This article was downloaded by: [The Aga Khan University]On: 25 November 2014, At: 07:12Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Criminal Justice Studies: A CriticalJournal of Crime, Law and SocietyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/gjup20

Deaths in Custody: The Utility of DataCollected from County CoronersWilliam V. Pelfrey Jr & Michele White CovingtonPublished online: 12 Mar 2007.

To cite this article: William V. Pelfrey Jr & Michele White Covington (2007) Deaths in Custody:The Utility of Data Collected from County Coroners, Criminal Justice Studies: A Critical Journal ofCrime, Law and Society, 20:1, 65-78, DOI: 10.1080/14786010701241358

To link to this article: http://dx.doi.org/10.1080/14786010701241358

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Criminal Justice Studies,

Vol. 20, No. 1, March 2007, pp. 65–78

ISSN 1478–601X (print)/ISSN 1478–6028 (online) © 2007 Taylor & Francis

DOI: 10.1080/14786010701241358

Deaths in Custody: The Utility of Data Collected from County CoronersWilliam V. Pelfrey, Jr. & Michele White CovingtonTaylor and FrancisGJUP_A_224048.sgm10.1080/14786010701241358Criminal Justice Studies1478-601X (print)/1478-6028 (online)Original Article2007Taylor & Francis201000000March 2007Dr. WilliamPelfreypelfrey@uwm.edu

The Deaths in Custody Act was recently expanded to require all states to collect data on

deaths which occur while in law enforcement custody. Traditionally, use of force data are

collected from law enforcement agencies. This research utilized a different approach:

county coroners were contacted and interviewed to collect data on deaths which occurred

while in the custody of law enforcement. The methods, logic, and findings of this research

are discussed as well as the efficacy of incorporating county coroners in data collection.

Keywords: Custody Act; County Coroners; Police Force; Reporting

Introduction

A substantial body of literature exists concerning the application, and misapplication,

of police force applied to suspects. Deaths of suspects while they are in police custody

occur relatively rarely, a fact that has been referred to as ‘an analytic misfortune but a

societal blessing’ (Geller & Scott, 1992, p. 22) but the costs of the deaths that do occur

are substantial and irreversible. A high profile death of a suspect (i.e. Nathaniel Jones

in Cincinnati, Adolph Archie in New Orleans, or Arthur McDuffie in Miami) can lead

to a deterioration or collapse of police/community relations, the indictment of officers,

or even rioting by disgruntled citizens (Skolnick & Fyfe, 1993). Reviewing these rare

events serves to foster trust between the police and the public as police are held

Professor Pelfrey is an Assistant Professor in the Department of Criminal Justice at the University of Wisconsin–

Milwaukee. He received his PhD in Criminal Justice from Temple University and his MS in Clinical Psychology

from Radford University. His primary research areas are: community policing, terrorism prevention, the spatial

analysis of crime, police use of force, and the psychology of the offender. His publications have appeared in Justice

Quarterly, The International Journal of Offender Therapy, and other venues. Michele White Covington earned her

master’s degree in criminal justice from the University of South Carolina. She is a former deputy sheriff at the

Greenville County Sheriff’s office and is pursuing doctoral studies in criminal justice. Correspondence to: William

V. Pelfrey, Jr., PhD, Department of Criminal Justice, PO Box 786, University of Wisconsin–Milwaukee, Milwau-

kee, WI 53201, USA. Tel: +1 414 229 6953; Email: pelfrey@uwm.edu

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

66 W. V. Pelfrey, Jr. & M. White Covington

accountable for deaths which should not have occurred. Although this determination

is outside the scope of this research, details of the cases reported here will evince the

nature of those scenarios which lead to the death of a subject while in police custody.

In statements made on the House floor in support of what would become the Death

in Custody Reporting Act, Senator Tim Hutchinson (Arkansas) stated:

This legislation will provide openness in government and will bolster public confidence

and trust in our judicial system. In addition, I believe that it will serve as a deterrent to

future misconduct by wrongdoers who will know that someone will be monitoring their

actions. (July 2000)

The Senator’s line of reasoning is consistent with the position of numerous criminal

justice researchers. Chevigny (1991) argues that closed investigations of police miscon-

duct, especially those instances involving use of force, lead to resentment and distrust

of the police, particularly in minority communities. Skolnick and Fyfe (1993) concur,

arguing for transparent reviews of police use of force incidents and open access to data

and the review process.

The Death in Custody Reporting Act was signed by President Bill Clinton in October

2000. This action (Public Law Number 106-297) amended the Violent Crime Control

and Law Enforcement Act of 1994 to require states to collect information on suspects

who die while in police custody, as described by the following language:

Section 20104(a) of the Violent Crime Control and Law Enforcement Act of 1994 (42 U.S.C.

13704(a)) is amended—

such State has provided assurances that it will follow guidelines established by the

Attorney General in reporting, on a quarterly basis, information regarding the death of

any person who is in the process of arrest, is en route to be incarcerated, or is incarcer-

ated at a municipal or county jail, State prison, or other local or State correctional facil-

ity (including any juvenile facility) that, at a minimum, includes—

(A) the name, gender, race, ethnicity, and age of the deceased;

(B) the date, time, and location of death; and

(C) a brief description of the circumstances surrounding the death.

Following the passage of the Deaths in Custody Reporting Act, states have reported

deaths in custody data for corrections and juvenile detention centers since the year

2000. Law enforcement organizations have been reporting deaths in custody only since

2003. Since many of the individuals who die while in the corrections system are victims

of illness, age, or other medical condition, the corrections data-sets are significantly

larger than the law enforcement data-sets. However, deaths which occur at the hands

of law enforcement usually receive greater interest. These deaths are often the direct

result of police actions (e.g. shooting a suspect) and therefore receive high levels of

public scrutiny.

Each state is responsible for its own data collection with few requirements as to the

process of securing those data. In South Carolina, researchers elected to contact county

coroners rather than law enforcement agencies. Although law enforcement agencies

represent the obvious contact point, contacting coroners is much more efficient (46

county coroners as opposed to approximately 300 law enforcement agencies) and

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

Criminal Justice Studies 67

removes the risk of a law enforcement agency obfuscating the circumstances surround-

ing a death to protect their reputation. Coroners represent a neutral component of

the criminal justice system, increasing the validity of the data. Thus, a census of the

46 county coroners in South Carolina was conducted during each quarter of 2003 and

the first quarter of 2004 to determine whether any suspects died while in police

custody.

The definition of a death in custody was developed by the Bureau of Justice Statistics

(BJS) and is as follows: A death of an arrested person or an intended arrestee that occurs

during the process of arrest, in the custody of law enforcement officers, or as the result of

lethal force by officers. BJS further specified the definition with the following stipulations:

We include deaths of persons arrested or subject to arrest who die:

● From any cause while in the physical custody or physical or practical restraint of law

enforcement officers (even if no arrest has yet occurred);

● At a crime scene, in transit, in a booking or intake center, or medical facility, or in a

police lock-up or jail prior to booking;

● Killed by any use of force (by officers).

We exclude deaths:

● Which result from accidents, suicides, or natural causes prior to being placed in custody

or physical restraint, or prior to the use of force by officers;

● Of persons in formal locked confinement after arraignment (jail, detention center, or

prison);

● Of persons not intended to be arrested (bystanders, etc.).

BJS developed a set of forms to be used in the data collection process, including a quar-

terly assessment sheet and a case-specific sheet which requests demographic and inci-

dent data.

Coroners represent an infrequently studied and often misunderstood component of

the criminal justice system (Combs, Parrish, & Ing, 1995; Walker, 1999). Most county

coroners are elected. Their responsibilities usually include: coordinating the paper-

work associated with deaths, determining or pronouncing the official cause of death,

and organizing autopsies. The Department of Justice reports that there are approxi-

mately 1,600 coroners or medical examiners in the USA (DOJ, 2000). Little research

attention has been focused on coroners despite their often-significant power and

discretion within the criminal justice system (Nolte, Yoon, & Pertowski, 2000).

Although law enforcement officers specialize in the investigation of the events

surrounding a death, coroners (and the pathologists or physicians who conduct autop-

sies) specialize in defining the cause of death. Coroners often incorporate portions of

the police investigation in their determination of the cause of death but are free to

augment those investigations with their own resources.

Since many coroners either investigate causes of death or, at a minimum, pronounce

the cause of death, they determine whether a death resulted from natural or unnatural

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

68 W. V. Pelfrey, Jr. & M. White Covington

causes. If a death resulted from unnatural causes, the coroner may assign responsibility

or determine whether the death was avoidable. As a public servant, the coroner may be

placed in a difficult, even compromising position, when a death occurs at the hands of

the state, such as a death while in the custody of law enforcement officers (Beckett,

1999). Incorporating county coroners in the research and data collection process not

only enhances the efficiency of the research but also increases our understanding of a

rarely considered component of the criminal justice system.

It should also be noted that the county coroner represents a relatively neutral compo-

nent of the criminal justice system. While law enforcement agencies are beholden to the

public, government officials, and the media for the production of crime control, coro-

ners face no such requirement. Thus the data and information provided by coroners

should be considered valid. The National Crime and Victimization Survey, one of the

most widely used and known databases in criminal justice, was created in part due to

concern over the validity of data from law enforcement agencies. Similarly, data from

coroners represent a relatively new and heretofore untapped resource within the crim-

inal justice system.

Because the deaths in custody data and the data collection process are new, they

should advance the understanding of the situational exigencies (Bittner, 1970)

surrounding deaths that occur while in police custody. Once the data were collected,

they were analyzed as individual cases and cross-referenced for similarities. As data are

collected over several years, a typology of deaths in custody scenarios may be developed.

Death at the hands of the police is often a result of one of the following: positional

asphyxia, alcohol or drug use, neurological impairments, obesity, oleoresin capsicum,

or some combination of the above (Ashcroft, Daniels, & Hart, 2003; Chan et al., 2001;

Dalton, 1998; Donahue & Horvath, 1991; Krosch, 1992; Reay, 1996). This is in addition

to the obvious threat of shootings. At this point there is only one statement about deaths

in custody cases with which most scholars in the field would agree: not all deaths in

police custody are avoidable (Leigh, 1999). The logical converse is that some deaths may

be preventable. A discussion of the research methodology employed to collect the data,

the results acquired during the research, and the utility and importance of the findings

follow.

Research Methodology

Design

This research employs a case study design, possibly the most misunderstood research

design in social science today. Traditionally, case studies have been viewed as weak

simply because of the lack of quantification involved in conducting them. This method

is still used, however, and has produced valid research from many well-respected schol-

ars in various fields (Yin, 1994). The principal drawback to case study research is the

difficulty in generalizing from the case to other scenarios. The case study ‘does not

represent a “sample”’ (Yin, 1994, p. 10) as other types of research may, and it is there-

fore less generalizable to other time periods, settings, and/or situations. In the instance

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

Criminal Justice Studies 69

of this particular research involving deaths in police custody, a case study design is the

most appropriate choice due to the relative rarity of the event.

Subjects

The units of analysis in this research and the population include all subjects who died

while in the custody of a police officer or law enforcement agency in South Carolina

during the time period of this research (all of 2003 and the first quarter of 2004). This

does not include subjects who died in the custody of jails or those who die in the

process of a police pursuit. The cases include only those who died under the actual

custody of police, from the time that the subject came under the control of the

officer(s) to the time of arraignment.1 The study includes deaths that occurred in the

calendar year of 2003 and the first quarter of 2004. This substantially limits the reach

of the research, but the collection of this type of data is new to South Carolina and most

other states, and it is the only data available. The small number of cases to be studied

does increase the level of specificity applied to each case. That is, there is no sampling

error because the entire population is specified in this research.

Data Collection Source

Data for this research were collected from all of the county coroners in South Carolina

through the use of a telephone survey. The county coroners were selected as respon-

dents for several reasons. First, contacting the county coroners, as opposed to contact-

ing all of the law enforcement agencies individually, is a much more efficient use of

limited research resources. This strategy meant contacting 46 county coroners rather

than about 300 (law enforcement agencies), making it possible to acquire detailed data

from all of South Carolina rather than using a sample (as opposed to summary data

from the FBI).

Second, the efficient use of resources was particularly important as data were

collected every quarter (a requirement imposed by BJS). Since data turnaround times

were so brief, a mail survey was deemed inappropriate. Additionally, many small, rural

law enforcement agencies do not have a formal email address. Thus a phone survey was

defined as the most appropriate data collection method in light of the quarterly collec-

tion requirement.

A third reason for utilizing the county coroners as the principal data source was their

objectivity. Since much of the data they collect is public information, they are viewed

as a reliable source of data. Although unlikely, it is conceivable that a law enforcement

agency might seek to protect its reputation by obfuscating the nature of a suspect’s

death (see books such as Kappeler, Sluder, & Alpert, 1994, or Skolnick & Fyfe, 1993, for

examples of this type of behavior). While many coroners’ offices are closely linked to

law enforcement agencies, and sometimes even populated by former law enforcement

officers, the coroner’s office is a distinct entity held accountable by both the county

government and the media. Thus they may be a more reliable source of data than local

law enforcement agencies.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

70 W. V. Pelfrey, Jr. & M. White Covington

Instrumentation

The instrument used to collect this data is a survey designed by the Bureau of Justice

Statistics for the purpose of collecting information on deaths in police custody (see

Appendix). Each respondent is first asked for the number of deaths in police custody

in his/her respective county between January 1, 2003 and December 31, 2004. An

explanation is provided of exactly what a death in police custody consists of for the

purposes of this research. For each death reported, an addendum is completed which

includes detailed information on the deceased and the characteristics of the incident

which led to his/her death, such as the weapon which caused the death (if any), the

manner and official cause of death, and whether the deceased exhibited any significant

or strange behavior (i.e. symptoms of illness, intoxication, drug use, threatening the

officer) before or during the time that the incident took place.

Data Collection Methodology

Prior to data collection, in lieu of a pre-sensitization letter, research staff attended one

of the annual meetings of the South Carolina Coroner’s Association. At this meeting

research staff described the purposes and methodology of the research, distributed

copies of the instruments to be used, and fielded questions concerning the research.

This meeting was edifying in that coroners suggested several hypothetical situations and

examples which were not included in the parameters of the DOJ instrument. Research

staff and the coroners then discussed how a death in custody was defined. The coroners

were told they would be contacted every quarter by research staff and to expect our call.

Data were collected during a series of phone calls to each county coroner’s office. A

list of coroners’ office phone numbers was acquired from the President of the South

Carolina Coroner’s Association. During this call, either the county coroner or a deputy

coroner was asked if any individual had died while in police custody during the past

quarter. If there was a death, the details of the event were discussed to insure that the

death fell within the definition of a death in custody (established by BJS). If the death

did in fact occur in custody, then the questions of the survey instrument were posed

and the data were collected. When the coroner or deputy coroner could not remember

all the details of the death, a follow-up call was arranged to finalize the data collection.

Several coroners asked if we could simply mail them a copy of the survey and they could

complete it if/when a death occurred. However, BJS precludes this approach, requiring

that research staff ask each question on the instrument.

With each quarterly wave of data collection, all 46 coroners’ offices were contacted.

Generally, research staff were able to speak to either the coroner or a deputy coroner

and ascertain whether a death occurred during the reporting period. On some occa-

sions, no one at the coroner’s office was available. When this happened, we either left

a message requesting that a representative from the office contact us, or, if no message

system was available, we simply called back. Each quarter there were a few counties

which we could not contact, despite every effort. Although we initially planned to make

only three repeat calls to a non-responsive county, we generally made 10–15 contact

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

Criminal Justice Studies 71

attempts before omitting that county from that quarter’s data-set. It should be noted

that these were usually the smallest, least populated counties. The largest counties were

relatively easy to contact as they operate with a higher degree of bureaucracy and staff-

ing. There were simply some counties where the coroner was frequently unavailable

(i.e. ‘He’s out fishing’) and the coroner was the only person who could answer ques-

tions (as there were no deputy coroners).

Results

Overview of Findings

The purpose of this research was to identify deaths that occurred in police custody in

South Carolina in 2003 and the first quarter of 2004. The cases described below were

collected by phone interviews with the state’s county coroners. While there are only six

cases represented (including one for which no details were released as the case is still

under internal and state investigation), this is not a sample but a census of all police

custody deaths for the entire calendar year. The data, of course, are limited in several

ways. It is difficult to draw conclusions from so few examples of a social phenomenon.

Also, the data only represent one specific state for one specific year. Obviously, the data

would be more generalizable and would possess greater reliability if there were more

events. Hopefully this methodology will be replicated. A narrative of each case is

presented to explain the exigencies surrounding each death.

Case 1

A single officer responded to an evening domestic violence call in September 2003.

When the officer attempted to question the subject, the subject fled across the yard.

After the officer gave chase, both subject and officer ended up back inside the dwelling.

At this point the officer realized that his handgun was missing. He assumed that the

subject had taken it at some point during the struggle and that consequently the subject

was now armed. At some point during the struggle, the subject lunged toward the

officer and threw him down onto the bed. It appeared to the officer that the subject was

going for his gun (the one that he had presumably taken from the officer), so the officer

pulled his back-up weapon and shot the subject, killing him. The cause of death was

ruled as gunshot wounds to the neck, fractured vertebrae and skull, and laceration of

the brain stem. The officer’s gun was later found in the subject’s yard, where it had

apparently been dropped at some point during the chase.2 The subject in this case was

a 24-year-old white male.

Case 2

This case began as the result of an officer’s suspicion that the subject was under the

influence of alcohol while driving. The officer, who was working by himself at the

time, proceeded to signal the subject to stop his vehicle. The subject complied, but

immediately exited the vehicle and began to run from the officer. The subject managed

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

72 W. V. Pelfrey, Jr. & M. White Covington

to get into a citizen’s back yard before the officer was able to catch up and attempt to

subdue him. At this point, the subject took the officer’s flashlight from his belt and

began beating the officer about the head with it. At this point the officer pulled his

handgun from his belt and shot the subject once in the chest. He died later at a medical

facility. Medical cause of death was ruled exsanguinations due to lacerations of the

heart and liver due to a gunshot wound to the chest. The officer sustained head inju-

ries serious enough to hospitalize him and prevent him from working for three

months. The deceased in this case was a 43-year-old black male.

Case 3

An officer arrived to a domestic dispute in progress and arrested the subject after an

extensive struggle. Upon handcuffing the subject, the officer found that the subject had

thumb injuries, which had occurred prior to the officer’s arrival. At this point, the

officer called emergency medical services (EMS) to the scene. When EMS arrived, the

subject was put onto a backboard with splints for his hand injuries. Shortly afterward,

the subject became unresponsive and could not be revived. The subject was

pronounced dead upon arrival at the local hospital’s emergency room around mid-

evening. The subject’s family reported that shortly before police arrived, the subject

had ingested a bottle of white powder. Upon autopsy, the cause of death was found to

be acute cocaine toxicity. The subject was a 36-year-old Hispanic male.

Case 4

On an evening in March 2003, officers arrived for a call to a burglary in progress to find

the subject inside the residence in question. The officers placed the subject under arrest,

at which time he offered resistance and was sprayed with pepper spray. The subject was

handcuffed and placed inside the officers’ vehicle, where he attempted to escape once

more and was pepper-sprayed three more times. As the officers proceeded towards the

jail, the subject began to have what the officers described as a seizure. At this point,

the officers began to monitor the subject’s condition. The officers then noticed that the

subject had become unresponsive and decided instead to transport him to a local

hospital. The subject was pronounced dead by attending emergency room physicians.

The subject in this case was a 34-year-old black male. Upon autopsy, the subject was

found to have cocaine in his system. The cause of death was ruled as cocaine-associated

agitated delirium syndrome due to cocaine toxicity.

Case 5

Around midday in January, officers arrived to arrest a 61-year-old black male on

outstanding warrants for aggravated assault, resisting arrest, and check fraud. The

officers took the subject to a holding facility where he was booked for these charges. At

this point, the subject began to act irrationally and display odd behavior such as shout-

ing and smearing fecal matter on the walls of the facility, after which the officers noted

that the subject became despondent. At this time, the officers transported the subject

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

Criminal Justice Studies 73

to a local emergency room where a medical examination showed that he had advanced

necrosis of his lower left leg (commonly known as gangrene). Attending physicians

placed a stint in the subject’s iliac artery in an attempt to alleviate this condition. The

stint became dislodged and the subject suffered extreme blood loss of two to three

liters, which was ruled in an autopsy as the cause of death. At the time of death, the

subject was also suffering from profound hypothermia with a core body temperature

of only 79°F. The subject had a history of schizophrenia. The only drugs found in his

system were consistent with his medical treatment at the hospital.

Case Characteristics

The five subjects described above were all males, three of which were black, one white,

and one Hispanic. Ages ranged from 24 to 61 years old (Table 1). A geographic review

indicates that the cases were not concentrated in one specific area of South Carolina but

were spread across much of the state and included both rural and urban areas (loca-

tions and names are withheld for confidentiality purposes).

The cause of death of the subjects, although varied, can easily be categorized. In two

cases, the deaths were directly attributed to drug use. Two of the deaths were due to

gunshot wounds. Only one case was due to medical trauma of some sort; this was Case 5

in which the subject died due to complications from his treatment in the emergency room.

All five of the subjects in this data-set, except that of Case 5, would have been charged

with either resisting arrest, assault, or some similar charge that reflected their struggle

with the officer(s) involved. The subjects from Case 1 and Case 3 would also have been

charged with criminal domestic violence. The subjects from Case 2 and Case 4 would

also have been charged with counts involving improper alcohol and drug use. Case 2

involved a charge of driving under the influence of alcohol as well as the lesser, related

charge of failure to stop for a blue light. Case 4 involved cocaine use. This case also

involved breaking and entering. The subject involved in Case 5, the only death to have

occurred after booking, was charged with aggravated assault, simple assault, and writ-

ing fraudulent checks.

Subject behavior appears to be a strong influence on the outcomes of these cases. In

all of the cases reported herein except Case 5 (which appears to be different from most

cases in many ways), the subjects appeared intoxicated, resisted being handcuffed or

arrested, tried to escape from custody, and grabbed, hit, or fought with the involved

officer(s). The subject from Case 2 also used a weapon to assault the officer who was

Table 1 Demographic Descriptions.

Case Gender Race Age

1 Male White 24

2 Male Black 43

3 Male Hispanic 36

4 Male Black 34

5 Male Black 61

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

74 W. V. Pelfrey, Jr. & M. White Covington

attempting to arrest him. In two of these cases, the subjects were handcuffed when the

events causing the deaths occurred. In the other two cases, the officers were unable to

handcuff the subjects before fighting ensued.

These four cases display striking similarities. Case 5, however, has little in common

with the others. This subject did not resist arrest, he was older than subjects in the other

cases, and he had a known history of mental disease. This is also the only case that led

to currently pending lawsuits. Furthermore, the events that led to the death of the subject

in this case seem to have been less under his control than in the other cases. Of course,

this situation could have been avoided had the subject not been involved in criminal

activity or if he had sought medical help for his gangrene before, but in the immediate

actions that led directly to his death, he seemed to have less control. Unlike the other

subjects, he did not attempt to flee or fight. He was simply arrested without incident,

which provides a strong dissimilarity between this and the other cases in this set.

Discussion

When the factors surrounding the individual deaths are considered, the accuracy of

Chevigny’s assertion (1991) is evident: resistance against police authority is likely to

lead to instances of violence and abuse. Examining the situational exigencies in each

case report indicates that the suspects precipitated the use of force against them. For

four of the five cases presented in the results section, resistance precipitated a degree of

force which ultimately led to the suspect’s death. Even in the single case where the death

appears to be a result of a medical failure, the suspect resisted the police as well. While

it is extremely difficult to determine whether the level of force was appropriate or

excessive, it appears that some degree of force was required to subdue the suspects and

either take them into custody or restrain them while in custody.

There were a number of important methodological limitations which must be

considered. Data were collected for a 15-month period in a single state. While we made

every attempt (usually at least a dozen calls) to contact every coroner’s office, each quar-

ter there was at least one of the 46 counties that was not responsive to our calls. Research

staff identified an inverse relationship between the population of a county and the ease

of locating the coroner. For each data collection period we were able to contact the most

populous counties with no more than three or four calls. However, in the lowest popu-

lation counties, the coroners were difficult, and sometimes impossible, to locate. It is

therefore possible that a death in custody was missed due to our inability to contact

some coroners. When we could not contact a coroner for a specific quarter, subsequent

data collections were retroactive. That is, if we could not reach a coroner for one quar-

ter, when we contacted them for the next quarter we asked whether there were any

deaths over the past six months. Thus we attempted to eliminate this possible confound.

Additionally, it is possible that some coroners erroneously omitted a death due to a

failure to recall it. Although a death at the hands of the police is a rare and high profile

event, it is conceivable, albeit unlikely, that a coroner would fail to remember it. In fact,

coroners were more likely to be overly inclusive than the converse. Research staff then

identified those cases which met the specifications established by BJS.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

Criminal Justice Studies 75

Despite these limitations, and those discussed in the research methods section, the

data collected from the coroners’ offices appears to have a degree of integrity and is

representative of the State of South Carolina. However, findings should be generalized

to other states with caution due to the short data collection period and the size and

demographic composition of the state.

Conclusions

The findings from this research indicate that several people died while at the hands of

the police, usually as a direct result of resisting arrest. This is unfortunately true in most

states for most years. Identifying the frequency of these deaths and the circumstances

around them represents valuable information for policy makers, researchers, and law

enforcement officials.

While the specific cases identified in this research are interesting, the most important

finding of this effort lies in the nature of the data collection. Incorporating this rare and

infrequently considered branch of the criminal justice system served as a valuable

contribution to the research: instead of contacting some 300 or so law enforcement

agencies, we contacted 46 county coroners every quarter. This approach was much

more efficient and likely removed a potential confound (the risk of law enforcement

agencies obfuscating an event to protect their reputation). We strongly encourage

other states to consider the utility of a coroner’s census approach as part of their data

collection effort.

One of the challenges of this research was the quarterly data collection requirement.

Each quarter there were several coroners who were unreachable—they were not avail-

able when we called them and they failed to return our calls. At some point, we became

a pest to them and they may have elected to ignore us. The Bureau of Justice Statistics

should certainly consider positing a request (to lawmakers) to reduce the data collec-

tion period to an annual or semi-annual requirement. Deaths in custody are rare

enough (in most states) that an annual or six month interval between contacts would

suffice. This period would decrease the burden on research staff, the agencies which

supply the information, and would probably increase the likelihood of accurate and

consistent reporting of events.

Future deaths in custody research will provide valuable data which will greatly

expand the utility of the current findings. As data for more years, and more states,

become available, researchers can identify demographic, geographic, and situational

patterns that will be of interest to many agencies and populations. Expanded data may

support comparisons of the efficacy of different laws or policies, the role of drugs in

leading to deaths, the utility of a typology in classifying deaths that occur while in police

custody, and various other research directions.

Acknowledgements

This research was sponsored by the Office of Justice Programs at the South Carolina

Department of Public Safety and conducted at the University of South Carolina. Points

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

76 W. V. Pelfrey, Jr. & M. White Covington

of view expressed in this work do not necessarily reflect the position or policies of the

granting agency.

Notes1

[1] The parameters defining ‘in custody’ were established by the Deaths in Custody Act and speci-

fied by BJS. For our purposes, a death in police custody includes any death that occurs while a

subject is under the control of law enforcement. This would include but not be limited to:

deaths occurring in arrest attempts, even if the subject had not been formally charged at time

of death, deaths occurring in the process of transport, and post-arrest deaths if the incident

occurs before arraignment. This particular definition does not include deaths occurring in

jails as under police custody.2

[2] Based on the information received regarding this case it is not clear whether the officer

dropped the gun during the chase (the most likely scenario) or the suspect acquired the gun

during the chase, then dropped or lost it in the yard.

References

Ashcroft, J., Daniels, D. J., & Hart, S. V. (2003). The effectiveness and safety of pepper spray (National

Institute of Justice: Research for Practice). Washington, DC: National Institute of Justice.

Beckett, C. (1999). Deaths in custody and the inquest system. Critical Social Policy, 19(2), 271–280.

Bittner, E. (1970). The functions of the police in modern society. Rockville, MD: National Institute of

Mental Health.

Chan, T. C., Vilke, G. M., Clausen, J., Clark, R., Schmidt, P., Snowden, T., et al. (2001). Pepper

spray’s effects on a suspect’s ability to breathe (National Institute of Justice: Research in Brief).

Washington, DC: National Institute of Justice.

Chevigny, P. (1991). Edge of the knife. New York: Macmillan.

Combs, D. L., Parrish, R. G., & Ing, R. (1995). Death investigation in the United States and Canada,

1995. Atlanta: Centers for Disease Control and Prevention.

Dalton, V. (1998). Australian deaths in custody & custody-related police operations, 1997. Trends

and Issues in Crime and Criminal Justice, 80, 1–8.

Department of Justice. (2000). Deaths in custody. Washington: DC: DOJ.

Donahue, M. E., & Horvath, F. S. (1991). Police shooting outcomes: Suspect criminal history and

incident behaviors. American Journal of Police, 10(3), 17–34.

Geller, W. A., & Scott, M. (1992). Deadly force: What we know; a practitioner’s desk reference on police-

involved shootings. Washington, DC: Police Executive Research Forum (PERF).

Kappeler, V. E., Sluder, R. D., & Alpert, G. P. (1994). Forces of deviance: Understanding the dark side

of policing. Prospect Heights, IL: Waveland Press.

Krosch, C. (1992, August–October). Task force report: Some in-custody deaths cited as preventable.

Law Enforcement Quarterly, 15–18.

Leigh, A. (1999). Deaths in police custody. Focus on Police Research and Development, 11, 8–10.

Nolte, K. B., Yoon, S. S., & Pertowski, C. (2000). Medical examiners, coroners, and bioterrorism.

Emerging Infectious Diseases, 6(5), 559–560.

Reay, D. T. (1996). Suspect restraint and sudden death. Retrieved from http://www.fbi.gov/publications

/leb/1996/may966.txt

Skolnick, J. H., & Fyfe, J. J. (1993). Above the law: Police and the excessive use of force. New York: Free

Press.

Walker, S. (1999). The police in America: An introduction. Boston, MA: McGraw-Hill College.

Yin, R. K. (1994). Case study research: Design and methods. Thousand Oaks, CA: Sage.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

Criminal Justice Studies 77

Appendix: Deaths in Custody Question FormFigure A1 Deaths in Custody Question Form.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014

78 W. V. Pelfrey, Jr. & M. White Covington

Figure A1 Deaths in Custody Question Form.

Appendix: (continued)

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

12 2

5 N

ovem

ber

2014