Resize Text: 
The shootings that took place last spring on the campus of Virginia Polytechnic Institute and State University, located in Blacksburg, Virginia, elicited a host of reactions, many deeply emotional. In groups of college and university presidents, the response was generally empathetic. “It could have happened on any of our campuses,” a number of them told me. And indeed, they were right to be put on alert by the random and unpredictable nature of a disaster like the loss of 33 lives at this large and generally peaceful land-grant university.
In the days following the mass killings, Virginia Governor Tim Kaine quickly established a panel to investigate the events leading up to that day, the incidents themselves, and their immediate aftermath. I was among the eight members of the panel, which was chaired by Col. Gerald Massengill, former Superintendent of the Virginia State Police. Some of what the panel ultimately reported to Governor Kaine pertains primarily to Virginia and to the university. But many of its recommendations are relevant to all 50 states and to most of the approximately 4,000 colleges and universities in the nation. There may not be a way to ensure that a mass killing like the one at Tech never happens again, but we can lessen the chance that one will.
The Events
On the morning of April 16, Seung Hui Cho, a senior at Virginia Tech, shot and killed two people in West Ambler Johnston residence hall shortly after 7:00 a.m. He then returned to his own residence hall, changed from his bloody clothes, and left again. University and city police, plus emergency rescue teams, quickly cordoned off the crime scene and began to search for the killer and any evidence that he might have left behind.
Cho blended in with the normal flow of students and staff for the next two hours, except for a trip to the Blacksburg post office. There he mailed to the NBC network a set of writings and videotapes expressing contempt for his fellow students as privileged, spoiled, and morally corrupted by a materialistic society. He also mailed a letter to the English Department, where he was a major, criticizing a faculty member for “holocausting” him.
Shortly after 9:00 a.m. classes began, Cho entered Norris Hall—which has a mix of classrooms, offices, and laboratories—carrying two semi-automatic handguns, about 400 rounds of ammunition, a hammer, and a knife. He chained the main doors shut from the inside and began entering classrooms on the second floor, shooting anyone he saw. According to survivors, he said nothing and showed no emotion.
A student placed a 9-1-1 call, and the first police were on the scene in three minutes. Five minutes later they had blasted the lock off an unchained door. Two police teams attacked; they didn’t know whether there was one shooter or several, because they could hear two different caliber guns being fired. Almost immediately on hearing the blast of the shotgun that took the lock off the door, Cho killed himself. Had the police not entered the building so quickly, more people would have been killed. Cho still had about 200 unused bullets when he died.
The police continued their search of the building, while emergency rescue and medical teams began to triage and evacuate the wounded. No one who was alive when they were triaged died, and several lives were saved by rapid emergency medical action. But by then 30 more people, 25 students and five faculty, were dead—plus, of course, Cho himself.
In little more than two hours, 33 people had died, and 17 were wounded. Still others were injured jumping from windows. An unknown number of people who were directly or indirectly involved will carry the experience with them for the rest of their lives.
The Panel’s Work
Our panel interviewed more than 200 people, including the president and other administrators at Virginia Tech, faculty, residence-hall advisors, student-affairs staff, mental-health professionals at the university, police, emergency-rescue and medical-team members, the staff of local hospitals who treated the wounded and injured, and experts in various dimensions of the incident and what led up to it. We met with the heads of national higher-education associations, Virginia community-college presidents, and presidents of Virginia’s independent colleges and universities. Panel members did not meet with the informal association of public-university presidents, whose chair told the panel in June that it was “not timely” to meet with them.
Expert members of the panel interviewed Seung Hui Cho’s mother, father, and sister. Panel members met privately with the spouses, parents, and siblings of many who were killed or wounded. At four public meetings, they heard from spouses and parents, government officials, experts, and deeply concerned residents of Virginia.
By executive order and by securing outstanding legal counsel for the panel, Governor Kaine ensured that the panel had access to all but a handful of records. Its members read thousands of pages, including Cho’s educational and health records from early childhood through April 2007.
At the end of this extensive review, we were left with three kinds of concerns:
1. Structural: The underlying systems of public health and public safety provided by state and federal governments;
2. Management by the university and state government: What was done or not done by top decision makers; and
3. Actions on the ground: What was done at the scenes of carnage, medical care, and victim-survivor services.
Structural Issues
Public Mental-Health Care
The mental-health system of Virginia and probably most other states is entirely inadequate to provide the services needed to prevent incidents of this sort. The systems have been under-funded for years, and states have not provided the out-patient resources needed to make de-institutionalization of persons with mental disabilities a viable policy. Beds for short-term detention are in short supply, patient assessments are done too quickly and without the benefit of ancillary information, and judicial processes can be haphazard—all for lack of resources.
Gun Laws
Gun-acquisition laws have loopholes, and there are inconsistencies between federal and state laws. For example, the Federal Gun Control Act of 1968 prohibits gun purchases by anyone who “has been adjudicated as a mental defective [i.e., a person determined to be a “danger to himself or to others”] or who has been committed to a mental institution.” But while federal law clearly includes court-ordered outpatient treatment as a form of involuntary commitment, Virginia law is ambiguous on this matter. Because of this ambiguity, Cho’s name was not entered into the federal system for firearms background checks, and he was able to buy two semi-automatic weapons from gun dealers.
This and other ambiguities in Virginia law were removed by Governor Kaine through an executive order issued soon after April 16. But ironically, Virginia is the state that complies most comprehensively with the federal law, through its Central Criminal Records Exchange (which includes persons ineligible to purchase guns for reasons of mental disability). Twenty-eight states do not enter the names of any mentally ill persons in the federal register that gun dealers must check before selling a gun.
Another problem in Virginia’s gun laws: While a mental disability or criminal check must be conducted for transactions with a registered gun dealer, guns can be purchased from private persons (at gun fairs, for instance) without a background check. Even had Cho been prevented from buying his guns from dealers, he could have obtained them privately. Maryland, a contiguous state, requires background checks for all gun purchases.
Finally, Virginia’s public colleges and universities are not clear about their right to set policy about guns on campus. A few years ago, Virginia Tech, with the approval of the state’s Attorney General, set tough constraints on the possession of guns on campus. Other institutions were not aware that they could ban guns if they chose to do so.
Privacy Laws
The federal education-privacy and health-privacy laws (the Family Educational Records Privacy Act, or FERPA, and the Health Insurance Portability and Accountability Act, or HIPAA) are confusing and inconsistent. For example, the records of the Cook Counseling Center at Virginia Tech come under FERPA; those of Carilion St. Albans Behavioral Health Center, where Cho was taken in December 2005 under a Temporary Detention Order because he was judged to be a danger to himself or others, come under HIPAA. But while Carilion St. Albans could and did share Cho’s records with Cook under the provisions of HIPAA, it could not be assured that Cook Counseling Center would reciprocate, since FERPA is unclear about institutions’ sharing records with outside entities.
Because of the difficulty in understanding the laws and fear of personal or professional liability, educators often err on the side of caution even in situations when public safety is at risk. In Cho’s case, a representative of Virginia Tech told the panel that FERPA prohibits the university from sharing disciplinary records with the campus police department. But the panel learned that the University of Virginia shares them, on the grounds that its chief of police is designated as an official with an educational interest in those records. Moreover, the University of Wisconsin, Madison, and several other institutions have policies requiring an administrator to call the parents of any student who is referred to “de-tox” because of drunkenness (New York Times, September 13, 2007).
FERPA allows much more freedom to share information than many in the higher-education community assume. Personal observations and conversations with a student, for instance, fall outside FERPA; teachers or administrators who observe troubling behavior are not restricted from telling other administrators, law enforcement, or parents what they observe.
Colleges and universities also would benefit from information about what support services a student may have had in high school. Cho, for example, had what is called an “Individualized Education Plan” (IEP) that took account of his unwillingness to speak and his extreme shyness. But no one at Virginia Tech knew anything about the special help he had received in high school. Suddenly he was at a large university, entirely alone and on his own, away from family and the support services of high school. It might have helped greatly had Tech staff known what he needed and what they were facing.
University Management
One of the metaphors that recurred during the panel’s discussions and in public testimony was that “no one connected all the dots.” It is true that there were dots all over the map, but the way Virginia Tech is organized virtually ensured that no one ever was in a position to see them all and intervene in a potentially dangerous situation that eventually spiraled into disaster.
Dealing With a Troubled Student
Virginia Tech has systems and procedures in place to address the problems of troubled students, but in dealing with Cho from fall 2005 through fall 2006, those systems failed in numerous ways. For instance, the university has a Care Team—composed of the dean of student affairs, legal counsel, and the directors of residence life, judicial affairs, student-health services, and occasionally others—which identifies and works with students who are having behavioral problems. The team was alerted by the English Department and residence advisors about Cho’s deviant behavior—stalking, taking cell-phone photos of female students during class, violent writing, and unwillingness to participate in classes—but it took no action.
Other university personnel who might have been expected to take action also failed to follow through. On November 30 and December 12, 2005, Cho contacted the Cook Counseling Center. Both times, a counselor collected personal information by telephone, but Cho never was diagnosed or treated. On December 12, a female student called the campus police to complain about Cho’s behavior. The next day, a police officer instructed him to have no further contact with the woman, upon which Cho sent an instant message to one of his suite mates saying “I might as well kill myself.” The suite mate notified the campus police, and Cho was detained at Carilion St. Alban’s Behavioral Health Center for assessment.
No one called Cho’s parents to seek mental-health records from prior years or did a complete toxicology analysis. But a licensed clinical social worker found him to be a danger to himself or others and requested that the Temporary Detention Order (TDO) be issued; thus he spent the night at St. Alban’s. The next day, December 14, the special justice agreed that he was a danger to himself and others and ordered that Cho receive treatment at Virginia Tech’s Cook Counseling Center. In the presence of staff from St. Albans, Cho made an appointment by telephone at the Cook Counseling Center. That afternoon, the staff at Cook collected information from him again, this time in person.
In 15 days, the counseling center had talked with Cho three times. But no one called his parents to tell them that he had been admitted to a mental-care facility under a TDO or ask them whether he had a history of mental problems—and, if he did, whether they would make his mental-health records available to staff at Virginia Tech. Had they done so, they might have learned, as the panel did, that he had been diagnosed with “selective mutism” while in middle school, that he had been fascinated by the Columbine High School shootings in 1999, and that he had fantasized about carrying out a similar mass killing.
Furthermore, the center never provided any services, because Cho made only one appointment for counseling and did not keep it. No one followed up, since Cho had been accepted as a voluntary patient—the only kind the Cook Center accepts—and therefore was deemed responsible for scheduling his own care. Nor did the Cook Center notify the court, St. Albans, Virginia Tech officials, or the campus police that Cho had not returned for diagnosis and treatment.
This haphazard procedure even lacked proper documentation: Records of the personal information collected by Cook Counseling Center were not available to the panel, having been either lost or destroyed.
Emergency Preparedness at Tech
When an emergency occurs at Virginia Tech, its emergency plan dictates that a Policy Group consisting of senior administrators (at that time not including the chief of campus police) be convened by the president to oversee the university’s response. On the morning of April 16, the Policy Group, anxious to avoid a panicked reaction, acted slowly to alert the campus to a dangerous situation. In the emergency message it sent out almost two hours after the first shootings at West Ambler Johnston Hall, the group said there had been a shooting but did not state explicitly that two people had been killed and that the killer had not been apprehended.
At two other research universities in Virginia, a senior administrator (the chief operating officer or equivalent) and the chief of campus police can unilaterally activate emergency-alert systems. At Virginia Tech at that time, the chief of police could ask that emergency-notification systems be activated but could not act independently.
As word of the shootings spread not only across Virginia but also across the nation, parents, family members, and others tried to contact loved ones at Tech. Despite its being called for by Tech’s emergency plan, the university did not establish an emergency-operations center, which would have been a central locus for all communications. The university did establish a family-assistance center, but it was plagued by difficulties for at least three reasons: lack of leadership, lack of coordination among service providers, and lack of training. Volunteers tried to step in but were not able to answer many questions or guide families to the resources they needed.
Despite these failings, Virginia Tech staff and others were deeply compassionate and caring as scores of people tried to learn the fate of loved ones and to comprehend what had happened. Many volunteers worked not just for a few days but for weeks with families assigned to their care.
State Government’s Role
The responses to the disaster by various agencies of Virginia state government were mixed: some were very good, others slow or flawed.
With support from the Department of Public Safety, the Virginia State Police played an essential role in the aftermath of the shootings. Officers arrived quickly and stayed through the week. Sometimes working with local law-enforcement officers, the State Police personally carried news of death or injury to families across the state. They helped families seeking advice about what to do and where to go.
Although Tech was in constant contact with the governor’s office in Richmond, other state agencies were slower to mobilize. Around noon on April 16, the Department of Social Services (DSS) asked the Commonwealth Victim Services section (VSS) of the Department of Criminal Justice Services (DCJS) to stand by to help at Virginia Tech. But DSS sent no further word that day. It did not instruct VSS and the victim advocates it had mobilized to go to Tech until the afternoon of April 17, and they did not arrive until April 18, two days after the incident. Meanwhile, the director of the Department of Criminal Injury Compensation Fund (CICF) arrived at Tech around midnight on April 17. CICF provided important support to many families from that time on.
Family members, bewildered and grieving, were besieged by media. But the state government, like the university, failed to establish a formal information center, staffed by experienced public-information professionals who could manage the deluge of press inquiries and answer questions. Some state-agency personnel attributed the lapse and delays to what they perceived to be the university’s reluctance to share control of the family-assistance center or to cede authority over functions on campus.
The Office of the Chief Medical Examiner performed its technical work (primarily identification of the slain and autopsies) as well and quickly as it could. But its communication with victims-survivors was poor, leaving many parents, spouses, and other family members frustrated and confused.
Actions on the Ground
The response of police, emergency-rescue squads, medics, and others at both West Ambler Johnston Hall and Norris Hall, as well as the response by local hospitals, was generally excellent. Many people performed courageously, sometimes at threat to their lives.
But the panel judged that the Virginia Tech Police probably made two errors that may have affected the way the Norris Hall incident played out. First, they conveyed the impression that the killer at the residence hall was no longer on campus, perhaps because they became too focused on one person as a likely suspect. They seem not to have given adequate consideration to the possibility that their “good lead” might be wrong. Second, they did not request the senior university administration, meeting as the Policy Group, to notify the entire campus that two people had been killed and that the killer remained at large.
Still, it is important to recognize that after the first two killings, Cho mailed a package of writings and videos to NBC. From that point on, he had identified himself as a killer committed to further violence. Had an all-campus alert been issued and classes been cancelled immediately after the first killings, he probably still would have struck somewhere.
The Virginia Tech Police worked extremely well with the Blacksburg police, probably because the two bodies train together and had done so quite recently. Similarly, the Virginia Tech Rescue Squad and the Blacksburg Volunteer Rescue Squad regularly train together and worked very well as a team. But communication among the various participants—police and rescue units, emergency-medical services, hospitals—was impeded initially because units were operating on their own radio frequencies and had trouble accessing other systems.
Police respose to the shootings at Norris Hall was extremely quick because so many campus and Blacksburg officers already were on campus in response to the killings at West Ambler Johnston Hall. They were at Norris three minutes after hearing of the shootings, gained entry to the building five minutes later, and were on the second floor within minutes after that.
Blacksburg and Tech police, the Virginia Tech Rescue Squad (consisting of students), and the Blacksburg Volunteer Rescue Squad provided essential triage and evacuation. Fourteen Emergency Medical Service teams from the region helped transport the wounded to local hospitals. Again, interagency training and drills probably helped the teams work well together.
Evacuation from Norris Hall of persons who were not wounded or injured was not handled as well, with many persons experiencing the terror of leaving a building under siege without escort and being confronted by armed police officers. But the police focus on ensuring that the shooter or shooters were dead was appropriate.
Crucial Lessons Learned
So what broadly relevant lessons can we learn from this tragedy? There are at least seven.
(1) States should provide sufficient outpatient mental-health services. De-institutionalizing many persons with mental-health problems was a humane and decent initiative almost 40 years ago. But Virginia, and probably other states, then entered into a prolonged period of under-funding outpatient mental-health facilities and providers. The result has been poor and superficial care, or no care at all.
In Virginia, a licensed clinical psychologist is paid $75 to perform an independent evaluation of a person facing a hearing to determine whether he is incompetent or poses a threat to himself or others. The psychologist usually spends 30 minutes or less doing the evaluation, which must be completed within 48 hours. The psychologist does not have enough time to gather information about the person’s mental-health history and usually cannot afford to wait around for the commitment hearing.
(2) States should comply with the Federal Gun Control Act. All states should enter the names of persons judged to be mentally disabled in the federal register, and the federal government should provide incentives to them to do so. All states should define key terms like “voluntary and involuntary commitment” and “threat to self or others” at least as stringently as the federal laws do. And all states should require background checks for all firearms sales, whether from registered gun dealers or in personal transactions.
(3) Congress and the state legislatures should review federal and state privacy laws, and universities should know what they do and do not permit. HIPAA (the Health Insurance Portability and Accountability Act), FERPA (the Family Educational Rights and Privacy Act), and state privacy laws should be reviewed to ensure that they are compatible and that they meet the real needs of our society. National higher-education associations might make training about FERPA an on-going service to member institutions. When college and university administrators (and faculty) assume that FERPA is more restrictive than it actually is, they can fail to act, with devastating consequences.
(4) Colleges and universities should communicate, both within themselves and beyond. Institutions need to break through current barriers to communication to ensure that information about potential threats is shared by everyone who needs to know.
The notion of “threat-assessment teams” seems foreign to institutions of higher learning. But something like them is necessary to evaluate a wide range of possible incidents: storms, toxic spills or leaks, pandemics, and, yes, active shooters. In conversations with numerous presidents and senior administrators and with the leaders of several national associations, members of our panel were assured that institutions do plan for most of these disasters, although “active shooter” was not widely included among the dangers for which plans had been developed. Tech has an emergency plan; it was out of date last spring, but it probably is current now. But if no one is in a position to assess threats, plans are useless.
When a critical incident does occur, it is essential that responders—police, fire, emergency-rescue and medical teams—be able to communicate quickly and effectively. Different communication frequencies were a problem, quickly surmounted, at Virginia Tech. But even initial disruptions can be avoided by advance planning among responders.
It also is essential to communicate quickly to the entire academic community and to notify law-enforcement agencies and local and regional emergency-rescue and medical teams. Ironically, Virginia Tech was installing a new text-messaging notification system on April 16. But even with the systems it had in place, it was able to send thousands of emails in just a few minutes. Institutions need multiple ways to send emergency notifications to students, faculty, and staff. Some of these should not depend on an electrical-power system that could be disrupted.
Finally, the process of communicating should be as free as possible of bureaucracy. In two Virginia research universities, one senior official (the chief operating officer or equivalent) and the chief of police can issue emergency notifications. In the absence of the senior administrative official, the police chief can act independently. At Virginia Tech, senior administrators had to meet to assess the situation and then decide whether to issue an emergency message and what to say in it. This cumbersome process may have contributed to the delay in issuing an emergency notification and to its lack of specificity once it was issued.
(5) Write a plan that fits. One size does not fit all. Large universities have the resources to maintain sworn police forces; small, rural community colleges and private institutions do not. There are colleges in places where cell phones do not work or that serve students who may not own them. Universities with medical schools and teaching hospitals have a rich array of health-care providers upon whom they can draw in an emergency; a small private college may not.
Different campuses are vulnerable in different ways. Colleges and universities in large cities often exist on a grid of city streets that are used by the general public. Research universities are more vulnerable to toxic spills or leaks than institutions that focus on undergraduate education.
(6) Make formal arrangements, and practice. Every institution, whatever its size and mission, needs to have formal arrangements with local law enforcement, emergency-rescue and medical teams, and hospitals. And every institution needs to practice for emergencies: pandemics, toxic spills or leaks, active shooters, dangerous weather. No matter how good the planning, whatever happens will not be exactly what had been anticipated. (As General Dwight D. Eisenhower once observed, “planning is essential, but plans are useless.”)
The fact that they trained together was crucial to the fact that law enforcement, rescue, and emergency-medical personnel responded so well to the emergency at Norris Hall on April 16.
(7) Develop a way to access students’ mental-health records. Records of immunization travel with us from early childhood through institution after institution. But a college or university does not get records about communicable diseases, not to mention serious mental-health problems, psychotropic medications (which a student may stop taking), or special-education programs that may have helped a student in high school.
This information clearly should not be used in admissions. But later, perhaps while choosing courses, students might be asked to sign waivers allowing the institution access to their health records. At the least, university staff should be expected to ask the parents of a student whose behavior causes concern for access to her or his health records.
Final Thoughts
The chances of such a disaster happening again are slim. On average, there are 16 killings a year at America’s colleges and universities. Still, every death is an incalculable loss to a mother or father who says, “She is dead, and my life has no meaning.” The poet is wrong; time does bring relief. Yet the loss remains.
But the rest of us forget quickly. When Virginia’s state-supported universities opened for the fall 2007 term, most offered text-messaging systems to alert students, faculty, and staff to emergency situations. But except for Virginia Tech, where 56 percent of the community registered for the system, the sign-up rates at the four largest universities were well below one-third.
This seems to be typical across the nation. When a text-messaging system was used to alert students at St. John’s University in New York to the presence of an armed man on campus, only 2,100 of about 20,000 persons in the community had signed up for the system (New York Times, September 28, 2007, C13). Nonetheless, the messaging was effective, and by the next day the number of subscribers had tripled to 6,652.
We cannot guarantee that an incident like that at Virginia Tech will not occur on some other campus. But we might reduce the probability of its happening again by attending to the lessons the Virginia Tech panel learned as it tried to comprehend what happened on April 16, 2007.
MEMBERS OF THE GOVERNOR’S VIRGINIA TECH INVESTIGATION PANEL
Col. Gerald Massengill, Chair, former Superintendent, Virginia State Police
Dr. Marcus L. Martin, Vice-Chair, emergency-medicine specialist, University of Virginia medical school
Gordon K. Davies, former director, State Council of Higher Education for Virginia
Dr. Roger Depue, former FBI agent and security specialist
Carroll Ann Ellis, victim’s advocate, Fairfax County Police Department
The Hon. Tom Ridge, former Secretary of Homeland Security and former Governor of Pennsylvania
Dr. Aradhana A. “Bela” Sood, child psychiatrist, Virginia Commonwealth University medical school
The Hon. Diane Strickland, former Virginia Circuit Court judge and co-chair, Boyd-Graves Conference on involuntary mental-commitment issues