Content uploaded by Sherry Hamby
Author content
All content in this area was uploaded by Sherry Hamby on May 01, 2017
Content may be subject to copyright.
EDITORIAL
On Defining Violence, and Why It Matters
Sherry Hamby
Life Paths Appalachian Research Center, Monteagle, Tennessee, and University of the South
Accurate definitions of phenomena are essential to any scientific enterprise. A definition of violence
should be fully capable of accounting for the exclusion of behaviors such as accidents and self-defense,
and the inclusion of behaviors such as child abuse, sexual offenses, and manslaughter. Violence research
has produced numerous and sometimes conflicting definitions of violence that can be organized into 4
general camps: the exemplars approach, the social psychology approach, the public health approach, and
the animal research approach. Each approach has strengths and limitations, but to fully distinguish
violence from other behaviors requires incorporating elements from all of them. A comprehensive
definition of violence includes 4 essential elements: behavior that is (a) intentional, (b) unwanted, (c)
nonessential, and (d) harmful. More sophisticated recognition of some elements is needed. For example,
shortened telomeres—a known consequence of child abuse—is a far more serious harm than a scratch or
bruise that will fully heal in a few days. Many problems in the field are due at least in part to insufficient
attention to definitions, such as minimization of sexual violence, bullying, and other behaviors that do not
map onto prototypical exemplars. More precise definitions of violence can improve surveillance, promote
more accurate identification of causes and consequences, enhance evaluation of treatment outcomes, and
guide development of prevention programs, among other benefits.
Keywords: violence, aggression, definitions, intent, injury
My friend’s 3-year-old daughter was a country girl. To her, dogs
meant shepherds and other large working dogs. One day, a visitor
brought a Chihuahua to her house, which soon started yapping.
Perplexed, the little girl asked her parents, “Why that cat barking?”
In developmental psychology, this is a classic assimilation error,
because of mistakenly incorporating size into her schema for
“dog” (Piaget & Cook, 1952). Definitions based on familiar ex-
emplars are just as problematic in science. For example, if you ask
a biologist to define “mammal,” one possible answer would be
“creatures such as dogs, cats, elephants, and lions.” However,
another biologist could offer a different, equally accurate response:
“Creatures such as whales, duckbilled platypuses, bats, and dogs.”
You may already be thinking, “Those are not very good biolo-
gists.” Indeed, both responses are very poor definitions of mam-
mals. The first repeats the assimilation errors of my little friend, by
implying that the definition of “mammal” might include charac-
teristics such as walking on four legs or living on land. The second
suggests that these two characteristics are not part of the definition,
but it is still unclear about the essential elements. Scientific defi-
nitions cannot rely only on examples, and especially not on pro-
totypical exemplars. Scientific definitions need to be precise
enough to distinguish one thing from another and to provide an
underlying rationale for the resulting classifications. For example,
biologists need to justify including whale and bat in the “mammal”
category, while excluding seahorse and ostrich. Biologists accom-
plish this by pointing to features that all mammals share, such as
being warm-blooded and producing milk for nourishing their
young.
The same scientific standards should apply to terms such as
“violence,” “aggression,” and “abuse” (Follingstad, 2017). Regret-
tably, it is seldom clear how acts called “violence” are like each
other, or distinct from other phenomena. Many operational defi-
nitions cannot distinguish violent behaviors from quite different
acts, such as accidents and horseplay (Hamby, 2016a,2016b;
Lehrner & Allen, 2014). Precise definitions are essential for any
scientific enterprise. Accurate definitions are needed for surveil-
lance, identification of causes and consequences, providing appro-
priate prevention and intervention, and conducting outcome stud-
ies.
Furthermore, many controversies in the field are driven in
whole or part by arguments related to definitions, such as
whether violent video games promote violence or aggression,
whether marital rape is as severe as stranger-perpetrated rape
(or even if such a concept as “marital rape” exists), the true
gender patterns of intimate partner violence (IPV), whether
microaggressions really count, and what kinds of peer victim-
ization are most serious (e.g., Calvert et al., 2017;Hamby,
2016b;Sue, 2017;Turner, Finkelhor, Shattuck, Hamby, &
Mitchell, 2015;Yllö & Torres, 2016). Our imprecision contrib-
Sherry Hamby, Life Paths Appalachian Research Center, Monteagle,
Tennessee, and Department of Psychology, University of the South.
My thanks to the following people for their comments on a draft of this
article: Victoria Banyard, Heather McCauley, David Sugarman, Zach
Blount, Annya Shalun, Alli Smith, and Elizabeth Taylor.
Correspondence concerning this article should be addressed to Sherry
Hamby, Life Paths Appalachian Research Center, 1016 West Main Street,
Monteagle, TN 37356. E-mail: sherry.hamby@lifepathsresearch.org
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychology of Violence © 2017 American Psychological Association
2017, Vol. 7, No. 2, 167–180 2152-0828/17/$12.00 http://dx.doi.org/10.1037/vio0000117
167
utes to major miscarriages of justice, such as when rape by an
acquaintance is not taken seriously by college campuses or law
enforcement, IPV victims experience “dual arrest” when they
seek help, or schools fail to protect children.
A precise definition of violence requires four elements. Vio-
lence is behavior that is (a) intentional, (b) unwanted, (c) nones-
sential, and (d) harmful. All four elements are necessary to prop-
erly include all acts that belong in the category and to properly
exclude similar acts that are not violence, such as self-defense (a
form of aggression but not a form of violence), accidents, and
horseplay. I elaborate on the rationale for each of the four defining
elements and why other elements are not needed, in much the same
way that “has four legs” is not relevant to the definition of
mammal. I also address some complexities in these distinctions,
including the need for more sophisticated recognition of harms not
visible to the naked eye. Surprisingly, although there are many
definitions of violence (even 10 years ago, Parrott and Giancola,
2007, estimated the number at over 200), there appear to be few
detailed rationales for the essential elements of a complete defini-
tion. This article presents such a rationale and describes some of
the advantages of a more precise definition of violence.
Common Approaches to Defining Violence and
Their Limitations
There are four common approaches to defining violence in the
scientific literature: the exemplar approach, the social psychology
approach, the public health approach, and the animal research
approach. Each has strengths and limitations; however, elements
of all are needed for a fully functional scientific definition, which
is provided in the section on essential elements of a violence
definition. As discussed in this section and the one on the essential
elements, this will also require some revisiting of the distinction
between aggression and violence.
Exemplar Approach
One common strategy is defining through exemplars. For ex-
ample, the Violence page on the American Psychological Associ-
ation (APA) website simply says, “Violence is an extreme form of
aggression, such as assault, rape or murder” (American Psycho-
logical Association, n.d.). This definition does not say what is
meant by extreme. The maliciousness of the intent? The extent of
injuries? If it includes murder, does it include or exclude man-
slaughter? Similarly, a recent review in a prominent medical
journal offered a confusing mix of exemplars: “The terms rape,
sexual violence, and sexual abuse encompass many forms of
violence, including sexual harassment and sexual trafficking”
(Abrahams et al., 2014, p. 1648). One cannot tell from this defi-
nition whether sexual harassment always falls under the definition
of rape or whether rape, sexual violence, and sexual abuse are
synonyms (correct answer in both cases: No).
Criminal law also often relies on an exemplar approach, by
specifying which crimes are considered violent, rather than what
elements these crimes have in common. For example, the National
Criminal Victimization Survey defines nonfatal violent crime as
“rape or sexual assault, robbery, aggravated assault, and simple
assault” (Truman & Morgan, 2016, p. 1). Like all exemplar ap-
proaches, this leaves many questions unanswered. For example,
why is robbery included in violent crime and not property crime?
(Robbery, unlike other forms of theft, requires a direct confronta-
tion with the victim, so there is a reason, but it is not often
specified in the literature.)
The main strength of the exemplars approach is that it suggests
a range of acts that should be considered violence, and by infer-
ence might suggest some shared characteristics. However, in gen-
eral, it is the weakest of the approaches and incapable of fully
defining the boundaries around any phenomenon.
Social Psychology Approach
A second general class of definitions is one common in social
psychology and other fields that tend to use the term “aggression”
more than “violence” (Parrott & Giancola, 2007). Although it
improves on the exemplar approach, the social psychology ap-
proach also has problematic elements. A typical version of this
approach and a common distinction between aggression and vio-
lence is offered by DeWall and colleagues. They define aggression
“as any behavior intended to harm another person who does not
want to be harmed,” and they define violence “as any aggressive
act that has as its goal extreme physical harm, such as injury or
death” (DeWall, Anderson, & Bushman, 2011, p. 246). This def-
inition was meant to improve on earlier efforts, such as Buss’s
influential definition, “a response that delivers noxious stimuli to
another organism” (Buss, 1961, p. 1). It was soon recognized that
Buss’s definition was insufficient, because it could not exclude
accidents or prosocial acts, such as the dentist causing mild pain to
promote better long-term health (Parrott & Giancola, 2007).
The strengths of the social psychology approach are that it
presents some version of three of four key elements: intent, un-
wantedness, and harm. An important limitation of the social psy-
chological approach is that it requires malicious intent—the intent
must be to harm, not simply to act. This would omit many offenses
that are categorized as violent crimes in most U.S. jurisdictions. Is
a bank robber who “accidentally” shoots a hostage guilty of
violence? Under the law, yes, but if they claimed lack of intent to
harm (just intent to steal), they might not be guilty of violence in
the DeWall et al. definition. This approach also does not distin-
guish self-defense from violence.
Further, the social psychology approach has limitations regard-
ing the harm element. Specifically, it (a) draws the line between
harm and extreme harm only by exemplars, leaving it undefined,
(b) appears to be limited to physical harm, and (c) suggests a focus
on immediate consequences. However, many of the most severe
harms caused by violence are long-term harms, such as the long-
term physical and mental health effects of child maltreatment,
which are far more damaging and costly— both to individuals and
to society—than a minor cut or bruise that will heal in a few days
(Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002). Many serious
harms are not visible in the same way that a scratch is visible,
including dysregulated stress responses, shortened telomeres, sui-
cidality, and chronic diseases (Felitti et al., 1998;Ouellet-Morin et
al., 2013;Shalev et al., 2013;Turner, Finkelhor, Shattuck, &
Hamby, 2012). The profound personal violation of forced pene-
tration is not well captured by examples that refer to “injury and
death,” and may inadvertently suggest such harms do not count as
violence.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
168 HAMBY
Finally, although definitions with these features are widespread
in aggression research with humans (Parrott & Giancola, 2007),
the social psychological approach is also inconsistent with com-
mon definitions of aggression and violence used in animal research
(Natarajan & Caramaschi, 2010). It should be clear how defini-
tions of violence used in human psychological and social science
research map onto definitions in criminal justice, animal research,
and other related fields (Hamby & Finkelhor, 2000). As discussed
below, the animal research field offers a more powerful approach
to distinguishing between aggression and violence.
Public Health Approach
The World Health Organization (WHO) has offered one of the
stronger existing definitions of violence that is an influential
example of the public health approach. Their definition is: “The
intentional use of physical force or power, threatened or actual,
against oneself, another person, or against a group or community,
that either results in or has a high likelihood of resulting in injury,
death, psychological harm, maldevelopment or deprivation” (Krug
et al., 2002, p. 5). As noted in the authors’ commentary on this
definition, this definition reflects the need “to include violence that
does not necessarily result in injury or death, but that nonetheless
poses a substantial burden on individuals, families, communities
and health care systems worldwide” (p. 5). This is perhaps the
most important strength of the definition, because it clearly estab-
lishes harmful health impact as a key defining feature without
regard to privileging one kind of health impact (immediate phys-
ical injuries) over others. The WHO definition also avoids exem-
plars and recognizes the probabilistic nature of the connection
between the act and the resulting harm.
It is important that the WHO definition also refers to intentional
acts that result in harm, and they clarify in the text that excludes
accidents (tripping and falling onto someone is not violence), but
includes reckless or negligent acts (such as shaking a baby, in their
example), regardless of whether harming the baby was specifically
intended. This is analogous to providing a definition of mammal
that captures “whale” and other animals that do not look like the
exemplars of the mammal category. They directly address some
types of incidents that present definitional challenges, and specify
their position.
However, the WHO definition is missing the elements of being
unwanted and nonessential. Also, although it is a strength of the
definition that it explicitly specifies that the use of physical force
is not the only way to commit a violent act, it is problematic that
they suggest that only the use of physical force or power (or their
threat) can lead to violence. Their commentary is more useful:
“The inclusion of the word “power,” in addition to the phrase “use
of physical force,” broadens the nature of a violent act and expands
the conventional understanding of violence to include those acts
that result from a power relationship, including threats and intim-
idation. The “use of power” also serves to include neglect or acts
of omission, in addition to the more obvious violent acts of
commission” (p. 5). This commentary is broad enough to include
some of the key offenses that should be deemed violent, such as
child neglect, acts of depraved indifference, and other intentional,
unwanted harms that often described as acts of omission (this issue
is revisited in more detail below). However, it is not clear whether
negligence is necessarily connected to a “power relationship.”
The Centers for Disease Control and Prevention (CDC) ap-
proach also falls within the public health realm. Surprisingly, the
CDC does not seem to offer an overarching definition of violence,
but they provide definitions for many specific types and subtypes
of violence that go beyond a mere list of exemplars (see https://
www.cdc.gov/violenceprevention/). For example, their definition
of child maltreatment reads “Child abuse and neglect is any act or
series of acts of commission or omission by a parent or other
caregiver (e.g., clergy, coach, teacher) that results in harm, poten-
tial for harm, or threat of harm to a child” (Leeb, Paulozzi,
Melanson, Simon, & Arias, 2008, p. 11). Some of the strengths are
the inclusions of so-called acts of omission, the emphasis on harm,
and the avoidance of defining-through-exemplars. However, al-
though elsewhere on the CDC website, intent is clearer (e.g., under
intimate partner violence), it is not specified here, and the un-
wanted and nonessential elements are missing.
Perhaps most notably, this definition also points to one of the
main problems in this area—that other terms, such as “abuse,” are
often used when violence is what is being described. This contrib-
utes to confusion, such as that seen in Abrahams et al., about
whether violence and abuse are synonyms or not. A better option
would be a term such as “caregiver violence,” which specifies that
it refers to violence by caregivers, or at least clarify that “mal-
treatment” means violence by caregivers, instead of substituting
multiple alternative words (maltreatment, abuse), without a clear
indication of what the substitution means.
Animal Research Approach
Animal researchers focus on aggression as an adaptive form of
social communication, which is aimed at functional endpoints such
as the acquisition of food, shelter, mates, and status. Intraspecies
aggression rarely causes significant harm, because it is typically
highly regulated and controlled by a range of inhibitory mecha-
nisms (Koolhaas, de Boer, & Buwalda, 2010). In contrast, violence
is defined as “a pathological form of aggressive behavior that is
not subjected to inhibitory control mechanisms and that has lost its
function in social communication” (Koolhaas et al., 2010, p 387).
Animal violence is often distinguished from animal aggression by
characteristics such as the following: (a) minimal preescalatory/
ritualistic behaviors; (b) longer attacks with low withdrawal rates,
even in the presence of opponent’s submissive or defeat postures;
(c) context-independent indiscriminate attacks; and (d) attack-bites
toward vulnerable body parts of the opponent resulting in severe
injury (Natarajan & Caramaschi, 2010). The strengths of the
animal research approach are the avoidance of exemplars and, like
the public health approach, the delineation of clear markers for
violent behavior. Some of the elements that are common in human
research can be seen here in some form, including a focus on
behavior that is intentional, nonessential, and harmful.
On first consideration, there may seem to be parallels between
“pathological” aggression and the idea of “extreme” harm in some
definitions of human violence. However, there are key differences.
Only the fourth of the Natarajan and Caramaschi criteria specifi-
cally refers to harm; the rest focus on the perpetrator’s behavior.
This is the sole approach that clearly emphasizes the nonessential
nature of violent acts, for example, an animal that continues to bite
even after an opponent has submitted. The first criterion also
embodies this idea because it is preferable (less risky) to use threat
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
169
ON DEFINING VIOLENCE, AND WHY IT MATTERS
displays (growls, beating on chest, spreading wings) instead of
aggression, when threat displays can accomplish the same goal.
This clear rationale for the operationalization of the distinction
between aggression and violence could benefit scientific research
with humans as well (Natarajan & Caramaschi, 2010).
The Four Essential Elements of a Definition
of Violence
The strengths of these four approaches can be incorporated into
a single definition that successfully distinguishes violence from
other actions. Many bad things happen that are not violence. Even
a great many acts that involve some level of physical force are not
violence. Fully distinguishing violence from other sorts of behav-
ior, including nonviolent aggression, accidents, horseplay and even
sadomasochism, requires four definitional components: nonessen-
tial, unwanted, harmful, intentional acts. Although there are, inev-
itably, fuzzy boundaries around these concepts as well (all scientific
concepts have fuzzy boundaries; Zwick, Carlstein, & Budescu, 1987),
the following focuses on distinctions that are most useful for advanc-
ing the science of violence. Table 1 provides examples of the way
these criteria can be applied to a range of behaviors. The goal, of
course, is not to define by exemplar, but rather to provide illustrations
of the application of these criteria to behaviors that fall within and
outside the construct of violence.
Nonessential Behavior: Revisiting the Distinction
Between Aggression and Violence
The distinction in animal research between aggression and vi-
olence can inform a more precise definition for the scientific study
of human violence. One element missing from most definitions of
human violence is its nonessential or gratuitous nature. As noted
previously, the term “maladaptive” is widely used in animal re-
search to distinguish violence from the broader concept of aggres-
sion (Koolhaas et al., 2010;Natarajan & Caramaschi, 2010).
Indiscriminate or unnecessarily harsh attacks seldom serve an
animal community or even a higher-ranking individual. For exam-
ple, attacking when a threat display (such as a growl) would be
effective is nonessential and risky, and hence maladaptive. Apply-
ing this criterion, most of what is studied as “aggression” with
human participants would qualify as violence. From bullying to
child abuse to intimate partner violence, most intentional, harmful,
unwanted behavior that is perpetrated by humans is not necessary
for survival or fitness (indeed, it often harms the fitness of both
perpetrators and victims). It does not serve a legitimate function,
or, perhaps more importantly for understanding human violence,
the behavior does not serve a legitimate function that could not
also be obtained by nonviolent means. Human behavioral reper-
toires (physical and verbal) are far broader than other animals. In
other words, such behaviors are “nonessential” for survival or
fitness.
For example, bullying might be said to have an “adaptive”
function, in that it is often perpetrated with the aim of improving
social status (like some animal aggression). Bullying can be a
successful means of attaining social status (Faris & Felmlee,
2011). However, it is well within the behavioral repertoire of even
school-age children to seek social status without resorting to
bullying, and not all high-ranking children bully. We now know,
contrary to some formerly widespread beliefs, that bullying is
extremely harmful (e.g., Lereya, Copeland, Costello, & Wolke,
2015). Bullying is intentional, unwanted, nonessential behavior
that results in harm. Bullying is violence.
Similarly, it is not necessary to use physical chastisement to
raise well-behaved children, and even legal (in the United States)
spanking has well-documented long-term harms (Gershoff &
Bitensky, 2007). Spanking and other physical assaults by caregiv-
ers on children are properly considered to be violence. Intimate
partner violence is a nonessential method of conflict resolution.
Even if the underlying goal is adaptive, humans are capable of
nonviolent means of achieving those goals. “Nonessential” also
helps avoid the pitfalls of labeling some horrific forms of human
behavior as “adaptive” (for good critiques of these pitfalls, see
Lloyd, 2003;Travis, 2003).
A good definition is clear about what is excluded as well as what
is included. Applying the criterion of nonessential provides more
insight into acts that are appropriately considered human aggression,
but not violence. For example, some humans, including infants, tod-
dlers, older adults with dementia, and those with significant cognitive
impairments, have a limited behavioral repertoire and may not be
capable of nonaggressive responses to some situations. Toddler tan-
trums, even when harmful, are aggression, not violence.
Essential behaviors would include self-defense and defense of
one’s children, which are appropriately considered aggression, but
not violence. As illustrated by the example of self-defense, one
advantage of incorporating the nonessential element is closer cor-
respondence to legal definitions of violence. Furthermore, this
criterion can help exclude a range of prosocial actions. A doctor
intentionally treating a sick child with chemotherapy is not engag-
ing in violence, even if the child does not “want” the treatment and
some harm is caused, albeit less harm than would be caused by
failing to treat the child. See Table 1 for the application of the
nonessential criterion to these and other behaviors.
In the cases of both self-defense and medical care, the question
of proportion has long been important. In traditional legal schol-
arship, acts of self-defense must meet three criteria that all pertain
to the essential nature of the force: the threat must be immediate,
there must not be an opportunity to withdraw, and the force must
be proportional to the threat (Hamby, 2014;Kopel, Gallant, &
Eisen, 2008). For example, you should not kill an assailant unless
the threat to you is potentially lethal. Although there are exceptions
to these criteria (the “Castle doctrine” defense and many versions
of “stand your ground” laws do not require retreating from your
own property), they all help justify the essential nature of aggres-
sion in the name of self-defense (for a more detailed discussion of
self-defense, see Hamby, 2014;Kopel et al., 2008). A similar
balance of judgments is made in medical care decision making.
The risk of harm from the treatment or secondary complications
must be deemed to be less than the risk from an untreated illness
or injury.
Unwanted Behavior
It may seem obvious that violence is unwanted, but this turns out
to be an important element for distinguishing violence from other
innocuous or even prosocial behavior. There are physically force-
ful and even injurious acts that are not unwanted (Parrott &
Giancola, 2007). Many medical interventions, ranging from sur-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
170 HAMBY
Table 1
Applying the Four Elements of the Violence Definition to Acts Inside and Outside the Boundaries of the Violence Construct
Act Violence? Nonessential? Intentional? Unwanted? Harmful? Comments
Acts that are not violence because they are essential
Self-defense Aggression, not
violence
Essential Intentional Unwanted Harmful In classic legal formulation,
self-defense must meet 3
criteria: (1) imminent
threat; (2) no opportunity
to withdraw;)3)
proportionate response to
threat (e.g., lethal force
only acceptable in response
to potentially lethal threat).
Toddler tantrums Aggression, not
violence
Essential Intentional Unwanted Occasionally
harmful
Toddlers do not have
sufficient cognitive or
behavioral capacity to
choose less aggressive
responses to frustration
Medical care of
children
Prosocial Essential Intentional Often
unwanted
Sometimes harmful Even risky procedures with
adverse side effects are
sometimes deemed the
safest and healthiest
alternative
Acts that are not violence because they are wanted (consensual)
Horseplay, joking
around, pillow
fights
Prosocial Nonessential Intentional Wanted Rarely harmful Usually carefully calibrated
to not cause any harm
Stunt fights and
other acting
Prosocial Nonessential Intentional Wanted Rarely harmful Usually significant
investment made to avoid
harm. If harm, often the
result of recklessness or
negligence
Contact sports Prosocial Nonessential Intentional Wanted Sometimes harmful,
but designed to
minimize harm
Multiple safety mechanisms
in place. Definitional
boundary issues related to
intentional fouls and other
instances of “lack of
sportsmanship.” Macho
culture and high salaries
can create atmospheres of
coercion and minimization
of true harms
Assisted suicide Debated Nonessential Intentional Wanted Harmful, although
also often relief
from pain and
suffering
Exemplars are clear cases of
desire after a careful
balancing of judgement by
terminally ill. However,
questions of coercion,
cognitive capacity, and
mental illness complicate
many individual cases
(especially given financial
gain or relief of caregiving
for family members).
Sadomasochism,
including
bondage
Not usually considered
violence, even
though violates
some social norms
Nonessential Intentional Wanted Sometimes,
although most
often minor
Safety steps often in place.
Hard to show true lack of
coercion in many instances
and “submissive” party
vulnerable to having “safe
word” ignored.
Acts that are not violence because of lack of harm
Daily hassles, such
as traffic tickets
Nonviolence Nonessential Intentional Unwanted Not harmful Harm standard requires a
lower bound to eliminate
transient pain or distress.
(table continues)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
171
ON DEFINING VIOLENCE, AND WHY IT MATTERS
Table 1 (continued)
Act Violence? Nonessential? Intentional? Unwanted? Harmful? Comments
Interpersonal
rejection
Nonviolence Nonessential Intentional Unwanted Not harmful Harm must be probabilistic
and usually based on
“reasonable person”
standard. Someone might
claim harm to getting
turned down for a date, but
that does not make it
violence.
Interpersonal
conflict
Nonviolence Nonessential Intentional Unwanted Not harmful Disagreements are not
violence.
Acts that are not violence because of lack of intent
Accidents Nonviolence Nonessential Unintentional Unwanted Sometimes Accidents are generally
unpredictable, unlikely
mishaps that could not
have been easily foreseen
or avoided
Negligence Nonviolence Nonessential Unintentional Unwanted Sometimes Negligence can rise to the
level of depraved
indifference and if
sufficiently reckless, can
rise to the level of
violence. However, many
errors, even if careless and
harmful, are not considered
violent or criminal
Acts that meet criteria for violence
Aggravated assault,
robbery, murder
Violence Nonessential Intentional, usually with
specific intent to
harm
Unwanted Harmful These are the most
prototypical exemplars of
violence
“Simple” assault Violence Nonessential Intentional, but not
always intent to cause
harm
Unwanted Less severe injuries,
psychological
harm, or high
potential for
harm
Surprisingly, one of the most
prototypical exemplars for
violence and a crime that
is classified as violent by
the FBI can be borderline
cases on these criteria.
Assaults that produce
injury will often be
charged as aggravated
assault and some simple
assaults do not lead to any
observable physical injury
and the lasting
psychological harm of
single simple assaults is
not well-established.
Manslaughter Violence Nonessential Intentional, but intention
is not to kill, but
rather the intent was
to commit an act so
reckless or dangerous
that the fatal outcome
could have been
foreseen
Unwanted Harmful (fatal) There are many forms of
manslaughter that all
involve the unlawful
killing of another.
Examples include the
consequence of another
felony such as arson (in
some jurisdictions called
felony murder), crimes of
passion, and death due to
depraved indifference.
Constructive and voluntary
manslaughter are always
considered violent crimes;
sometimes involuntary or
negligent manslaughter is
not.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
172 HAMBY
gery to chemotherapy, are wanted. In contact sports, individuals
willingly subject themselves to physically forceful and potentially
harmful acts by others. More exotically, perhaps, but just as
relevant, there is sadomasochism and other sexual behaviors that
include elements of intent and harm, even some extreme enough
they have led to accidental death. In these examples, the perceived
benefits of the acts are deemed to outweigh the harms or risks of
harm (albeit rather idiosyncratically in some cases), and consent to
participate is given by all actors.
One important but underappreciated type of wanted and con-
sensual behavior that should also be excluded is horseplay or
joking around. Horseplay, such as faux “wrestling,” is common
among youth. Physical horseplay serves many adaptive social
functions, especially during adolescence and young adulthood
(Maccoby, 1998). It allows youth to explore physical intimacy
outside the family under the cover of “nothing serious” and “just
kidding.” It allows them to initiate nonsexual touch with prospec-
tive romantic partners, and use the response as a guide to whether
further advances might be welcome. Many couples engage in this
kind of horseplay too (Follingstad, 2017). This kind of physical
play is also seen in the animal kingdom and is not considered to be
a form of violence but rather, as it is in humans, a type of practice
in needed social skills. It is also important to note that all of these
examples of excluded acts have inhibitory mechanisms in place to
minimize risk.
As with other definitional elements, one can describe a contin-
uum of wantedness (Hamby & Koss, 2003). Sometimes people get
talked into participating in behaviors that they may have legally
consented to, but are reluctant participants. In most cases, that
would not meet the unwantedness criterion for violence. There are
many innocuous examples, such as getting talked into helping a
friend move. Constructs such as statutory rape point to develop-
mental complexities, when a youth may believe they “want” a
sexual encounter, but society has deemed that they are too young
to give informed consent. In most legal jurisdictions, threats of
force represent unlawful coercion, but acceding to something
based on threats to end a relationship would not. The latter would
be considered a balance of judgment. In social psychology, re-
searchers refer to noxious stimuli that one would otherwise be
motivated to avoid (Parrott & Giancola, 2007), and that is a helpful
threshold for this boundary that can be applied across many social
and cultural settings.
One might ask if both “unwanted” and “nonessential” are
needed, especially because unwanted is mentioned in many defi-
nitions (e.g., DeWall et al., 2011;Parrott & Giancola, 2007) and
nonessential seldom is. They are both necessary because some
Table 1 (continued)
Act Violence? Nonessential? Intentional? Unwanted? Harmful? Comments
Rape and other
sexual offenses
Violence Nonessential Intentional Unwanted Harmful Another violent act that is
intentional, but the intent
may be to coerce, exploit,
or even impress one’s
peers, and may not
explicitly intend to harm.
Some cases have elements
of depraved indifference
Bullying, including
relational
aggression
Violence Nonessential Intentional Unwanted Harmful Formerly minimized, but now
recognized as a source of
serious, lasting
psychological harm.
Also note that a lot of
bullying involves “malice
aforethought”: It is not just
blurting out an angry
response, it is a
coordinated, planned
attempt to harm someone
else. For example, bullying
often involves lying in
wait, or coordinated
attempts to spread rumors
or exclude a peer.
Bullying peaks during the
middle school years, when
the cognitive skills to plan
such attacks are first
attained.
Child neglect Violence Nonessential Intentionally reckless or
depraved indifference
Unwanted Harmful Some offenses that are
commonly treated as “acts
of omission” could more
accurately be considered
reckless endangerment.
Note. Rather than simply offering a list of exemplars without specifying why they should all be considered violence, a scientific definition names the
essential features of violence and then systematically applies these criteria to distinguish violence from other acts. These criteria are further elaborated in
the text.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
173
ON DEFINING VIOLENCE, AND WHY IT MATTERS
nonessential acts are still wanted and should not be considered
violence (dental care, contact sports, S&M). Likewise, some es-
sential acts are unwanted by the recipient (self-defense, lifesaving
medical care of children). Many common everyday behaviors
cannot be accurately categorized without reference to both con-
structs.
Harmful Outcomes
Harm is defined quite differently in the social psychology and
public health approaches. The public health approach, as exempli-
fied by the WHO definition (Krug et al., 2002), has more scientific
justification. Harm is defined in that definition as: “results in or has
a high likelihood of resulting in injury, death, psychological harm,
maldevelopment or deprivation.” Social psychology definitions
tend to refer just to “extreme harm.” Definitions that suggest that
harm requires visible physical injury or death are defining harm
too narrowly. Any act that is still having significant, measurable
adverse health impacts a half century after its occurrence—such as
caregiver violence toward children—should qualify as “extreme”
harm. It is borderline irrational to consider a scratch or a bruise
evidence of sufficient harm to meet the definition of violence,
while less visible but more impactful harms, such as posttraumatic
stress, suicidality, shortened telomeres, dysregulated stress re-
sponses, and chronic diseases are omitted from the definition
(Felitti et al., 1998;Ouellet-Morin et al., 2013;Shalev et al., 2013;
Turner et al., 2012). Focusing on typical exemplars has proved
particularly problematic here and has led to extensive scientific
and public neglect of some of the most harmful forms of violence,
such as child neglect and rape.
As with other elements of the violence definition, it is important
to determine a threshold for what constitutes notable harm. Tran-
sient moments of physical or emotional pain are not usually
considered to meet the harm threshold. In the Juvenile Victimiza-
tion Questionnaire, our lower bound threshold for physical injury
is pain that can still be felt a day later (Hamby, Finkelhor, &
Turner, 2013). For psychological disorders, the Diagnostic and
Statistical Manual typically sets a minimum time period of dis-
tress, such as one month for posttraumatic stress disorder (PTSD)
symptoms, and also a minimum degree of impact, such as signif-
icant distress or functional impairment (American Psychiatric As-
sociation, 2013). These obviously require a degree of judgment,
but the goal is to focus on harms that have some lasting impact.
They need not be “extreme” to have lasting impact, and that
undefined term is not a good substitute for these more specific
operationalizations.
It is also important to recognize the probabilistic nature of the
link between an act and the resulting harm. A single harmful
outcome is not sufficient to redefine a whole class of behaviors as
violence. For example, some obsessive or delusional individuals
may experience a romantic rejection as harmful, and may even
develop suicidality or some other highly negative outcome. How-
ever, that does not mean that everyone who turns someone down
for a date is committing an act of violence. On the other side, a
common example is shooting and missing. Someone who attempts
to shoot and kill someone, but misses, has perpetrated an act of
violence (and the felony of attempted murder). That is the source
of the phrase “high likelihood of resulting” in the WHO definition.
Focusing on health impacts (both physical and mental), and not
merely transitory emotional distress or momentary physical pain,
is also helpful for distinguishing harm that merits the label of
violence from other unpleasant events.
Harm is insufficient as a criterion in and of itself, because
bruises can be inflicted and bones can be broken in a wide array of
activities that are not violence. The element of harm is necessary
because there are many unpleasant events that are nonessential,
unwanted, and intentional that do not meet criteria for violence,
because of the lack of harmful health impacts. All sorts of daily
hassles, interpersonal conflicts, and annoying interactions fall into
this category, ranging from junk mail to ending romantic relation-
ships (Follingstad, 2017). Over the years, some behaviors have
been reconsidered, such as spanking. For centuries, spanking was
widely considered harmless or even helpful for child rearing, but
is now known to cause many serious, long-term adverse impacts
(Gershoff & Bitensky, 2007). New data on long-term brain dam-
age from football (Gavett, Stern, & McKee, 2011) may be leading
to a redefinition of acceptable force in contact sports.
Intent
Scientists have also put forth different definitions of intent. In
the social psychology approach, most definitions specify “intent to
harm” to omit accidents or prosocial acts such as medical care. The
public health approach also omits accidents and many prosocial
acts of force from the category of violence, but also specifies that
intentional behaviors that result in harm can also be violence. Most
difficult discriminations concern the question of whether reckless
or negligent behavior should also count, as promoted by the public
health approach. Although few definitions specify it, there are also
cognitive preconditions for meeting the criterion of intent. Small
children and some with serious cognitive impairment are generally
incapable of the capacity to understand and plan their behavior.
This is embodied in many laws that minimize liability for young
and cognitively impaired actors.
Recognizing that many aspects of intent are continuous, it can
still be helpful to focus on three general ranges of intent. At one
extreme of the intent continuum, many acts of violence involve
clearly malicious intent (Finkelhor, 2008). Everyone agrees that
acts that are intended to cause harm meet the intent criterion for
violence (American Psychological Association, n.d.;Cook & Par-
rott, 2009;DeWall et al., 2011;Krug et al., 2002;Leeb et al., 2008;
Parrott & Giancola, 2007). This includes actions such as premed-
itated murder (“with malice aforethought”), which is a more seri-
ous felony than various forms of manslaughter (described in more
detail below). Most prototypical exemplars of violence involve
malicious intent—the barroom brawl, the drunken batterer. Few
people “accidentally” throw a punch, or, for that matter, “acciden-
tally” stalk, bully, or harass.
Motive is different than the intent to harm. The intent to harm
usually arises from a motive to attain status or resources, and a
willingness to pursue those goals illegitimately. For example,
(borrowing prototypical exemplars from fiction plots), the motive
might be to seize an inheritance prematurely, or revenge. Per-
ceived slights to one’s honor are common motives for barroom
brawls (Graham et al., 2013). The goals for social status or
resources could be pursued through nonviolent means, and many
motives are similar to those shared by almost everyone. Some
motives are illegitimate or delusional, such as having total control
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
174 HAMBY
over one’s spouse, or attempting to enact a fantasy about a rela-
tionship with a celebrity. Pursuing illegitimate or delusional goals
will often lead to violence, but the motives could remain in the
fantasy realm, and hence never be connected to violence.
At the other extreme of the intent continuum, many acts that
lead to harmful outcomes are unintentional. Someone who trips
and falls onto someone else is not intentionally falling. If a child
darts into the street at the last second and a careful driver cannot
avoid her, that is an accident. It is not violence and it is not a crime.
The prototypical exemplars of incidents in this range of intention-
ality are unavoidable results of unlikely mishaps that are clearly
not the fault of the person causing the injury. Other circumstances
where the intent criterion is not met include incidents such as
injuries from contact sports, if the players were abiding by the
rules at the time of the incident. Pillow fights and horseplay are
other forms of prosocial, positively intended behaviors. Part of the
positive intent in sports and horseplay includes considerable effort
to calibrate those interactions to minimize the risk of harm. Again,
these are inevitably probabilistic judgments. No behavior is guar-
anteed to be safe, and some of these behaviors, such as contact
sports, carry nontrivial risks of harm, but they are not reckless.
Scholars do not treat these sorts of incidents as meeting the intent
criterion for violence.
However, these clear-cut examples do not come close to ex-
hausting the range of possibilities regarding intent. Especially,
there are a variety of reckless and negligent behaviors. The public
health and legal worlds have addressed these nuances more than
the fields of social psychology or animal research, and in general,
these scholars make the case that many forms of reckless or
negligent behaviors are also properly considered forms of vio-
lence. Criminal law has developed a range of offenses to handle
nuances which are not always well recognized in academic vio-
lence research, with particular attention to acts that lead to an
unlawful killing.
The crime of manslaughter lacks the specific intent to kill, but
involves behavior so dangerous or reckless that a reasonable
person could have foreseen the fatal outcome, and many forms are
treated as a violent crime (FBI, 2011). Manslaughter helps illus-
trate that violence cannot be defined solely by a specific malicious
intent to enact the obtained result. There are several types, and
while all the complexities cannot be summarized here, a few key
distinctions and categories should inform the scientific understand-
ing of violence. The most criminal of these types of incidents is
known as “constructive manslaughter” or “felony murder” (Ber-
man, 2013). In these incidents, the intentional misbehavior is a
felony, such as a burglary or arson, where a fatality occurs that was
not the original goal of the criminal. The robber cannot put a box
around some malicious intent—stealing, threatening with a
weapon, and hostage-taking—and claim blamelessness when these
dangerous behaviors lead to a fatality. In most jurisdictions, not
only the shooter, but anyone else on the robbery crew, would also
be guilty of constructive manslaughter or felony murder (Berman,
2013).
Voluntary manslaughter is often the charge in “crimes of pas-
sion.” Two examples (and plots of endless TV episodes) are a man
walking in on his wife with her lover, or a financially desperate son
getting cut out of a will, who then strike the lover or parent. There
is intent to assault, but not kill. The provocation and emotional
intensity of the situation are considered mitigating factors, but
these are still violent crimes. At a still lower level of culpability is
involuntary manslaughter, such as drunk driving that leads to a
fatal accident or a carnival worker who did not properly secure all
riders. In many jurisdictions, negligent manslaughter is a felony
but not a crime of violence (FBI, 2011), unless the degree of
recklessness or neglect reaches a level of depraved indifference.
Depraved indifference. Some criminal negligence can be
charged as a violent crime, if it is perceived that the degree of
indifference to others’ well-being is sufficiently extreme. Other
terms such as “callous disregard” are common in this body of law.
For example, if the drunk driver has had her license taken away for
previous convictions, that could add an element of criminal neg-
ligence that would elevate the charge to voluntary manslaughter or
perhaps even second-degree murder (Berman, 2013). A drive-by
shooting would also represent depraved indifference, even if the
explicit intent to kill is not present.
Child neglect. Public health experts have been keen to ad-
dress the role of negligence because of the problem of child
neglect. Child neglect is often described as an “act of omission,”
but many forms of child neglect might be better construed as acts
of recklessness. It is callous disregard or depraved indifference to
leave a toddler alone so the parent can go out drinking, or to spend
their entire income on Oxycontin and leave no food in the house.
Those are reckless choices, and they are also nonessential, un-
wanted, and harmful. As suggested in the public health model,
child neglect is properly treated as violence.
Sexual violence. Sexual violence is often treated as complex
with regard to intent—more complex than is probably warranted in
many cases— but nonetheless the pervasive misunderstanding of
sexual violence warrants some special consideration. Sexual vio-
lence is typically motivated by a desire for power and status,
including being perceived as sexually experienced and highly
sexually active. The “audience” for sexual violence is often not the
victim, but the perpetrator’s peers, because of offenders’ percep-
tions about peer pressure to have frequent sex, perceptions about
the sexual offending of their peers, and the culture of some male
peer groups (Hipp et al., 2017;McCauley et al., 2014;Thompson,
Kingree, Zinzow, & Swartout, 2015). Other sexual offenders have
delusional motives, such as beliefs that sexual violence is a path to
intimacy (Hartley, 2001). These numerous cognitive distortions
are used to justify intentional violent behavior. Rapists are not
“accidentally” penetrating their victims or “accidentally” not no-
ticing their victim is passed out or trying to get away. Like a
murderer trying to gain premature access to an inheritance, sexual
offenders are willing to use illegitimate, violent means to attain
common goals for power and status. Even when the intent to harm
is repressed or denied, sexual perpetrators demonstrate depraved
indifference to the foreseeable harms of their behavior. The failure
to understand how the elements of violence apply to sexual of-
fenses is one obstacle to justice for victims.
In summary intent cannot be limited to incidents when the
perpetrator admits to a desire to cause the specific resulting harm.
Intent to engage in reckless and dangerous behavior, where the
harmful outcome could have been foreseen if not explicitly in-
tended, has long been recognized to also form intent sufficient to
define an act as violent. You cannot shoot into a crowd without
the “intent” to kill, and be absolved of a violent crime when
someone dies. Many acts that are sometimes defined as acts of
omission are better conceived as acts of recklessness or depraved
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
175
ON DEFINING VIOLENCE, AND WHY IT MATTERS
indifference, including many neglectful parenting behaviors,
which are often defined as lack of supervision or lack of providing
basic necessities, but could be more accurately defined as choosing
to abandon one’s children.
Constructs That Do Not Belong in the Definition
of Violence
It is just as important to be able to identify extraneous constructs
for the definition of violence. Any element that is not needed to
fully distinguish violence from other behaviors should be ex-
cluded, even if it is true of most violent acts, just as land-based
does not belong in the definition of mammal. Some that have been
mentioned and deserve some discussion are violations of social
norms, anger and hostility, the form of the act, and personal and
social characteristics.
Violations of Social Norms
As Krug and colleagues (2002) have noted, some scholars have
proposed that violations of social norms are an important aspect of
understanding violence and victimization. A representative defini-
tion that refers to social norms is: “harm that comes to individuals
because other human actors have behaved in ways that violate
social norms” (Finkelhor, 2008, p. 23). However, social norms are
not a reliable guide to what is, and is not, violence, and can even
be used in attempts to excuse violence (Krug et al., 2002). Family
violence was tolerated for many years. Some forms of violence
against family members have even been explicitly excluded from
criminal law, such as the marital rape exemption that only fully
ended in the United States in 1994. Social norms are imperfect and
variable. Near the time of this writing, Russia voted to decrimi-
nalize some forms of domestic violence, but that does not mean
domestic violence is less harmful in Russia than it is elsewhere.
Instead, it is better to consider why such social norms exist and
what they represent. Many of those who point to social norms are
child abuse researchers, who are seeking some way to include
neglect and sexual molestation in definitions of victimization or
violence. The problem with caregiver neglect and child sexual
molestation is not that many people disapprove of them, the
problem is that they are developmental insults that overwhelm
children’s capacities and cause damage that frequently lasts until
old age. Another construct is not needed if harm is accurately
recognized as more than bruises or broken bones. Further, al-
though I recognize the many challenges in reaching— or even
wanting—a universal or pan-cultural definition, the reference to
observable health impacts is one way to emphasize the human
commonalities across sociocultural groups. Many so-called “cul-
tural” differences are mislabeled. People do not want to be vic-
timized. Like other justifications offered by perpetrators, many
claims about cultural endorsements of violence are false excuses
meant to protect illegitimate uses of power.
Anger and Hostility
As noted by Parrott and Giancola (2007), some definitions and
many measures confound aggression and/or violence with anger
and hostility. It is important to keep these distinct. Anger is a
common emotion that will be experienced by virtually all people,
and may be the most appropriate response to an injustice. It is the
task of self-regulation to keep us from inappropriately acting on
angry impulses. When we restrain ourselves, we have avoided
violence, no matter how angry we feel. At the other extreme, it has
often been noted in the domestic violence and sociopathy litera-
tures that anger is not a prerequisite for violence, either. The
classic distinction between instrumental and expressive aggression
also recognizes that emotions do not drive all violence. There is no
need to refer to the emotions of perpetrators in a definition of
violence.
The Form of the Act
Sometimes definitions of violence focus only on physically
forceful acts, and there has been some discussion about whether
psychological or verbal assaults should count as violence (Folling-
stad, 2017). Others have documented the wide range of actions that
can fall within the realm of violence and aggression, including not
only physical and verbal behaviors, but also “postural” threats,
vandalism, and other forms of assault (Cook & Parrott, 2009;
Parrott & Giancola, 2007). The public health approach also em-
phasizes that some acts of violence are not expressed as physical
assault. Although it is valuable to note the range of violent behav-
iors, it should be understood that acts of violence need not “look
like” prototypical exemplars. Furthermore, many behaviors that
“look like” physical assaults are not violence, including tackling in
football, jostling in a crowd, and performing stunts in a movie or
play. An excessive focus on the form of the act is one reason that
major types of violence such as rape and other sexual offenses go
underrecognized. Malicious actors can cause substantial, long-
lasting, even devastating harm without ever throwing a punch. We
need to move beyond stereotypes in our scientific approach to
violence. Just as mammals do not need four legs, violence does not
require punches and slaps.
Other Personal and Situational Characteristics
It is well established that many personal characteristics of the
victim or the perpetrator can affect perceptions of violent inci-
dents. These include sociodemographic characteristics such as
gender, race, and sexual orientation; behavioral characteristics
such as recent alcohol use; and situational and relational charac-
teristics, such as the familiarity of the victim and perpetrator (e.g.,
Hamby & Jackson, 2010;Sommer, Reynolds, & Kehn, 2016). In
some cases, people may be using personal or situational charac-
teristics as imprecise proxies for other elements of the construct of
violence. For example, it is well-established that male-perpetrated
violence is more injurious than female-perpetrated violence. As a
result, in the absence of precise information on harm, people
appear to use perpetrator gender to estimate the likely harm of an
incident (Hamby & Jackson, 2010). Understanding the impact of
these factors is important for understanding victim blame, help-
seeking, health care and criminal justice responses, and a host of
other important research topics. However, despite the correlation
of gender and other characteristics with elements of the violence
definition (such as harm), they are conceptually orthogonal and
should be kept separate.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
176 HAMBY
Implications
The scientific definition of violence matters because it is the
bedrock on which all nomenclature, assessment, theory, and inter-
vention should rest. Some key implications are described below,
with highlights summarized in Table 2.
Implications for Terminology
Given the foregoing, many terms can be more precisely defined
and standardized for research with humans. Violence is nonessen-
tial, unwanted, intentional, and harmful behavior. Violence is a
subset of aggressive behavior that is most usefully distinguished
from aggression by its nonessential nature, following the animal
research model. Aggression includes self-defense, toddler tan-
trums, and other unwanted, intentional, harmful behaviors that are
either necessary to protect oneself or loved ones, or occur because
of the limited behavioral repertoire of the perpetrator. The concept
of “extreme” regarding harm or maliciousness of intent does not
add to the utility of the definition of violence and can be dropped.
Instead, a more sophisticated recognition of the true health impacts
and the full range of intents are needed. Terms such as “abuse,”
“maltreatment,” or “battering” have never been systematically
defined in relation to violence or aggression, and they obscure the
commonalities and distinctions among forms of violence more
than they illuminate them. Terms that make specific reference to a
class of perpetrators, such as caregiver violence, would better
identify the distinctive feature of what is now known variously as
child abuse or child maltreatment.
The use of multiple vague synonyms or not-quite-synonyms
does not advance science. It is not widely known in psychology,
but it took centuries to fully address this issue in chemistry and
physics. The transition from the idiosyncratic terminology of al-
chemists to the standardized nomenclature in use today did not
begin in earnest until the late 19th century, and was not formalized
until the creation of the International Union of Pure and Applied
Chemistry in 1919 (International Union of Pure and Applied
Chemistry, n.d.). This ensures that one researcher does not refer to
an element as “hydrogen” while another researcher calls the same
element “sherrinagen.” The field of violence research, and indeed
all of psychology, needs to get behind a similar effort.
Implications for Measurement
Once you think about the key elements of a violence definition,
it is surprising how few violence questionnaires measure whether
a behavior is nonessential, unwanted, intentional, or harmful.
Simply reporting a “hit” does not allow one to distinguish a
football tackle from a pillow fight from an assault. I have been told
by numerous colleagues that survey participants “know what they
mean” when they intend to ask about violence, no matter how
imprecise their questions are. First, that is no way to conduct
science, even if it were true. Second, there are volumes of data
indicating that this is not true. Several experimental or quasi-
experimental studies show that the operationalization of violence
has a large impact on reported rates, gender patterns, and other
factors (Finkelhor, Shattuck, Turner, & Hamby, 2016;Fisher,
Cullen, & Turner, 2000;Hamby, 2016b;Lessne & Cidade, 2017;
Rothman & Xuan, 2012). Furthermore, many people in the general
population do not view child maltreatment, domestic violence,
sexual assault, and other forms of violence in the same way that
most scientists view them. Many people endorse (and commit)
Table 2
Summary of Implications for Terminology, Measurement, Theory, Prevention, and Intervention
(1) Violence is nonessential, unwanted, intentional, and harmful behavior. The animal research model suggests the most scientifically useful
distinction between violence and the broader construct of aggression, by focusing on violence as nonessential aggression (for humans, not in the
interest of self-defense or protecting others, and not due to limited behavioral repertoires). Using this distinction, most human aggression meets
criteria for violence.
(2) Harms not visible to the naked eye need more recognition, especially because many of them have far more lasting and damaging effects than
minor cuts and bruises. Bullying, corporal punishment, child neglect, intimate partner violence, and sexual offenses all meet all four criteria for
violence given the current state of scientific information about them.
(3) Terms such as “abuse,” “maltreatment,” and “battering” should either be systematically defined in relation to these four elements or dropped. A
proliferation of synonyms or near-synonyms for the same class of behavior does not advance science. Better terms would focus on the
distinguishing features of subtypes, such as “caregiver violence” versus “child abuse.”
(4) Survey measures of violence should clearly incorporate all four definitional elements (nonessential, unwanted, intentional, and harmful), ideally in
the text of the questionnaire (either in the items, a preamble, or both).
(5) It should be recognized that measures that simply ask about behaviors such as hitting and pushing are incapable of distinguishing violence from
accidents, horseplay, contact sports, and many other unintended or prosocial behaviors.
(6) Laboratory analogs should specify the extent to which they can model all four elements of violence (recognizing that, by definition, analogs will
not fully model all elements). Analogs should not merely present noxious stimuli.
(7) Measures of harm should include measures of harm not visible to the naked eye, including psychological harm and long-term neurological
damage, or at least specify the limitations of scope.
(8) Theories of violence and theories of change should specify how their proposed mechanisms are related to the production of violent behavior.
Theories of violence and theories of change need to pay more attention to motive, depraved indifference, and the precursors to intent to harm or
callous disregard for likely harms.
(9) Programs to prevent or intervene against violence also need to understand these elements. They should know whether their target problem is more
likely to be intentionally perpetrated harm, or due to recklessness and callous disregard for others, for example. Instead of calling for general
principles of respect and fellowship, a clearer focus on definitional elements would call for a more central focus on the elements of wantedness
and consent. This is increasingly common in sexual assault prevention, but might be extended to other forms of violence as well.
(10) Similarly, people who work with victims should be aware of the full range of harms, and not minimize incidents that do not produce visible
bodily injuries. This includes not only specialized victim advocates, but psychologists, health care providers, and criminal justice personnel.
Note. See text for elaboration of these points.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
177
ON DEFINING VIOLENCE, AND WHY IT MATTERS
severe corporal punishment, justify domestic violence in cases of
infidelity or even “talking back,” and believe that women are
“asking for it” when they are sexually assaulted (e.g., De Puy,
Hamby, & Lindemuth, 2015). Indeed, there are whole subdisci-
plines of violence research devoted to these issues. There is also
evidence of underreporting, because of social desirability, shame,
or other factors (e.g., Sugarman & Hotaling, 1997).
These factors probably contribute to the incredibly wide range
of prevalence estimates that are common in our field. For example,
a recent meta-analysis found rates of physical teen dating violence
ranging from 1% to 61% (Wincentak, Connolly, & Card, 2017).
That is an unacceptable level of imprecision. Scientists need to rise
to the challenge of improving the assessment of all constructs, not
stubbornly sticking to claims that everything that has been done in
the past is “good enough.” Although underlying this reluctance for
more precision may be concerns about depressing reporting, ex-
isting evidence suggests that better and clearer operationalizations
enhance disclosure (Hamby, 2016a).
A clearer definition can also help with other controversies in
violence and aggression research. For example, the suitability of
laboratory analogs of aggression or violence has been a topic of
dispute (Calvert et al., 2017;Tedeschi & Quigley, 1996). Analogs,
by definition, need not exactly match the phenomenon they are
trying to model and are used when ethics or other considerations
keep researchers from creating the full phenomenon in the labo-
ratory. Laboratory analogs for violence typically use a “noxious
stimuli” (to use Buss’s term), such as an unpleasant electric shock,
loud noise, or hot sauce exposure. They cause unpleasantness or
even pain, but so transiently that they do not meet the harm
criterion for violence. Given that one element of violence cannot
ethically be included, it is even more important to be clear about
where these analogs stand on the other three elements. They all
require the participant to act intentionally. On the matter of non-
essential, some paradigms present challenges. In the classic Mil-
gram paradigm (Milgram, 1963), the experimenter insists that the
shock is “necessary,” and required the participant to dispute that.
Some analogs more clearly establish unwantedness than others.
For example, in the hot sauce paradigm, researchers explicitly
communicate that the confederate really dislikes spicy food. Im-
proving the specification of all four key elements should reduce
error in violence research data and aid interpretation of generaliz-
ability.
Implications for Theory
Many psychological and sociological theories of violence as-
sume, to a surprising degree, that the precise question of what
phenomena they apply to does not need stating. A theory of
violence should focus on mechanisms, and how it can come to pass
that someone would intentionally commit unwanted, nonessential,
harmful acts. The presence of the element of intent requires an
understanding of motive, or in criminological terms, “state of
mind” at the time the violence is committed. The feminist model
of patriarchy is one of the few that explicitly addresses the ques-
tion of motive, by pointing to the use of power and control in a
desire to maintain the patriarchal structure of society. However, in
its simplest form, the focus on a single motive limits the patriarchal
model. Other single-factor models, such as self-control theory
(Gottfredson & Hirschi, 1990), suffer the same problem. The
General Aggression Model (DeWall, Anderson, & Bushman,
2011) and its many social– cognitive cousins cover more risk
factors for aggressive acts, but these models also pay insufficient
attention to the essential elements. For example, our lack of
understanding of how motive develops and leads to intent is one
reason why we are still so limited in our ability to explain why one
child with an extensive history of trauma becomes a perpetrator,
while another mistreated child does not. By not attending to the
underlying goals of behavior, our models of change tend to focus
on “just say no” instead of teaching children how to use their full
nonviolent behavioral repertoire to meet their legitimate needs for
social status and resources.
The public health approach, despite better clarity in definitions
than most other approaches, has focused mostly on general risk
factors, and has not integrated their definition of violence into an
analysis of the causes of violence. The Haddon matrix, the dom-
inant conceptual framework in public health, calls for incorporat-
ing an analysis of human, agent (such as weapons), and environ-
mental factors, and was originally focused on minimizing the
harms of natural disasters, accidents, and other unintentional acts
(Haddon, 1970). The matrix has been applied to violence (e.g.,
Hemenway, 2009), and certainly agents and environmental factors
affect the risk of a violent act, but the public health approach needs
greater recognition that only human factors define whether an act
is violent or not, and a complete approach to violence prevention
must include human factors.
Implications for Practice and Policy
A lack of consensus on the definition of violence is the under-
lying problem in a huge number of challenges to practice and
policy (Hamby & Finkelhor, 2000). Reports of rape and IPV are
still rarely prosecuted—with under 5% of reports to law enforce-
ment resulting in a conviction, even in recent years (Hamby,
Finkelhor, & Turner, 2015). Disbelief in the constructs of rape and
battering are still one of the main reasons for this tepid response
(Shaw, Campbell, Cain, & Feeney, 2016). Minimization of the
harms caused by sexual assault is an everyday problem in the
news. We need a more systematic approach to specifying the
reasons that bullying, sexual assault, child neglect and other major
public health problems should be treated as acts of violence. We
should be explicitly seeking to establish a definitional consensus as
an essential step to resolving these program and policy issues.
Conclusion
The main problem with most measurement in psychology is that
we focus on exemplars. We think about “mammal,” and then we
write measures with images of cats and dogs dancing in our heads.
But the tricky thing to defining mammal is not describing cats and
dogs, it is justifying why whales and bats are mammals too, and
not fish or birds. Not to belabor the metaphor, but it is equally
important to understand why ostriches still count as birds and why
kangaroos are mammals, but seahorses are not. To advance the
science of violence, we need to study the whales and bats and
ostriches of interpersonal behavior—incidents that fall near the
boundary of the violence construct—and justify why some should
be called violent and others nonviolent.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
178 HAMBY
References
Abrahams, N., Devries, K., Watts, C., Pallitto, C., Petzold, M., Shamu, S.,
& García-Moreno, C. (2014). Worldwide prevalence of non-partner
sexual violence: A systematic review. Lancet, 383, 1648 –1654. http://
dx.doi.org/10.1016/S0140-6736(13)62243-6
American Psychiatric Association. (2013). Diagnostic and statistical man-
ual of mental disorders (5th ed.). Washington, DC: Author.
American Psychological Association (APA). (n.d.). Violence. Retrieved
from http://www.apa.org/topics/violence/index.aspx
Berman, S. J. (2013). Homicide: Murder and manslaughter. Retrieved from
http://www.nolo.com/legal-encyclopedia/homicide-murder-manslaughter-
32637-2.html
Buss, A. H. (1961). The psychology of aggression. New York, NY: Wiley.
http://dx.doi.org/10.1037/11160-000
Calvert, S. L., Appelbaum, M., Dodge, K. A., Graham, S., Nagayama Hall,
G. C., Hamby, S.,...Hedges, L. V. (2017). The American Psycholog-
ical Association Task Force assessment of violent video games: Science
in the service of public interest. American Psychologist, 72, 126 –143.
http://dx.doi.org/10.1037/a0040413
Cook, S., & Parrott, D. (2009). Exploring a taxonomy for aggression
against women: Can it aid conceptual clarity? Aggressive Behavior, 35,
462– 476. http://dx.doi.org/10.1002/ab.20321
De Puy, J., Hamby, S. L., & Lindemuth, C. (2015). Teen dating violence
in French-speaking Switzerland: Attitudes and experiences. [IJCV]. In-
ternational Journal of Conflict and Violence, 8, 305–315.
DeWall, C. N., Anderson, C. A., & Bushman, B. J. (2011). The general
aggression model: Theoretical extensions to violence. Psychology of
Violence, 1, 245–258. http://dx.doi.org/10.1037/a0023842
Faris, R., & Felmlee, D. (2011). Status struggles network centrality and gender
segregation in same-and cross-gender aggression. American Sociological
Review, 76, 48 –73. http://dx.doi.org/10.1177/0003122410396196
FBI. (2011). Violent crime. Retrieved from https://ucr.fbi.gov/crime-in-
the-u.s/2010/crime-in-the-u.s.-2010/violent-crime
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M.,
Edwards, V.,...Marks, J. S. (1998). Relationship of childhood abuse
and household dysfunction to many of the leading causes of death in
adults: The Adverse Childhood Experiences (ACE) Study. American
Journal of Preventive Medicine, 14, 245–258. http://dx.doi.org/10.1016/
S0749-3797(98)00017-8
Finkelhor, D. (2008). Childhood victimization: Violence, crime, and abuse
in the lives of young people. New York, NY: Oxford University Press.
http://dx.doi.org/10.1093/acprof:oso/9780195342857.001.0001
Finkelhor, D., Shattuck, A., Turner, H., & Hamby, S. (2016). A behavior-
ally specific, empirical alternative to bullying: Aggravated peer victim-
ization. Journal of Adolescent Health, 59, 496 –501. http://dx.doi.org/10
.1016/j.jadohealth.2016.05.021
Fisher, B. S., Cullen, F. T., & Turner, M. G. (2000). The sexual victim-
ization of college women (Research report). Washington, DC: U.S.
Department of Justice.
Follingstad, D. (2017). The challenges of measuring violence against
women. In C. Renzetti, J. Edleson, & R. Bergen (Eds.), Sourcebook on
violence against women (3rd ed., pp. 57–77). Thousand Oaks, CA: Sage.
Gavett, B. E., Stern, R. A., & McKee, A. C. (2011). Chronic traumatic
encephalopathy: A potential late effect of sport-related concussive and
subconcussive head trauma. Clinics in Sports Medicine, 30, 179 –188.
http://dx.doi.org/10.1016/j.csm.2010.09.007
Gershoff, E. T., & Bitensky, S. H. (2007). The case against corporal
punishment of children: Converging evidence from social science re-
search and international human rights law and implications for U.S.
public policy. Psychology, Public Policy, and Law, 13, 231–272. http://
dx.doi.org/10.1037/1076-8971.13.4.231
Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime.
Redwood City, CA: Stanford University Press.
Graham, K., Bernards, S., Osgood, D. W., Parks, M., Abbey, A., Felson,
R.B.,...Wells, S. (2013). Apparent motives for aggression in the social
context of the bar. Psychology of Violence, 3, 218 –232. http://dx.doi
.org/10.1037/a0029677
Haddon, W., Jr. (1970). On the escape of tigers: An ecologic note. Amer-
ican Journal of Public Health and the Nation’s Health, 60, 2229 –2234.
http://dx.doi.org/10.2105/AJPH.60.12.2229-b
Hamby, S. (2014). Battered women’s protective strategies: Stronger than
you know. New York, NY: Oxford University Press.
Hamby, S. (2016a). Advancing survey science for intimate partner vio-
lence: The Partner Victimization Scale and other innovations. Psychol-
ogy of Violence, 6, 352–359. http://dx.doi.org/10.1037/vio0000053
Hamby, S. (2016b). Self-report measures that do not produce gender parity
in intimate partner violence: A multi-study investigation. Psychology of
Violence, 6, 323–335. http://dx.doi.org/10.1037/a0038207
Hamby, S. L., & Finkelhor, D. (2000). The victimization of children:
Recommendations for assessment and instrument development. Journal
of the American Academy of Child & Adolescent Psychiatry, 39, 829 –
840. http://dx.doi.org/10.1097/00004583-200007000-00011
Hamby, S., Finkelhor, D., & Turner, H. (2013). Perpetrator and victim
gender patterns for 21 forms of youth victimization in the National
Survey of Children’s Exposure to Violence. Violence and Victims, 28,
915–939. http://dx.doi.org/10.1891/0886-6708.VV-D-12-00067
Hamby, S., Finkelhor, D., & Turner, H. (2015). Intervention following
family violence: Best practices and helpseeking obstacles in a nationally
representative sample of families with children. Psychology of Violence,
5, 325–336. http://dx.doi.org/10.1037/a0036224
Hamby, S., & Jackson, A. (2010). Size does matter: The effects of gender
on perceptions of dating violence. Sex Roles, 63, 324 –331. http://dx.doi
.org/10.1007/s11199-010-9816-0
Hamby, S., & Koss, M. P. (2003). Shades of gray: A qualitative study of
terms used in the measurement of sexual victimization. Psychology of
Women Quarterly, 27, 243–255. http://dx.doi.org/10.1111/1471-6402
.00104
Hartley, C. C. (2001). Incest offenders’ perceptions of their motives to sexually
offend within their past and current life context. Journal of Interpersonal
Violence, 16, 459 – 475. http://dx.doi.org/10.1177/088626001016005005
Hemenway, D. (2009). While we were sleeping: Success stories in injury
and violence prevention. Berkeley, CA: University of California Press.
Hipp, T. N., Bellis, A. L., Goodnight, B. L., Brennan, C. L., Swartout,
K. M., & Cook, S. L. (2017). Justifying sexual assault: Anonymous
perpetrators speak out online. Psychology of Violence, 7, 82–90. http://
dx.doi.org/10.1037/a0039998
International Union of Pure and Applied Chemistry. (n.d.). Our history.
Retrieved from https://iupac.org/who-we-are/our-history/
Koolhaas, J. M., de Boer, S. F., & Buwalda, B. (2010). Neurobiology of
offensive aggression. In G. Koob, M. l. Moal, & R. Thompson (Eds.),
Encyclopedia of behavioral neuroscience (pp. 387–392). San Diego,
CA: Elsevier Press. http://dx.doi.org/10.1016/B978-0-08-045396-5
.00086-5
Kopel, D. B., Gallant, P., & Eisen, J. D. (2008). The human right of
self-defense. BYU Journal of Public Law, 22, 43–178.
Krug, E. G., Dahlberg, L. L., Mercy, J. A., Zwi, A. B., & Lozano, R.
(2002). World report on violence and health. Retrieved from http://
www.who.int/violence_injury_prevention/violence/world_report/en/
FullWRVH.pdf http://dx.doi.org/10.1016/S0140-6736(02)11133-0
Leeb, R., Paulozzi, L., Melanson, C., Simon, T., & Arias, I. (2008). Child
maltreatment surveillance: Uniform definitions for public health and
recommended data elements, Version 1.0. Atlanta, GA: Centers for
Disease Control and Prevention.
Lehrner, A., & Allen, N. E. (2014). Construct validity of the Conflict
Tactics Scales: A mixed-method investigation of women’s intimate
partner violence. Psychology of Violence, 4, 477– 490.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
179
ON DEFINING VIOLENCE, AND WHY IT MATTERS
Lereya, S. T., Copeland, W. E., Costello, E. J., & Wolke, D. (2015). Adult
mental health consequences of peer bullying and maltreatment in child-
hood: Two cohorts in two countries. The Lancet Psychiatry, 2, 524 –531.
http://dx.doi.org/10.1016/S2215-0366(15)00165-0
Lessne, D., & Cidade, M. (2017). Split-half administration of the 2015
School Crime Suppl. to the National Crime Victimization Survey meth-
odology report (NCES 2017– 004). Retrieved from http://nces.ed.gov/
pubsearch
Lloyd, E. A. (2003). Violence against science: Rape and evolution. In C. B.
Travis (Ed.), Evolution, gender, and rape (pp. 235–261). Cambridge,
MA: MIT Press.
Maccoby, E. E. (1998). The two sexes: Growing up apart, coming together.
Cambridge, MA: Harvard University Press.
McCauley, H. L., Jaime, M. C. D., Tancredi, D. J., Silverman, J. G.,
Decker, M. R., Austin, S. B.,...Miller, E. (2014). Differences in
adolescent relationship abuse perpetration and gender-inequitable atti-
tudes by sport among male high school athletes. The Journal of Ado-
lescent Health, 54, 742–744. http://dx.doi.org/10.1016/j.jadohealth.2014
.01.001
Milgram, S. (1963). Behavioral Study of obedience. Journal of Abnormal
and Social Psychology, 67, 371–378. http://dx.doi.org/10.1037/
h0040525
Natarajan, D., & Caramaschi, D. (2010). Animal violence demystified.
Frontiers in Behavioral Neuroscience, 4, 9.
Ouellet-Morin, I., Wong, C. C., Danese, A., Pariante, C. M., Papadopoulos,
A. S., Mill, J., & Arseneault, L. (2013). Increased serotonin transporter
gene (SERT) DNA methylation is associated with bullying victimization
and blunted cortisol response to stress in childhood: A longitudinal study
of discordant monozygotic twins. Psychological Medicine, 43, 1813–
1823. http://dx.doi.org/10.1017/S0033291712002784
Parrott, D. J., & Giancola, P. R. (2007). Addressing “The criterion prob-
lem” in the assessment of aggressive behavior: Development of a new
taxonomic system. Aggression and Violent Behavior, 12, 280 –299.
http://dx.doi.org/10.1016/j.avb.2006.08.002
Piaget, J., & Cook, M. T. (1952). The origins of intelligence in children.
New York, NY: International University Press. http://dx.doi.org/10
.1037/11494-000
Rothman, E., & Xuan, Z. (2012, March). Assessing the prevalence of
dating violence using national and state YRBS data from 1999 –2009:
What we’ve learned, and how we might improve the YRBS question
going forward. Paper presented at the Futures Without Violence Health-
care and Domestic Violence Conference, San Francisco, CA.
Shalev, I., Moffitt, T. E., Sugden, K., Williams, B., Houts, R. M., Danese,
A.,...Caspi, A. (2013). Exposure to violence during childhood is
associated with telomere erosion from 5 to 10 years of age: A longitu-
dinal study. Molecular Psychiatry, 18, 576 –581. http://dx.doi.org/10
.1038/mp.2012.32
Shaw, J., Campbell, R., Cain, D., & Feeney, H. (2016). Beyond surveys
and scales: How rape myths manifest in sexual assault police records.
Psychology of Violence. Advance online publication. http://dx.doi.org/
10.1037/vio0000072
Sommer, S., Reynolds, J. J., & Kehn, A. (2016). Mock juror perceptions of
rape victims: Impact of case characteristics and individual differences.
Journal of Interpersonal Violence, 31, 2847–2866. http://dx.doi.org/10
.1177/0886260515581907
Sue, D. W. (2017). Microaggressions and “evidence.” Perspectives on
Psychological Science, 12, 170 –172. http://dx.doi.org/10.1177/1745
691616664437
Sugarman, D. B., & Hotaling, G. T. (1997). Intimate violence and social
desirability a meta-analytic review. Journal of Interpersonal Violence,
12, 275–290. http://dx.doi.org/10.1177/088626097012002008
Tedeschi, J. T., & Quigley, B. M. (1996). Limitations of laboratory
paradigms for studying aggression. Aggression and Violent Behavior, 1,
163–177. http://dx.doi.org/10.1016/1359-1789(95)00014-3
Thompson, M. P., Kingree, J. B., Zinzow, H., & Swartout, K. (2015).
Time-varying risk factors and sexual aggression perpetration among
male college students. Journal of Adolescent Health, 57, 637– 642.
http://dx.doi.org/10.1016/j.jadohealth.2015.08.015
Travis, C. B. (Ed.), (2003). Evolution, gender, and rape. Cambridge, MA:
MIT Press.
Truman, J., & Morgan, R. (2016). Criminal victimization, 2015. Retrieved
from https://www.bjs.gov/content/pub/pdf/cv15.pdf
Turner, H. A., Finkelhor, D., Shattuck, A., & Hamby, S. (2012). Recent
victimization exposure and suicidal ideation in adolescents. Archives of
Pediatrics & Adolescent Medicine, 166, 1149 –1154. http://dx.doi.org/
10.1001/archpediatrics.2012.1549
Turner, H. A., Finkelhor, D., Shattuck, A., Hamby, S., & Mitchell, K.
(2015). Beyond bullying: Aggravating elements of peer victimization
episodes. School Psychology Quarterly, 30, 366 –384. http://dx.doi.org/
10.1037/spq0000058
Wincentak, K., Connolly, J., & Card, N. (2017). Teen dating violence: A
meta-analytic review of prevalence rates. Psychology of Violence, 7,
224 –241. http://dx.doi.org/10.1037/a0040194
Yllö, K., & Torres, M. G. (2016). Marital rape: Consent, marriage, and
social change in global context. New York, NY: Oxford University
Press. http://dx.doi.org/10.1093/acprof:oso/9780190238360.001.0001
Zwick, R., Carlstein, E., & Budescu, D. V. (1987). Measures of similarity
among fuzzy concepts: A comparative analysis. International Journal of
Approximate Reasoning, 1, 221–242. http://dx.doi.org/10.1016/0888-
613X(87)90015-6
Received March 8, 2017
Accepted March 8, 2017 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
180 HAMBY