma; hypervigilance is required to spot the omens in the future. None of this is necessary if the self is not to blame for the past trauma. In this sense, PTSD is more about preventing future trauma than it is about past trauma. It is more a pretraumatic stress disorder than a posttraumatic stress disorder.

Conclusion

These are some thoughts and suggestions for how to help suicidal combat veterans make a commitment to life, and to forgive themselves for the crimes they never committed. The treatment is a blend of cognitive, systems, and psychodynamic principles. I have not attempted to provide outcome data, epidemiological evidence, or external validation for these ideas and therapeutic strategies. Nor have I attempted to review the treatment literature. My goal has been simply to offer some tools that clinicians can consider putting in their tool boxes.