Hartpury Student Research Journal

Home » Uncategorized » The Graduation Hypothesis: does juvenile animal abuse lead onto adult interpersonal violence

The Graduation Hypothesis: does juvenile animal abuse lead onto adult interpersonal violence

Author Name: Heather Watts, BSc (Hons) Animal Behaviour and Welfare



Substantial research has supported the theory that children who abuse animals may become desensitised to the abuse, and therefore, are likely to move onto violence to humans in adulthood. This is termed ‘the graduation hypothesis’. However, much of this research is methodically limited and uses retrospective and cross-section approaches. More longitudinal studies are needed, with control groups, moving away from the focus on prison populations and self-report methods. This may improve accuracy of results and our understanding of juvenile animal cruelty and its relation to adult interpersonal violence. Alternative research has produced evidence in support of the ‘deviance generalisation hypothesis’, whereby violence to animals and violence to humans is generally related throughout the offenders’ lifetime. It is possible that results supporting the graduation hypothesis have produced ‘false positives’, as mainly the relationship between juvenile animal cruelty and adult interpersonal violence has been investigated, with general offending not included. Research into the methods of juvenile animal cruelty and their relation to adult interpersonal violence, has revealed bestiality as a risk factor for future human violence. However, the motivations for bestiality should be further investigated to confirm the risks to humans. Research into the motivations for animal cruelty and how they relate to future human violence needs more consideration of the full array of motives and social circumstances. Understanding juvenile animal cruelty and its relation to future interpersonal violence is of great interest to criminologists, sociologists, psychologists, legal scholars and social workers. Research may shape attitudes and laws, and reveal potential risk factors for future human violence.


1.0 Introduction

Throughout history, those with violent tendencies towards animals, have been said to likely be a risk to humans also (Flynn, 2011). Animal cruelty is often related to many aggressive and antisocial behaviours, such as assault, rape and domestic violence (McPhedran, 2009). Interest has increased in the putative relationship between childhood animal cruelty and its progression onto adult interpersonal violence, often termed ‘the graduation hypothesis’ (Walters, 2013). This hypothesis postulates that abusing and killing animals provides an opportunity to experiment with abuse, desensitising the child, and serving as a stepping stone or starting point to later progress onto adult human violence (Walters, 2014). There are a wealth of case studies involving infamous murderers, who are said to have abused animals in their youth (Petersen and Farrington, 2007). Carroll Cole, Jeffrey Dahmer, Edmund Kemper, Henry Lee Lucas, and Arthur Shawcross are among the list of serial killers who abused and killed animals during their troubled childhoods (Wright and Hensley, 2003). Psychiatric interest began in the 1960’s (Macdonald, 1967) and much support for this hypothesis has arisen since the 1990’s (Patterson-Kane and Piper, 2009). However, conflicts exist in published research regarding the validity of the ‘link’ (Hensley, Tallichet, and Dutkiewicz, 2009). Research is largely underpinned by anthropocentric approaches, with animal abuse often labelled as a ‘red flag’ for interpersonal violence (Taylor and Signal, 2008). Critics suggest that rather than concentrating on abuse at a psychological or personal level, social and cultural levels should be more thoroughly investigated (Taylor and Signal, 2008). The treatment of animals and what is socially acceptable may vary between cultures (Flynn, 2011), therefore harming animals may not necessarily indicate underlying psychosis, if it is a societal norm (Patterson-Kane and Piper, 2009). Accurately understanding animal abuse and interpersonal violence is important to identify risk factors that may cause possible future violence, facilitating prevention (Lockwood and Arkow, 2016). Preventing animal cruelty in juveniles may also provide positive benefits to humans through reducing the likelihood of violence escalating to adult human violence (Petersen and Farrington, 2007).


2.0 The Graduation Hypothesis

Substantial research produced since the 1980’s has supported the graduation hypothesis (e.g. Hensley and Tallichet, 2004; Ressler et al., 1986; Wright and Hensley, 2003). However, it has been suggested that this research has not been supported by enough sound evidence (Ascione and Shapiro, 2009). For example, Wright and Hensley (2003) analysed 5 case studies of infamous serial murderers who abused animals in their youth to gather evidence for the graduation hypothesis. The validity of this can be questioned as there are also a vast number of serial killers that did not abuse animals as children (Arluke et al., 1999), and it is problematic to assume all those who harm animals will develop into serial killers (Patterson-Kane and Piper, 2009). A study by Simons, Wurtele, and Durham (2008) investigating the childhood experiences of child sexual abusers and rapists revealed that 38% of child sexual abusers participated in sexual activities with animals during their developmental period. In addition, 68% of rapists reported regular experiences of participating in animal cruelty during childhood (Simons, Wurtele, and Durham, 2008). However, the animal offences committed were combined with other detrimental developmental experiences, such as experiencing sexual and physical abuse, therefore it cannot be concluded that animal cruelty in childhood alone caused future interpersonal violence. The study was also reflective of adult sexual offenders only, hence the prevalence of juvenile animal abuse in the general population is unknown, as is the relation of juvenile animal abuse to other forms of non-sexual adult interpersonal violence. Hensley, Tallichet, and Dutkiewicz (2009) investigated the relationship between recurrent childhood animal cruelty and adult recurrent interpersonal violence using a survey sample of 180 male prisoners. The results indicated that those who participated in a higher number of childhood animal abuse acts were more likely as adults to have committed repeated interpersonal crimes. Demographic factors were not significant. However, the findings were taken from prison inmates, which are an extreme group, and not representative of the general population (Flynn, 2011). These kinds of participants may even be inclined to exaggerate answers to keep up a ‘tough’ persona, which may produce inaccurate results (Flynn, 2011). The reliability of self-report research can be questioned (Kimberlin and Winetrstein, 2008), especially involving subjects where people may not want to admit to harming an animal or human (Patterson-Kane and Piper, 2009). This may apply more to members of the general population who may wish to avoid criminal convictions. However, prison inmates may be given reduced sentences for good behaviour (Couture and Scheller, 2009), therefore, may not want to confess to extra crimes. The questionnaire received a 10% response rate, hence, may not be a fair representation of the prison population. The use of paper questionnaires may have excluded illiterate inmates, therefore, conducting interviews may be a more effective way to make sure that illiterate inmates are not excluded from the data collection (Hensley, Tallichet, and Dutkiewicz, 2009). The sample consisted of all male inmates, thus, gender differences in human-animal interactions were excluded. However, there is evidence that boys may commit more animal cruelty than girls (Herzog, 2007). The definition of cruelty, ‘how many times have you ever hurt or killed animals, other than hunting’ may also be problematic. A participant may have killed an animal as an act of mercy, or have worked in an abattoir, but this would be counted as abuse.

Walters (2014) tested the relationship between childhood animal cruelty and subsequent aggressive and non-aggressive offending. A larger sample size of 1,336 inmates, including 182 females was collected, compared to the Hensley, Tallichet, and Dutkiewicz (2009) study which had 180 males. The sample is still highly biased in favour of men though, which may not fully represent gender differences. The participants used for the study were the same participants from the pathways to resistance study (Mulvey, 2013) who were judged delinquent between the ages of 14-18. Participants were asked ‘Did you ever physically hurt an animal on purpose?’. Findings revealed that juvenile animal abuse was a precursor to non-aggressive and aggressive offending. This does not support the graduation hypothesis as violence to humans did not always follow, rather general offending. This is more in line with the ‘deviance generalisation’ hypothesis, whereby animal abuse and many other forms of general offending positively correlate, making no postulation of time-order, but based on the idea that those who abuse animals are likely to commit other forms of crime also (Arluke et al., 1999). Volant et al., (2008) state that it is more likely that human and animal violence are present throughout the offenders’ life time, rather than children ‘graduating up’. There was no evidence that age, sex, race and early onset of behavioural problems had an effect on the animal cruelty following offending relationship. The findings by Walters (2014) may be more accurate than the study by Hensley, Tallichet, and Dutkiewicz (2009), as prospective data was gathered from a longitudinal study. Hensley, Tallichet, and Dutkiewicz (2009) also only included interpersonal crimes in their study, and did not account for general offending as did Walters (2014), hence, the probability of the deviance generalisation hypothesis was excluded. This may have produced a false positive result for the graduation hypothesis. Merz-Perez, Heide, and Silverman, (2001) also investigated aggressive and non-aggressive offending in incarcerated individuals and the link to juvenile animal abuse. Their results differ to Walters (2014) such that they found that juvenile animal abuse was predictive of adult human violence, but not non-violent offending, thus supporting the graduation hypothesis. However, a small sample of 90 participants were used, which may make the results less accurate, compared to the 1,336 in the study by Walters (2014).


2.1 Childhood Animal Cruelty Methods and Adult Interpersonal Violence

Hensley and Tallichet (2009) investigated juvenile animal cruelty methods (drowned, hit or kicked, shot, choked, burned, and had sex) and their relation to adult interpersonal violence (assault, murder and rape), through a survey of 261 inmates. Results revealed that participants who had drowned or had sex with an animal, were more likely to have committed adult interpersonal violence. Hensley and Tallichet (2009) propose that this may be because these types of abuse are more ‘hands on’, and involve overpowering an animal, which may result in the same treatment of humans. Abuse such as bestiality may be severely underreported by juveniles (Schenk et al., 2014), which may be similar for adults, possibly understating results. However, as this is a retrospective study it may be flawed by recall bias, which could create inaccuracies in the results through incorrect recollections (Holoyda and Newman, 2016). Similarly, to Hensley, Tallichet, and Dutkiewicz (2009), the study used incarcerated individuals, and the questionnaire yielded a low response rate (10%) hence not providing a fair representation of the prison population. A prospective study, involving a group of juvenile animal abusers and a non-abusing control group and the outcomes of their adult behaviour may provide more accurate longitudinal research (Holoyda and Newman, 2016). Hensley, Tallichet and Dutkiewicz (2010) further investigated childhood bestiality and its link to adult interpersonal crime. A small sample size of 180 incarcerated individuals were used, with again only a 10% response rate to the survey, questioning the validity of the sample. Similarly, to the study by Hensley and Tallichet (2009), the results revealed that childhood bestialics were more likely to have committed interpersonal violence than those who were not. Hensley, Tallichet and Dutkiewicz (2010) suggest that this is due to lack of empathy and violent and sexual behaviours being joined during development, resulting in the enjoyment of overpowering victims. However, unlike Hensley and Tallichet (2009), drowning was not a significant predictor for future interpersonal violence. Hensley, Tallichet and Dutkiewicz (2010) added robbery and aggravated assault to violent history options, and this may have had an effect on results. Although race was accounted for, only two groups were presented ‘white’ or ‘other’. Cultural differences may have an impact on how animals are treated (Knight and Herzog, 2009), therefore, a broader option of ethnicities may have been more useful in determining if culture has an impact on results. The type of animal may also be a valid measure to record, relating to socially acceptable cultural differences (Patterson-Kane and Piper, 2009). The participants involved mostly had non-bestiality convictions, therefore, were not likely to have zoophilic tendencies, and may have been generally violent to humans, which may restrict the applicability of results (Holoyda and Newman, 2016). It is possible that those who have zoophilic tendencies, do not do so out of aggression, but sexual attraction, and wanting to express affection towards the animal (Holoyda and Newman, 2014). These ‘pure zoophiles’ may not necessarily pose a risk to humans (Holoyda and Newman, 2014). Using a classification scheme based on the motivations for bestiality, may help to elucidate the risks to humans, as those doing so out of violence or cruelty may be more of a threat (Holoyda and Newman, 2016). In 2011, Henderson, Hensley, and Tallichet, replicated the study by Hensley and Tallichet (2009) to further research animal cruelty methods and their link to adult interpersonal violence. The results revealed animal cruelty starting at a younger age, drowning animals and bestiality to be significant predictors for adult interpersonal violence, results consistent with their previous study (Hensley and Tallichet, 2009). This potentially provides more support for childhood bestiality being linked to adult interpersonal violence.


2.2 Childhood Animal Cruelty Motives and Adult Interpersonal Violence

Research into the motives for childhood animal cruelty linking to adult interpersonal crime has proved contradictory (Holoyda and Newman, 2016). Hensley and Tallichet (2008) conducted a study to determine if particular motives for animal cruelty had a link to later interpersonal violence. A sample of 261 inmates completed a survey on the types of violent interpersonal convictions obtained, and reasons for juvenile animal abuse (fun, anger, dislike, imitation). The results indicated that abusing animals for fun was significant for predicting adult interpersonal crime. Control, punishment, cultural prejudice, attempting to impress others, revenge, and displaced aggression are some of the common reasons for animal cruelty (Lockwood and Arkow, 2016), which were not included in the study by Hensley and Tallichet (2008). The motives for animal abuse may be far more complex, and depend on an array of circumstances (Lockwood and Arkow, 2016), therefore, four motivational categories do not seem robust enough to give this study meaningful results. Children from violent families that witness animal abuse (DeGue and DiLillo, 2009), and juveniles who are victims of, or carry out physical bullying are more likely to commit animal abuse themselves (Henry and Sanders, 2007). Animal abuse may have occurred through fear of the animal, or ignorance, such that possibly the child did not realise the animal could suffer (Pagani, Robustelli, and Ascione, 2010). Therefore, it would be beneficial to ask the participants’ personal history to make motives clearer. When investigating frequencies of crimes, rather than asking the inmates if they had been convicted of a crime, it may have been more beneficial to ask them if they had committed the crime, as they may not necessarily have been convicted. The presumption that inmates would also understand the legalistic terms included in the survey may also be problematic. In 2012, Overton, Hensley, and Tallichet replicated the study by Hensley and Tallichet (2008), investigating childhood animal cruelty motives and their link to adult interpersonal violence. The same survey was distributed to a different prison, robbery was added to the choice of violent crime and 180 inmates responded. The study produced contradictory results to Hensley and Tallichet (2008), with none of the motives being a significant predictor of adult interpersonal violence. Only recurrent childhood animal abuse was predictive of later recurrent violent crime. However, robbery was not defined to the participants; directly stealing possessions from an individual would be considered interpersonal violence, but stealing from a house or shop would not (Perreault and Brennan, 2010). Participants may have been included into the ‘robbery’ category, but possibly have not committed interpersonal crime, affecting the accuracy of results. It is also possible that the complexity of a full range of motives was not investigated thoroughly enough. Using a different prison and receiving contradictory results may highlight the need for a larger sample size when investigating these populations.


3.0 Conclusion

Research has supported the theory of childhood animal abuse being likely to lead onto adult human violence. However, some major methodological issues are present that are occurring throughout most of the studies. Small samples of male incarcerated individuals are repeatedly used, which are atypical of the prison population and certainly of the general population. Clearer definitions of animal abuse are needed, with investigation into personal social history and cultural factors. Future studies should move away from incarcerated participants and include those who may have abused animals but did not graduate to human violence. Self-report, cross-sectional and retrospective research may be limiting results and more longitudinal studies are needed. Most research is correlational in nature which does not help to identify the causation or time-order. Groups of juvenile animal abusers could be studied, with a control group of non-abusers, their adult behaviour could be recorded to see the likely progression onto human violence. It is possible that this kind of research has not been conducted due to the ethically sensitive methods involved, or has commenced but the results will not be available for a considerable time. Classification schemes looking at the motives for childhood bestiality are needed to realise the true threat to humans. The complexity of animal abuse motives during childhood may be the reason for contradictory results, and including detrimental social experiences to the participants may improve understanding. There is substantial evidence supporting that whenever humans are at risk, animals are too, and when animals are at risk, so are humans. It is possible that theories such as the deviance generalisation hypothesis are more sound. However, the graduation hypothesis should not be dismissed. Continuously replicating studies without improving on their weaknesses is not advancing research. Only once the major limitations are addressed, can we truly begin to understand and draw conclusions regarding the relationship between childhood animal cruelty and adult interpersonal violence.



Arluke, A., Levin, J., Luke, C., and Ascione, F. (1999) The relationship of animal abuse to violence and other forms of antisocial behavior. Journal of Interpersonal Violence [online]. 14 (9), pp. 963-975. [Accessed 18th October 2017].

Ascione, F.R. and Shapiro, K. (2009) People and animals, kindness and cruelty: Research directions and policy implications. Journal of Social Issues [online]. 65 (3), pp. 569-587. [Accessed 15th October 2017].

Couture, G.A. and Scheller, F. (2009) How Should the Bureau of Prisons Calculate the Credits that Federal Inmates Receive for Good Conduct While in Prison (09-5201). Preview US Sup [online]. 37, pp. 290. [Accessed 7th November 2017].

DeGue, S. and DiLillo, D. (2009) Is animal cruelty a “red flag” for family violence? Investigating co-occurring violence toward children, partners, and pets. Journal of Interpersonal Violence [online]. 24 (6), pp.1036-1056. [Accessed 30th October 2017].

Flynn, C.P. (2011) Examining the links between animal abuse and human violence. Crime, Law and Social Change [online]. 55 (5), pp. 453-468. [Accessed 14th October 2017].

Henderson, B.B., Hensley, C. and Tallichet, S.E. (2011) Childhood animal cruelty methods and their link to adult interpersonal violence. Journal of interpersonal violence [online]. 26 (11), pp. 2211-2227. [Accessed 20th October 2017].

Hensley, C. and Tallichet, S.E. (2004) Exploring the Link between Recurrent Acts of Childhood and Adolescent Animal Cruelty and Subsequent Violent Crime. Criminal Justice Review [online]. 29 (2), pp. 304-316. [Accessed 20th October 2017].

Hensley, C. and Tallichet, S.E. (2008) The Effect of Inmates’ Self-Reported Childhood and Adolescent Animal Cruelty: Motivations on the Number of Convictions for Adult Violent Interpersonal Crimes. International Journal of Offender Therapy and Comparative Criminology [online]. 52 (2), pp. 175-184. [Accessed 20th October 2017].

Hensley, C. and Tallichet, S.E. (2009) Childhood and adolescent animal cruelty methods and their possible link to adult violent crimes. Journal of interpersonal violence [online]. 24 (1), pp. 147-158. [Accessed 15th October 2017].

Hensley, C., Tallichet, S.E. and Dutkiewicz, E.L. (2009) Recurrent childhood animal cruelty: Is there a relationship to adult recurrent interpersonal violence? Criminal Justice Review [online]. 34 (2), pp. 248-257. [Accessed 14th October 2017].

Hensley, C., Tallichet, S.E. and Dutkiewicz, E.L. (2010) Childhood bestiality: A potential precursor to adult interpersonal violence. Journal of interpersonal violence [online]. 25 (3), pp. 557-567. [Accessed 15th October 2017].

Henry, B.C. and Sanders, C.E. (2007) Bullying and animal abuse: Is there a connection? Society and Animals [online]. 15 (2), pp. 107-126. [Accessed 23rd October 2017].

Herzog, H.A. (2007) Gender differences in human–animal interactions: A review. Anthrozoös [online]. 20 (1), pp. 7-21. [Accessed 15th October 2017].

Holoyda, B. and Newman, W. (2014) Zoophilia and the law: legal responses to a rare paraphilia. Journal of the American Academy of Psychiatry and the Law Online [online]. 42 (4), pp. 412-420. [Accessed 22nd October 2017].

Holoyda, B.J. and Newman, W.J (2016) Childhood animal cruelty, bestiality, and the link to adult interpersonal violence. International Journal of Law and Psychiatry [online]. pp. 129. [Accessed 15th October 2017].

Kimberlin, C.L. and Winetrstein, A.G. (2008) Validity and reliability of measurement instruments used in research. American Journal of Health-System Pharmacy [online]. 65 (23). [Accessed 15th October 2017].

Knight, S. and Herzog, H. (2009) All creatures great and small: New perspectives on psychology and human–animal interactions. Journal of Social Issues [online]. 65 (3), pp. 451-461. [Accessed 20th October 2017].

Lockwood, R. and Arkow, P. (2016) Animal Abuse and Interpersonal Violence: The Cruelty Connection and Its Implications for Veterinary Pathology. Veterinary Pathology [online]. 53 (5), pp. 910-918. [Accessed 22nd October 2017].

Macdonald, J.M. (1967) Homicidal threats. American Journal of Psychiatry [online]. 124 (4), pp. 475-482. [Accessed 22nd October 2017].

McPhedran, S. (2009) A review of the evidence for associations between empathy, violence, and animal cruelty. Aggression and Violent Behavior [online]. 14 (1), pp. 1-4. [Accessed 14th October 2017].

Merz-Perez, L., Heide, K.M. & Silverman, I.J. (2001) Childhood Cruelty to Animals and Subsequent Violence against Humans. International Journal of Offender Therapy and Comparative Criminology [online]. 45 (5), pp. 556-573. [Accessed 16th November].

Mulvey, E. P. (2013) Research on pathways to resistance Maricopa County, AZ and Philadelphia County, PA Subject Measures, 2000‐2010. ICPSR29961‐v2. Ann Arbor, MI Inter‐University Consortium.org Political and Social Research distributor, 2013‐01‐07.

Overton, J.C., Hensley, C. and Tallichet, S.E. (2012) Examining the Relationship Between Childhood Animal Cruelty Motives and Recurrent Adult Violent Crimes Toward Humans. Journal of Interpersonal Violence [online]. 27 (5), pp. 899-915. [Accessed 22nd October 2017].

Pagani, C., Robustelli, F. and Ascione, F.R. (2010) Investigating animal abuse: Some theoretical and methodological issues. Anthrozoös [online]. 23 (3), pp.259-276. [Accessed 30th October 2017].

Patterson‐Kane, E.G. and Piper, H. (2009) Animal abuse as a sentinel for human violence: A critique. Journal of Social Issues [online]. 65 (3), pp. 589-614. [Accessed 14th October 2017].

Perreault, S. and Brennan, S. (2010) Criminal victimization in Canada, 2009. Juristat: Canadian Centre for Justice Statistics [online]. 30 (2), pp. 1G. [Accessed 7th November 2017].

Petersen, M. and Farrington, D.P. (2007) Cruelty to animals and violence to people. Victims and Offenders [online]. 2 (1), pp. 21-43. [Accessed 23rd October 2017].

Ressler, R.K., Burgess, A.W., Hartman, C.R., Douglas, J.E. and McCormack, A. (1986) Murderers who rape and mutilate. Journal of Interpersonal Violence [online]. 1 (3), pp. 273-287. [Accessed 22nd October 2017].

Schenk, A.M., Cooper‐Lehki, C., Keelan, C.M, and Fremouw, W.J. (2014) Underreporting of Bestiality Among Juvenile Sex Offenders: Polygraph Versus Self‐Report. Journal of forensic sciences [online]. 59 (2), pp. 540-542. [Accessed 7th November 2017].

Simons, D.A., Wurtele, S.K. and Durham, R.L. (2008) Developmental experiences of child sexual abusers and rapists. Child Abuse and Neglect [online]. 32 (5), pp. 549-560. [Accessed 15th October 2017].

Tallichet, S.E. and Hensley, C. (2004) Exploring the link between recurrent acts of childhood and adolescent animal cruelty and subsequent violent crime. Criminal Justice Review [online]. 29 (2), pp. 304-316. [Accessed 20th October 2017].

Taylor, N. and Signal, T. (2008) Throwing the Baby out with the Bathwater: Towards a Sociology of the Human-Animal Abuse ‘Link’? Sociological Research [online]. 13 (1), pp. 2. [Accessed 30th October 2017].

Volant, A.M., Johnson, J.A., Gullone, E. and Coleman, G.J. (2008) The relationship between domestic violence and animal abuse: An Australian study. Journal of interpersonal violence [online]. 23 (9), pp. 1277-1295. [Accessed 15th October 2017].

Walters, G.D. (2014) Testing the specificity postulate of the violence graduation hypothesis: Meta-analyses of the animal cruelty–offending relationship. Aggression and violent behaviour [online]. 18 (6), pp. 797-802. [Accessed 14th October 2017].

Wright, J. and Hensley, C. (2003) From Animal Cruelty to Serial Murder: Applying the Graduation Hypothesis. International Journal of Offender Therapy and Comparative Criminology [online]. 47 (1), pp. 71-88. [Accessed 14th October 2017].