Modern Rejection of the Notion of the “Criminal Mind”
Jean-Baptiste Lamarck was among the first, if not the first, to put forth a theory of evolution, and his name might be better remembered today had his theories not contained the element known as “soft inheritance.” One example that is given of this notion is that giraffes had to extend their necks to reach higher foliage, and that as a particular giraffe did this throughout its life, this characteristic was passed on to its offspring. Given the state of knowledge on the subject in general at the time, it is easy to see how this theory could have come about.
Charles Darwin and others rejected soft inheritance in favor of the version that we know today which relies solely on the concept that successful traits lead to better chances of reproduction than do unsuccessful traits, an idea that we refer to as “natural selection.”
Ironically, however, some aspects of Lamarck’s model are being reconsidered in modern science. I’ll leave that for another time. But, it does illustrate how, in science, discredited concepts can make a comeback even if the basis for the comeback is entirely unrelated to the basis for the original theory. 
Sometimes erroneous theories persist not because there is remaining evidence that they are correct, but because they fill some psychological need that we have, or match a pet theory or paradigm that we have grown accustomed to. History is full of examples of man's inhumanity to man based on the notion that one group of people were inherently inferior to another, falling into categories such as lesser human, subhuman, or even nonhuman.
While the modern world has largely discarded such thinking, it still lingers in more primitive cultures and even in pockets of ignorance within and across the modern world. We see evidence of this every day on the news, and unfortunately, it occasionally finds its way into our entertainment, political discourse, and even the manners wherein we raise our children. Hate begets hate.
These appeals to the degree to which we are powerless over our “human nature” take almost comical forms in some cases, such as the notions that men must be forgiven for their sexual urges over which “they have no control,” while women must be castigated for giving in to their own supposedly inbuilt propensities to “use” men. That both notions can coexist within a single human brain serves as stark testimony to human subjectivity. Each view of women and men, though, is almost certainly devoid of substance.
For the greater half of the last century, the belief was common that the descendants of slaves imported to the western hemisphere were so inherently inferior that they must be relegated to second-class status. We do not have to look far for evidence that such notions persevere among some people to this day, and that for many of said individuals, it becomes something of a self-fulfilling proposition.
In a prior piece entitled “Monoamine Oxidase A: The War over the “Warrior Gene,” I presented both sides of the story on how just one genetic marker can be shown not to control behaviors, but to influence them in mostly subtle, but taken to the extreme, not so subtle ways. 
Fortunately, the world of real science does not have to wait for the popularizers, pundits, and particularly, the con artists and demagogues, to catch up as it builds on theories that have proven out while whittling away aspects of those which have not. While madmen of the twentieth century sought to use superficial characteristics as a way to “purify” the human race even before the genome had been sequenced, let alone thoroughly understood, contemporary science has breathed new life into the idea that “all men are created equal” by demonstrating that our environment plays the more important role in how we turn out as adults.
In what follows, I want to trace an example of this area of research, noting that, as a related collection of concepts are examined, rejected, resurrected in another form, improved, and, let us hope someday, perfected, we can not only learn what causes certain behaviors, but also what environmental factors might produce optimal results.
The Science of Nurture over Nature
Beginning around the last half of the twentieth century, sociologists began to suspect that criminal behavior was not, for the most part, an inborn characteristic. Rather, it was learned behavior that, if fully understood, could be compensated for to attack crime at its roots. Because theories at the time postulated that events in earliest childhood might result in criminal behavior throughout life, only expensive longitudinal studies were likely to shed light on these root causes. Fortunately, a married couple, the Gluecks, provided a solid foundation for study that is still bearing fruit today. We might, however, be at a stage where every last result has been wrung out of these original data sets, and only a fresh round of data gathering will refine our insights. Additionally, the world has changed so significantly since the Gluecks conducted their work that older notions regarding ideals in family life are difficult or impossible to achieve today.
During the 1940s, Sheldon and Elenor Glueck studied 500 delinquent and 500 non-delinquent children in the Boston area of Massachusetts. This work resulted in the book Unraveling Juvenile Delinquency, published in 1950.  It was a monumental effort in terms of the amount of data collected, and the fact that it was a longitudinal study. The Gluecks spent the rest of their lives enhancing their interpretations. As documented by researchers Sampson and Laub, the Gluecks’ work was widely criticized for a number of reasons, some of which were due to the Gluecks’ emphasis on physiology (for example, mesomorphy) as a major factor, and others having to do with their analysis of their own data. [4 pp. 357-361] The Gluecks were not the first, nor the last, to place nature and nurture on a similar level, but as Sampson and Laub demonstrated, it is possible to factor these prejudices out and still have an abundance of useful data to work with.
The difficulty in performing such a large study over a long period of time has served to increase the value of the Gluecks’ work throughout the years, as Sampson and Laub and others have found it more expedient to redo the analysis of the Glueck data, rather than justifying the time and cost of having to replace it. A primary, and for the most part lasting, outcome of these studies has been an emphasis on the importance of the family in shaping a child’s future. The details of that emphasis, though, have changed over time with each new refinement of older data, such as that of the Gluecks, or more modern, if less ambitious collections of data that have been gathered since.
The Sampson and Laub work largely rescued the Glueck data from what might have become an ultimate obscurity due to the criticisms that they received, and they found that there was good reason to support many of the Gluecks’ conclusions, particularly when it came to the importance of establishing a strong family basis. The factors of supervision, attachment, and discipline, as identified by the Gluecks, were the most important (negatively) correlating factors with serious and persistent delinquency. A second major finding was that, with the exception of residential mobility, none of the other structural background factors were significant. Family processes mediated 80% of these structural background factors. Previously neglected factors such as alcoholism and criminality of parents were found to be important. Finally, the Sampson and Laub study showed that the Glueck data could still be used to provide additional insights into the causes of delinquency. 
Following the work of Sampson and Laub, others sought to confirm, deny, or extend these findings. By way of illustration, in 1990, Mednick, Baker, and Carothers showed that divorce alone could not explain delinquent behavior if the subsequent family environment was stable.  As we will see, other studies confirmed this and improved on the specificity of the findings with respect to which secondary family structures produced the best results. 
Around a year later in 1991, Wells and Rankin, conducting a survey of 50 family studies, a process known as meta-analysis, found that one thing which stood out as a predictor of criminal behavior was divorce. They also concluded that even though the span of time was significant for the 50 studies that they looked at the results of, where they were comparable remained relatively consistent.  We will witness later work, though, which might indicate that some changes have taken place in our social structure which will alter this notion and provide a reason to call for fresh data on a scale similar to, or larger than, the Glueck data.
McCord, likewise in 1991, made a refinement to the importance of family structure by determining that a competent mother could offset the negative effects of a broken home. The impact of the father was less, but could serve as a positive influence if he played a cooperative role.  This was confirmed by Barber and Eccles through survey data in 1992, and in fact stressed that in a marriage beset with conflict, the children had better chances if a divorce happened sooner rather than later. 
Then, Amato, in 1994, took the meta-analysis concept to an extreme, looking at over 90 prior studies to draw new conclusions from old data. Unlike Wells and Rankin in one regard, Amato noted that there had been a societal shift which, presumably, had made the stigma associated with divorce less over time as single-parent households became more commonplace. Previously, the death of one parent had been the primary cause of single-parent households coming into being, but with longer lifespans and an increasing divorce rate, this situation reversed itself.  Coupled with the McCord results, we might deduce that the stress of an unstable two-parent household is more detrimental than one consisting of a single parent, and further, that as such situations became more commonplace, the main influence became stability rather than the number of parents.
A slight shift back to the primacy of marriage is seen in a 1999 work by Simons, Lin, Gordon, Conger, and Lorenz, who noted that ongoing contact with a divorced father, especially for boys, was important, as was the financial stability of the custodial mother, along with her parenting skills. A situation in which the divorced parents did not continue to contend with one another was seen to have better results. In this case, continued contact with the father served as a role model for boys, at least if the father was willing to play a supportive role. This study thereby introduced the idea that divorce can have significantly greater impact on boys than girls. Considering that many of the earlier studies consisted mostly, if not entirely, of boys, it is not surprising that this possibility was missed. 
I must note here that, in several of these studies that did include children of both sexes, differences were often observed. If there was any point at which to suggest that genetic differences between the sexes might find expression in later behavioral characteristics, this would likely be it. As my earlier post indicated, the placement of that one genetic marker (MAOA) on the Y chromosome meant that its impact was muted in girls, while having full force in boys. Furthermore, studies show that the genetic marker alone is not sufficient to produce a measurable difference in behavior, but must be accompanied by some form of childhood “trauma” (for lack of a better word). These studies may well be considered a survey of such traumatic conditions. 
Because of its large sample size of over 21,000 Swiss army recruits, a study by Haas in 2004 might not appear to correlate well with the others, but the results were fairly congruent with the others in most important regards. Because it consisted only of males, it can do no more to illuminate the differences in impact between the sexes of the children. It does confirm the importance of a skilled mother, though, and adds to that the finding that continued care by an extended family member could also lessen the negative effects of loss of one or both parents. 
Murray and Farrington in 2008 summarized several other studies involving parental imprisonment that added to the variety of family disruptors, such as divorce or death, which could have an impact on future delinquent and criminal behaviors. The authors claimed that this form of disruption was as significant, if not more so, than death of one parent or divorce. Once again, the impact for an imprisoned mother was greater than if the father was the one absent from the home. This study might be used to justify the often lesser sentences meted out to women versus men for similar crimes, even if that is not the typical rationale relied upon. Mitigating factors were found to be higher IQ for the child as well as welfare provisions for a single parent, typically the mother, who had to keep close to home and had no income from an imprisoned father. The stigma of having a parent in prison might additionally be a factor that makes this scenario potentially worse than divorce or the death of one parent. 
Finally, Wildeman and Western, in 2010, constructed an impassioned case (if not so much data-driven) for prison reform to avoid an avalanche effect of poverty and criminality that resulted from generation after generation of imprisoned parents. This offers a probable explanation like no other study has for why certain racial and ethnic groups fail to respond to various social programs. Among concerns cited were the negative impacts on employment for parolees, issues with reincarceration over minor technical parole violations, and limited educational opportunities that might allow truly reformed inmates to re-enter society successfully. In their closing few sentences, Wildeman and Western assert:
Taking full account of the negative social effects of incarceration shows that the costs of mass imprisonment are far higher than correctional budgets suggest. More fundamentally, criminal justice agencies are only residual sources of social order. The primary sources of order and stability—public safety in its wide sense—are the informal social controls of family and work. 
By working through these studies chronologically, it can be discerned that the contradictions are actually few. Instead, each new study bolsters some of the results of those that have come before. Moreover, we see that greater precision is achieved whether adding fresh data to the mix or reanalyzing previous data from a new perspective.
Although it was at one time concluded that such results were little affected by the passage of time and societal shifts, there are now reasons to believe that that stability has passed as we observe a greater variety of family lifestyles, first coming into existence, and then becoming more common, and finally more acceptable and accepted. The importance of stigma can be noted as well, both in the sense that old causes of it such as divorce have faded, and new causes such as imprisonment have emerged.
Also emerging are new challenges to the family, whatever forms they take, as an increasing percentage of the population is ill-prepared for a twenty-first century workplace; a workplace that is smaller in many realms due to automation, and thus less dependent on physical labor. Assurances that a new workforce would be based on a service economy model have not taken up the slack, as not only have labor-intensive industries become more efficient, but so have white-collar jobs such as accounting, engineering, manufacturing, and others become more computerized. Guaranteeing each citizen the chance of a productive life, even without the issues of divorce, imprisonment, and dislocation, seems to be growing more and more difficult.
Nevertheless, it is not true that we are short of fresh ideas to address the causes of poverty and family breakdown. For instance, sociologist Julius Wilson closes his book When Work Disappears: The World of the New Urban Poor with recommendations that superficially resemble the New Deal’s Works Progress Administration (WPA). Acknowledging that these ideas would probably cost more than the initial benefits could justify, he instead puts the focus on the long-term societal benefits that could, at least to some extent, provide the framework for a more inclusive society. In his final chapter he states:
My framework for long-term and immediate solutions is based on the notion that the problems of jobless ghettos cannot be separated from those of the rest of the nation. Although these solutions have wide-ranging application and would alleviate the economic distress of many Americans, their impact on jobless ghettos would be profound. Their most important contribution would be their effect on the children of the ghetto, who would be able to anticipate a future of economic mobility and share the hopes and aspirations that so many of their fellow citizens experience as part of the American way of life. 
The political class may continue to prescribe the same old solutions for a time, but in stronger doses, yet paradigm shifts that are well underway will eventually only respond to radically new treatments. Studies on these subjects should continue to place a sharper focus on what these new solutions might be, and should likewise strive to convince both the public and the political class that they must be tried, and with greater attention to which prescriptions produce truly desirable results and which ones don’t.
Analyzing and recycling family structure data from the post-industrial revolution might well have run its course, with every last drop of meaning now wrung out of it. New studies must factor in the impacts of the internet, working mothers, the influence of fathers on girls and boys in different roles, commoditization of information content, education, relevant changes in social policy, and more. In a world that is evolving faster than ever before, at an ever more rapid pace, our understanding of it must evolve faster as well.
Author: Krista [@Femitheist]
NOTE: If you liked this post, feel free to share it, and if you have any thoughts on it or the general subjects discussed within, or even merely semi-related topics, please leave them in the comments below. I always enjoy reading the feedback of others whether they liked and agreed with what I had to say or not.
References (Last Accessed on July 30, 2015):
 Jean-Baptiste Lamarck | French biologist (Encyclopedia Britannica Online/Wikipedia)
 Krista | Philanthropy. Egalitarianism. Humanism. Humanitarianism. (Monoamine Oxidase A: The War over the “Warrior Gene” ~)
 Glueck, S., & Glueck, E. (1950). Unraveling Juvenile Delinquency. New York: Commonwealth Fund.
 Laub, J., & Sampson, R. (1988). Unraveling Families And Delinquency: A Reanalysis Of The Gluecks’ Data*. Criminology, 26(3), 355-380.
 Mednick, B., Baker, R., & Carothers, L. (1990). Patterns of family instability and crime: The association of timing of the family's disruption with subsequent adolescent and young adult criminality. Journal of Youth and Adolescence, 19(3), 201-220.
 Wells, L., & Rankin, J. (1991). Families And Delinquency: A Meta-Analysis Of The Impact Of Broken Homes. Social Problems, 38(1), 71-93.
 McCord, J. (1991). Family Relationships, Juvenile Delinquency, And Adult Criminality*. Criminology, 29(3), 397-417.
 Barber, B., & Eccles, J. (1992). Long-term Influence Of Divorce And Single Parenting On Adolescent Family- And Work-related Values, Behaviors, And Aspirations. Psychological Bulletin, 111(1), 108-126.
 Amato, P. (1994). Life-Span Adjustment of Children to Their Parents' Divorce. The Future of Children, 4(1), 143-164.
 Simons, R. (1999), Kuei-Hsiu Lin, Leslie C. Gordon, Rand D. Conger and Frederick O. Lorenz. Explaining the Higher Incidence of Adjustment Problems among Children of Divorce Compared with Those in Two-Parent Families. Journal of Marriage and Family, 61(4), 1020-1033.
 Haas, H. (2004). The Impact of Different Family Configurations on Delinquency. British Journal of Criminology, 520-532.
 Murray, J., & Farrington, D. (2008). The Effects of Parental Imprisonment on Children. Crime and Justice, 37, 133-206.
 Wildeman, C., & Western, B. (2010). Incarceration in Fragile Families. The Future of Children, 20(2), 157-177.
 Wilson, W. (1996). Chapter 8. In When Work Disappears: The World of the New Urban Poor (p. 267). New York: Knopf.
Extended References with More Detail and Web Links Where Available
The following notes with links to the original papers, where not behind paywalls, are my original notes on each, and in the original order that I found them, which was somewhat random. Thinking at first that it was a mass of contradictions, I decided to try re-reading them in chronological order (as presented above) so as to better grasp the backwards references in the various studies. It was after doing this that I realized that there was a certain trajectory to these findings which made sense. Hence the ordering in my article here.
1) Simons, R. (1999), Kuei-Hsiu Lin, Leslie C. Gordon, Rand D. Conger and Frederick O. Lorenz. Explaining the Higher Incidence of Adjustment Problems among Children of Divorce Compared with Those in Two-Parent Families. Journal of Marriage and Family, 61(4), 1020-1033. Retrieved on February 14, 2015, from http://www.jstor.org/discover/10.2307/354021
This paper reviews several previous studies on the impact of divorce on subsequent successful social adjustment in children. At the time it was written, the consensus was that there was a relationship between divorce and adjustment problems, but the more specific causes of this had only been the subject of speculation.
Factors viewed as important in this study were the financial well-being of the custodial mother, her parenting skills, and the extent of continued involvement by the father. It was also shown that divorce is more emotionally disturbing to boys than to girls.
The study’s results support the notion that children of divorce are more at risk for adjustment problems than those of married couples. Couples can mitigate this risk by engaging in practices which avoid hostile exchanges while in the presence of children. Unfortunately, this mitigating effect applies primarily to girls and less so to boys, who are apt to experience depression anyway due to the loss of the father in the home (in cases of maternal custody).
2) Laub, J., & Sampson, R. (1988). Unraveling Families And Delinquency: A Reanalysis Of The Gluecks’ Data*. Criminology, 26(3), 355-380. Retrieved on February 14, 2015, from http://scholar.harvard.edu/files/sampson/files/1988_crim_laub.pdf
This study seeks to review the findings of Sheldon and Eleanor Gluecks’ 1950 work, Unraveling Juvenile Delinquency (UJD), which had been criticized for lack of rigor. The Glueck study was drawn from 500 delinquent and non-delinquent children in the cities of Westboro and Boston, Massachusetts during the 1940s.
A summary of objections to the Glueck study is presented, including concerns about causal ordering (establishing a cause and effect relationship), and sloppy data analysis. This paper attempts to reanalyze the original Glueck data and rescue it from these criticisms. The results are only based on a partial reanalysis, but conclude that further work can make productive use of the raw data that was gathered in the Glueck research, and that, in fact, this data was better in some ways than the then current (1988) data from other sources.
Family process variables are directly related to serious and persistent delinquency. Results support author’s version of social control theory. The factors of supervision, attachment, and discipline, as identified by the Gluecks, were the most important correlating factors with serious and persistent delinquency. This research confirms the findings of the Gluecks. Its second major finding was that, with the exception of residential mobility, none of the other structural background factors were significant. Family processes mediated 80% of these structural background factors. Previously neglected factors such as alcoholism and criminality of parents were found to be important. Finally, this study shows that the Glueck data can still be used to provide additional insights into the causes of delinquency.
3) Haas, H. (n.d.). The Impact of Different Family Configurations on Delinquency. British Journal of Criminology, 520-532. Retrieved on February 14, 2015, from
This is based on a sample of 21,314 Swiss male army recruits who completed a survey at age 20 in 1997. The large sample size and concentration on male impacts distinguished it from many earlier studies of the type. Its findings were that the loss of the mother as caregiver was a larger impact than the loss of father. Care by extended family members, however, was found to lessen this impact to some extent.
As with other studies, there were shown to be individuals that were exceptions to the rule who were able to avoid adjustment problems in spite of just the loss of one or both parents alone. The suggestion was made that there may be earlier factors that influence these children before the disruption occurs that need to be studied.
This study confirms that a warm and loving mother can mitigate the impacts of family break-up. Comparisons were made between post break-up habitation with mother, father, or other family members, with the best outcomes produced by continued involvement of the mother.
4) Murray, J., & Farrington, D. (2008). The Effects of Parental Imprisonment on Children. Crime and Justice, 37, 133-206. Retrieved on February 14, 2015, from
This paper claimed that this cause for family disruption is both understudied, yet significant, roughly tripling the risk for antisocial behavior. It also showed that the impact is worse when it is the mother that is the missing parent. It speculates that higher IQ, hopefulness, and social support may be moderating factors, as well as liberal prison policies and strong welfare provisions.
Parental imprisonment, particularly of the mother, was found to be an important risk factor in adverse outcomes for children. Numerous implications for policy and practice were discussed: Trauma Theories, Strained Caregiving, Economic Strain, and Stigma.
5) McCord, J. (1991). Family Relationships, Juvenile Delinquency, And Adult Criminality*. Criminology, 29(3), 397-417. Retrieved on February 14, 2015, from
This study seeks to eliminate variables of poverty, social disorganization, and poor neighborhood conditions by isolating families in similar economic circumstances. As with other studies, it found that the parenting skills of the mother can offset negative outcomes of broken families. The father’s relationship with the family can play an important role too, though, with the good or bad behavior of the father when interacting with the mother producing a corresponding impact on sons as they mature. Another finding was that there can be differences in the causative influence for juvenile delinquency versus adult criminal behavior.
This re-affirmed the value of a competent mother in insulating children from criminogenic influences, even in bad neighborhoods. Competent was defined as possessing self-confidence and providing leadership in an affectionate and non-punitive manner. The father’s influence was found to be less, especially in early years, but increased as children got older. Best results were obtained when cooperation existed between parents. Fathers who undermined the mother’s efforts yielded worst results. The impact of the father setting a good example of behavior was particularly important for boys.
6) Wells, L., & Rankin, J. (1991). Families And Delinquency: A Meta-Analysis Of The Impact Of Broken Homes. Social Problems, 38(1), 71-93. Retrieved on February 14, 2015, from http://socpro.oxfordjournals.org/content/socpro/38/1/71.full.pdf
This paper lamented the fact that there was so little correlation among the results of studies of the effects of broken homes on criminality. It demonstrated this by examining 50 such studies, and attempting to, where possible, find patterns of agreement which would suggest more reliable results in that area. For example, several of the studies included data for families broken up by either divorce or the death of one parent. Notably, a stronger correlation with crime was detected for the case of divorce.
Another useful result was that, given the wide range in time for the various studies, it could be concluded that there was no apparent change in the model between the earliest studies and the later ones. It also confirmed that the impact is stronger for minor juvenile offenses than for later adult criminality.
7) Mednick, B., Baker, R., & Carothers, L. (1990). Patterns of family instability and crime: The association of timing of the family's disruption with subsequent adolescent and young adult criminality. Journal of Youth and Adolescence, 19(3), 201-220. Retrieved on February 14, 2015, from
This study was based on 410 males aged 19-21 over an 18-year period and demonstrated that divorce alone could not explain subsequent criminality. The existence of a stable replacement family structure could compensate for the broken home effects. On the other hand, disruptive changes to the family structure after a divorce were found to be more likely the cause of the subsequent criminal behavior.
8) Amato, P. (1994). Life-Span Adjustment of Children to Their Parents' Divorce. The Future of Children, 4(1), 143-164. Retrieved on February 14, 2015, from http://futureofchildren.org/futureofchildren/publications/docs/04_01_08.pdf
This paper reviewed a number of studies on the topic, noting that at one time, death of a parent was the number one cause of broken homes, but that at the beginning of the twentieth century, it was overtaken by divorce. This was due to the combination of longer life spans and an increasing divorce rate. It covered the primary sampling techniques that had been used up to that point.
It also covered the method of meta-analysis which takes data from previous studies and combines them using different methods. As an example, the author references a study of his own that used data from 92 previous studies to draw new conclusions. Contrary to Wells, L., & Rankin, J. (1991), there appears to be a difference based on the time of the study, with more recent studies showing a decreased impact of broken homes on the likelihood of antisocial behavior. A possible explanation for this is that, with divorce becoming more common, the social stigma is lessened, and the increased number of children in this situation can provide support for one another. Additionally, the introduction of no-fault divorces leading to less acrimony between parents may also be a factor.
This study’s results somewhat contradict those of earlier studies on which is it based, and so it constitutes a reanalysis (meta-analysis) of some of that data, showing that the impact of divorce on children may be lessening as it becomes more common. This can be partially attributed to reduced stigmatization.
9) Wildeman, C., & Western, B. (2010). Incarceration in Fragile Families. The Future of Children,20(2), 157-177. Retrieved on February 14, 2015, from
This paper presented strong support for sentencing reform to avoid the impact that lengthy incarceration, or incarceration at all, for minor drug offenses has on family break-ups and the resulting impact on children. It pointed out that imprisonment had increased by a factor of five times since the mid-70s and suggested that it has a greater impact on poor families that are already at a disadvantage.
Solutions, however, must include solving problems that lead up to imprisonment in the first place, such as joblessness, untreated addictions, and mental illness. Comparisons are made between the U.S. prison system and that of Western Europe. That has had a cascading effect that first showed up in the U.S. starting in the 80s.
10) Barber, B., & Eccles, J. (1992). Long-term Influence Of Divorce And Single Parenting On Adolescent Family- And Work-related Values, Behaviors, And Aspirations. Psychological Bulletin, 111(1), 108-126. Retrieved on February 14, 2015, from
This survey and summary of other research concluded that it cannot be demonstrated that keeping a two-parent structure in place will always produce the best outcome in a particular case. Rather, in cases where there may be a continued conflict if the parents stay together, it might actually produce a better outcome for them to separate early. This echoes other suggestions that it is not the absence of two parents at all which causes increased criminality among children, but rather, the stress of animosity between parents that is a more important factor.