10 Three Questions For Your Just Culture

Let us now put your organization together with the country, the legal system, and the society in which it is situated. The previous chapter shows that creating a just culture in your organization cannot be done in a vacuum. The outside world can impose all kinds of rules and expectations on you. These may relate to your recording and keeping of safety data, your handling of the aftermath of incidents, and your organization's compliance with a variety of laws and regulations. Your ability to create just culture internally is also affected by the legal climate of the country in which you operate.

Here you will see how organizations from a number of different countries (whether they know it or not) seem to settle on pragmatic solutions that allow them to retain some balance between learning and accountability. These solutions also help them deal with pressures from the outside world which may just interfere dramatically with their ability to create a just culture on the inside. When you look at these "solutions" a little more closely, you can see that they really boil down to answers to three central questions:

1. Who in the organization or society gets to draw the line between acceptable and unacceptable behavior?
2. What and where should the role of domain expertise be in judging whether behavior is acceptable or unacceptable?
3. How protected against judicial interference are safety data (either the safety data from incidents inside of the organization or the safety data that come from formal accident investigations)?

The differences in the directions that countries or organizations or professions are taking toward just cultures come down to variations in the answers to these three questions. Some work very well, in some contexts, others less so. Also, the list of solutions is far from exhaustive, but it could inspire others to think more critically about where they or their organization may have settled (and whether that is good or bad).

In general, though, we can already see this for the three questions:

On question 1) The more a society, industry, profession, or organization has made clear, agreed arrangements about who gets to draw the line, the more predictable the managerial or judicial consequences of an occurrence are likely to be. That is, practitioners will suffer less anxiety and uncertainty about what may happen in the wake of an occurrence, as arrangements have been agreed on and are in place.

Anxiety that results from uncertainty about what may happen is really bad, and in many ways unnecessary. For example, while the real risk of being sued for medical malpractice is low, doctors perceive it to be high. And it is perception that creates anxiety, not reality. The anxiety leads to defensive medicine, unnecessary tests, rather than high-quality care. Even if you, as an organization, have to tell your people that the aftermath of an incident may be unpleasant for them, that is still better than not telling them anything at

On question 2) The greater the role of domain expertise in drawing the line, the less practitioners and organizations may be likely to get exposed to unfair or inappropriate judicial proceedings.

There is actually no research that suggests that domain experts automatically prevent the biases of hindsight slipping into their judgments of past performance. Hindsight is too pervasive a bias. It takes active reconstructive work, for everyone, to even begin to circumvent its effects. Domain experts, however, do have an easier time forming an understanding of the situation as it looked to the person at the time, as they probably know such situations from their own experience. Here is how that may influence their ability to make a fairer judgment of the controversial action:

• It is easier for domain experts to understand where somebody's attention was directed. This is one area where domain experts may have an easier time avoiding the hindsight bias: even though the outcome of a sequence of events will reveal (in hindsight!) what data was really important, domain experts can make better judgments about the perhaps messy or noisy context of which these, now critical, data were part and understand why it was reasonable for the person in question to be focusing on other tasks and attentional demands at the time.
• It is likely easier for domain experts to understand the various goals that the person in question was pursuing at the time, and whether these were reasonable given the circumstances, and whether and how these goals may have conflicted with each other (e.g. safety versus efficiency, production versus protection). Domain experts can also form a better judgment than outsiders about the reasonability of goal priorities in cases of goal conflicts, especially since the system's preference for one goal over another may have been expressed tacitly, without explicitly stating it. Outsiders would not likely get access to that kind of information.
• For domain experts it is also easier to assess whether any unwritten rules or norms may have played a role in people's behavior. All professions have unwritten rules or norms, to which members of the profession are essentially supposed to perform. Without conforming to these tacit rules and norms, people often could not even get their work done. The reason, of course, is that written guidance and procedures are always incomplete as a model for practice in context. That means that practitioners need to bridge the gap between the written rule and the actual work-in-practice, which often involves a number of expert judgments. Outsiders often have no idea about the existence or subtlety of these norms, and would perhaps not appreciate their importance or relevance for getting work done.

That said, domain experts may have other biases that work against their ability to fairly judge the quality of another expert's performance. There is, for example, the issue of psychological defense: If experts were to affirm that the potential for failure is baked into their activity and not unique to the practitioner who happened to inherit that potential, then this makes them vulnerable too. Sometimes it can be more comforting to think that the errors made by a fellow practitioner would not happen to you; that they really are unique to that other person.

On question 3) The better protected safety data is from judicial interference, the more likely practitioners could feel free to report. The protection of this safety data is connected, of course, to how the country or profession solves questions 1) and 2). For example, countries or professions that do protect safety data typically have escape clauses, so that the judiciary can gain access "when crimes are committed," or in "justified cases when duly warranted," or "for gross negligence and acts sanctioned by the criminal code." It is very important to make clear who gets to decide what counts as a "crime," or "duly warranted" or "gross negligence," because any uncertainty there (or the likelihood of non-domain experts making that judgment) can once again hamper practitioners' confidence in the system and their willingness to report or disclose.

Local Solutions to the Three Questions

Local solution 1: Do nothing to actively handle the three questions

This is a solution that a number of country or professions apply, perhaps because they have not yet been confronted by the consequences of judicial action against practitioners or have themselves seen the difficulty of acting in the wake of failure. This may, of course, be just a matter of time.

1. Who gets to draw the line: In most likelihood a prosecutor who has become inspired by media reports or other triggers that made her or him look more closely into an occurrence. General risk statutes, or other laws, can be used to accuse practitioners of, for example, endangering the lives of other people. Access to data to build a criminal case should be relatively easy if the country or profession has not done much or anything to prevent such judicial intrusions into their safety data. The prosecutor draws the line in the first instance, and then the judge (or jury) gets to decide.
2. The role of domain expertise is likely minimal in judging whether a line of acceptability was crossed or not. A prosecutor, for example, has none or limited domain expertise, yet she or he gets to demonstrate whether professional judgments are culpable or not. The judge is not likely to have any domain expertise either.
3. Protection of safety data is not likely to exist, and even if it does, then a country or profession that goes by local solution 1 probably has caveats in its protection so that a prosecutor can open up databases upon suspicion of a crime (and the prosecutor is often the one who decides when that is the case).

Consequences: Practitioners may feel uncertain and anxious about whether "they will be next" because the rules of criminalization are left unclear and open to interpretation. Who gets criminalized for what seems random. A just culture is a long way off, and open and honest reporting could be difficult.

Local solution 2: The volatile safety database

Some countries or professions who do not actively handle the three questions in legislation or cross-disciplinary arrangements (e.g. between their departments of transportation or health on the one hand and justice on the other) spontaneously call for the creation of a next local solution: The destroyable safety database. This means that the safety data that organizations gather themselves are stored in a form that is very easy and quick to destroy. Some safety departments have seriously considered this idea, so as to immunize themselves against prosecution. This is especially the case in countries where the organization's personnel are themselves government employees (such as some hospital workers or air traffic controllers) and can thus be forced, through various statutes and laws, to hand over anything that belongs to the State.

1. Who gets to draw the line: Same as local solution 1.
2. The role of domain expertise: Same as local solution 1.
3. Protection of safety data could be guaranteed, as the data will vanish when prosecutorial pressure is applied. The cost, of course, is huge: The disappearance of an organization's safety database (which can in turn violate other statutes).

Consequences: This is not really a practical solution because of the consequences of destroying a database. But that it is being considered in several countries or professions in the first place should serve as an indication of the lack of trust necessary for building a just culture. The relationship between the various stakeholders may be troubled or underdeveloped. The suspicious climate sustained by this solution will not be good for the growth of a just culture.

Local solution 3: Formally investigate beyond the period of limitation

In almost all countries, prosecutors have a limited number of years to investigate and prosecute crimes. In one country, the investigation of an accident took so long that the so-called period of limitation for any possible charges (seven years in this case) expired. Practitioners sighed in relief. Inspired by such apparently legitimate delaying tactics, stakeholders in other countries and professions have considered deliberately stalling an investigation so that the judiciary could not get access until the period of limitation has expired. This solution works only, of course, if the judiciary is legally limited in beginning its probe of an occurrence while the formal investigation is still ongoing. In some countries or professions this is indeed the case.

1. Who gets to draw the line: While prosecutors and judges would still be left to draw the line eventually, other parties can withhold from them both the data and the opportunity to do so.
2. The role of domain expertise is interesting in this solution, as those with more expertise of the domain (safety investigators) make a judgment of the potential culpability of the acts they are investigating. If they judge these acts to be potentially (but unjustifiably and counterproductively) culpable, they may stall an investigation until the period of limitation has expired. In this sense, investigators introduce domain expertise into the judgment of whether something is acceptable or not, but they apply this expertise in advance—anticipating how the judiciary would respond to the data they have. Investigators may of course lack the domain expertise in the legal area to really make an accurate ex ante judgment, but previous experiences or the general climate in the country or profession may give them a good basis for their conjecture.
3. Protection of safety data is pretty strong, but of course hinges on the strength of the laws and statutes that prevent the judiciary from getting access to investigation data before the period of limitation has expired. Any legal opportunities that allow the judiciary to get into the formal investigation will directly undermine this solution.

Consequences: A climate of distrust and competition between stakeholders remains strong with this solution. Rather than resolving issues on merit, stakeholders may engage in legal gaming to try to get access (or retain privileged access) to safety data for their own purposes. The climate is not encouraging for the emergence of a just culture.

Local solution 4: Rely on lobbying, prosecutorial, and media self-restraint

A solution that is different from the previous ones relies almost entirely on trust between stakeholders. It has been achieved in a few countries (often after intense lobbying of lawmakers and other government officials by industry stakeholders). It has succeeded particularly in countries with strong freedom of information acts that leave their safety data exposed to both media and judiciary.

This local solution depends entirely on the extent of the trust developed and maintained, not on legal protection for any of the stakeholders. Thus, these countries typically have no protection in place for either reporters or safety data, and the judiciary has unfettered access to investigations—in principle. In practice, no prosecutor has dared to be the first to breach the trust built up. Interestingly, this solution seems to work in smaller countries that are culturally inclined toward homogeneity, trust, coherence, and social responsibility.

1. Who gets to draw the line: Prosecutors would in principle get to draw the line, but they have so far not dared to draw anything. The proscription against them doing so is not a legal one, but rather cultural or political: Going in and upsetting the delicate trust developed between parties is "not done" or politically not wise. But that does not mean it cannot be done. In fact, rules in countries with this solution still make exceptions for the kinds of "crimes" or "gross negligence" that prosecutors should still prosecute. The problem is of course chicken-and-egg: How is a prosecutor to find out whether a line was crossed without drawing one?
2. The role of domain expertise has been considerable in building the necessary trust between stakeholders, particularly in convincing other stakeholders (the media, the judiciary) of the value of their self-restraint, so that the entire society can benefit from safer professional systems.
3. Protection of safety data is not legally guaranteed but achieved by cultural convention and/or political pressure.

Consequences: At first sight, this solution comes across as a fraud, and as extraordinarily brittle. After all, there is nothing "on paper:" The entire contract between stakeholders to not interfere with each others' business is left to consensual agreements and trust. Practitioners may feel free to report because historically there is no threat (but can history be a guarantee for the future in this case?). On deeper inspection, though, this solution is as robust as the culture in which it is founded. And cultures can be very robust and resistant to change. This, at the same time, creates a high threshold for entry into such an arrangement: Without the right cultural prerequisites, this solution may be difficult to achieve.

Local solution 5: Judge of instruction

In the wake of prosecutions of practitioners that were widely seen as counterproductive to safety, some countries have moved ahead with installing a so-called judge of instruction. Such a judge functions as a go-between, before a prosecutor can actually go ahead with a case. A judge of instruction gets to determine whether a case proposed by a prosecutor should be investigated (and later go to trial). The judge of instruction, in other words, can check the prosecutor's homework and ambitions, do some investigation him- or herself, and weigh other stakeholders' interests in making the decision to go ahead with a further investigation and possible prosecution or not.

1. Who gets to draw the line: Initially (and most importantly) it is the judge of instruction who gets to draw the line between acceptable and unacceptable (or: between worthy of further investigation and possible prosecution or not). Other considerations can make it into the drawing of the line too (e.g. the interests of other stakeholders).
2. The role of domain expertise is supposed to be considerable in this solution. In the solution of one country, the judge of instruction is supported by a team from the aviation industry to help determine which cases should go ahead and which not. The make-up of this team and their interaction with the judge of instruction are crucial of course. For example, if unions or professional associations are not sufficiently represented, industry representatives may decide that it is in their interest to recommend to the judge to go ahead with prosecution, as it may help protect some of their concerns.
3. Protection of safety data is managed through the judge of instruction. If prosecutors want access to safety data, they will have to go via the judge of instruction, but there are (as usual) exemptions for serious incidents and accidents.

Consequences: A judge of instruction could function as a reasonable gate-keeper—weighing the various interests before a case can even be investigated by a prosecutor. It means, though, that such a judge needs a fair representation of all stakes, and not be susceptible to asymmetric lobbying by certain parties or interests over others. Since it is a rather new solution to the criminalization of human error, there is not a lot of data yet to see whether it works well or not.

Local solution 6: Prosecutor is part of the regulator

A solution that takes domain expertise right up to prosecutor level is one in which the prosecutor him- or herself has a history in or affiliation with the domain, and the office of prosecutor for that particular domain is inside of the regulator.

1. Who gets to draw the line: The prosecutor gets to draw the line (to be confirmed or rejected by a judge or jury), and the prosecutor is a person from the domain and employed by a major stakeholder in the domain.
2. The role of domain expertise is considerable, since the prosecutor comes from the domain and is employed by one of its large safety stakeholders. It is thus likely that the prosecutor is better able to balance the various interests in deciding whether to draw a line, and better able to consider subtle features of the professional's performance that non-domain experts would overlook or misjudge,
3. Protection of safety data is managed as an effect of this arrangement. The regulator has interests in protecting the free flow of safety information (not only as data for its oversight, but particularly for the self-regulation of the industry it monitors).

Consequences: The integration of prosecutor and regulator can prevent unfair or inappropriate prosecution, not only because of the tight integration of domain expertise, but also because of the greater relevance of the laws or regulations that will likely be applied (as the prosecutor works for a body that makes and applies the laws for that particular domain). The risk in this solution, of course, is that the regulator itself can have played a role (e.g. insufficient oversight, or given dispensation) in the creation of an incident and can have a vested interest in the prosecution of an individual practitioner so as to downplay its own contribution. There is no immediate protection against this in this local solution, except for regulatory self-restraint and perhaps the possibility of appeals higher up in the judiciary.

Local solution 7: Disciplinary rules within the profession

A large number of professional groups (everything from accountants to physicians to hunters to professional sports players) have their own elaborate system of disciplinary rules that are meant foremost to protect the integrity of a profession. Usually, a judiciary delegates large amounts of legal authority to the boards that credibly administer these professional disciplinary rules. Professional sanctions can range from warning letters (which are not necessarily effective) to the revocation of licenses to practice. The judiciary will not normally interfere with the internal administration of justice according to these disciplinary rules. There is, however, great variation in the administration of internal professional justice and thus a variation in how much confidence a country can have in delegating it to an internal disciplinary board.

1. Who gets to draw the line: The professional's peers get to draw the line between acceptable and unacceptable. There may be pressures, of course, that go outside the actual situation considered, so as to guarantee society's (and the judiciary's!) continued trust in the profession and its ability to manage and rectify itself. This may make it necessary to sometimes lay down the line more strictly so that a message of "we are doing something about our problems " clearly gets communicated to the outside—to the detriment of justice done to an individual professional. Who gets to draw the line for criminally culpable actions is an even larger problem: Internal rules are not equipped to handle those, so somewhere there needs to be a possibility for judging whether outside legal action is necessary. This can be the prosecutor's initiative (but then he or she needs enough data to trigger action) or the disciplinary board (yet they likely lack the legal expertise to make that judgment).
2. The role of domain expertise is total. Domain expertise is the basis for making the judgment about the right or wrong of somebody's actions, not some externally dictated law or statute. Domain expertise is also used to consider whether to forward a case to the formal judiciary (as there will always be an escape hatch for cases of "gross negligence" and so forth). But it is at least largely domain expertise that gets to draw that line here too.
3. Protection of safety data is likely to be independent of professional disciplinary rules and would need additional legislation for formal protection. However, with a functioning (and trustworthy) internal professional disciplinary system in place, people inside a professional may feel freer to report incidents and concerns.

Consequences: The total integration of domain expertise in the administration of justice makes a solution based on professional disciplinary rules attractive. Not only do domain experts judge whether something is acceptable or unacceptable, they also draw largely from the domain the "rules," written or unwritten, on which basis they make their judgment.

But there is a possible paradox in the justness of professional disciplinary rules. Because disciplinary rules exist for the maintenance of integrity of an entire profession, individual practitioners may still get "sacrificed" for that larger aim (especially to keep the system free from outside interference or undue scrutiny). To remain trustworthy in the eyes of other stakeholders, then, the disciplinary rules may have to wreak an occasional internal "injustice" so as to outwardly show that they can be trusted. This does not necessarily enhance the basis for just culture, as practitioners could still feel threatened and anxious about possible career consequences.

I came across an interesting, spontaneous variant of solving incidents and safety matters internally. An air traffic control center bad essentially agreed with some of the airlines frequenting its airspace that they would send reports on near misses and other problems directly to them. This was of course a very short-loop way to learn: A problem was seen and reported directly to the air traffic control center that could do something about it. It also prevented outsiders from making their own judgments about the performance of those involved and meting out any consequences (legal, regulatory). But those other parties (e.g. the regulator) felt that accountability was being shortcut—reporting routines established earlier had ensured that reports would go through them. Lessons learned could also be relevant to the wider industry, something that was missed by this local, internal solution.

A Staggered Approach to Building Your Just Culture

Where do you go from here? Building a just culture starts at home, in your own organization. But your organization does not live in a vacuum. The approach I would like to suggest is a staggered one (see Figure 10.1). It allows you to match your organization's ambitions to the profession's possibilities and constraints, the culture of your country and its legal traditions and imperatives. Each step in the staggered approach is already a contribution to the creation of a just culture. Each subsequent step is probably more difficult, as it draws in more parties with different backgrounds and persuasions, and larger stakeholder groups and their perspectives and interests. But each step already goes a little bit of the way to reconcile the agendas of different stakeholders. Each step may contribute a little bit to the building of trust between them.

Figure 10.1 A staggered approach to building a just culture

Step 1: Start at home, in your own organization

To lay the basis for the emergence of just culture in your profession or country, nothing is as important as starting at home, in your own organization. This will allow you to begin building relationships and trust between the first parties that matter: practitioners and their managers. Trust in management is not necessarily wide-spread among practitioners in all industries, which may have a number of reasons. One can be that managers sometimes come from backgrounds other than the profession, but even if they are ex-professionals, managers can be seen as "outsiders" (or worse: turncoats). Trust that was lost in management because of their positions on industrial or social issues (e.g. the application of work time regulations, vacation time) can also spill over into safety issues. So even if management has not acted negatively in relation to an incident before, its behavior elsewhere (or perception thereof) can affect the trust that practitioners will have in management handling of safety matters.

You can do a number of things immediately (if you have not already done so today):

• An incident must not be seen as a failure or a crisis, neither by management, nor by colleagues. An incident is a free lesson, a great opportunity to focus attention and to learn collectively.
• Abolish all financial and professional penalties in the wake of an occurrence. Suspending practitioners after an incident should be avoided at all cost. These measures serve absolutely no purpose other than making incidents into something shameful, something to be kept hidden. If your organization has these kinds of rules in place, you can count on losing out on a lot of valuable safety information.
• Monitor and try to prevent stigmatization of practitioners involved in an incident. They should not be seen as a failure, or as a liability to work with by their colleagues. This is not only devastating for them, but for every practitioner and by extension the organization, as incidents are once again seen as something to be kept concealed, out of view. Reintegrate these practitioners into the operation smoothly and sensitively, being aware of the possibility for stigmatization by their own colleagues.
• Implement, or review the effectiveness, of any debriefing programs or critical incident/stress management programs you may have in place to help practitioners after incidents (and if you don't have any in place, start building such programs pronto!). Such debriefings and support form a crucial ingredient in helping practitioners see that incidents are "normal," that they can help the organization get better, and that they can happen to everybody.
• Build a staff safety department, not part of the line organization, that deals with incidents. The direct manager (supervisor) of the practitioner should not necessarily be the one who is the first to deal with that practitioner in the wake of an incident (other than perhaps relieving him or her temporarily to deal with the stress and aftermath). Aim to decouple an incident from what may look like a performance review of the practitioner involved. Any retraining of the practitioner involved in the incident will quickly be seen as punishment (and its effects are often quite debatable), so this should be done with utmost care and only as a last resort.
• Start with building a just culture at the very beginning: During basic education and training of the profession. Make trainees aware of the importance of reporting incidents for a learning culture, and get them to see that incidents are not something individual or shameful but a good piece of systemic information for the entire organization. Convince new practitioners that the difference between a safe and an unsafe organization lies not in how many incidents it has, but in how it deals with the incidents that it has its people report.
• Be sure that practitioners know their rights and duties in relation to incidents. Make very clear what can (and typically does) happen in the wake of an incident. One union had prepared little credit-sized cards on which it had printed the practitioner's rights and duties in the wake of an occurrence (e.g. to whom they were obliged to speak (e.g. investigators) and to whom not (e.g. the media)). Even in a climate of anxiety and uncertainty about the judiciary's position on occurrences, such information will give practitioners some anchor, some modicum of certainty about what may happen. At the very least this will prevent them from withholding valuable incident information because of misguided fears or anxieties.

Starting at home, in your own organization, will allow you to lay the basis for a just culture. Without you laying that basis first, don't count on anybody else (like your judiciary or legislature) to do it for you.

Step 2: Decide who draws the line in your organization

One important decision for an organization is not only who gets to handle the immediate aftermath of an incident (the line organization: supervisor/manager, or a staff organization such as safety department). It is also how to integrate practitioner peer expertise in the decision on how to handle this aftermath, particularly decisions that relate to the individual practitioner's stature.

Whether a practitioner should undergo retraining, for example, is something that should be discussed not only with the practitioner in question (rather than just handed down from above), but also checked with a group of peers who can consider the wider implications of such a measure in the wake of an incident (e.g. on the reputation of that practitioner, but also on the way incidents will be seen and treated by colleagues as a result).

Empowering and involving the practitioner him- or herself in the aftermath of an incident is the best way to maintain morale, maximize learning, and reinforce the basis for a just culture.

Step 3: Protect your organization’s data from undue outside probing

Protecting your organization's data from outside probing should not be left to chance (i.e. the prosecutor has not previously shown interest, so why would he or she now), and probably not just to cultural convention or political pressure either.

The creation of trust between stakeholders is of course very important, and in this case it means that the judiciary will be willing to let an organization handle its own data when it has been given the assurance and confidence that the organization will come to it if a case is really likely to be culpable. This in turn hinges on the decisions made in step 2: Who draws the line between acceptable and unacceptable inside the organization? This person or group will likely be the one that has to give the judiciary the confidence that it knows what it is doing and that it will treat the organization's data with integrity and fairness.

None of this, however, is likely to automatically inspire practitioners to freely report without fear. It is best to try to enshrine the protection of the organization's critical data in law.

As said in Chapter 3, however, you have to think through the consequences of this step. One problem is that it can lock information up even for those who rightfully want access to it, and who have no vindictive intentions. Remember the patient, for example, or the victim of a transportation accident (or the family), whose main aim is to find out something specific about what happened. The protection of reporting can make such disclosure to such parties more difficult.

Step 4: Decide who draws the line in your country

Having non-domain experts draw the line between acceptable and unacceptable practitioner performance is fraught with risks and difficulties. This is where the advantages of local solutions come in that somehow meaningfully integrate domain expertise in the drawing of the line nationally. The use of expert witnesses during a trial is not likely to do this meaningfully, as that role is always rather constrained and testimony limited. In fact, this may be too late a stage to wait with bringing in domain expertise.

It could be profitable to start a discussion with the prosecuting authority in your country on how to help them integrate domain expertise (to support them in making better judgments about whether something is worthy of further investigation and prosecution). This may require that previous mistrust is overcome and may seem difficult in the beginning. In the end, however, it may tremendously benefit all parties, as it may also create a better understanding of each other's point of view and interests.

Local solutions that rely on internal professional disciplinary rules (to which the judiciary delegates its legal oversight) achieve a total integration of domain expertise in the determination of where the line is drawn. These solutions are currently not very common. Local solutions that otherwise integrate domain expertise (such as a integrating prosecutor and regulator in one organization, or having a judge of instruction supported by a team of domain experts) have advantages too. But they do not go so far as to really create a "jury of peers" that is able to judge performance in context. Any delegation to a greater degree of domain expertise, of course, does require that the judiciary can be confident that cases will be handled fairly and without prejudice in favor of colleagues (who may be seen to try to protect one another).

Not Bad Practice, But Bad Relationships

Unjust responses to failure are almost never the result of bad performance. They are the result of bad relationships. You can see this in almost any situation where we want to talk of just culture.

The strongest predictor, for example, of whether a physician will be sued is the extent to which patients feel they are treated with respect, honesty, and personal interest. The nature and graveness of the injury matters much less, if much at all. Injuring a patient during medical care is a severe breach of the fiduciary relationship between care-giver and patient. Patients typically feel betrayed, and angry. Restoring that relationship, or at least managing it wisely, is often the most important ingredient of a successful response.

Managing relationships between patients and doctors, if not restoring them, is one major aim of mediation, a form of alternative dispute resolution (ADR) in medicine. Mediation restores communication between the two parties, often (if not always) with the help of a mediator. What is said is kept confidential by law, thus making mediation a safe place for showing remorse, for introspection and the exploration of corrective actions without it being seen as admitting liability. In what is called interest-based mediation, In contrast to litigation or criminal-legal action, mediation allows apology, expressions of regret, compassion to occur much more naturally. Mediation is also much more flexible in allowing different outcomes. Compensation does not have to be money (indeed, it most often is not in ADR). In addition to agreeing to care for the injured party in whatever way necessary (medical or otherwise), mediation can inspire changes to procedures, augmenting of education, or other changes that respond to a patient's desire to never see this happen again.¹

Here is another example of the importance of relationships. Whether employees will see management responses to failure as just depends not so much on the response (or on the bad performance that triggers it). Rather, it depends to a great extent on the existing relationship between management and employees.

We did extensive field work among firefighters, to see how they learn from failures that occur during their emergency responses. If firefighters felt that they could come forward with their errors, then it was due largely to the relationship with their supervisors and their managers. In one station firefighters worked in close concert with their management, allowing an atmosphere where reporting errors and suggesting changes was normal, expected, and without jeopardy for any of the parties. Conversely, at a larger urban fire station with a distrustful industrial relationship, there was less bottom-up participation in decisions involving work-context, less firefighter involvement in learning, and much greater suspicion that any reported errors would not be responded to fairly.

Here is a final example: The relationship between the judiciary (or even one particular prosecutor) and a profession can be a strong determinant of legal action in the wake of an incident. A prosecutor may suspect that there is deliberate stonewalling, that legally accessible information is being deliberately withheld (remember: omertá). This, however, is often a response to earlier action by the prosecutor. Both are evidence of a relationship gone sour, and one way forward is to simply go and talk together. In the wake of several cases mentioned in this book, I have seen professional representatives propose exactly that. Talking, of course, is about the possible restoration of a relationship.

If bad relationships are behind unjust responses to failure, then good relationships should be seen as a major step toward just culture. Good relationships are about openness and honesty, but also about responsibilities for each other, and bracketing (this is yours, this is mine). Good relationships are about communication, about being clear about expectations and duties, and about learning from each other.

Perhaps this can come as somewhat of a relief. "Justice" and "culture" are two huge concepts. They are both essentially contested categories: What either means will forever be open to debate and controversy. They are basically intractable, unmanageable. A relationship, on the other hand, is manageable. At least half of it is in your hands. So if you want to do something about just culture, that is probably where to start.