Impersonal Rules vs Discretion and Kadi Justice
Kahneman & Sunstein 2021 on Interpersonal Noise, plus Heller
Daniel Kahneman, Olivier Sibony, Cass R. Sunstein wrote:
In the 1970s, the universal enthusiasm for judicial discretion started to collapse for one simple reason: startling evidence of noise. In 1973, a famous judge, Marvin Frankel, drew public attention to the problem … Frankel called on Congress to end this “discrimination,” as he described those arbitrary cruelties … Writing in the early 1970s, he did not go quite so far as to defend what he called “displacement of people by machines”. But startlingly, he came close. He believed that
“the rule of law calls for a body of impersonal rules, applicable across the board, binding on judges as well as everyone else”.
He explicitly argued for the use of “computers as an aid toward orderly thought in sentencing.” He also recommended the creation of a commission on sentencing. Frankel’s book became one of the most influential in the entire history of criminal law— not only in the United States but also throughout the world …
… In the 1970s, Frankel’s arguments, and the empirical findings supporting them, came to the attention of Edward M. Kennedy, brother of the slain president John F. Kennedy, and one of the most influential members of the US Senate. Kennedy was shocked and appalled …
… Responding to the evidence of unjustified variability, Congress enacted the Sentencing Reform Act of 1984. The new law was intended to reduce noise in the system by reducing “the unfettered discretion the law confers on those judges and parole authorities responsible for imposing and implementing the sentences.” …
… [The] guidelines ran into a firestorm of criticism. Some people, including many judges, thought that some sentences were too severe — a point about bias, not noise. For our purposes, a much more interesting objection, which came from numerous judges, was that guidelines were deeply unfair because they prohibited judges from taking adequate account of the particulars of the case. The price of reducing noise was to make decisions unacceptably mechanical. Yale law professor Kate Stith and federal judge José Cabranes wrote that “the need is not for blindness, but for insight, for equity,” which “can only occur in a judgment that takes account of the complexities of the individual case.”
This objection led to vigorous challenges to the guidelines, some of them based on law, others based on policy. Those challenges failed until, for technical reasons entirely unrelated to the debate summarized here, the Supreme Court struck the guidelines down in 2005. As a result of the court’s ruling, the guidelines became merely advisory. Notably, most federal judges were much happier after the Supreme Court decision. Seventy-five percent preferred the advisory regime, whereas just 3% thought the mandatory regime was better.
… What have been the effects of changing the guidelines from mandatory to advisory? Harvard law professor Crystal Yang investigated this question, not with an experiment or a survey but with a massive data set of actual sentences, involving nearly four hundred thousand criminal defendants. Her central finding is that by multiple measures, interjudge disparities increased significantly after 2005. When the guidelines were mandatory, defendants who had been sentenced by a relatively harsh judge were sentenced to 2.8 months longer than if they had been sentenced by an average judge. When the guidelines became merely advisory, the disparity was doubled. Sounding much like Judge Frankel from forty years before, Yang writes that her “findings raise large equity concerns because the identity of the assigned sentencing judge contributes significantly to the disparate treatment of similar offenders convicted of similar crimes.”
After the guidelines became advisory, judges became more likely to base their sentencing decisions on their personal values. Mandatory guidelines reduce bias as well as noise. After the Supreme Court’s decision, there was a significant increase in the disparity between the sentences of African American defendants and white people convicted of the same crimes. At the same time, female judges became more likely than male judges were to exercise their increased discretion in favor of leniency. The same is true of judges appointed by Democratic presidents. Three years after Frankel’s death in 2002, striking down the mandatory guidelines produced a return to something more like his nightmare: law without order …
… In 1973, Judge Marvin Frankel was right to call for a sustained effort to reduce noise in criminal sentencing. His informal, intuitive noise audit, followed by more formal and systematic efforts, uncovered unjustified disparities in the treatment of similar people. Those disparities were outrageous. They were also startling.
Much of this book can be understood as an effort to generalize Frankel’s arguments and to offer an understanding of their psychological foundations.
To some people, noise in the criminal justice system seems uniquely intolerable, even scandalous. But in countless other contexts, it is not exactly tolerable, as supposedly interchangeable people in the private and public sectors make different judgments on the job. In insurance, recruitment and evaluation of employees, medicine, forensic science, education, business, and government, interpersonal noise is a major source of error. We have also seen that each of us is subject to occasion noise, in the sense that supposedly irrelevant factors can lead us to make different judgments in the morning and in the afternoon, or on Monday and Thursday.
But as the intensely negative judicial reaction to the sentencing guidelines suggests, noise-reduction efforts often run into serious and even passionate objections. Many people have argued that the guidelines are rigid, dehumanizing, and unfair in their own way … The rules in question may seem stupid and even cruel, but they may have been adopted for a good reason: to reduce noise (and perhaps bias as well).
Even so, some efforts to reduce noise raise serious concerns, perhaps above all if they make it difficult or impossible for people to get a fair hearing. The use of algorithms and machine learning has put that objection in a new light. No one is marching under a banner that says “Algorithms now!” …
… If the goal is to reduce noise or decide how and whether to do so (and to what degree), it is useful to distinguish between two ways of regulating behavior: rules and standards. Organizations of all kinds often choose one or the other or some combination of the two …
[Our] examples illustrate the central distinction between rules and standards. Rules are meant to eliminate discretion by those who apply them; standards are meant to grant such discretion. Whenever rules are in place, noise ought to be severely reduced …
… Under rules, the enterprise of fact-finding may itself involve judgment and so produce noise or be affected by bias. We have encountered many examples. But people who design rules aim to reduce those risks, and when a rule consists of a number (“no one may vote until they reach the age of eighteen” or “the speed limit is sixty-five miles per hour”), noise should be reduced. Rules have an important feature: they reduce the role of judgment. On that count, at least, judges (understood to include all those who apply rules) have less work to do. They follow the rules. For better or worse, they have far less room to maneuver.
Standards are altogether different. When standards are in place, judges have to do a lot of work to specify the meaning of open-ended terms. They might have to make numerous judgments to decide what counts as (for example) “reasonable” and “feasible.” In addition to finding facts, they must give content to relatively vague phrases. Those who devise standards effectively export decision-making authority to others. They delegate power.
The kinds of guidelines discussed [earlier] might be rules or standards. If they are rules, they dramatically constrain judgment. Even if they are standards, they might be far from open-ended. Apgar scores are guidelines and not rules. They do not forbid some exercise of discretion. When guidelines are tightened so as to eliminate that discretion, they turn into rules. Algorithms work as rules, not standards …
… In business and in government, the choice between rules and standards is often made intuitively, but it can be made more disciplined. As a first approximation, the choice depends on just two factors: (1) the costs of decisions and (2) the costs of errors.
With standards, the costs of decisions can be very high for judges of all kinds, simply because they have to work to give them content. Exercising judgment can be burdensome …
… [I]t’s complicated. Rules may be straightforward to apply once they are in place, but before a rule is put in place, someone has to decide what it is. Producing a rule can be hard. Sometimes it is prohibitively costly …
… The costs of errors refer to the number and the magnitude of mistakes. A pervasive question is whether agents are knowledgeable and reliable, and whether they practice decision hygiene …
… Of course there is a continuum from perfect trust to complete distrust. A standard might lead to numerous errors by less-than-trustworthy agents, but if those errors are minor, they might be tolerable. A rule might lead to only a few mistakes, but if they are catastrophic, we might want a standard. We should be able to see that there is no general reason to think that the costs of errors are larger with either rules or standards. If a rule is perfect, of course, it will produce no errors. But rules are rarely perfect …
… Outlawing Noise? In many situations, noise should be a scandal. People live with it, but they should not have to do that. A simple response is to shift from open-ended discretion or a vague standard to a rule or something close to it. We now have a sense of when the simple response is the right response. But even when a rule is not feasible or not a good idea, we have identified an assortment of strategies to reduce noise.
All this raises a large question: should the legal system outlaw noise? It would be too simple to answer yes, but the law should be doing much more than it now does to control noise. Here is one way to think about the problem. The German sociologist Max Weber complained of “Kadi justice”, which he understood as informal, ad hoc judgments undisciplined by general rules. In Weber’s view, Kadi justice was intolerably case by case; it was a violation of the rule of law. As Weber put it, the judge “precisely did not adjudicate according to formal rules and ‘without regard to persons.’ Just the reverse largely obtained; he judged persons according to their concrete qualities and in terms of the concrete situation, or according to equity and the appropriateness of the concrete result.”
This approach, Weber argued, “knows no rational rules of decision”. We can easily see Weber as complaining about intolerable noise that Kadi justice ensured …
… We would not go so far as to say that noise reduction should be part of the Universal Declaration of Human Rights, but in some cases, noise can be counted as a rights violation, and in general, legal systems all over the world should be making much greater efforts to control noise.
The Source:
Daniel Kahneman, Olivier Sibony, Cass R. Sunstein, Noise: A Flaw in Human Judgment, Harper Collins 2021 [pp. 12-13, 14, 198, 213, 217, 218]
Further readings on Rules vs Discretion, and, recently, on complementary Intuitions:
#1 Michael G. Heller, Capitalism, Institutions and Economic Development, Routledge 2009, 2011
My book is almost entirely about the dangers of discretion and the rationale for beneficial general rules. It has a whole chapter on the evolution of law in that context. The book introduces the same core argument/insight as Kahneman et al above but from the perspectives of sociology and political economy rather than psychology.
#2 Michael G. Heller, ‘Institutions of Capitalism And Their Decay’ June 21, 2012
Too much discretion at the margin and inadequate boundary rules at the centre make capitalism an unhappy place to be. This … hints at solutions. Radically simplify governance. Automate governance wherever possible. Replace plethoras of regulations with constitution-level rules. Take unnecessary welfare burdens off representational politics. Foster private insurance innovations that devolve risk back to individuals.
#3 Michael G. Heller, ‘Hellerian Institutional Interactions’ July 13, 2014
The institutional spheres are separate, and they need to understand each other in order to work together. They must, in other words, read each other’s minds. The activated impersonal procedural norm is the communication code that enables functional mindreading by stipulating that all governance relationships are between large impersonal function-categories of office, occupation, citizenship, and so on. All categories are to be treated equally …
Expectations are grounded in universal criteria of impersonality. There is no emotion, no personal favour, minimal discretion. These rules apply to every person without exception. My description is in need of further thought and fine tuning, but I hope it indicates a convergence between the ordoliberal and neoliberal visions of institutional subsystem interaction …
In my current work I expect to be able to show why some of the most important reform initiatives, which were not conceptualised as ‘depersonalising’, were nevertheless plainly driven by demands to remove people or people-dependent decisions — especially the personal entitlement to arbitrary discretion — from governance systems. In turn, the great revolutions of the 17th century were stepping stones in a longer series of radical disequilibrium transformations that further depersonalised institutions all the way down to the present day …
Through coping repeatedly with complexity and uncertainty, the institutions that provide state services of law, administration and representation progressively reduce the human element of management and control by simplifying the interaction rules and subtracting discretional decisions.
#4 Michael G. Heller 'Autistic Institutions (Part 2 Double Cognition)’ August 05, 2014
Another historian perceptively examines the “depersonalisation” of state administration, starting in the 16th century and advancing thereafter with periodic bursts of energy led by named systemising individuals in counsels that claimed to bear disinterested moral logic and authority. It is no irony or coincidence that key persons figured prominently in each drive to replace persons with laws, to, in effect, automate some decisions. “Routinisation”, “codification” and “rule-making”, and “formalisation” of “elaborate repetitive procedure” eliminated arbitrariness, discretion and caprice. Bureaucratisation, if viewed autistically, had the objective of removing uncertainty from social interactions of governance. As the state expanded its functions it became more complex. Order and predictability had to be created anew with every change in financing, recruitment, remuneration, and discipline.
#5 Michael G. Heller, ‘Learning Code from Wittgenstein [Part 2]’ May 28, 2021
We could well imagine that the trainee state clerk finally experiences a flash of insight. Intuitive insight about the subliminal rule is converted into something like an impersonal bureaucratic routine, individualised in the mind of the person, while corresponding correctly to a system-impelled routine of state machinery in action.
The state clerk has always been offered collegial advice, in training or on the job, about how to interpret and respond to the sudden complexities presented to clerks in a seemingly random manner. The trick he performs is to put particular kinds of situation in one of a limited range of general conceptual boxes that correspond to a familiar universal sort of decision: “Look, I always do the same thing”, meaning that in similar circumstances I will make the decision impersonally rather than use my discretion. I follow the rule.
#6 Michael G. Heller, ‘Learning Code from Wittgenstein [Part 3]’ May 29, 2021
What about the operational aspect? In language there are many nuances to be considered in order to establish understanding of meaning. Coded responses to interactions in governance are likely to be less modulated than in language. The decision choice is more either-or; habituation to an order that one was trained to obey without much need for personal deliberation. The intuition lies in knowing in a flash where and when to repeat the rule.
This does not eliminate subtle distinctions. There are programable elements in governance where exceptions can and should be made without subverting the rule, and where the response will not be automatic or instinctive. Wittgenstein’s conception of language rules take us only some of the way. Much more needs to be learned about programatic discretion, shading, or even blunt exceptions to the rule, and where the limits lie.
What about origins? During novel events, which present more or less tabula rasa conditions, the role of instinct, intuition and rule-following can be detected not in sheer repetition but in a post-hoc inference from the events themselves. Even then the rule decision accords with the actor’s experience of previous contexts, rather than unique insight. The history must take account not only of the events, individuals, and stated intents, but also the underlying philosophies, the long run structural or environmental context, catalytic processes, and system dynamics. The observer has the task of interpreting the actions and reactions of individuals who experienced the intuition and recognised the rule.
The Files Category/Series:
Rules vs discretion and impersonal coding in transmutations from foraging societies onwards and eventually to the post-modern age of artificial intelligence.