The Predictive Value of Childhood Animal Cruelty Methods on Later Adult Violence

Abstract

The present study seeks to replicate Tallichet, Hensley, and Singer’s research on childhood animal cruelty methods by using a sample of 180 male inmates surveyed at both medium- and maximum-security prisons in a southern state. The purpose of the current study was to first reexamine the relationship between demographic and situational factors and specific methods of childhood animal cruelty. Second, the correlation between an abuser’s chosen method(s) of childhood animal cruelty on later recurrent acts of adult violent crimes was reinvestigated. Regression analyses revealed that respondents who engaged in frequent animal cruelty were more likely to have drowned, shot, kicked, or had sex with animals. Those who had grown up in urban areas and those who did not become upset after abusing animals were more likely to have kicked animals. Respondents who covered up their abuse were more likely to have had sex with animals. Sex with animals was the only method of childhood animal cruelty that predicted the later commission of adult violent crimes.

Keywords

animal cruelty violence animal cruelty methods

Public concern over the potential relationship between childhood animal cruelty and later adult violence toward humans continues to increase, perhaps due in part to the media’s portrayal of the more horrendous acts along the continuum of the animal cruelty spectrum. Nearly 40 years of research into the complexities of childhood animal cruelty and its potential relationship to adult violent crimes have produced inconsistent and controversial results. In fact, Felthous and Kellert (1987) conducted a meta-analysis of 15 studies from the 1960s and 1970s that examined the link between childhood animal cruelty and adult violence toward humans. Five of the studies found a link between the two behaviors, whereas the other 10 did not. The studies that found a link between childhood animal cruelty and adult violence toward humans used direct interviews, defined the terms animal cruelty and personal aggression, used recurrent acts of animal cruelty rather than a single act, and used recurrent acts of personal aggression rather than a single act. The studies that failed to find a link mainly used chart reviews, did not define the behaviors, used one act of animal abuse as animal cruelty, and used one act of violence toward humans as personal aggression.

Some recent research has revealed either no association at all or no apparent time order between acts of childhood animal cruelty and adult violence toward humans (Arluke, Levin, Luke, & Ascione, 1999; Miller & Knutson, 1997). Other studies have supported such association, including the time order (Flynn, 1999; Hensley, Tallichet, & Dutkiewicz, 2009; Merz-Perez & Heide, 2004; Merz-Perez, Heide, & Silverman, 2001; Tallichet & Hensley, 2004), with some of these studies finding a moderate to strong relationship between the two behaviors.

Perhaps more conclusive evidence supporting or discrediting the potential link between childhood animal cruelty and adult violence toward humans can be found if contributing factors are considered individually. According to Merz-Perez and Heide (2004), such factors may include the method of animal abuse, the type of animal abused, and the motivation for the abuse. The current study seeks to find additional support (as found in the study by Tallichet, Hensley, and Singer, 2005) for the potential relationship between adolescent animal cruelty and later violent crimes by following the ideas of Merz-Perez and Heide (2004) to individually analyze these contributing factors, specifically the method of animal cruelty. This exploratory research that examines methods of animal cruelty is critical in our attempt to gain insight into the complex relationship between childhood animal cruelty and later adult violence. If research can determine that specific demographic and/or situational factors related to the methods of animal cruelty are more reliable than others to the prediction of an abuser’s propensity for later violence against humans, clinicians could more easily identify which childhood animal perpetrators were in most critical need of intervention. Such an understanding could prove invaluable to society in preserving the lives of innocent animal and human victims.

Literature Review

Very few studies have examined the relationship between childhood animal cruelty and later violence toward humans in the context of the specific methods of cruelty used by the perpetrator(s) (Felthous, 1980; Kellert & Felthous, 1985; Merz-Perez et al., 2001; Merz-Perez & Heide, 2004; Miller & Knutson, 1997; Tallichet et al., 2005). In his study of aggression against, cats, dogs, and people, Felthous (1980) investigated animal cruelty methods by interviewing 346 male psychiatric patients who were categorized according to their level of aggressiveness and the presence or absence of childhood animal cruelty. Of the 71 patients labeled aggressive, 18 were also classified as recurrent animal abusers. In discussing the methods of childhood animal abuse committed, Felthous discovered that all but one of the 18 recurrent animal abusers had victimized cats, nearly triple the number of respondents who had victimized dogs. He further found that four subjects reported killing cats by hanging and eight had burned or exploded cats and/or dogs. Other methods of animal abuse included “limb amputation, decapitation, choking, brutal beatings, fracturing bones, and scalding with hot water” (p. 161). Finally, Felthous noted that three subjects reported tying the tails of animals together over a clothesline in order to observe the animals destroy each another.

Further research was conducted by Kellert and Felthous in 1985 when they interviewed 32 aggressive inmates, 18 moderately aggressive inmates, 52 nonaggressive inmates, and randomly selected 50 nonincarcerated males from Kansas and Connecticut. Respondents admitted to a wide range of animal abuse during childhood. While the purpose of their research was to examine the relationship between childhood animal cruelty and adult aggressiveness, their documentation of the subjects’ specific methods of animal abuse opened the door further for future research. Among the most common methods of childhood animal abuse identified were beating (n = 18), shooting (n = 14), stoning (n = 11), and throwing from heights (n = 10). Other methods identified included “dismemberment, exploding, stabbing, burning or electrocuting, breaking bones, and entering into fights” (p. 228).

Using a self-report questionnaire to examine the methods of childhood animal cruelty engaged in by a sample of 314 inmates, Miller and Knutson (1997) found that the methods used by subjects to kill animals included poison (n = 17); drowning (n = 5); hitting, beating, or kicking (n = 43); shooting (n = 77); strangling/smothering (n = 6); stabbing (n = 6); burning (n = 5); throwing against an object (n = 9); exploding (n = 7); accidental (n = 16); and other methods (n = 6). These figures only included childhood animal abuse in which the subject was the prime perpetrator and excluded abuse that did not result in the death of the animal.

Although early studies discussed the perpetrators’ methods of abuse, few used these data in any qualitative or quantitative analyses in order to describe the potential relationship between childhood and adolescent animal cruelty and adult violent crimes. More recently, Merz-Perez and Heide (2004) and Merz-Perez et al. (2001), as part of a larger study on adolescent animal cruelty and adult violence toward humans, examined methods of animal cruelty by conducting interviews with a random sample of 45 violent and 45 nonviolent male offenders incarcerated in a maximum-security Florida penitentiary. Participants were identified as violent or nonviolent based on the offense for which they had been convicted and additional screening for offense history during the offender interview. They found that 56% of the violent offender sample reported committing acts of animal cruelty, whereas only 20% of the nonviolent offender sample reported engaging in such acts.

Qualitative data in this study revealed that nonviolent offenders were more likely than violent offenders to have used methods of animal cruelty that could be categorized as “distanced” acts of abuse because their acts could be committed without close physical contact with the animal. Nonviolent offenders reported having committed childhood animal cruelty through the following methods: shooting (n = 6), forced fighting (n = 3), and fear (n = 1). Violent offenders, on the other hand, reported having committed acts of cruelty that required actual physical contact between the perpetrator and the animal. These offenders committed direct acts of violence by beating, kicking, or stomping (n = 5); engaging in sexual activity with the animal (n = 3); pouring chemical irritants on the animal (n = 2); dismembering (n = 2); burning (n = 1); and stabbing (n = 1).

In 2005, Tallichet and colleagues surveyed a sample of 261 male inmates at three prisons in a southern state to examine what methods inmates used while engaging in childhood or adolescent animal cruelty. The study examined the relationship between demographic attributes (race, education, and residence while growing up), situational factors (was the abuse committed alone, did abuser try to conceal the act, was abuser upset by the abuse, what was the perpetrator’s age at initial animal cruelty, how frequent was the animal abuse), and specific methods of animal cruelty (shooting, drowning, hitting/kicking, choking, burning, and sex). In addition, the authors investigated the link between childhood animal cruelty methods and later violence against humans (as assessed by three of the four index violent crimes). The 112 inmates who had engaged in animal cruelty reported using a variety of methods in their acts: shot (n = 72), hit or kicked (n = 50), choked (n = 24), burned (n = 17), drowned (n = 16), and sex with animals (n =16).

Logistic regression analyses revealed that inmates who had sex with animals were less likely to have done so alone, but were more likely to have covered up their actions. However, those who had shot animals were less likely to have covered up their abuse. Inmates who shot animals were more likely to be white and had committed recurrent acts of animal cruelty but were less likely to become upset by their actions. Despite these findings, the authors failed to find support for a potential link between the methods used during acts of childhood animal cruelty and future adult violent crimes (Tallichet et al., 2005).

In light of the often contradictory findings of previous research into the potential relationship between childhood and adolescent animal cruelty and later violence toward humans, and the insufficient quantitative data regarding the methods used in the commission of such acts, this study was designed to replicate the Tallichet et al. (2005) study and further the understanding of this complex phenomenon. Given their potential for identifying individuals at greater risk for violence, this is an important research question. Therefore, the current study reexamines the following: (a) methods by which animal cruelty was committed during childhood and adolescence as related to subjects’ race, education, and residential location while growing up; (b) situational factors including whether the abuse was committed alone, whether the perpetrator attempted to conceal the act, and whether the perpetrator was upset by the act; (c) the age when they first committed cruelty toward animals; and (d) the frequency of animal abuse during childhood and adolescence. In addition, this study reanalyzes the potential predictors of childhood animal cruelty methods on future adult violent crimes.

However, the current study is different in four important ways. First, the methods hit and kicked have been separated in the following analyses. Second, the inmates were asked whether they had committed adult violent crimes and how often rather than being asked if they had been convicted of those crimes. Third, a measure for robbery, another index for violent crime, was included. Robbery was excluded in the study by Tallichet et al. (2005). Finally, the sample in the current study is from a different southern state.

Method

Participants

In March 2007, all inmates housed in one medium- and one maximum-security southern correctional facility for men were requested to participate in a study of childhood animal cruelty. Of the 1,800 inmates incarcerated in the two prisons, a total of 180 agreed to participate in the study, yielding a response rate of 10% (as each inmate received a questionnaire). Although this response rate appears low, most prison studies dealing with sensitive issues attract 25% or fewer respondents (Hensley et al., 2009; Tallichet & Hensley, 2004; Tourangeau, Rips, & Rasinski, 2000). After obtaining approval from the state department of corrections and the university’s Institutional Review Board, the principal researcher drove to the facilities and delivered the questionnaires and informed consent forms. The informed consent stated that the questionnaires were confidential. In addition, the state department of corrections agreed not to open any of the surveys prior to their being mailed back to the authors.

Correctional counselors at each prison distributed self-administered questionnaires to each inmate. Inmates were informed that it would take approximately 20 min to complete the 26-item questionnaire. Inmates were asked to return their completed questionnaires and signed informed consent forms in a stamped, self-addressed envelope within 1 month of distribution. No incentives were given for completion of the survey. The researchers contacted the two prisons after the 30-day period to make sure all completed surveys had been mailed.

Table 1 displays the characteristics of the state prison population and the sample. A comparison of the racial composition, type of offense committed, and age distribution of the respondents and the state prison population revealed no significant differences. Thus, the sample appears to be representative of the state prison population in terms of these variables.

Table 1.

Population and Sample Characteristics

	Prison population		Sample
Characteristic	n	%	n	%
Race
White	9,333	51.9	91	50.6
Other	8,667	48.1	89	49.4
Type of offense
Violent crime	11,283	62.7	115	63.9
Other crime	6,717	37.3	65	36.1
Median age (years)	35		33.5

Measures

The primary goals of the present study were to investigate the demographic and situational correlates of animal cruelty methods used by children and to examine the predictive value of animal cruelty methods with respect to later adult violence. The variables were derived from previous studies that examined the link between childhood animal cruelty and adult violence (Ascione, Thompson, & Black, 1997; Boat, 1994; Merz-Perez et al., 2001; Merz-Perez & Heide, 2004; Tallichet et al., 2005; Tallichet & Hensley, 2004).

It should be noted that animals were defined as “pet or stray” (with the inmate listing the type of animal hurt or killed for both categories). Animal cruelty included any action where the respondent hurt or killed animals as children (other than for hunting). This is consistent with the most frequently used definition of animal cruelty by social scientists which states that animal cruelty is “socially unacceptable behavior that intentionally causes unnecessary pain, suffering, or distress to and/or death of an animal” (Ascione, 1993, p. 228). This definition is preferred because it omits behaviors that may be socially and culturally acceptable or condoned in other contexts, such as hunting.

Inmates were asked to indicate “what they did to hurt or kill animals” by circling each of the methods that were listed on the survey. These included drowned, hit, shot, kicked, choked, burned, and/or had sex with animals (Boat, 1994; Tallichet et al., 2005). Each response for each method was coded so that zeros indicated that the inmate had not used that method and ones indicated that the offender had used that method.

Inmates were also asked a series of questions regarding their commission of violent crimes. They included the following: (a) “Have you ever committed murder or attempted murder?” (b) “Have you ever committed rape or attempted rape?” (c) “Have you ever committed aggravated or simple assault?” and (d) “Have you ever committed robbery?” These questions were coded 0 = no and 1 = yes. They were asked how many times they had committed each of these crimes. To develop a cumulative score of recurrent violent crimes, we added the number of times each inmate had committed these crimes. The scores ranged from 0 to 16, with an average of 3.46 times and a standard deviation of 4.00.

Demographic characteristics (race, educational level, and residence while growing up) were recorded from the survey and used as predictor variables. Respondents were also asked if they had hurt or killed the animals alone, if they had covered up what they had done to the animals, and if hurting or killing the animals had upset them (Ascione et al., 1997; Merz-Perez et al., 2001; Merz-Perez & Heide, 2004; Tallichet et al., 2005). Finally, inmates were asked how many times they had hurt or killed animals and how old they were when they first committed animal cruelty (Tallichet et al., 2005).

Data Analysis

In order to achieve the two main goals of the study, frequencies and percentages of inmates who committed childhood animal cruelty and their methods while engaging in these acts were first run. Second, zero-order correlations between the demographic and situational factors, as well as the methods used to engage in childhood animal cruelty, were assessed. Correlations were examined because some of these variables were ordinal or interval in nature. Third, because one of the primary goals was to examine the effect that demographic and situational factors had on each of the methods used to commit animal cruelty, logistic regression was performed. The dependent variables for these models (each of the methods of animal cruelty) were dichotomous and the independent variables (the demographic and situational factors) were simultaneously entered into each model. Finally, ordinary least squares regression was calculated to determine the effects that the methods of animal cruelty had in predicting later adult violent crimes.

Results

One hundred and three respondents of the 180 engaged in some form of childhood animal cruelty at least once. Of those who engaged in animal cruelty, more than 82% reported that they had hit animals; approximately 36% had kicked animals and 33% had shot animals (see Table 2). Approximately 22% reported that they had engaged in sex with animals and approximately 17% had either drowned or choked animals. Finally, approximately 15% had burned animals as children. Inmates could select more than one method, resulting in a total cumulative percentage considerably higher than 100%.

Table 2.

Frequencies and Percentages of Inmates Who Committed Childhood Animal Cruelty and Their Methods While Engaging in These Acts (n = 103)

Method	n	%
Drowned	18	17.5
Hit	85	82.5
Shot	34	33.0
Kicked	37	35.9
Choked	18	17.5
Burned	16	15.5
Sexual	23	22.3

Table 3 presents the zero-order correlation matrix between the demographic and situational factors, as well as the methods of animal cruelty. White respondents were more likely to have grown up in rural areas and to have hurt or killed animals alone. Those who hurt or killed animals alone were less likely to be upset by their actions and were older when they first hurt or killed animals. Inmates who had covered up their actions were less likely to be upset, but were more likely to have had sex with animals. Those who felt upset after committing animal cruelty committed fewer acts of animal cruelty and were less likely to have kicked them. Respondents who engaged in frequent animal cruelty were more likely to begin at a younger age. In addition, those who committed frequent animal abuse were more likely to have drowned, shot, kicked, choked, burned, and had sex with animals.

Table 3.

Zero-Order Correlation Matrix—Demographic and Situational Factors

	X₁	X₂	X₃	X₄	X₅	X₆	X₇	X₈	Y₁	Y₂	Y₃	Y₄	Y₅	Y₆
X₂	.04
X₃	−.35*	.02
X₄	.30*	−.17	−.11
X₅	.12	.08	−.01	−.12
X₆	−.13	−.12	.17	−.20*	−.61*
X₇	.18	−.11	−.14	.07	.17	−.26*
X₈	.19	.04	.09	.26*	−.07	.07	−.23*
Y₁	−.10	−.08	.01	.01	.09	−.17	.40*	−.25*
Y₂	.10	.03	−.01	−.11	.18	−.02	.12	−.09	.01
Y₃	−.02	.01	.05	.07	.02	−.05	.26*	−.06	.28*	−.06
Y₄	−.09	−.13	.15	.15	−.02	−.23*	.24*	−.02	.46*	.03	.16
Y₅	.01	−.03	.06	.11	.09	−.11	.20*	−.12	.53*	.08	.28*	.40*
Y₆	−.04	.02	.01	−.09	−.01	−.08	.24*	−.22*	.44*	.20*	.27*	.29*	.37*
Y₇	.10	−.01	−.11	.14	.27*	−.16	.35*	.12	.06	.01	.17	.08	.18	.09

Coding of Independent Variables: (X₁) Race (0 = non-White, 1 = White); (X₂) Education (0 = 8th grade or less, 1 = some high school, 2 = completed high school, 3 = some college, 4 = completed college, 5 = graduate or professional school after college); (X₃) Residence (0 = rural area; 1 = urban area); (X₄) Hurt or Kill Animals Alone (0 = no, 1 = yes); (X₅) Cover Up Hurting or Killing Animals (0 = no, 1 = yes); (X₆) Committing Animal Cruelty Upset You (0 = no, 1 = yes); (X₇) Number of Times Hurt or Killed Animals (1 = once, 2 = twice, 3 = more than twice); (X₈) Age When First Hurt or Killed Animals (continuous variable). Coding of Independent Variables: (Y₁) Drowned (0 = no, 1 = yes); (Y₂); Hit (0 = no, 1 = yes); (Y₃) Shot (0 = no, 1 = yes); (Y₄) Kicked (0 = no, 1 = yes); (Y₅) Choked (0 = no, 1 = yes); (Y₆) Burned (0 = no, 1 = yes); (Y₇) Sex (0 = no, 1 = yes).

p < .05.

Inmates who hurt or killed animals at a younger age were more likely to have drowned or burned their animal victims. Those who drowned animals were also more likely to have shot, kicked, choked, and burned animals. Those who hit animals were also more likely to have burned them. Respondents who shot and kicked animals during childhood were also more likely to have choked and burned them. Finally, inmates who had choked animals were also more likely to have burned them.

In sum, most of 27 significant correlations discussed above were weak to moderate in strength (.20-.39). However, six correlations were strong (.40 or higher). They included the following: Inmates who covered up their animal cruelty were less likely to report feeling upset about doing it; respondents who frequently engaged in animal cruelty were more likely to drown their victims; those who drowned animals were also more likely to have kicked, choked, and burned them; and respondents who kicked animals were also more likely to have choked them.

Because the dependent variables were dichotomous, logistic regression analyses were performed to test what, if any, influence the demographic and situational variables had on each method of childhood animal cruelty. According to Table 4, only 7 of a possible 56 findings were statistically significant. The most statistically salient variable in the seven models was the number of times the respondents had hurt or killed animals as children. Inmates who reported frequent childhood animal cruelty were more likely to have drowned, shot, kicked, or had sex with animals. Two of the seven statistically significant findings were indeed strong. Respondents who grew up in urban areas were almost 3 times more likely to have kicked animals. Inmates who covered up their animal cruelty were more than 17 times more likely to have had sex with animals. The last significant finding, that those inmates who did not become upset after committing animal cruelty were more likely to have kicked animals, in contrast to the others, was weak. The independent variables accounted for between 11% and 37% of the total variance in the seven models. Four of the seven models were weak (hit, shot, choked, and burned), one was moderate (kicked), and two were moderately strong (drowned and sex with animals); however, all models were significant.

Table 4.

Summary of Logistic Regression Beta Weights

	Drowned		Hit		Shot		Kicked		Choked		Burned		Sex
Variable	B	Odds ratio	B	Odds ratio	B	Odds ratio	B	Odds ratio	B	Odds ratio	B	Odds ratio	B	Odds ratio
X₁	−1.19	0.30	0.61	1.84	−0.42	0.66	−0.57	0.57	−0.28	0.76	−0.15	0.86	−0.65	0.52
X₂	0.04	1.04	0.03	1.03	0.24	1.27	−0.22	0.81	0.17	1.18	0.11	1.12	0.23	1.25
X₃	0.34	1.40	0.19	1.21	0.36	1.44	1.06	2.88*	0.55	1.74	0.32	1.38	−0.84	0.43
X₄	0.36	1.43	−0.52	0.60	0.59	1.81	0.35	1.42	1.17	3.21	−0.64	0.53	1.16	3.20
X₅	−0.04	0.97	1.00	2.72	0.12	1.13	−1.17	0.31	0.50	1.65	−0.83	0.44	2.87	17.57*
X₆	−1.70	0.18	0.66	1.93	0.30	1.35	−2.45	0.09*	−0.04	0.96	−1.15	0.32	1.87	6.45
X₇	0.30	1.35*	0.08	1.08	0.19	1.20*	0.16	1.17*	0.13	1.14	0.17	1.18	0.31	1.37*
X₈	−0.20	0.82	−0.07	0.93	−0.02	0.98	0.02	1.03	−0.13	0.88	−0.15	0.86	0.20	1.22
Pseudo R²	.35		.11		.12		.28		.13		.17		.37

p < .05.

Using ordinary least squares regression, we also investigated the predictive value of animal cruelty methods, both individually and collectively, on later recurrent violent crimes. According to Table 5, one individual method of childhood animal cruelty (having sex with animals) predicted later adult violence against humans. Those inmates who reported having childhood sex with animals were more likely to have engaged in adult recurrent violent crimes. Collectively, approximately 10% of the total variance in the model was explained by the seven methods of childhood animal cruelty.

Table 5.

OLS Regression Solutions Predicting Recurrent Violent Crimes

	B	SE	β
Drowned	0.61	1.25	.06
Hit	0.25	0.99	.02
Shot	0.27	0.84	.03
Kicked	0.69	0.87	.09
Choked	1.45	1.19	.14
Burned	−0.21	1.18	−.02
Sex	2.66	0.90	.29*
Adjusted R²			.10
F value			2.64
Significance			.02

Coding of Independent Variables: (X₁) Drowned (0 = no, 1 = yes); (X₂); Hit (0 = no, 1 = yes); (X₃) Shot (0 = no, 1 = yes); (X₄) Kicked (0 = no, 1 = yes); (X₅) Choked (0 = no, 1 = yes); (X₆) Burned (0 = no, 1 = yes); (X₇) Sex (0 = no, 1 = yes). OLS = ordinary least squares.

p < .05.

Discussion

The purpose of this study was to replicate research that had examined the relationships between demographic (race, education, and residence while growing up) and situational factors (was the abuse committed alone, did the abuser attempt to conceal the act, was the abuser upset by the act, the abuser’s age at initial animal cruelty, and the frequency of their animal abuse) and specific methods of childhood acts of animal abuse (drowned, hit, shot, kicked, choked, burned, and sex). We also examined the predictive value of these different methods on adult acts of human violence. As noted earlier, this study represented a more refined version of a previous study (Tallichet et al., 2005) because it more clearly specified two methods of animal cruelty (hit and kicked) and included another form of human violence (robbery). In addition, the present study also measured the frequency of both animal abuse and the commission of adult violent crimes, not convictions of the latter, in our examination of the role methods may play in understanding the link between childhood animal abuse and later adult violence against humans.

We found that respondents who admitted to having abused animals during their childhood reported hitting an animal most frequently, whereas less than half of the time they admitted to kicking an animal, demonstrating that a distinction between the two methods exists among animal abusers in our sample. Based on these data, it can also be noted that acts such as kicking, hitting, or shooting were reported more frequently and may constitute the relatively more common and therefore “more understandable under certain circumstances” methods of abusive acts taken against a pet or stray animal. Drowning, choking, burning, or having sex with animals are less excusable as means of punishing a pet or stray animal, which may account for their less frequent occurrence and potentially more pathological nature. Rather, by cultural standards the latter set of methods could be seen as more “true” acts of animal cruelty for which there is little or no understandable or tolerable justification or rational explanation. In sum, both the frequencies of methods used to abuse animals and the patterns that emerged from the correlations between demographic and situational factors and methods suggest that some methods are more commonly used. This is not to suggest that any form of animal cruelty is normal or acceptable. Rather, certain individuals within an inmate population may consider certain forms of animal cruelty as less severe and other forms as completely unacceptable regardless of the circumstances. This is an area that needs to be further investigated in a qualitative study. Researchers could ask people from various populations how they view different types of animal cruelty and if they could imagine someone doing certain things under certain conditions. In such a study, researchers could control for those who have engaged in acts of animal cruelty and those who have not to see if, even among offenders, differences are found.

Respondents who reported that they had kicked animals as children were less likely to be upset by their actions than those animal abusers who reported using any of the other methods under study. These same respondents who had kicked animals were also more likely to have grown up in an urban area. Because many people who reside in cities live in apartments and thus may not have their own animals, it is possible that these people are less tolerant of animals and are more likely to kick animals in order to “shoo” them away if animals approach them. It is also possible that rural residents’ socialization toward and exposure with animals may make them less likely to kick animals as compared to their urban counterparts. Moreover, although respondents who abused animals with greater frequency either shot, kicked, drowned, or had sex with an animal, only those who had sex with an animal also reported concealing their actions. Finally, having sex with an animal during childhood or adolescence was the only method of abuse that predicted later recurrent adult violent crimes in our sample.

Similar to the Tallichet et al. (2005) study that used a different sample of inmates, we replicated the finding that individuals who reported having sex with an animal as children tended to conceal their actions, indicating the taboo nature of this form of animal cruelty. However, dissimilar to the previous study of animal cruelty methods and after refining the previous measures of human and animal violence, we did find support for the animal-to-human violence link, specifically that bestiality is an apparent precursor for later recurrent violent crimes in our present sample. The current study used measures that included a broader spectrum of human violence (i.e., the addition of robbery and the use of commission of violent crimes as opposed to conviction for violent crimes), further elaborating the link between the two behaviors. More specifically, this relationship replicates one of the findings by Hensley, Tallichet, and Singer (2006) in which bestiality was the sole method of animal cruelty under study. Moreover, the results of this study revealed that respondents with less education and those who had been convicted of committing crimes against people on one or more occasions were more likely to have had sex with animals during their childhood or adolescence as compared with other respondents in the sample.

In a generalized way, this finding regarding bestiality and later human violence complements the finding by Merz-Perez and Heide (2004) that more violent offenders used more severe acts of animal cruelty, requiring prolonged and close physical contact with their animal victims. It is important to keep in mind that our results also lend a measure of support to other studies showing that bestiality is most often found among violent offenders, sex offenders, and those individuals who have themselves been sexually abused. Our present study suggests that recurrent childhood acts of animal cruelty that involve prolonged and intense physical contact with an animal and that more than likely elicit a similarly intense and painful response on the part of the animal may portend the commission of subsequent human intraspecies violence. Therefore, both the 2006 bestiality study by Hensley et al. and our present study lend support to the sexual polymorphous theory, which purports that during their development certain individuals fuse aggressive tendencies with sexual ones (Merz-Perez & Heide, 2004). Infantile sexuality can assume a number of different forms. In certain individuals, “perverse sexual activities” are violent ones that fall outside the parameters of acceptable behavioral norms (Merz-Perez & Heide, 2004, p. 66). These activities, including bestiality, occur when “sexuality and aggression have become developmentally fused, and the two are mutually inclusive in the psyche of the offender” (Merz-Perez & Heide, 2004, p. 66). For these individuals, the only means for experiencing a release of these pent-up drives is to act on their violent and sexually related tendencies simultaneously. In sum, it is possible that inflicting pain on an animal and later on humans provides the same outlet and resulting relief.

Overall, earlier studies linking childhood animal cruelty and later adult violence were relatively inconclusive. However, more recent studies, without the limitations of these earlier ones, have demonstrated that recurrent acts of youthful animal cruelty are associated with subsequent acts of aggression against humans (Hensley et al., 2009; Merz-Perez et al., 2001; Merz-Perez & Heide, 2004; Tallichet & Hensley, 2004). Based on the current literature and at the very least, we may conclude that some individuals who do commit animal cruelty as children or adolescents do graduate to adult violence. The use of greater specificity regarding the methods by which animal cruelty is committed could help researchers identify the pathway by which this link is more likely to occur. However, we have yet to systematically study the effect of methods on identifying those who are so inclined. In the present study, relatively speaking, demographic variables, namely race, education, and residence while growing up, were not as important as situational factors when predicting animal cruelty methods. This finding suggests that future studies should focus on the social context in which animal abuse occurs and the perpetrators’ own experiences resulting from its commission.

Understanding the link between animal and human violence apparently requires researchers to further refine our attempts to define and study these phenomena. Specifically, using a sample with demographic characteristics that closely mirrored the prison population of the southern state from which it was drawn, the present study has furthered the examination of the link by measuring the frequency of both animal and human violence and sought to identify various methods commonly associated with those individuals who do graduate from animals to humans. The present study is not without limitations. First, by using a survey technique, the study was able to analyze a relatively larger sample than previous studies have done, but it also meant relying on paper-and-pencil self-reports. By doing so, illiterate inmates were excluded. Moreover, by relying on inmates’ self-reports, this study may have potentially compromised the validity of the sample delineation of violent and nonviolent inmates (Merz-Perez & Heide, 2004). Inmates may not have accurately reported their involvement in violent behavior. Second, although other prison studies dealing with sensitive topics have yielded relatively low response rates, our 10% return rate is very low for survey-based methodology. These conditions could possibly affect the generalizability of the study to the larger population of inmates both state- and nationwide.

Future studies could avoid some of the survey-related pitfalls by using direct interviews, which would yield much richer data, such as those conducted by Merz-Perez and Heide (2004). In addition, future studies should check self-reports against record data or official measures. These studies would also be desirable if the number of cases are sufficiently large to see if childhood animal cruelty is more strongly related to or predictive of specific types of violent crimes (i.e., homicide, rape, robbery, and aggravated assault). Either interview or survey-based studies could also include nonoffender and nonincarcerated populations and children who are known to have been physically or sexually mistreated, or both (Deviney, Dickert, & Lockwood, 1983), or children who have become emotionally and psychologically scarred from witnessing severe or repeated acts of animal cruelty (Ascione, 1988). Given what we know from the present study, identifying children at greatest risk for committing the type of animal cruelty that predisposes them to later adult violence could be problematic depending on the extent and nature of their concealment of their abuse of animals. Nonetheless, if researchers persevere, they may eventually be able to develop a model for identifying the early development of the animal-to-human progression in order to devise intervention and prevention programs that effectively utilize therapeutic and educational strategies to enhance the well-being of all species.

Footnotes

The author(s) declared no conflicts of interests with respect to the authorship and/or publication of this article.

The author(s) received no financial support for the research and/or authorship of this article.

References

Arluke

Levin

Luke

Ascione

F. R.

(1999). The relationship of animal abuse to violence and other forms of antisocial behavior. Journal of Interpersonal Violence, 14, 963-976.

Ascione

F. R.

(1988). Battered women’s report of their partner’s and their children’s cruelty to animals. Journal of Emotional Abuse, 1, 1-12.

Ascione

F. R.

(1993). Children who are cruel to animals: A review of research and implications for developmental psychopathology. Anthrozoos, 6, 226-247.

Ascione

F. R.

Thompson

T. M.

Black

(1997). Childhood cruelty to animals: Assessing cruelty dimensions and motivations. Anthrozoos, 10, 170-197.

Boat

(1994). Boat Inventory on Animal-Related Experiences. Cincinnati, OH: University of Cincinnati, Department of Psychiatry.

Deviney

Dickert

Lockwood

(1983). The care of pets within child abusing families. International Journal of the Study of Animal Problems, 4, 321-329.

Felthous

A. R.

(1980). Aggression against cats, dogs, and people. Child Psychiatry and Human Development, 10, 169-177.

Felthous

A. R.

Kellert

S. R.

(1987). Childhood cruelty to animals and later aggression against people: A review. American Journal of Psychiatry, 144, 710-717.

Flynn

C. P.

(1999). Animal abuse in childhood and later support for interpersonal violence in families. Society & Animals, 7, 161-172.

10.

Hensley

Tallichet

S. E.

Dutkiewicz

E. L.

(2009). Recurrent childhood animal cruelty: Is there a relationship to adult recurrent interpersonal violence? Criminal Justice Review, 34, 248-257.

11.

Hensley

Tallichet

S. E.

Singer

S. D.

(2006). Exploring the possible link between childhood and adolescent bestiality and interpersonal violence. Journal of Interpersonal Violence, 21, 910-923.

12.

Kellert

S. R.

Felthous

A. R.

(1985). Childhood cruelty toward animals among criminals and noncriminals. Human Relations, 38, 1113-1129.

13.

Merz-Perez

Heide

K. M.

(2004). Animal cruelty: Pathway to violence against people. Lanham, MD: Rowman & Littlefield.

14.

Merz-Perez

Heide

K. M.

Silverman

I. J.

(2001). Childhood cruelty to animals and subsequent violence against humans. International Journal of Offender Therapy and Comparative Criminology, 45, 556-573.

15.

Miller

K. S.

Knutson

J. F.

(1997). Reports of severe physical punishment and exposure to animal cruelty by inmates convicted of felonies and by university students. Child Abuse and Neglect, 21, 556-573.

16.

Tallichet

S. E.

Hensley

(2004). Exploring the link between recurrent acts of childhood and adolescent animal cruelty and subsequent violent crime. Criminal Justice Review, 29, 304-316.

17.

Tallichet

S. E.

Hensley

Singer

S. D.

(2005). Unraveling the methods of childhood and adolescent cruelty to nonhuman animals. Society & Animals, 13, 91-107.

18.

Tourangeau

Rips

L. J.

Rasinski

(2000). The psychology of survey response. Cambridge, UK: Cambridge University Press.