From Animal Abuse to Interhuman Violence? A Critical Review of the Progression Thesis

Piers Beirne

ABSTRACT
This paper reviews evidence of a progression from animal abuse to interhuman violence. It finds that the “progression thesis” is supported not by a coherent research program but by disparate studies often lacking methodological and conceptual clarity. Set in the context of a debate about the theoretical adequacy of concepts like “animal abuse” and “animal cruelty,” it suggests that the link between animal abuse and interhuman violence should be sought not only in the personal biographies of those individuals who abuse animals but also in those institutionalized social practices where animal abuse is routine, widespread, and socially acceptable.

Key words: animal abuse, assaultive children, cruelty, institutionalized abuse, longitudinal analysis

Impassioned claims of a significant relationship between nonhuman animal abuse and interhuman violence have been made by such diverse thinkers as Pythagoras, Thomas Aquinas, Immanuel Kant, Mary Wollstonecraft, Mahatma Gandhi, and Margaret Mead. Espoused by its holders at a high level of abstraction and disseminated in the mantra-like catchphrase “The Link,” it nowadays is advanced most prominently by members of state agencies and philanthropic organizations who work with abused animals and/or at-risk families. It also implicitly appears in the writings of moral philosophers and feminists on animal welfare and animal rights. By the mass media and by numerous practitioners and activists in the animal protection community, moreover, knowledge claims about The Link are projected as indisputable scientific fact with urgent policy ramifications.

This paper, too, assumes that animal abuse and interhuman violence are linked in a concatenation of sites, but it reviews evidence of only one aspect of this “animal abuse web” (Solot, 1997), namely, whether there is a progression from animal abuse to interhuman violence. The chronological causal relationship posited between animal abuse and interhuman violence I term, “the progression thesis.” As an embryonic idea about human-animal interaction, the progression thesis originated in the 1980s, but, as a more focused object, it has appeared only in the last decade, chiefly in the United States. Recently, it has garnered interest in some other countries--including Australia, Belgium, Canada, England, Italy, Scotland, and Wales.

Among scholars of human-animal interaction, most assessments of the progression thesis, extended or brief, currently lie on a continuum between possible disconfirmation (Miller & Knutson, 1997; Arluke, Arnold, Levin, Luke, & Ascione, 1999) and a cautious attitude of wait-and-see (Dadds, Turner, & McAloon, 2002). Few would subscribe to the ironclad determinism embedded in the view of Farrington (2002), past president of the American Society of Criminology that [p]eople graduate from hyperactivity at age two to cruelty to animals at age six, shoplifting at ten, burglary at fifteen, robbery at twenty, and eventually spouse assault, child abuse and neglect, alcohol abuse, and employment and later health problems later on in life. (p. 58)

Confirmation of the progression thesis ultimately depends on the successful combination of two quite separate propositions. Chronologically and causally, one proposition looks forward, the other, backward. In the one, those who abuse animals must be more likely than those who do not subsequently to act violently toward humans. In the other, those who act violently toward humans must be more likely than those who do not previously to have abused animals. Logically, these propositions need not entail strict Humean causality. Robust and persistent statistical association will suffice. If the association is found to be robust, then how is it explained? What is its direction? Might some other variable(s) influence it?

In reviewing the merits of the progression thesis, I begin with its originating site, which is commonly lodged in the social dynamics of families in crisis. Among the chief dysfunctional qualities of these families is their propensity for interpersonal violence, to whose stated links with animal abuse I now turn.

Family Violence and Companion Animal Abuse

It is well established that different forms of family violence tend to coexist (Widom, 2000). If a male is battering his spouse, then children in that household also are more likely to be abused or neglected there, either by adults or by siblings.
Does this mean that companion animals are more likely to be abused in a household experiencing interhuman abuse? Unfortunately, this question cannot be addressed in the way that studies of other crimes routinely are, namely, through analysis of official (government-generated) data on intra-familial animal abuse. There are none. Indeed, no technologically advanced society has generated large-scale, police-based data on the incidence and prevalence of animal abuse. There are no large-scale self-report studies on animal abuse, no household victimization surveys.

These absences can be overcome partially, however, with the aid of sometimes focused, sometimes incidental data generated in social science research. These suggest that in situations of intra-familial conflict animals often are used as instruments of psychological and physical terror by one human against another or as objects against whom humans vent aggression, whether pent-up or learned. Precisely because the several forms of family violence tend to cluster and because companion animals usually are regarded as family members, in families where any given form of family violence exists, animal abuse also is more likely to exist. Empirical evidence indicates that companion animal abuse often occurs disproportionately in diverse situations of family violence, including the following:

1. heterosexual partner abuse (Baldry, 2003; Flynn, 2000a, 2000b; Ascione, 1998; Ascione, Weber & Wood, 1997; Arkow, 1996; Boat, 1999; Arkow, 1994; Gelles & Straus, 1988 pp. 68, 119; Browne, 1987, p. 157; Donley, Patronek & Luke, 1999; Patronek, 1997).
2. lesbian partner abuse (Renzetti, 1992, p. 21).
3. child physical abuse (Deviney, Dickert & Lockwood, 1983; Munro, 1996; Guymer, Mellor, Luk & Pearse, 2001; Munro & Thrusfield, 2001);
4. child sexual abuse at home (Friedrich, Urquiza, & Beilke, 1986; Hunter, 1990, pp. 214-216; Boat, 1999) and in day care centers (Finkelhor, Williams, & Burns, 1988; Kelley, 1989, p. 508); Faller, 1990, pp. 199-201; and
5. sibling abuse (Wiehe, 1990, pp. 44, 45).

One study found that 60% of families with abused children also had pets abused there; fathers had abused two-thirds of the animals, children the remainder (Deviney, et al., 1983). In a study of lesbian partner abuse 38% of respondents with companion animals reported that their partners had abused their pets (Renzetti, 1992, p. 21). These findings have been supported by several studies of battered women seeking refuge in shelters. Ascione (1998) asked abused women arriving in a Utah shelter about the incidence of animal abuse committed by their partners and their children. Here, 71% of the pet-owning women said that their partners had killed or mistreated one or more of their pets or that they had threatened to do so. Women with children reported in 32% of the cases that one or more of their children had abused or killed companion animals (Ascione, 1998; Ascione, et al., 1997). Moreover, in Flynn’s (2000b) study of battered women in a South Carolina shelter, of 43 women with pets, 20 (46.5%) reported that their male abuser had threatened to harm, or actually had harmed, their pets.

Does the finding that companion animal abuse disproportionately occurs with other forms of family violence lend credence to the progression thesis? Before answering this question, I must emphasize that a strength of this particular finding is the diversity of its data sources, including not only structured interviews with battered women and abused children but also self-report studies and information from veterinarians, animal control officers, animal shelters, women’s shelters, and police. Clearly, family violence, including animal abuse, is a multifaceted phenomenon in which various forms of abuse often occur together and in which the presence of one form might signify the existence of others. It is likely, too, that some of the key sociological dimensions of animal abuse mirror those of interhuman violence. Besides the predominance of males in animal abuse by adults, young males very likely commit animal abuse far more frequently than do young females--when they do, their abuse often is considerably more egregious.
However, there are nagging gaps and inconsistencies in existing research. Though households with animal abuse are more likely also to suffer from interhuman violence, nothing precise is known about the prevalence of animal abuse among young males and young females. To mention one area of uncertainty, Miller and Knutson (1997, p. 77) found that 20.5% of a sample of 308 Iowan undergraduate psychology students (with a slight over-representation of females) reported that they actually had engaged in one or more acts of animal cruelty. However, from a sample of undergraduate psychology and sociology students at a southeastern university in the United States, Flynn (1999, pp. 165, 166) found that 34.5% of males and 9.3% of females admitted that they had abused animals during childhood. Moreover, much higher rates of animal abuse than these have been reported by Baldry (2003). In her study of animal abuse and exposure to interparental violence among Italian youth aged 9 to 17, Baldry (p. 272) found that 50.8% of the 1,392 youth in her study had abused animals at least once; 66.5% were boys.

How can we explain such discrepant findings? Do they mean Italian youth are more abusive than American youth? Probably not, though without more information this cannot be dismissed. Are the discrepancies purely random? With so few studies, there is no way of knowing. What the different findings probably mean is that data about the prevalence of animal abuse in any given population reflect not the actual prevalence of animal abuse there as much as they do other factors such as the nature and sensitivity of the survey instrument and subjects’ variable willingness to self-report animal abuse.

In respect of this last possibility, as Baldry (2003, p. 274) herself is well aware, most earlier studies, including Miller and Knutson (1997) and Flynn (1999), have tended to focus on relatively extreme forms of animal abuse, whereas her broader operationalization of the concept included any form of hitting, tormenting, bothering, harming, or cruelty. Baldry’s discovery of higher rates of prevalence among Italian youth flowed directly, therefore, from her extended definition of animal abuse. To the question, “how much animal abuse is there?” one is tempted to answer, “as much as you are willing to find.”

Baldry (2003) also persuasively shows how, in trying to discover what factors precipitate children’s animal abuse, one must examine the nature of the family violence to which children are exposed. Have they witnessed family violence without being its direct victims? Were the victims humans or animals? Or, have the children been actually the direct victims of family violence and, if so, were the offenders male or female? Examination of such questions in the area of human-animal interaction is important, moreover, because it is widely known in criminology that juvenile victims of violence risk developing a variety of psychological difficulties in interpersonal relationships and, soon afterwards, are themselves more likely to act violently both against other humans and themselves (Shaffer & Ruback, 2002).

Existing research on how often, how seriously, and in what ways companion animal abuse exists with other forms of family violence therefore tends neither to confirm nor disconfirm the progression thesis. Although there is no good reason to suppose that the etiology of companion animal abuse differs markedly from that of the abuse of human family members, nothing systematic is known about the direction of abuse. Though animal abuse and interhuman violence are linked because they occur disproportionately in the same households, this tendency reveals nothing about a possible developmental movement from one to the other. It is unclear whether men who batter their spouses tend previously to have abused animals. Do they perhaps begin a cycle of violence by abusing animals and their partners concurrently? Do they abuse their partners first and later abuse animals? We can similarly question the misbehavior of children. Do young boys witness their father abusing their mother and then later abuse an animal? Are they more likely to do this if, instead of witnessing violence, they are its direct victims? Is this process one of social learning motivated by anger? What of older siblings--do they abuse their younger siblings first and abuse animals afterward, or do they begin by abusing animals?

These questions must be addressed before the undoubted propensity of animal abuse to coexist with other forms of family violence can be inserted into a full assessment of the merits of the progression thesis. At present, therefore, this segment of the evidence about the progression thesis is inconclusive.

Animal Abuse and the Futures of Assaultive Children

The first proposition of the progression thesis is that those who abuse animals are more likely subsequently to act violently towards humans. It has been suggested by the mass media that some young male animal abusers later commit homicide.
Yet, how might one test whether adolescents who abuse animals are more likely than those who do not subsequently to act violently toward humans? Because longitudinal analysis never has been applied to the progression thesis, any current assessment of its status must settle for a re-working of cross-sectional research on children and adolescents that has been generated in a hodge-podge of intellectual and social milieux (Ascione, 2001). Three main avenues of research have been directed to children who abuse animals (“assaultive children”). Some studies have claimed to detect, first, mental and characterological defects in assaultive children (Patterson, DeBaryshe, & Ramsey, 1989; Ascione, 1993, pp. 233-235; Ascione, Thompson, & Black, 1997; Achenbach, Howell, Quay, & Conners, 1991; Reber, 1996). Assaultive children sometimes are described as having multiple personality and dissociative disorder; indeed, the American Psychiatric Association (2000, p. 99) identifies physical cruelty to animals as a diagnostic criterion for conduct disorder. Lack of proper modeling, peer reinforcement, post-traumatic play, hostility displacement, and suicidal tendencies have all been described variously as the personality characteristics of assaultive children (Boat, 1999).

Following the lead of Hellman and Blackman (1966), some researchers have found, second, that assaultive children disproportionately display other anti-social tendencies, including nonproductive firesetting (Slavkin, 2001) and enuresis, though this triad has been disputed and contradicted (Justice, Justice, & Kraft, 1974; Wooden, & Berkey, 1984, pp. 35-37, 56,57; Youssef, Attia, & Kamel, 1999). Yet, even Macdonald (1968, pp. 109, 110), an early popularizer of the triad, cast doubt on its utility in predicting homicide. It would be interesting to know what proportion of “violence-prone” or “at-risk” youth also engage in animal abuse, yet no large-scale studies of such youth have examined the possible significance of animal abuse.

Third, assaultive children also have been found to be overwhelmingly young, male, and of normal intelligence (Flynn, 1999, p. 165; Tapia, 1971; Felthous, 1981; Felthous & Yudowitz, 1977); often sexually abused at home (Friedrich et al., 1986; Hunter, 1990, p. 214-216) or physically abused and neglected there (Deviney et al., 1983; and often living in a family situation of spouse abuse. However, these findings reveal nothing about the claimed chain of causation from animal abuse to interhuman violence and serve only to open up an array of other, equally unresolved questions. Why are assaultive children overwhelmingly male? After polite nods to concepts like the socialization process and defective personality, existing explanations are either too individualist or prone to biological reductionism, including vague assertions about innate male aggressiveness. In abusing the most available living beings who are unable to offer resistance to them--dogs, cats, fish, birds and reptiles--perhaps young boys are mimicking their fathers’ violence against their mothers and sisters. Does this mean that their witnessing of others’ interhuman violence precedes some children’s animal abuse? How does this affect the progression thesis? Moreover, if the original tendencies that propel some children to abuse animals are so ironclad, then why do most young males desist eventually from abusing them?

Given the importance of these unanswered questions, existing research on the futures of assaultive children cannot be regarded, even generously, as a functional, if lesser, equivalent of the would-be findings of longitudinal studies. Even if it is true that youthful animal abusers tend to have more psychosocial health problems than do nonabusers and also to engage in other antisocial acts, these facts alone shed no light on the question of whether they are more likely subsequently to engage in interhuman violence.

Animal Abuse and the Histories of Violent Adults

The second proposition of the progression thesis is that those who act violently toward humans are more likely than those who do not previously to have abused animals. In this regard, and with varying degrees of methodological sophistication, most research has proceeded with the use of questionnaires and/or structured interviews asking violent adults to recall the frequency and intensity of their childhood violence against animals. Among the findings tending to support the progression thesis are the following:-

1. Seven female serial killers suffered abuse, abandonment and instability as children, and each of them tortured or killed animals, especially cats (Schurman-Kauflin, 2000, pp. 119-124); and
2. A comparison of the frequency of animal abuse by aggressive and non-aggressive male inmates in federal penitentiaries in Connecticut and Kansas and a control group of randomly-selected noncriminals in New Haven and Topeka found that 25% of the aggressive group reported having abused animals five or more times during childhood, compared with only 5.8% of the nonaggressive group and 0% of the noncriminals (Kellert & Felthous, 1985).

The above findings emerged from information provided by convicted criminals or by psychiatric patients who were interviewed when they were incarcerated. However, such comparisons between incarcerated and non-incarcerated populations should be viewed with skepticism. For one thing, it is a mistake to assume that comparisons of the behavior and characteristics of those who are incarcerated with those who are not will enable us confidently to isolate differences between those who commit crimes and those who do not. Rather, incarcerated populations, comprised only of those unfortunates who have been charged with crimes and convicted of them, are not representative of all those who commit crimes. Similarly, those who never have been incarcerated cannot represent the law-abiding citizenry--among the never-incarcerated are numerous citizens who have committed crimes and who have avoided detection, arrest, conviction, and incarceration.

This segment of the progression thesis also faces two major counterfactual cases (Miller & Knutson, 1997; Arluke et al., 1999). Using self-report questionnaires given to 314 inmates in the Iowan Department of Corrections and to 308 college students, Miller and Knutson found either a modest association or none among abusive childhood environments, witnessing or committing animal cruelty, and subsequent violent behavior. Arluke et al. (1999), who compared the criminal records of 153 animal abusers with 153 neighborhood control participants in Massachusetts, found that although animal abusers also were more likely to commit a range of offences, including those associated with property, drugs and public disorder, no progression existed from animal abuse to interhuman violence. Although it would seem, therefore, that their study tends to disconfirm the progression thesis, instead, they suggest, it reveals the presence of “deviance generalization” (Arluke et al., p. 970).

However, whether the progression thesis really is damaged by these counterfactual cases is uncertain. This is so not because philosophers of science disagree about how many counterfactuals are required for disconfirmation but because both the above studies contain methodological difficulties that impair their ability adequately to test the progression thesis. Thus, Miller and Knutson’s (1997) methodology does not permit a determination of the key question of whether those particular felons, who as children or youth either engaged in acts of animal abuse or witnessed such acts, subsequently committed interhuman violence. Indeed, Miller and Knutson (p. 74) caution that their data allow no inference whatsoever about a causal or temporal sequence between animal cruelty and interhuman violence. Their methodology permits findings about animal abuse and interhuman violence that are, at best, tangential to the progression thesis.

Consider also the study by Arluke et al. (1999). This was devised as a direct test of the progession or “violence graduation” thesis and concluded that no graduation existed from animal abuse to interhuman violence. This conclusion should be treated cautiously. First, because Arluke et al. were legally barred from obtaining any criminal records in the state of Massachusetts for those aged 16 years and younger, their study was unable to test whether there is a progression from animal abuse to interhuman violence during the period from childhood to adulthood. Yet, it is precisely this lengthier age span that commonly is asserted to lie at the heart of the progression thesis by all existing researchers--and probably rightly so. Second, in consciously trying to avoid the pitfalls of self-report data, the Arluke et al. solution succumbs to a different set of difficulties. In their study, they rely on official crime data that derive from reports of animal abuse to the Massachusetts Society for the Prevention of Cruelty to Animals (SPCA), on the one hand, and, on the other, from reports of adult crime to state and local police. All the problems in using official crime data as a measure of the seriousness and frequency of crime cannot be rehearsed here, but it must be said that whatever animal abuse data are lodged in official SPCA and police records are social constructions rather than an objective social reality; as such, their meaning is quite problematic. Each act of animal abuse in official SPCA records is the result of complicated social processes that include the following:
1. a potential complainant who must perceive an animal who is capable of being abused;
2. a potential complainant who must perceive an act of commission or of omission as animal abuse;
3. a perceived case of animal abuse, which must somehow come to the attention of MSPCA officers;
4. formal recognition by an MSPCA official that a report of animal abuse has correctly identified an illegal act of animal abuse and that the act is worthy of their attention; and
5. a given case of animal abuse that has negotiated steps (1-4) must then accurately be entered into official SPCA records.

Clearly, official records of animal abuse do not speak for themselves. They measure no objective or meaningful social reality.
Put differently, only a tiny fraction of animal abuse cases is recorded in official data. However, though Arluke et al. (1999) are keenly aware of social constructivist objections to the meaning and accuracy of official data on animal abuse, this cannot overcome the awkward fact that what can properly be inferred from official records of animal abuse is unclear. Do animal abusers whose acts eventually enter official records typify animal abusers as a whole? Not necessarily; they perhaps simply are less adept at avoiding detection. Perhaps the acts of those who commit greater abuse or who commit it more regularly are somehow less likely to be recognized, detected, and recorded. Or, perhaps, for numerous possible reasons, the lives of those whose acts enter official records are subject to more surveillance than are those of other citizens.

Just as it is important to understand who enters official records of animal abuse--and why--so too, we need to know whether the acts of animal abuse that enter official records are representative of animal abuse as a whole. On this note, therefore, I should stress, as Baldry (2003) and, especially, Agnew (1998) have, that the detection of acts of animal abuse by scholars, by police, and by members of the public very much hinges on how “animal abuse” is defined. “More” animal abuse undoubtedly would have been detected by Arluke et al. (1999) if their concept of it had been broader than “cruelty,” which they operationalized as any investigated case where an animal had been intentionally harmed physically (“beaten, stabbed, shot, hanged, drowned, stoned, poisoned, burned, strangled, driven over, or thrown”) (p. 966). Acts of animal cruelty like those in this definition actually are more dramatic and visible than everyday cases of animal abuse, approximately half of which are acts of neglect and some unknown amount of which involve verbal and emotional abuse.

In trying to assess the merits of the progression thesis, therefore, it is prudent not to rely too much on the two counterfactual cases above. Indeed, what these two studies very usefully point to is the pressing need for careful investigation of the relationship between official data on animal abuse and the unrecorded, otherwise socially invisible character of much animal abuse.
A second avenue of potential retrospective support for the progression thesis lies in several mass-media anecdotes of multiple murder. These suggest that mass murderers and serial murderers tend disproportionately, as children, to have committed animal abuse. This allegedly was the case with serial murderers Patrick Sherrill, Ted Bundy, Alberto DeSalvo , and Jeffrey Dahmer. Consider, in particular, the case of James Hicks who, aged 48, was convicted in 2000 of killing three women in Maine between 1977 and 1996. Though the voices of Hicks’ victims and Hicks, himself, were conspicuously absent from contemporary media accounts of the murders, the excerpt below from a lengthy newspaper account of Hicks’ life illustrates the genre’s explanatory structure. In the excerpt, an investigative journalist recounts his interview with Denise Clark (Hicks’ childhood friend and the murdered woman’s sister) (Wolfe, 2000):
The saga of Jimmy Hicks can begin almost 30 years ago with four cold words that haunt Denise Clark still “I killed your cat,” she said Hicks, then 18, told her in a calm voice, a few days after she’d said something that he didn’t like. Clark, 15 at the time, told him she didn’t believe him. But Hicks insisted, explaining that he had wrapped a wire around the cat’s neck, hooked it to his bumper, and dragged the helpless animal along the roadway. “He didn’t blink an eye,” she recalled. Clark and a friend later found the cat, dead, with a wire still around its neck. (p. A12)

This narrative invites its audience to ponder how and why local boy Jimmy Hicks could have become a serial murderer. It does so by suggesting that this process of becoming a serial murderer (“[t]he saga of...”) was a fairly straightforward series of events that logically preceded and prepared the way for Hicks to commit multiple murders. Readers are informed that the relevant events in this chain of causation “can begin” 30 years earlier when Hicks calmly told his friend Denise Clark that he had tortured and killed her cat.

What do such anecdotes signify? Prised from their cultural matrix to illuminate with dramatic effect some aspect of a larger and more compelling story, anecdotes are flimsy constructions whose narrative truth is less important than either the discursive functions they are asked to serve or the interests of those who wield them. Unsurprisingly, the anecdotally constructed generalization that multiple murderers are more likely as adolescents to have committed animal abuse is vulnerable to simple counterfactual cases. Its applicability in her own case has been strenuously and credibly denied by the English “Moors murderess” (Hindley, (1995). If the possibly relevant facts in the prior histories of serial murderers are to include anecdotes, then one anecdote, of course, may be legitimately countered with another.

Moreover, some aspects of the anecdotal evidence about multiple murderers’ histories are clearly more complex than their dramatic presentation indicates. Consider the reportage of Patrick Sherrill, a postal worker who killed 14 co-workers in 1986 and who is said to have stolen local pets and then allowed his own dog to mutilate them (IACP, 1989, p. 2).

Suppose it is true that at some time before he killed 14 colleagues, Sherrill had allowed his dog to mutilate neighborhood companion animals--from this it would not follow that those who allow their dogs to mutilate neighborhood companion animals have a greater propensity subsequently to engage in interhuman violence. Even if these facts had fit Sherrill’s case, we also would need to inquire of Sherrill’s life history not only how the earlier form of violence led to the later one but also whether other aspects of his life might have been even more proximate or more influential. Did Sherrill commit other forms of violence before his mass-murder spree? Had he been abused at work? Passed over for promotion? Was he suicidal and, if so, why?

Also, consider the anecdote that the young Jeffrey Dahmer (the “Milwaukee cannibal”) used to roam his neighborhood for roadkill and that he had a little graveyard with animals buried in it (Dvorchak, 1991). Add, for good measure, the report that as a boy Dahmer had impaled or staked frogs and cats to trees (Goleman, 1991). Nothing in either description indicates that Dahmer ever tortured or killed live animals, and his father (Dahmer, 1994) has stated that his adolescent son even rescued several at-risk animals. If a given adolescent is fascinated with dead animals, why should we infer that the adolescent is a serial killer in-the-making rather than a budding zoologist or forensic scientist?

Summary

Especially in popular discourse, the lack of subtlety with which the complex relationship between animal abuse and interhuman violence sometimes is asserted tends to make it appear more the brittle product of sloganeering than of hard evidence and logic. Worrisome evidentiary weaknesses currently beset the progression thesis, chief among which are the paucity of focused empirical data, the absence of longitudinal studies, and, as I have hinted, the uncritical constitution and employment of such concepts as “animal abuse” and "cruelty." In concert, these weaknesses suggest that current generalizations about a progression from animal abuse to interhuman violence are, at best, premature. Indeed, rather than emanating from a coherent research program, support for the progression thesis comprises little more than pro-animal sloganeering.

In particular, although the several forms of family violence undoubtedly are strongly associated, existing knowledge of how, and how often, companion animal abuse exists with other forms of family violence tends neither to confirm nor to disconfirm the progression thesis. Crucially, it is not known whether animal abuse precedes and signifies other forms of violence or whether it follows them. Whichever the case, we need additionally to know under what circumstances it is so, and why. What currently is known about their futures actually sheds little light on the likelihood that assaultive children subsequently will engage in interhuman violence. Moreover, given the largely anecdotal and contradictory nature of the evidence in this regard, it is not yet clear if those who act violently toward humans are more likely than those who do not previously to have abused animals.

A Plea for Longitudinal Analysis

Suppose that eventually it is confirmed that assaultive children are more likely later to commit interhuman violence. We then would need to inquire if this heightened disposition derives from their prior animal abuse, and why. Do factors other than animal abuse influence assaultive children subsequently to commit interhuman violence? What influence is exerted by gender, age, race, ethnicity, social class, conduct disorder, and intervention strategies?

The best method for examining the chronological causal sequence in the progression thesis is to combine sensitive retrospective analyses of violent adults with a prospective longitudinal study. With the careful use of self-report studies, in-depth interviews, ethnography and official crime records, a longitudinal study of a random sample of the youth population could be done that measured animal abuse and interhuman violence at two or more points in time. The study could begin with a newly born population and then examine each individual’s significant events over 25 years. The effect of prior animal abuse on subsequent interhuman violence then could be estimated over the life course, with controls for prior interhuman violence and other independent or interactive variables known, or thought, to be correlated with animal abuse and interhuman violence.
Besides the financial cost of a longitudinal study, it would take a generation to execute. However, it might be possible to design an accelerated study of only 6-7 years (Farrington & Coid, 2003, pp. 361-363). Additional time could be saved by tagging onto existing longitudinal surveys of delinquent youth such as the Cambridge Study in Delinquent Development and the Pittsburgh Youth Study (Loeber & Farrington, 2001). However, even longitudinal analysis would not prove conclusively that committing animal abuse causes animal abusers subsequently to engage in human violence; problematically, it would be comparing the subsequent interhuman violence of individuals who had different degrees of prior animal abuse. For conclusive proof, a randomized experiment would be needed; for ethical and other reasons, however, this would be extraordinarily difficult to execute. It would, however, substantially increase our confidence that engaging in animal abuse exerts an independent causal effect on interhuman violence.

Expanding the Scope of the Progression Thesis: From Abusive Individuals to Institutionalized Abuse

The progression thesis lacks coherent empirical evidence partly because there has been insufficient theoretical attention to key concepts like “animal abuse” and “animal cruelty.” Each is highly contentious. At what level in hierarchies of consciousness and sentience must animals be positioned for them to be included in the concept “animal abuse”? Am I abusive, for example, if I swat the bloodsucking mosquito on my arm? Similarly, what counts as “abuse”? Should the concept of animal abuse be expanded from the purely physical domain to include emotional and psychological dimensions? Should it include neglect? Why are most existing studies of one-on-one or face-to-face situations of “intentional cruelty” to companion animals rather than, say, to feral animals or to animals used in agriculture, laboratories, and entertainment?

The field of human-animal studies has no scientific warrant to deploy societal definitions of acceptable and unacceptable behavior--these often are anthropocentric, arbitrary, and capricious. In the study of animal abuse, the dominant focus on “socially unacceptable behavior that intentionally causes unnecessary pain, suffering, or distress to and/or death of an animal” (Ascione, 1993, p. 228) deliberately and uncritically excludes exploration of less visible, even more pervasive, ways in which the abusive situation of one species might lead to a situation of violence against another. The link between animal abuse and interhuman violence surely must be sought not only in the personal biographies of those individuals who abuse or neglect animals but also in those institutionalized social practices where animal abuse is routine, widespread, and often defined as socially acceptable.
The multiple sites of violence condoned in slaughterhouses perfectly exemplify these practices. Consider how these might lead, or “progress,” to extra-institutional violence. There is, first, the abrupt, unnatural, and sometimes painful death of billions of terrified animals. Less acknowledged, second, is the awesome physical and psychic toll on slaughterhouse workers who, among all private sector U.S. industries, suffer the highest annual rate of nonfatal injuries and illnesses and repeated-trauma disorders (U.S. Department of Labor, 1999, p. 2). Less documented, third, is the violence visited on those beings--human and animal--with whom slaughterhouse workers interact outside their work sites. Whenever human-animal relationships are marked by authority and power, and thus by institutionalized social distance, there is an aggravated possibility of extra-institutional violence. Eisnitz (1997) graphically uncovers this human toll. One worker interviewed by Eisnitz--Van Winkle--believed that it is “not uncommon” for slaughterhouse workers to be arrested for having assaulted humans. Describing the mental attitude developed from “sticking” hogs (slitting hogs’ throats in the often-botched attempt to kill them), he divulged that “[M]y attitude was, its only an animal. Kill it.”(Eisnitz, p. 88). Then:
I’ve had ideas of hanging my foreman upside down on the line and sticking him. I remember going into the office and telling the personnel man I have no problem pulling the trigger on a person--if you get in my face I’ll blow you away....Every sticker I know carries a gun, and every one of them would shoot you. Most stickers I know have been arrested for assault. A lot of them have problems with alcohol. They have to drink, they have no other way of killing live, kicking animals all day long. (Eisnitz, p. 88)
For such narratives to be intelligible, animal abuse as a complex of social practices must be understood and explained theoretically. Some explanatory power might be afforded by a research program that combined ethnography with existing sociological theories of violence, especially if they are mindful of the role of subjective states such as empathy, caring, and compassion. If compassion involves an understanding of others and others’ suffering and the desire to ameliorate it, then compassion for animals and compassion for humans, respectively, probably are strongly linked. Thus, whatever their social situation and motivation, both assaultive children and slaughterhouse workers might be so desensitized by the act of animal abuse that subsequently they have lesser compassion for the suffering and welfare of many other beings (including humans). In reducing abusers’ compassion, animal abuse might be found to increase tolerance or acceptance of pro-violent attitudes and, thereby, to foster interhuman violence. Indeed, a plausible corollary of the progression thesis, if found to be true, is that children who have, or who are taught to have, compassion for animals might be more likely to become adults who act more sensitively and more gently toward humans.

* Piers Bierne, University of Southern Maine

References

Achenbach, T. M., Howell, C. T., Quay, H. C., & Conners, C. K. (1991).National survey of problems and competencies among four- to sixteen-year-olds, Monographs of the Society for Research in Child Development, 56 (3), 225.

Adams, C. J. (1995). Woman-battering and harm to animals. In C. J. Adams & J. Donovan (Eds.), Animals and women: Feminist theoretical explorations (pp.55-84). Durham: Duke University Press.

Agnew, R. (1998). The causes of animal abuse: A social psychological analysis. Theoretical Criminology, 2 (2), 177-209.
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders. Washington, DC: Author.

Arkow, P. (1994). Child abuse, animal abuse, and the veterinarian. Journal of the American Veterinary Medical Association, 2047, 226-247.

Arluke, A., Levin, J., Luke, C., & Ascione, F. (1999). The relationship of animal abuse to violence and other forms of antisocial behavior. Journal of Interpersonal Violence, 14 (9), 963-975.

Arluke, A., & Sax, B. (1992). Understanding Nazi animal protection and the Holocaust. Anthrozöos, 5 (1), 6-31.

Ascione, F. R. (2001). Animal abuse and youth violence. Washington, DC: U.S. Department of Justice.
----- (1998). Battered women's reports of their partners' and their children's cruelty to animals. Journal of Emotional Abuse, 1 (1), 119-133.
----- (1993). Children who are cruel to animals: A review of research and implications for developmental psychopathology. Anthrozoös, 6 (4), 226-247.
-----, & Randall Lockwood (2001). Cruelty to animals: Changing psychological, social, and legislative perspectives. In D. J. Salem and A. N. Rowan (Eds.), The state of the animals 2001 (pp. 39-53). Washington, DC: Humane Society of the United States.

Ascione, F. R., Weber, C. V., & Wood, D. S. (1997). The abuse of animals and domestic violence: A national survey of shelters for women who are battered. Society & Animals, 5 (3), 205-218.

Ascione, F. R., Thompson, T. M., & Black, T. (1997). Childhood cruelty to animals: Assessing cruelty dimensions and motivations. Anthrozoös, 104,170-177.

Baldry, A. C. (2003). Animal abuse and exposure to interparental violence in Italian youth. Journal of Interpersonal Violence, 18 (3), 258-281.

Beirne, P. (1995). "The use and abuse of animals in criminology: A brief history and current review. Social Justice, 22 (1), 5-31.

Bell, L. (2001). Abusing children--abusing animals. Journal of Social Work, 1 (2), 223-234.

Boat, B. (1999), Abuse of children and abuse of animals. In F. R. Ascione & and P. Arkow (Eds.), Child abuse, domestic violence, and animal abuse (pp.83-100). West Lafayette, IN: Purdue University Press.

Browne, A. (1987).When battered women kill. New York: Free Press.

Cazaux, G. (2002). Anthropocentrism and speciesism regarding animals other than human animals in contemporary criminology. Unpublished doctoral dissertation, Faculty of Law, University of Ghent.
----, & Beirne, P. (2001), Animal abuse. In E. McLaughlin & J. Muncie (Eds.), Sage dictionary of criminology (pp.8-10). London: Sage.

Coston, C., & Protz, C. M. (1998). Kill your dog, beat your wife, screw your neighbor’s kids, rob a bank? A cursory look at an individual’s vat of social chaos resulting from deviance, Free Inquiry in Creative Sociology, 26 (2), 153-158.

Dadds, M. R., Turner, C. M., & McAloon, J. (2002). Developmental links between cruelty to animals and human violence. Australian and New Zealand Journal of Criminology, 35 (3), 363-382.

Dahmer, L. (1994). A father’s story. New York: William Morrow.

Daniell, C. (2001). “Ontario SPCA’s women’s shelter survey shows staggering results. The Latham Letter, xxii, 16-17.

Deviney, E. Dickert, J., & Lockwood, R. (1983). The care of pets within child abusing families. International Journal for the Study of Animal Problems, 4 (4), 321-329.

Donley, L., Patronek, G J., & Luke, C. (1999). Animal abuse in Massachusetts: A summary of case reports at the MSPCA and attitudes of Massachusetts veterinarians. Journal of Applied Animal Welfare Science, 2 (1), 59-73.

Dvorchak, R. (1991, August 11). Dahmer’s troubled childhood offers clues but no simple answers. Dallas Times Herald, pp.A6/A8.

Eisnitz, G. A. (1997). Slaughterhouse: The shocking story of greed, neglect, and inhumane treatment inside the U.S. meat industry. Amherst, NY: Prometheus Books.

Elander, J., Rutter, M., Siminoff, E., & Pickles, A. (2000). Explanations for apparent late onset criminality in a high-risk sample of children followed up in adult life. British Journal of Criminology, 40 (3), 497-509.

Faller, K. C. (1990). Understanding child sexual maltreatment. Newbury Park, CA.: Sage.

Farrington, D. P. (2002). In M. Maguire, R. Morgan, & R. Reiner (Eds.), The Oxford handbook of criminology (3rd ed., pp.657-701.). Oxford: Clarendon Press.
-----, & Coid, J. W. (2003). Conclusions and the way forward. In D. P. Farrington & J. W. Coid (Eds.), Early prevention of adult antisocial behaviour (pp.356-368). Cambridge: Cambridge University Press.

FBI. (2002). Crime in America: The uniform crime reports. http://www.fbi.gov/ucr/ucr.htm

Felthous, A. R. (1981). Childhood cruelty to cats, dogs and other animals. Bulletin of the American Academy of Psychiatry and Law, 9, 48-53.

Felthous A. R., & Kellert, S. R. (1987). Childhood cruelty to animals and later aggression against people: A review. American Journal of Psychiatry, 144 (6), 710-717.
-----, & Yudowitz, B. (1977). Approaching a comparative typology of assaultive female offenders. Psychiatry, 40 (6), 270-276.

Finkelhor, D., Williams, L. M., & Burns, N. (1988).Nursey crimes: Sexual abuse in day care. Newbury Park, CA: Sage.

Flynn, C. P. (1999). Animal abuse in childhood and later support for interpersonal violence in families. Society & Animals, 7 (2), 161-172.
----- (2000a). Battered women and their animal companions: Symbolic interaction between human and nonhuman animals. Society & Animals, 8 (2), 100-127.
----- (2000b). Woman’s best friend. Violence Against Women, 6 (2), 162-177.

Friedrich, W. N., Urquiza, A. J., & Beilke, R. L. (1986). Behavior problems in sexually abused young children. Journal of Pediatric Psychology, 11 (1), 47-57.

Gelles, R. J., & Straus, M. (1988). Intimate violence. New York: Simon and Schuster.

Gest, E. (1999, August 12). City seeks poisoners of Central Park birds. Daily News.

Goleman, D. (1991, August 7). Clues to a dark nurturing ground for one serial killer. New York Times, A8.

Government of Canada (1972). Commission of inquiry into the non-medical use of drugs. Ottawa: Canadian Government Information.

Guymer, E. C., Mellor, D., Luk, E. S. L., & Pearse, V. (2001). The development of a screening questionnaire for childhood cruelty to animals. Journal of Childhood Psychology and Psychiatry, 42, 1057-1063.

Hellman, D. S., & Blackman, N. (1966). Enuresis, firesetting and cruelty to animals: A triad predictive of adult crime. American Journal of Psychiatry, 122 (12), 1431-1435.

Hindley, M. (1995, December 18). The moors murderess. The Guardian.

Hunter, M. (1990). Abused boys: The neglected victims of sexual abuse. New York: Lexington.

Hutton, J. S. (1983), Animal abuse as a diagnostic approach in social work: A pilot study. In A. Katcher & A. Beck (Eds.), New perspectives on our lives with companion animals (pp. 444-447). Philadelphia: University of Pennsylvania Press

IACP (1989). Cruelty to animals and human violence. Training Key No.392, International Association of Chiefs of Police.

Itzin, C. (1997). Pornography and the organization of intra-and extrafamilial child sexual abuse. In G. K. Kantor & J. L. Jasinski (Eds.), Out of the darkness: Contemporary perspectives on family violence (pp.58-79). Thousand Oaks, CA: Sage.

Justice, B, Justice, R., & Kraft, I. A. (1974). Early warning signs of violence: Is a triad enough? American Journal of Psychiatry, 131 (4), 457-459.

Kellert, S., & Felthous, A. (1985). Childhood cruelty toward animals among criminals and noncriminals. Human Relations, 38 (12), 1113-1129.

Kelley, S. J. (1989). Stress responses of children to sexual abuse and ritualistic abuse in day care centers. Journal of Interpersonal Violence, 4 (4), 502-513.

Langevin, R., Paitich, D., Orchard, B., Handy, L. & Russon, A. (1983). Childhood and family background of killers seen for psychiatric assessment. Bulletin of the American Academy of Psychiatry and the Law, 11 (4), 331-341.

Lasik, A. (1998). Rudolf Höss: Manager of crime. In Y. Gutman & M. Berenbaum (Eds.), Anatomy of the Auschwitz death camp. Bloomington: Indiana University Press.

Levin, J., & Fox, J. A. (2001). Dead lines: Essays in murder and mayhem. Needham Heights, MA: Allyn and Bacon.

Loeber, R., & Farrington, D. P. (2001) (Eds.). Child delinquents: Development, intervention and service needs. Thousand Oaks, CA: Sage.

Macdonald, J. M. (1968). Homicidal threats. Springfield, IL: Charles C. Thomas.

McFarlane, J., Campbell, J., Wilt, S., Sachs, C., Ulrich, Y., & Xu, X (1999). Stalking and intimate partner femicide. Homicide Studies, 3 (4), 300-316.

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, 45 (5), 556-573.

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 & Neglect 21, (1), 59-82.

Munro, H. (1996). Battered pets. Irish Veterinary Journal, 49 (12), 712, 713.
----, & Thrusfield, M. V. (2001). Battered pets: Features that raise suspicion of non accidental injury. Journal of Small Animal Practice, 42, 218-226.

Patronek, G. J. (1997). Issues for veterinarians in recognizing and reporting animal neglect and abuse. Society & Animals, 53, 267-280.

Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989). A developmental perspective on antisocial behavior. American Psychologist. 44 (2), 329-335.

Paul, E. S. (2000). Empathy with animals and with humans: Are they linked? Anthrozoös, 13 (4), 194-202.

Randal, J., & Boustany, N. (1990). “Children of war in Lebanon. In C. Moorehead (Ed.), Betrayal: A report on violence toward children in today’s world (pp.59-82). Garden City: Doubleday.

Reber, K. (1996). Children at risk for reactive attachment disorder: Assessment, diagnosis and treatment. Progress: Family Systems Research and Therapy, 5 (3), 83-98.

Renzetti, C. (1992). Violent betrayal: Partner abuse in lesbian relationships. Newbury Park, CA: Sage.

Ressler, R. K., Burgess, A. W., & Douglas, J. E. (1988). Sexual homicide: Patterns and motives. Lexington, MA: Lexington Books.

Sauder, J. (2000). Enacting and enforcing felony animal cruelty laws to prevent violence against humans. Animal Law, 6, 1-21.

Sack, K. (1997, October 15,). Grim details emerge in teen-age slaying case. New York Times, p. A10.

Santillo, P., & Haapasalo, J. (2001). Neurological and psychological risk factors among young homicidal, violent and nonviolent offenders in Finland. Homicide Studies, 1 (3), 234-253.

Schofield, M. (1971). The strange case of pot. Middlesex: Pelican.

Schurman-Kauflin, D. (2000). The new predator:Women who kill. New York: Algora.

Serpell, J. A., & Paul, E. S. (1994). Pets and the development of positive attitudes to animals. In A. Manning & J. Serpell (Eds.), Animals and human society: Changing perspectives (pp.127-144). London: Routledge.

Shaffer, J. N., & Ruback, R. B. (2002). Violent victimization as a risk factor for violent offending among juveniles. [NCJ 195737]. Washington, DC: U.S. Department of Justice.

Skrapec, C. (1996). The sexual component of serial murder. In T. O’Reilly-Fleming (Ed.), Serial & mass murder. Toronto: Canadian Scholars Press.

Slavkin, M. L. (2001). Enuresis, firesetting, and cruelty to animals: Does the ego triad show predictive validity? Adolescence, 36 (143), 461-466.

Solot, D. (1997). Untangling the animal abuse web. Society & Animals, 5 (3), 257-265.

Tapia, F. (1971). Children who are cruel to animals. Child Psychiatry and Human Development, 2 (2), 70-77.

Tingle, D., Barnard, G. W., Robbins, L., Newman, G., & Hutchinson, D. (1986). Childhood and adolescent characteristics of pedophiles and rapists. International Journal of Law and Psychiatry, 9, 103-116.

U.S. Department of Labor (1999). Occupational injuries and illnesses: counts, rates, and characteristics, 1997. [Bulletin 2518]. Washington, DC: Bureau of Labor Statistics.

Widom, C. S. (2000). Childhood victimization: Early adversity, later psychopathology. National Institute of Justice Journal, [NCJ 180077] January, 2-9.

Wiehe, V. R. (1990). Sibling abuse. Lexington, MA: Lexington Books.

Wolfe, J. (2000, October 8). Suspect’s deal holds promise of closure. Maine Sunday Telegram, pp. A1, A12.

Wooden, W. S., & Berkey, M. L. (1984). Children and arson: America's middle class nightmare. New York: Plenum.

Yates, Roger (2003). Developing the sociology of animal abuse. Unpublished doctoral dissertation, University of Wales, Bangor.

Yates, R., Powell, C., & Beirne, P. (2001). Horse maiming in the English countryside: Moral panic, human deviance, and the social construction of victimhood. Society & Animals, 9 (1), 1-23.

Youssef, R. M., Attia, M. S., & Kamel, M. I. (1999). Violence among schoolchildren in Alexandria. Eastern Mediterranean Health Journal, 5 (2), 282-298.

For Abstracts of all issues, including the most current, click Article Abstracts

To order Society & Animals Journal, go to our secure online ordering page

You can Search the online issues of Society & Animals, as well as the entire Society & Animals Forum (formerly PSYETA) website,
for topics and keywords of your interest: