Making judicial decisions with a purely random selection of information from an offender's file

Eddy M. Elmer

Simon Fraser University, March 2002

Foreword

        Every day, judges, parole board members, and other professionals in the criminal justice system are required to make a large number of decisions as to the sentencing and/or early release (parole) of criminal offenders. Under most circumstances, professionals have only a limited amount of time in which to make these decisions. The challenge, therefore, is to make as many high-quality decisions as possible in the shortest period of time possible.

        The following is my response to a decision-making exercise that is regularly used in training legal professionals. In this exercise, trainees are given a box containing 60 cards of information on all aspects of an offender's life (e.g., any previous crimes, medical and mental health history, family circumstances, education and school history, etc.). One card contains information about the current offence. The information in the file is taken from an actual case file (the name of the offender has, of course, been changed). The goal of the exercise is to make the best decision possible using the fewest number of cards possible. Any card(s) can be chosen, using any criteria or "system". One can use no cards, all 60 cards, or any number in between.

        I completed this training exercise two times: once in 1999 and again in 2002. In this response, I outline the specific steps I followed in both years, the reasoning behind the steps, and what I see as the benefits and disadvantages of making decisions the way I did in each of the two years. The system I used in the second year is of particular interest, because it involved selecting a specific number of information cards at random.

Introduction

        The Robert Côté* exercise provided an excellent opportunity to actively reflect upon the biases and prejudices that inevitably shape and direct the judgements and decisions I make regarding other people's lives. Granted, I am not a judge or parole board member, so I do not make decisions which determine other people's freedom and livelihood. However, during any given day, I make a multitude of smaller decisions which in some way affect the people around me. Taken together, the sum of these smaller decisions could very well be as powerful as the decisions that a judge or parole board member makes in the case of an offender.

        Regardless of the type of decision being made, I think the dynamics of an individual's decision-making process remain fairly consistent across situations. We are human, and, for better or for worse, we are continuously influenced by bias, prejudice, attitudes, emotions, memories, and our own personal experiences. But we have also been blessed with the capacity to be aware of these influences (which are usually unconscious) and to devise ways to prevent them from clouding our better judgement. In this journal assignment, I would like to discuss the unusual decision-making method I employed when determining an appropriate disposition in the Robert Côté case. In particular, I would like to illustrate how I feel this method helped me to avoid becoming overly-influenced by subjective factors[1].

        For sake of organisation, the first part of this response will concern the actual decision I made in this case (including how I chose information cards). The second part will be a detailed discussion of the decision-making process itself.

How I see my role as a decision-maker

        I think the most important role of the judge or parole board member is to determine the disposition that will be of greatest benefit to both community and offender. I base this opinion on the fact that the offender affects his community, the community affects the offender, and both must live together. To make a disposition that will be optimally beneficial to everyone involved, I need to use the best decision-making process I can come up with. Whether or not this method is seen as unorthodox by my peers must remain of little concern to me. Later in this paper, I will discuss some of the reactions my peers had to the process I used in this exericse.

How I chose information in this case

        The process by which I selected information in this matter is quite simple. First, I wanted to determine the nature of the crime at hand. In the interests of public safety, this is one piece of information I felt I absolutely needed. If the crime at hand were of a violent nature, my disposition might have required incarceration or hospitalisation in order to prevent further violence. Thus, I chose card 33: "Description of delinquency and of surrounding circumstances". Fortunately, Mr. Côté's delinquency was not of a significantly violent nature. Second, I chose 10 additional cards at random-a sample representing approximately 15% of all 60 cards available for my use. These 10 cards I used to determine any extenuating circumstances in the commission of the crime, and, thus, the direction my disposition would take. I chose 15% arbitrarily: 15% seems to be enough to yield a sample fairly representative of all the available cards. The premise for this process will be discussed shortly.

Cards used

My current disposition

1. probation up to 1 year (I felt 3 years was unmanageable, because Mr. Côté had a high probability of re-offending, and to send him back to court for re-offending would only discourage whatever progress he might make)
2. an order of compensation-in-kind or by way of personal service to the victim (whatever the victim feels is appropriate)
3. placement in group home until Mr. Côté can find a job and his own housing
4. mandatory weekly career and personal counselling
5. report to me, the judge, every 3 months (to demonstrate that somebody actually cares about his progress)

The primary bases for the current disposition are as follows:

1. giving back to the community (i.e., restorative justice)
2. treatment of offender's underlying emotional problems
3. addition of structure to the offender's life (by way of weekly counselling sessions, group home residence, and regular progress reporting to the judge); such structure may be effective in giving the offender some direction in his life, which may in turn prevent a return to a chaotic, insecure life of delinquency

Premises upon which my decision-making method is based

        When I completed this exercise a year ago, the decision-making process I used strongly reflected my own biases on a number of issues. The main bias was that those who commit crimes usually have some kind of underlying emotional or family-related problem. Such a problem must be resolved if we are to minimise the likelihood of recidivism. Firmly entrenched in this bias, I systematically chose all those cards which I felt contained some kind of information about Mr. Côté's mental health. The result was a decision that saw Mr. Côté released on probation and placed in mandatory counselling. Even though my dispositions in both years were basically the same, I want to point out how the method used a year ago could have led to a very poor decision. Then I would like to show how the method I used a few weeks ago circumvented these potential pitfalls.

        When I completed this same actively 15 months ago, several things went over my head:

1. It did not occur to me[2] that offenders can easily lie about their mental health and deceitfully paint themselves as troubled individuals more in need of treatment than of punishment. Many offenders are good criminals precisely because they are good liars-they can lie to commit their crimes, and they can lie to avoid the consequences.
2. Psychological tests, personality inventories, and medical experts' opinions are not 100% valid and reliable. In fact, they are only modestly valid and reliable, and need to be taken in the context of the offender's entire history.
3. Not all offenders have mental disorders.
4. While having a difficult, disorganised, or abusive family increases the likelihood that one will not behave in prosocial ways, not all individuals coming from bad homes commit crimes.
5. Not all offenders are amenable to counselling and psychotherapy.

        If, hypothetically, Mr. Côté had feigned mental illness and fabricated stories of familial abuse, I still might very well have given him probation under condition of mandatory counselling. Meanwhile, he might have happily gone on to commit additional crimes-perhaps more serious than automobile theft, and perhaps including violence-with little reason, motivation, or incentive to stop. While it will probably always remain a question whether criminals are psychologically "fit" and whether a crime can be committed simply because an individual wants to commit it or is "evil", to allow my decisions to be based primarily on psychological factors reflects a bias on my part-a bias that can encourage me to ignore other realities.

        When I did this exercise a few weeks ago, I wanted to find a way to minimise my biases. Thus I considered different pieces of information at random. My primary belief was that if there were any extenuating circumstances in this case (psychological, familial, financial, whatever), the circumstances would somehow be reflected in the 10 cards I chose at random. This is based upon my steadfast belief that all aspects of our lives affect one another. Thus, if Mr. Côté suffers a from significant emotional difficulties or comes from a very difficult home, I do not necessarily have to read about it in a psychologist's report. I can see these circumstances reflected in his family's openly hostile emotions towards him (e.g., card 60); the lack of support from his family (card 6); his relative isolation from the community (card 36); his institutional experience (card 17); and so forth. The same would apply to non-psychological circumstances. For instance, if finances were the primary problem[3], then Mr. Côté might have few clean clothes and appear unkempt (card 8); might have problems with accommodation (cards 13, 16); and might be having arguments with family about financial support (card 6). Clearly, it is difficult to determine which circumstances affect which pieces of information-and not all pieces of information in Mr. Côté's file might reflect the same extenuating circumstances-but I think the benefits of this procedure somewhat outweigh these pitfalls. Before these pitfalls are discussed, I would first like to expand on the benefits:

1. In choosing only 15% of all available pieces of information, I do not have to read an entire file. Thus, I save myself much time.
2. A random sampling of information gives me a greater chance of discovering various extenuating circumstances. Certainly, I may miss a critical piece of information, but I think the chances of this happening are far greater when I focus myself exclusively on one or two types of information.
3. A random sampling of information helps me to get to know the offender better-which in turn helps me to create the most personalised corrections plan possible. It also gives me the background information I need I order to "connect" with the offender; to establish some sense that someone cares for him; and to encourage him to be genuine and honest with me.
4. Most importantly, a random sampling prevents me from following only one train of thought. It increases the chances that I will consider numerous perspectives in the case, and it prevents me from over-relying on information that may be false or misleading.

Limitations of this method

        A primary concern with my method may be that I am not assessing enough information. A random selection of cards may not capture the full extent of various extenuating circumstances. While I believe that it might be wise to increase the percentage of cards chosen from 15% to 25%, I still feel that a random selection offers better quality of information than a subjectively-driven selection.

        Many people will argue against my method on cause-and-effect grounds. If some of the cards drawn at random reflect Mr. Côté's psychological difficulties, can we reasonably assume that the circumstances and factors described on those cards were caused by psychological difficulties? The answer is that no, we probably cannot. However, I do not think it is important to ask this question, because whatever the causes, the circumstances and factors represented by those cards are appearing over and over. And factors which are appearing repeatedly are probably the ones that should be addressed first in any corrections plan. Indeed, helping an offender involves breaking the cycle of delinquency at any point in the cycle. If psychological factors are a big part of the cycle (which to me they are, based upon my random selection of cards), then they should be addressed immediately. They may not be what caused the delinquency in the first place, but they are clearly playing a big role.

        It may also be argued that if certain extenuating circumstances inevitably affect all aspects of an individual's life, then any extenuating circumstances which are fabricated can easily affect information relating to all other parts of the offender's life. Indeed, as the Scottish novelist, Sir Walter Scott, said, "O, what a tangled web we weave, // When first we practise to deceive!" However, lies do not consistently affect all aspects of our lives. Inevitably, there will be inconsistencies in the facts, and a random selection of information will give anyone plenty of opportunities to cross-check the facts in order to determine the overall cohesiveness of an offender's file. In fact, reviewing verbatim transcripts of an individual's actual discourse with another individual (regardless of the content of the discourse) can provide a good opportunity to determine lack of factual cohesiveness (e.g., Grice, 1989).

        One could also suggest that even though I am controlling for bias in the information-gathering phase of decision-making, there is still bias in terms of what disposition I make based upon the information gathered. With this I cannot argue. But I can say that if a significant amount of bias occurs during information-gathering phase and we can eliminate it to some extent, then, overall, we are making less biased decisions than if we were relying on our usual methods. My actual disposition is clearly biased-towards helping Mr. Côté overcome his emotional and familial issues. This reflects my own philosophy that many offenders are troubled individuals and require more than mere punishment. However, this philosophy is based upon a thorough fact-gathering method that I feel is less biased than alternative methods. Indeed, I think it would be worse to base such a philosophy on misleading of false information.

        It may also be argued that the very notion of choosing information at random is a biased affair. Clearly, it is. The bias is that random selection somehow reflects truth more effectively than our own intuition and reasoning. But I would like to think that this bias is based on observations and research which repeatedly demonstrate the ineffectiveness of our own mental shortcuts, intuition, and emotions (e.g., Gilovich, 1991). In any event, when the day comes that we can determine truth with 100% accuracy, I will gladly abandon this strange decision-making method!

        I am sure there are many other problems with my random selection procedure, but these are the ones that come to mind at present.

Conclusions: Group vs. individual decision-making

        A question I keep asking myself is, Did I make a "better" decision than did my classmates? It is very difficult to answer that, because few people told the class how they made their decisions. They told us what their decisions were, and they made it clear what their philosophies are with respect to punishing/treating offenders, but few of them explained their rationale for choosing the information they did. I would have to say that whatever ultimate decision they made with respect to Mr. Côté's correctional plan, the best decision-makers were probably those who made a conscious effort to either challenge their biases, or to at least recognise and "hold them at bay".

        Personally, I feel that there was an increased probability that my classmates' biases were challenged because they were working individually. I think that spending time with their own thoughts gave them more opportunity to actively question themselves and experiment with different decision-making processes. A group setting would have:

1. distracted them from their own thought processes
2. encouraged narrow or highly-biased lines of thought (e.g., focussing exclusively on experts' reports)
3. created pressure to conform to a particular decision-making style in order to save time and to create group cohesiveness (e.g., Janis, 1982)
4. encouraged more discussion about the disposition per se, and not the process of deliberations[4]

        When we were driving up to the Kent Institution last week, our carpool discussed the Robert Côté exercise. In particular, we asked one another whether we preferred to work alone or to work as a group. The general consensus was that both situations had their benefits, and were equally preferable. I was in the minority when I stated that I generally prefer doing such exercises on my own. One of my classmates asked, "What does that say about you that you don't like to work in a group?" I responded that all it said was that I prefer not to work in a group when doing such exercises.

        Once again, the implication was that there is something wrong with experimenting and trying bold new things, especially when it would not be well-received by a group. After some period of silence, I commented that my preference for solitary work was not indicative of a dislike for people. In actuality, my preference for solitary decision-making stems from a concern for other people. Even though the decisions I make on a day-to-day basis do not affect people as significantly as do decisions which judges make, they do have some impact-no matter how small. If I can encourage myself to be more cognisant of my biases, I think I can avoid doing thoughtless things which negatively affect the people around me.

Notes

*The name of the offender has been changed.

1. I should note that it is impossible to completely divorce ourselves from our biases, emotions, attitudes, and so forth. To do so would turn us into machines-and machines have consistently proven to be poor decision-makers! If we cannot experience our own biases and emotions, it becomes difficult to think about those of the offender. And when we can no longer see an offender as having his/her own subjective states, we often inadvertently see that person as just another body, to whom we can carelessly and frivolously apply any decision whatsoever-whether appropriate or not. Indeed, as the American senator , J. William Fulbright, once said, "We have the power to do any damn thing we want to do, and we seem to do it about every ten minutes." Furthermore, to completely divorce ourselves from our biases and emotions is a tremendously time-consuming and energy-robbing experience-because it is a battle that cannot be won. It seems to me more prudent to simply be aware of our biases and emotions and temporarily channel them in the appropriate directions (one of which may be towards the back of our mind).
2. Or I did not want to believe this because it was inconsistent with my attitudes and values. In other words, it would have made me feel uncomfortable.
3. In Mr. Côté's case, finances are a big problem, but they seem to be more the result of emotional and family difficulties than anything else.
4. It seems to me that people are more inclined to debate personal values regarding crime and punishment than they are to debate each other's fundamental thought processes. I think this happens because we accept that values are relative-but we do not want to accept that thought processes are relative. When we accept that thought processes are relative, we feel lost in determining what is "normal" and "sane" thinking. And when that happens, we sometimes start to question our own sanity!
   
   

References

Gilovich, T. (1991). How we know what isn't so: The fallibility of reason in everyday life. New York: The Free Press.

Grice, P. (1989). Studies in the way of words. Cambridge, MA: Harvard University Press.

Janis, I.L. (1982). Groupthink. Boston: Houghton Mifflin.